SOIL 6 179ndash194 2020httpsdoiorg105194soil-6-179-2020copy Author(s) 2020 This work is distributed underthe Creative Commons Attribution 40 License

SOIL

Variation of soil organic carbon stable isotopes andsoil quality indicators across an erosionndashdeposition

catena in a historical Spanish olive orchard

Joseacute A Goacutemez1 Gema Guzmaacuten2 Arsenio Toloza3 Christian Resch3 Roberto Garciacutea-Ruiacutez4 andLionel Mabit3

1Institute for Sustainable Agriculture-CSIC Coacuterdoba Spain2Applied Physics Dept University of Coacuterdoba Coacuterdoba Spain

3Soil and Water Management amp Crop Nutrition Subprogramme Joint FAOIAEA Division of NuclearTechniques in Food and Agriculture International Atomic Energy Agency Vienna Austria

4Animal and Plant Biology and Ecology Dept Ecology sectionCenter for Advanced Studies in Olive Grove and Olive Oils

University of Jaeacuten Jaeacuten Spain

Correspondence Joseacute A Goacutemez (joseagomeziascsices)

Received 29 August 2019 ndash Discussion started 9 October 2019Revised 4 April 2020 ndash Accepted 15 April 2020 ndash Published 15 May 2020

Abstract This study compares the distribution of bulk soil organic carbon (SOC) its fractions (unprotectedand physically chemically and biochemically protected) available phosphorus (Pavail) organic nitrogen (Norg)and stable isotope (δ15N and δ13C) signatures at four soil depths (0ndash10 10ndash20 20ndash30 and 30ndash40 cm) between anearby open forest reference area and a historical olive orchard (established in 1856) located in southern SpainIn addition these soil properties as well as water stable aggregates (Wsagg) were contrasted at eroding anddeposition areas within the olive orchard previously determined using 137Cs SOC stock in the olive orchard(about 40 t C haminus1) was only 25 of that in the forested area (about 160 t C haminus1) in the upper 40 cm of soil andthe reduction was especially severe in the unprotected organic carbon The reference and the orchard soils alsoshowed significant differences in the δ13C and δ15N signals likely due to the different vegetation compositionand N dynamics in both areas Soil properties along a catena from erosion to deposition areas within the oldolive orchard showed large differences Soil Corg Pavail and Norg content and δ15N at the deposition weresignificantly higher than those of the erosion area defining two distinct areas with a different soil quality statusThese overall results indicate that the proper understanding of Corg content and soil quality in olive orchardsrequires the consideration of the spatial variability induced by erosionndashdeposition processes for a convenientappraisal at the farm scale

1 Introduction

Research on soil organic carbon (SOC) sequestration andthe potential of agricultural soils to store carbon has in-creased since the declaration of the 4 per 1000 program (Lal2015) which seeks to increase global soil organic carbonstocks by 04 per year as compensation for global anthro-pogenic C emissions Under this program special empha-sis is given to combating soil degradation due to its strong

impact on the global carbon cycle because of the deple-tion of SOC stock For instance in European agriculturalsoils Lugato et al (2016) reported that erosion-induced SOCfluxes were of the same order as the current gains from im-proved management and they must be reduced to maintainsoil health and productivity Lal (2003) estimated the globalerosion-induced displacement of SOC at 57 Pg C yrminus1 ap-proximately 70 of which is redistributed and redepositedover the landscape and the remaining 30 is transported by

Published by Copernicus Publications on behalf of the European Geosciences Union

180 J A Goacutemez et al Organic carbon across an olive orchard catena

rivers into aquatic ecosystems SOC is the most importantindicator of soil quality (Rajan et al 2010) and erosion-induced loss of SOC affects on-site soil fertility and off-site environment quality (Lal 2019) However the effects ofsoil erosion and the fate of the specific SOC fraction trans-ported by erosion in specific agricultural systems such asolive groves remain poorly understood therefore agroenvi-ronmental impacts of SOC dynamics and variability requiremore site- and crop-specific research

Olive trees one of the most important crops in theMediterranean region which account for approximately97 Mha (FAOSTAT 2019) have been linked to severe envi-ronmental impacts including the acceleration of erosion andsoil degradation (eg Beaufoy 2001 Scheidel and Kraus-mann 2011) In fact soil degradation is common in olive or-chards as they have been traditionally cultivated under rain-fed conditions on sloping land at relatively low tree densi-ties with limited canopy size due to pruning and under bare-soil management to optimize water use by the tree under thesemiarid conditions which characterize the Mediterraneanclimate (Goacutemez et al 2014) Indeed there are many stud-ies which have measured high erosion rates in olive orchardson sloping areas (eg Goacutemez et al 2014) although thesehigh erosion rates are not necessarily a direct consequence ofcurrent management In a study of historical erosion rates inseveral ancient olive orchards of Montefriacuteo (southern Spain)Vanwalleghem et al (2011) reported unsustainable erosionrates in the range of 23 to 68 Mg haminus1 yrminus1 during the 19thand early 20th centuries when these orchards were managedunder the same slope and rainfall conditions with bare soilalbeit based on animal plowing Vanwalleghem et al (2011)also reported a further increase in the erosion rates whenbare-soil management started to be implemented in these or-chards by mechanization and herbicides in the late 20th cen-tury In the last 5 decades (Ruiacutez de Castroviejo 1969) therehas been an attempt to control soil degradation while main-taining a favorable soilndashwater balance for the tree throughthe gradual development of temporary cover crops (grownduring the rainy season) (Goacutemez et al 2014) These higherosion rates have also been linked to the degradation of soilproperties observed in olive orchards For instance Goacutemezet al (2009b) measured in a 5-year experiment and in anolive grove a decrease in SOC aggregate stability and in-filtration rates with bare soil as compared to temporary covercrops Such scientific evidence which links changes in soilproperties to different erosion rates in olive orchards un-der controlled conditions is rarely reported in the literatureIndeed most of the studies aimed to connect soil proper-ties with different types of soil management in olive grovescome from surveys of soil properties in orchards with simi-lar soil types but with different soil management Examplesof these studies are those of Aacutelvarez et al (2010) and Sori-ano et al (2014) who found an improvement in soil prop-erties ndash particularly aggregate stability SOC and biologicalactivity ndash in organic olive orchards with cover crops com-

pared to those with bare soil In recent years these stud-ies have started to deepen our understanding of investigatingkey properties such as SOC For instance Vicente-Vicente etal (2017) evaluated the impact of cover crops in the distri-bution of unprotected and protected SOC in the top 15 cm ofthe soil Typically field studies take samples in a represen-tative area of the slope which is a common assumption inmany soil quality studies (eg Andrews and Carroll 2001)Although there is a limited number of experiments on thespatial variability of soil properties in olive orchards theysuggest the existence of significant in-field variability (egGargouri et al 2013 Huang et al 2017) Moreover Goacutemezet al (2012) suggested that regarding organic carbon part ofthis on-site variability of soil properties might be related toerosionndashdeposition processes

In-field variability associated with erosionndashdeposition pro-cesses is relatively well documented for organic carbon con-tent in field crops (eg De Gryze et al 2008 Mabit andBernard 1998 2010 Van Oost et al 2005) While thehuman-induced acceleration of soil erosion has depleted theSOC stock of agroecosystems the fates of SOC transportedacross the landscape and that deposited in depressional sitesare not fully understood despite the fact that these transfersmight explain a high proportion of the on-site variability ofsoil properties

Most of the erosion rates recorded or established in oliveorchards come from runoff plots or small catchment exper-iments (eg Goacutemez et al 2014) The use of the 137Cs ap-proach has demonstrated its potential in establishing long-term soil erosion rates with this specific land use An exam-ple of these studies is that of Mabit et al (2012) in whicherosion as well as deposition rates since the 1950s was de-termined in one ancient olive orchard in the municipality ofMontefriacuteo showing an average annual rate in the eroding partof the slope of 123 t haminus1 yrminus1 and an average depositionrate in the lower section of the hillslope much shorter thanthe eroding section of 131 t haminus1 yrminus1 This study involveda reference area for establishing precisely the initial 137Csinventory a natural undisturbed area located 200 m from theorchard As reported by Mabit et al (2012) based on 13 in-vestigated soil profiles the local reference 137Cs inventoryin this undisturbed area was evaluated at 1925plusmn250 Bq mminus2

(meanplusmn 2 standard error) with a coefficient of variation (CV)of 23

To complement andor to circumvent some limitation as-sociated with the use of this anthropogenic radioisotope (seeMabit et al 2008) and to maintain the capacity to deter-mine erosion and deposition rates without the need to use di-rect measurements other natural radioisotopes such as 210Pb(eg Mabit et al 2014 Matisoff 2014) or stable isotopessuch as δ15N or δ13C (eg Meusburger et al 2013) havebeen proposed

In this study we hypothesized that the contribution of thelong-term erosionndashdeposition processes to the in-field vari-ability of soil properties in olive orchards (or other woody

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 181

crops) at a medium-steep slope is relevant and should betaken into account when analyzing the effects of specificstrategies on SOC sequestration or on other soil propertiesIn addition we exploited the advantage provided by theunique location of an ancient olive orchard near an undis-turbed reference area and the previous information on thissite from studies on historical erosion rates (Vanwalleghemet al 2011 Mabit et al 2012) to fulfill the following objec-tives

1 to quantify the long-term variability in soil total organiccarbon and in their different fractions as well as soilquality indicators in relation to erosion and depositionareas in a historical olive orchard

2 to evaluate these differences in relation to the referencevalues found in an undisturbed natural area

3 to evaluate differences in stable isotope signatures (δ13Cand δ15N) and explore their potential for identifying de-graded areas within the olive orchard

2 Materials and methods

21 Description of the area

The study area is located in the municipality of Montefriacuteosouthwestern Spain (Fig 1) The municipality extension isaround 220 km2 of which 81 is cultivated mostly witholive trees The climate in the region is continental Mediter-ranean with a long-term (1960ndash2018) average annual pre-cipitation of 630 mm a mean annual evapotranspiration of750 mm and a yearly average temperature of 152 C It is amountainous area with elevation ranging between 800 and1600 m asl at the highest point (Sierra de Parapanda) Soilsampling took place in two areas around the archeologicalsite ldquoPentildea de los Gitanosrdquo where the soil is classified as Cal-cic Cambisol according to the FAO classification (Deckerset al 2004) The undisturbed reference area was inside anarcheological site (Fig 1) This undisturbed area is coveredby open Mediterranean forest interspersed with shrubs andannual grasses on limestone material (calcarenites) The sta-tus of this protected site guarantees that no anthropogenicactivities have impacted it for a long period of time approx-imately since the end of 16th century This area is coveredby natural vegetation typical of the Mediterranean regionmainly bushes like Pistacia lentiscus and Retama sphaero-carpa and herbaceous species such as Anthemis arvensisCalendula arvensis Borago officinalis Brachypodium sppBromus spp and Medicago spp The area studied was anolive orchard established in 1856 located within tens of me-ters of the reference area (Fig 1) Both areas were describedin detail in previous studies on historical erosion rates in theregion (Vanwalleghem et al 2011 Mabit et al 2012) Thisolive orchard is rainfed and soil management in the decadesbefore the sampling was based on bare soil with pruning

residues (trees pruned every 2 years) being chopped and lefton the soil surface Olive trees are fertilized annually with5 kg of 15 NPK spread below the tree canopy area

22 Soil sampling

The reference site adjacent to the olive orchard belongsto an archeologically protected site and is therefore a non-cropped area where neither erosion nor sedimentation pro-cesses take place (Fig 1) This site was sampled at 13 pointsacross a transect spaced at an average distance of 6 m withonly four of these sampling points used in this study (Fig 1)We use only four soil samples because not enough soil for theanalysis was collected in the other samples At each samplingpoint the excavation method was used Soil samples werecollected at 5 cm depth intervals until bedrock was reached(between 20 and 60 cm) The four samples used in this studyreached 40 cm depth or more Composite soil samples at10 cm intervals were prepared at the laboratory to performthe chemical analysis of the reference area as described be-low

In the olive orchard a hydraulic mechanical core samplerwas used It gently rotates and pushes an 8 cm diameter coreto sample eight points along a 452 m long catena (Fig 1) Tominimize soil disturbance soil sampling was made when soilwater content was between 40 and 80 of water-holdingcapacity Precautions were taken to assure that four rangesof soil depth (0ndash10 10ndash20 20ndash30 and 30ndash40 cm) were col-lected at each sampling point in the orchard soil

In a previous study soil erosion and deposition rates weredetermined at each sampling point comparing the 137Cs in-ventory among these points and that of the undisturbed refer-ence area (Mabit et al 2012) The positions of all samplingpoints were recorded by a real-time kinematic global posi-tioning system (RTK-GPS) at submeter resolution (Table 1)Overall 12 points were sampled 4 in the reference area and8 along the catena across the olive orchard with all of themreaching the bedrock below 40 cm depth

23 Physicochemical analysis

Soil samples were passed through a 2 mm sieve and homog-enized The fraction under 2 mm was determined and ex-pressed as a percentage of total soil on a dry mass basisSeparation of the four soil organic carbon (Corg) pools wasperformed by a combination of physical and chemical frac-tionation techniques through a three-step process developedby Six et al (2002) and modified by Stewart et al (2009)summarized here This three-step process isolates a total of12 fractions and it is based on the assumed link between theisolated fractions and the protection mechanisms involvedin the stabilization of organic C First a partial dispersionand physical fractionation of the soil are performed to ob-tain three size fractionsgt 250 mm (coarse nonprotected par-ticulate organic matter POM) 53ndash250 mm (microaggregate

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

182 J A Goacutemez et al Organic carbon across an olive orchard catena

Table 1 Location of the sampling points along the transect and associated soil redistribution rates derived from the 137Cs technique (adaptedfrom Mabit et al 2012) Negative values indicate net erosion and positive values net deposition

Point no Code Distance in transect (m) Elevation (m) Erosion deposition rate (t haminus1 yrminus1)

1 Cs1 0 1044 minus522 Cs3 664 10328 minus1783 Cs5 1250 10178 minus714 Cs7 1790 10068 minus1635 Cs12 3122 9868 minus1526 Cs13 3381 9848 597 Cs15 3888 9818 2478 Cs17 4295 9798 88

Figure 1 Site location and associated sampling scheme (a b c) Location map of the sampling area in Montefriacuteo southern Spain Referencesite delineated by the white line within a protected archeological site (yellow line) Yellow markers in (b) indicate the sampled transect withinthe olive orchard Numbering starts in the points at higher elevation see (d) for the elevation change in the transect in the orchard Yellowmarkers in (c) indicate sampling point in the reference area The map of Europe was designed by Freepik and the aerial images are takenfrom Google Earth (copy Google Earth 2018)

fraction) and lt 53 mm (easily dispersed silt and clay) Thisphysical fractionation is done on air-dried 2 mm soil sievedthrough a 250 mm sieve Material greater than 250 mm re-mained on the sieve Microaggregates were collected on a53 mm sieve that was subsequently wet-sieved to separate the

easily dispersed silt- and clay-sized fractions from the water-stable microaggregates The suspension was centrifuged at127 g for 7 min to separate the silt-sized fraction This su-pernatant was subsequently separated flocculated and cen-trifuged at 1730 g for 15 min to separate the clay-sized frac-

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 183

tion All fractions were dried in a 60 C oven and weighedAfterwards there was a second step involving a further frac-tionation of the microaggregate fraction isolated in the firststep A density flotation with sodium polytungstate was usedto isolate fine nonprotected POM (LF) after removing thefine nonprotected POM the heavy fraction was dispersedovernight by shaking and passed through a 53 mm sieveto separate the microaggregate-protected POM (gt 53 mm insize iPOM) from the microaggregate-derived silt- and clay-sized fractions The resulting suspension was centrifuged toseparate the microaggregate-derived silt-sized fraction fromthe clay-sized fraction as described above A final third stepinvolved the acid hydrolysis of each of the isolated silt- andclay-sized fractions The silt- and clay-sized fractions fromboth the density flotation and the initial dispersion and phys-ical fractionation were subjected to acid hydrolysis The un-protected pool includes the POM and LF fractions isolatedin the first and second fractionation steps respectively Thephysically protected SOC consists of the SOC measured inthe microaggregates It includes not only the iPOM but alsothe hydrolyzable and nonhydrolyzable SOC of the intermedi-ate fraction (53ndash250 microm) The chemically and biochemicallyprotected pools correspond to the hydrolyzable and nonhy-drolyzable SOC in the fine fraction (lt 53 microm) respectivelyIn all cases SOC fractions and in the bulk soil organic car-bon concentrations were determined by using the wet oxida-tion sulfuric acid and potassium dichromate method of An-derson and Ingram (1993)

Inorganic carbon was removed prior to stable isotope anal-ysis by acid fumigation following the method of Harris etal (2001) Moistened subsamples were exposed to the exha-lation of HCl in a desiccator overnight Afterwards the sam-ples were dried at 40 C before the stable isotope ratio wasmeasured The N measurements were done with unacidifiedsamples and the stable N isotope ratios and the C and N con-centrations were measured by isotope ratio mass spectrome-try (Isoprime 100 coupled with an Elementar Vario IsotopeSelect elemental analyzer both instruments supplied by El-ementar Langenselbold Germany) The instrumental stan-dard deviation for δ15N is 016 and 011 for δ13C Sta-ble isotopes are reported as delta values (permil) which are therelative differences between the isotope ratios of the samplesand the isotope ratio of a reference standard

In addition available phosphorus (Pavail) was determinedby the Olsen method (Olsen and Sommers 1982) and or-ganic nitrogen (Norg) was determined by the Kjeldahl method(Stevenson 1982) Water-stable aggregates (Wsagg) weremeasured using the method of Barthes and Roose (2002)Soil particle size distribution was determined using the hy-drometer method (Bouyoucos 1962) for the topsoil (0ndash10 cm) of the reference area and the olive orchard Two bulkdensity values were calculated for the whole profile (a) oneconsidering the mass of soil finer than 2 mm and (b) oneconsidering the mass of soil finer than 2 mm as well as thestone content which was determined from the excavation

and core sampling described above Additionally the bulkdensity of the topsoil (0ndash10 cm depth) was measured usinga manual soil core sample with a volume of 100 cm3 Soilcarbon stocks were calculated for the fine soil fraction af-ter discounting rock or stone fragments larger than 2 mm andconsidering bulk density They were then presented on equiv-alent soil mass as described by Wendt and Hauser (2013) Asproposed by Hassink and Whitmore (1997) theoretical val-ues of carbon saturation were calculated from the soil par-ticle analysis Finally the soil degradation index developedby Goacutemez et al (2009a) was calculated from the Corg Pavailand Wsagg

24 Statistical analysis

The overall effect of depth and area (reference site vs oliveorchard or eroded vs deposition area within the olive or-chard) was evaluated using a two-factor analysis of vari-ance (ANOVA) (p lt 005) Additionally for some compar-ison at similar soil depths values of soil properties betweentwo different areas were assessed using a one-way ANOVAtest (p lt 005) In both situations data were log-transformedwhen necessary to fulfill ANOVA requirements Exploratoryanalysis using principal component analysis (PCA) was per-formed in the olive orchard area using the variables and sam-pling depths showing significant differences in the ANOVAThis PCA was complemented by determining the linear cor-relation coefficient variables showing the highest load on thePCA and erosion deposition rates using the Pearson corre-lation test The statistical software package Stata SE141 wasused for these analyses

3 Results

31 Organic carbon concentration and distributionamong fractions

Table 2 shows the significance of the differences in bulk soilCorg and the various Corg fractions between reference andolive orchard plots at different soil depths and due to the in-teraction between both factors (Table 2a) It also shows theeffects of the erosion deposition ratio (Table 2b) Overallbulk soil Corg was always significantly higher in the referencearea as compared to the olive orchard (00001lt p lt 00013depending on depth) (Table 2a and Fig 2) and this was inde-pendent of the soil sampling depth Corg values on the refer-ence site were between 2 to 5 times higher than those of theolive orchard for a given depth with the greater differencesin the top 10 cm of the soil Soil depth has a significant ef-fect on bulk Corg and Corg fractions with values typically de-creasing with depth in both areas Corg concentrations in theunprotected and physically chemically and biochemicallyprotected fractions were significantly higher (p lt 00001)(Table 2a) in the reference site as compared to the olive or-chard and across the different depths (00001lt p lt 00007)

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

184 J A Goacutemez et al Organic carbon across an olive orchard catena

(Fig 3) Corg values were between 2 to 6 times higher for theunprotected and chemically protected fractions and between2 to 35 times higher for the physically and biochemicallyprotected fractions with differences tending to decrease withthe soil depth

Within the olive orchard there were statistically signifi-cant differences between the erosion and deposition areas forbulk Corg values (p = 00198) (Table 2b) Higher Corg val-ues (11 to 06 ) were observed in the deposition arealocated downslope whereas lower values (085 to 055 )were measured in the areas with net erosion in the upper andmid sections of the catena It is worth noting that these differ-ences between erosion and deposition areas are detected forthe overall analysis using a two-way ANOVA (Table 2b) al-though an individual analysis at each depth (Fig 4) does notdetect statistically significant differences probably due to themoderate number of replications Significant differences be-tween the deposition and eroding area were also found forthe unprotected (p = 004) and the physically (p = 00077)and chemically protected fractions (p = 00299) (Table 2b)However differences for the biochemically protected frac-tions (Table 2b Fig 4) were not significant

The percentage distribution of SOC among fractions wasalso significantly different between both areas (reference vsolive orchard) except for the biochemically protected fraction(p lt 00001) (Table 3a) and depths (p lt 0011) (Fig 5) Inthe reference area most of the Corg was unprotected (between50 and 65 approximately) with no significant trend withdepth (Table 3a Fig 5) followed in relative importance bythe chemically and physically protected fractions which con-tributed between 18 and 30 as well as 10 and 20 ofthe bulk soil Corg respectively The biochemically protectedfraction represents a very low percentage (between 4 and6 ) In the olive orchard Corg is stored predominantly inthe physically and chemically protected fractions which ac-counted for about 38 to 27 and 34 to 28 respec-tively followed by the pool of unprotected fractions (be-tween 22 to 32 ) (Fig 5) The biochemically protectedfraction represents between 4 and 11 of the organic car-bon stored in the olive orchard There are no clear differencesin the organic carbon distribution among the different frac-tions between the erosion and deposition areas in the oliveorchard with the exception of the physically protected frac-tions at 10ndash20 cm depth (p = 0024) (Fig 6)

32 Organic carbon stock

SOC stock in the reference area is approximately 160 t haminus1making it significantly higher at p lt 005 for an equivalentmass than that of the olive orchard (Fig 7) which stores be-tween 38 and 41 t haminus1 in the eroded and deposition areasrespectively There were no significant differences betweenthese two orchard areas Similar results were obtained acrossthe top 40 cm soil profile Clay silt and sand contents of thetopsoil (0ndash10 cm) along the catena in the olive orchard av-

eraged 41 37 and 22 respectively with similar av-erage values in the eroding and deposition areas Variabilitywas relatively low (average coefficient of variation of 17 )and there were no significant changes between the erosionand deposition areas In the reference area the soil has anaverage clay silt and sand content of 30 31 and 39 respectively also with a homogeneous distribution across thesampling area (coefficient of variation of 10 ) According tothe Hassink and Whitmore (1997) model the percentages oforganic carbon of maximum soil stable Corg are 324plusmn011 and 363plusmn019 in the reference site and olive orchard re-spectively So protected Corg in the reference and olive or-chard areas account for 498plusmn 115 and 2049plusmn 52 ofthe maximum soil stable Corg respectively in the topsoil

33 δ15N and δ13C isotopic signal

Figure 8 and Table 4a compare stable isotope delta values be-tween the reference site and the overall olive orchard accord-ing to depth There are significant statistical differences inδ15N δ 13C and the δ13C δ15N ratio between the two areas(p lt 0002 Table 4a) although in the case of δ 15N the dif-ferences at individual depths were not significant Soil depthalso had a significant effect (00026lt p lt 0029 Table 4a)When comparing differences between the erosion and depo-sition areas within the olive orchard we detected statisticallysignificant differences only in δ15N and the δ13C δ15N ra-tio (p = 001 Table 4b) and they were mostly marked in theupper 20 cm of the soil (Fig 8)

34 Soil quality of topsoil across the catena variabilitybetween eroded and deposition areas in the orchard

Figure 9 depicts the comparison of the Pavail Norg Wsaggand the soil degradation index (SDI Goacutemez et al 2009a)in the top 10 cm of the soil between the erosion and deposi-tion areas of the olive orchard and Table 5 shows a similarcomparison for Norg Pavail and bulk density at the differentsoil depths Pavail in the deposition area is significantly higherthan that of the erosion area in the topsoil (p = 0005 Fig 9)and for the whole profile (p = 001 Table 5) whereas no sig-nificant differences in individual soil layers were found forNorg and Wsagg The SDI which combines these three vari-ables was about 3 times higher in the eroded area than in thedeposition area

Table 6 shows the loads in the first three principal com-ponents (PCs) of the principal component analysis (PCA)for the variables used in this analysis More than 70 ofthe variance was explained by the first two PCs Soils ofthe eroded and deposition area were clearly separated in thespace defined by the two PCs (Fig 10) The variables withhigher contributions in PC 1 and 2 were related to Pavail con-centration in the 0ndash10 and 0ndash40 cm soil depths to δ15N andδ13C in the 10 to 30 cm soil depths and to Corg concentra-tion and distribution in some fractions also in soil depths be-

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 185

Figure 2 Comparison of average soil organic carbon concentration in bulk soil depending on soil depth distinguishing between the referencesite the whole olive orchard the eroded area of the olive orchard and the deposition area in this orchard The labels of the bars for eachdepth indicate the p value according to a one-way ANOVA for the reference area (dark blue) vs the whole olive orchard (dark green) andfor the eroding area (beige) vs the deposition area (light green lower label in italics)

Figure 3 Organic carbon concentration in the different soil organic carbon fractions at each depth comparing the reference site vs oliveorchard The labels of the bars indicate the p value according to a one-way ANOVA comparing treatments for the same soil depth and carbonfraction between the reference area and olive orchard

tween 10 and 30 cm (Table 6) The deposition area tendedto show higher values in PC 1 and PC 2 and there wasno clear tendency in the erosion area along the catena Thelinear correlation coefficient between the variables and theerosion deposition rate was rather significant (r gtplusmn0731)for most of the variables (Table 7) Interestingly Pavail was

highly positively correlated in the whole soil layer Amongthe two most robust correlations with erosion depositionrates were that with Pavail concentration across the 40 cm soildepth y = 03975x+ 98364 (r2

= 0907) and δ13C at 10ndash20 cm depth y =minus00094xminus 25318 (r2

= 0632)

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

186 J A Goacutemez et al Organic carbon across an olive orchard catena

Table 2 Results of the two-way ANOVA of soil organic carbon concentration Corg () in different fractions and in bulk soil In (a) arearefers to the reference site vs the olive orchard and in (b) area refers to eroded vs deposition areas in the olive orchard NS stands for notsignificant

(a) Model Bulk soil Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00000 00000 00000 00000 00000Depth (D) 00023 00022 00061 00190 NSAtimesD NS NS NS NS 00300

(b) Model Bulk soil Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00198 00400 00077 00299 NSDepth (D) 00081 00070 00058 00055 00847AtimesD NS NS NS NS NS

Figure 4 Organic carbon concentration in the different soil organic carbon fractions at each depth comparing the eroded vs the depositionarea within the olive orchard The labels of the bars indicate the p value according to a one-way ANOVA comparing treatments for the samesoil depth and carbon fraction between the reference area and olive orchard

4 Discussion

41 Organic carbon concentration and distributionamong fractions

After approximately 175 years of contrasted land use be-tween the undisturbed reference site and the olive orchardbulk soil organic carbon concentration and its fractions havebeen dramatically reduced in the olive orchard Current lev-els of Corg concentration in the soil profile are approximately20 to 25 of that found in the reference area covered bynatural vegetation in the area adjacent to the orchard This ra-

tio is similar albeit in the lower range to the comparison ofCorg in topsoil among olive orchards with different types ofmanagement and natural areas reported for the region (Mil-groom et al 2007) The increased soil disturbance the lowerannual rate of biomass returned to the soil and the highererosion rate in the olive orchard explain this difference Inboth areas the Corg is clearly stratified indicating that de-spite the different mechanisms involved there is a periodicinput of biomass from the olive trees (eg fall of senescenceleaves and tree pruning residues) plus the annual ground veg-etation Vicente-Vicente et al (2017) estimated this biomasscontribution in the range of 148 to 056 t haminus1 yrminus1 It is

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 187

Figure 5 Contribution () of the different fractions with respect to total soil organic carbon by depth comparing the reference site vs theolive orchard The labels of each fraction and depth are the p value according to a one-way ANOVA comparing the reference area vs theolive orchard for the same carbon fraction and soil depth

Figure 6 Fraction of total organic carbon stored in the different fractions by depth comparing the reference site vs the olive orchard Thelabels of each fraction and depth are the p value according to a one-way ANOVA comparing the reference area vs the olive orchard for thesame carbon fraction and soil depth

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

188 J A Goacutemez et al Organic carbon across an olive orchard catena

Table 3 Results of the two-way ANOVA of the distribution of the total soil organic carbon content in the soil among the different fractionsof soil organic carbon Corg In (a) area refers to the reference site vs the olive orchard and in (b) area refers to eroded vs deposition areasin the olive orchard NS stands for not significant

(a) Model Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00000 00000 00000 NSDepth (D) NS NS 00640 NSAtimesD NS 00059 NS NS

(b) Model Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 0091 00881 NS NSDepth (D) 0051 00214 NS 0033AtimesD NS NS NS NS

Figure 7 Cumulative soil organic carbon (SOC) stock across thesoil profile in terms of cumulative soil mass on the vertical axisDifferent letters for similar soil mass refer to statistically significantdifferences (KruskalndashWallis test at p lt 005) For this analysis cu-mulative soil organic carbons were interpolated linearly to the aver-age cumulative soil mass corresponding to all the points in the threeareas

worth noticing that the decrease in Corg as compared to thenatural area is much higher than the reported rates of increasein Corg in olive orchards using conservation agriculture (CA)techniques such as cover crops and incorporation of organicresidues from different sources In a meta-analysis Vicente-Vicente et al (2016) found a response ratio (the ratio of Corgunder CA management as compared to Corg in orchards withbare-soil management) of 11 to 19 suggesting that underCA management which combines cover crops and organicresidues Corg doubled as a maximum

Table 4 Results of the two-way ANOVA of the stable isotope sig-nal In (a) area refers to the reference site vs the olive orchard andin (b) area refers to eroded vs deposition areas in the olive orchardNS stands for not significant

(a) Model δ13C δ15N δ13C δ15N

Area (A) 00001 0002 0002Depth (D) 00026 00175 0029AtimesD NS NS NS

(b) Model δ13C δ15N δ13C δ15N

Area (A) NS 001 001Depth (D) NS NS NSAtimesD NS NS NS

Table 5 Results of the two-way ANOVA of some physical andchemical soil properties comparing eroded vs deposition areas inthe olive orchard NS stands for not significant

Model Norg Pavail Bulk density

Area (A) 00000 001 NSDepth (D) 00009 NS NSAtimesD NS NS NS

Combining all Corg data of the olive orchard the variabil-ity was about 35 which is similar to what has been re-ported so far in the few studies found on soil Corg variabilityin olive orchards For instance Gargouri et al (2013) indi-cated a 24 coefficient of variation (CV) in a 34 ha oliveorchard in Tunisia while Huang et al (2017) reported an av-erage CV of 41 in a 62 ha olive orchard in southern SpainNeither of these two studies reported clear trends in the distri-bution of Corg depending on topography Huang et al (2017)

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 189

Figure 8 13C and 15N isotopic signal of soil by depth distinguish-ing among reference site the whole olive orchard the eroded areaof the olive orchard and the deposition area in this orchard Thelabels of the bars for each depth indicate the p value according toa one-way ANOVA between the reference area (dark blue) vs thewhole olive orchard (dark green) and the eroding area (beige) vsthe deposition area (light green lower label in italics)

pointed out the additional difficulties in the determination ofCorg due to the topography heterogeneity although this wascompounded by the fact that within the orchards there weretwo areas with different planting dates for the trees Goacutemezet al (2012) reported a CV of 49 with higher Corg in ar-eas where there was a change in the slope gradient from thehillslope to a central channel draining into the catchment al-though they could not find a simple relationship between theincrease in content and the topographic indices Despite thefact that a lot of work has been done on the correlation be-tween erosion and deposition and the redistribution of soilCorg (eg Van Oost et al 2005) our study is to our knowl-edge the first attempt to quantify this in detail under the

Table 6 Loads of selected variables in the PCA for the first threeprincipal components (PCs) The values in brackets below PC 1 2and 3 indicate the percentage of variance explained by this PC Vari-ables in bold are those with a load higher than 90 of the variablewith the maximum load for this PC Conc refers to Corg concen-tration for this fraction and Frac means the relative contribution ofthis fraction to the total Corg for this soil depth

Variable at each depth interval (cm) PC 1 PC 2 PC 3(558) (176) (132)

Pavail 0ndash10 02298 008765 minus007278Pavail 0ndash40 02271 0101 minus008455δ13C 0ndash10 minus003385 02861 minus03302δ15N 0ndash10 01941 01677 01836Norg 0ndash10 02147 minus01375 minus00336δ13C δ15N 0ndash10 01594 0249 02211δ13C 10ndash20 minus02329 01461 004296δ15N 10ndash20 01856 007054 03121Norg 10ndash20 02317 minus008699 minus01404δ13C δ15N 10ndash20 01637 008512 03309δ13C 20ndash30 minus02316 006018 009789δ15N 20ndash30 01748 minus02656 01812Norg 20ndash30 02176 009979 minus007458δ13C δ15N 20ndash30 01684 minus02982 01774Corg conc 10ndash20 02421 minus002951 minus006407Corg unpr conc 10ndash20 02116 00331 00385Corg unpr Frac 10ndash20 minus009663 minus00509 0364Corg Phys Pro conc 10ndash20 02371 minus005728 minus01183Corg Phys Pro Frac 10ndash20 0123 0163 minus002953Corg Chem Pro conc 10ndash20 02072 003097 minus02335Corg Chem Pro Frac 10ndash20 minus004095 009461 minus04678Corg conc 20ndash30 02326 minus01108 minus005737Corg unpr conc 20ndash30 02002 minus02372 minus005071Corg unpr Rel 20ndash30 cm minus0132 minus03675 minus01379Corg Phys Pro conc 20ndash30 02257 005677 minus001036Corg Phys Pro Frac 20ndash30 009518 03528 01294Corg Chem Pro conc 20ndash30 02323 minus006802 minus01366Corg Chem Pro Frac 20ndash30 004565 04437 minus001278

agroenvironmental conditions of an olive orchard The vari-ability induced by the combined effects of water and tillageerosion in this olive orchard was similar to that described inother agroecosystems For instance Van Oost et al (2005)measured a clear correlation between the erosion and depo-sition rates and the topsoil Corg concentration which rangedbetween 08 of the erosion to 14 of the deposition sitesin the top 25 cm of the soil at two field crop sites under atemperate climate Besides this Bameri et al (2015) mea-sured a higher Corg in the lower part of a field crop site undersemiarid conditions where deposition of the eroded soil fromthe upper zones took place Overall in such landscapes culti-vated for a long time the cumulative effect of tillage and wa-ter erosion on the redistribution of soil across the slope hasbeen observed (Dlugoszlig et al 2012) These processes alsoproduce a vertical redistribution of Corg resulting in a rela-tively homogeneous profile in the tilled layer (top 15ndash20 cm)and a gradual decline below this depth as noted in our study

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

190 J A Goacutemez et al Organic carbon across an olive orchard catena

Figure 9 Soil available phosphorus (Pavail) organic nitrogen (Norg) aggregate stability (Wsagg) and soil degradation index (SDI) by depthcomparing eroded vs deposition areas within the olive orchard The labels of the bars indicate the p value according to a one-way ANOVAcomparing areas for the same soil depth

Figure 10 Scores on principal components 1 and 2 (PC 1 PC 2)for sampling points in the eroded and deposition areas in the oliveorchard

42 Organic carbon stock

The differences in soil organic carbon stock between the ref-erence site and the olive orchard are similar to those de-scribed previously when comparing cropland and forestedareas with the latter presenting a higher concentration ofCorg most of which is in the unprotected fraction whilethe cropland presented a higher fraction of the carbon inthe physically and chemically protected fractions (eg Poe-plau and Don 2013) This is likely due to the fact that un-der soil degradation processes such as water erosion and

low annual organic carbon input as is the case under oliveorchard land use most of the unprotected Corg decomposesrelatively quickly and a great proportion of the remaininglow SOC is protected In addition the mobilization of theunprotected Corg is expected to be reduced in the protectedforested area because of the canopy and the existing veg-etation on the ground that protects the soil against runoffand splash erosion processes In fact the protected Corg con-centration in the topsoil of the olive orchard in the erodedarea accounted for 186plusmn 39 of the upper limit of pro-tected Corg (364plusmn 023 ) according to the model of Has-sink and Whitmore (1997) Therefore the low unprotectedSOC concentration found in the olive orchard is an elementof concern in the increase in SOC stocks This is because pro-tected fractions are fueled from recently derived partially de-composed plant residues together with microbial and micro-meso- and macrofaunal debris (unprotected organic carbon)through processes like SOC aggregation into macro- andormicroaggregates (physically protected SOC) and complexSOC associations with clay and silt particles (chemicallyprotected SOC) which are disrupted in the cropland area incomparison with the reference area The distribution amongsoil Corg fractions in the orchard of this study was similarto the result obtained by Vicente-Vicente et al (2017) whomeasured Corg fraction distribution in olive oil orchards withtemporary cover crops with the exception of the unprotectedSOC which was much higher in soils under cover crops thanthat of our study under bare soil Also interesting is the dif-ficulty obtaining statistically significant differences in SOC

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 191

Tabl

e7

Pear

son

corr

elat

ion

coef

ficie

nts

oflin

ear

corr

elat

ion

betw

een

vari

able

sat

each

dept

hin

terv

al(c

m)

with

the

high

est

load

son

prin

cipa

lco

mpo

nent

s(s

eeTa

ble

6)lowast

Mea

nsst

atis

tical

lysi

gnifi

cant

atαlt

005

Ero

sde

pP a

vail

0ndash10

P ava

il0ndash

40δ

13C

10ndash2

0N

org

10ndash2

0δ

13C

20ndash3

0C

org

conc

C

org

Phys

C

org

conc

C

org

Phys

C

org

Che

m

Cor

gC

hem

10

ndash20

Pro

conc

20

ndash30

Pro

conc

Pr

oco

nc

Pro

Frac

10

ndash20

20ndash3

020

ndash30

20ndash3

0

Ero

sde

p1

P ava

il0ndash

100

746lowast

1P a

vail

0ndash40

095

3lowast0

857lowast

1δ

13C

10ndash2

0minus

082

3lowastminus

076

6lowast1

Nor

g10

ndash20

073

1lowast0

898lowast

083

9lowastminus

096

5lowast1

δ13

C20

ndash30

minus0

792lowast

minus0

893lowast

minus0

891lowast

088

8lowastminus

093

8lowast1

Cor

gco

nc1

0ndash20

075

7lowast0

843lowast

089

9lowastminus

092

9lowast0

920lowast

minus0

911lowast

1C

org

Phys

Pro

con

c10

ndash20

075

5lowast0

908lowast

088

2lowastminus

092

90

967lowast

minus0

985lowast

095

2lowast1

Cor

gco

nc2

0ndash30

078

6lowast0

799lowast

minus0

965lowast

092

7lowastminus

084

6lowast0

9453lowast

089

42lowast

1C

org

Phys

Pro

con

c20

ndash30

083

7lowast0

756lowast

minus0

846lowast

081

7lowastminus

071

4lowast0

8819lowast

079

28lowast

090

1lowast1

Cor

gC

hem

Pro

con

c20

ndash30

074

2lowast0

855lowast

085

9lowastminus

096

2lowast0

980lowast

minus0

896lowast

095

15lowast

094

34lowast

095

0lowast0

853lowast

1C

org

Che

mP

roF

rac

20ndash3

01

stock between the eroding and deposition areas (Fig 7) de-spite the apparent clear differences between the two areas insome other soil properties such as Pavail or δ15N and δ 13Cisotopic signatures in the subsoil (see below)

43 δ15N and δ13C isotopic signal

Changes in vegetation types induced differences in δ13C be-tween the olive orchard and the reference area but as ex-pected no differences in δ13C were detected between the ero-sion and deposition areas in the olive orchard given the sameorigin of vegetation-derived organic matter ie C3 plantsInterestingly within the olive orchard significant differencesin δ15N were detected between the erosion and deposition ar-eas especially for the top 20 cm of soils (Fig 9) This sug-gests the potential of using δ15N as a variable for identifyingdegraded areas in olive growing fields as has been proposedfor other eroding regions in the world (eg Meusburger etal 2013) This might provide an alternative when other con-ventional or isotopic techniques are not available Neverthe-less further studies exploring this potential are necessary inorder to also consider the influence of the NndashP fertilizer mod-ifying the δ15N in relation to the reference area (eg Bate-man and Kelly 2007) The source of N in soil is multifariousand subject to a wide range of transformations that affect theδ15N signature therefore we can only speculate on the rea-sons for this difference in δ15N in a relatively homogeneousarea Bulk soil δ15N tended to be more positive (eg moreenriched in δ15N) as the N cycling rate increases soil micro-bial processes (eg N mineralization nitrification and deni-trification) resulting in products (eg nitrate N2O N2 andNH3) depleted in 15N while the substrate from which theywere formed becomes slightly enriched (Robison 2001) Thehigher δ15N signature of the soil at the deposition locationsuggests that the rates of processes involved in N cycling arehigher than in the erosion area and that it is in accordancewith the higher bulk Corg and Pavail contents and lower SDIof the deposition site The relatively lower soil δ15N signa-ture at the reference site could be partially due to the input oflitter N from the natural legumes and to the closed N cyclingwhich characterize natural forest ecosystems The trend in15N enrichment with soil depth as found at the referencesite is a common observation at forest and grassland sitesand has been related to different mechanisms including 15Nisotope discrimination during microbial N transformationsdifferential preservation of 15N-enriched soil organic mattercomponents during N decomposition and more recently tothe build-up of 15N-enriched microbial necromass (Huygenset al 2008) However there still remains the need for a care-ful calibration to an undisturbed reference site and a betterunderstanding of the influence of different vegetation in thereference and the studied area on the change in the δ15N sig-nal for its further use as an additional tool to determine soildegradation

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

192 J A Goacutemez et al Organic carbon across an olive orchard catena

44 Soil quality of topsoil across the catena variabilitybetween eroded and deposition areas in the orchard

This horizontal distribution due to tillage and water erosionalso simultaneously affected other soil properties and hasbeen described previously in other cultivated areas For in-stance De Gryze et al (2008) described how for a field croparea under conventional tillage in Belgium Pavail almost dou-bled (229 vs 122 mg kgminus1) in the depositional area as com-pared to the eroding upper part They also reported that inhalf of the field under conservation tillage these differencesin Pavail between the upper and lower areas of the field dis-appeared We have observed in our sampled orchard a pro-nounced increase in topsoil Pavail in the deposition area ofaround 400 as compared to the eroding part of the orchardwhich can be attributed to the deposition of enriched sedi-ment coming from the upslope area The cumulative effectsof the differences in Corg and Pavail and the trend towardshigher although nonsignificant Wsagg in the deposition arearesulted in two areas within the orchard with marked differ-ences in soil quality the eroded part which is within therange considered to be degraded in the region (Goacutemez et al2009a) and the depositional area representing 20 of theorchard transect length (Table 1) which is within the non-degraded range according to the same index Topographyand sediment redistribution by erosive processes introducea gradient of spatial variability that questions the conceptof representative area when it comes to describing a wholefield In fact several studies (eg Dell and Sharpley 2006)have suggested that the verification of compliance of envi-ronmental programs such as those related to Corg seques-tration should be based preferentially at least partially onmodeling approaches Our results raise the need for a care-ful delineation of subareas when analyzing soil quality indi-cators andor SOC carbon stock within the same field unitThey also warn about the difficulty of drawing hypothesesfor quantifying the differences between these delineated ar-eas in relation to specific soil properties For instance in ourstudy case erosionndashdeposition processes had a major impacton Pavail and soil quality but the impact on Corg concentra-tion and stock was moderate and extremely difficult to detectstatistically using a moderate number of samples

Our PCA and regression analyses confirmed the relativelyhigh variability of Corg and stock in relation to other soilquality indicators related to erosionndashdeposition processessuch as Pavail as discussed in the previous section While inthe catena studied in the current research the P cycle seemsto be mostly driven by sediment mobilization the Corg and Ncycles seem to be much more complex The moderate differ-ences in Corg and the homogeneity in δ15N and δ13C isotopicsignals between the eroding and deposition areas may be dueto several processes some of which were discussed abovesuch as spatial variability of carbon input due to biomassin the plot surface soil operations in the orchard and fer-tilization We found it both interesting and worth exploring

in future studies that significant correlations between erosionand deposition rates and Corg- δ15N- and δ 13C-related vari-ables were found for samples from the 10ndash20 and 20ndash30 cmlayers indicating that short-term disturbance by surface pro-cesses can mask experimental determination of the impact oferosion deposition processes in this olive orchard for thesevariables

5 Conclusions

1 The results indicate that erosion and deposition withinthe investigated old olive orchard have created a signifi-cant difference in soil properties along the catena whichis translated into different soil Corg Pavail and Norg con-tent δ15N and thus contrasting soil quality status

2 This variability was lower than that of the natural areawhich indicated a severe depletion of SOC as comparedto the natural area and a redistribution of available or-ganic carbon among the different SOC fractions

3 The results suggest that δ15N has the potential for beingused as an indicator of soil degradation although moreinvestigation in different agroecosystems would be re-quired for confirming this statement on a larger scale

4 This research highlights that proper understanding andmanagement of soil quality and Corg content in olive or-chards require consideration of the on-site spatial vari-ability induced by soil erosionndashdeposition processes

Data availability The basic data used in this paper are freelyavailable from the CSIC digital repository at httpsdoiorg1020350digitalCSIC12513 (Goacutemez Calero et al 2020)

Author contributions JAG coordinated and designed the over-all study conducted the field sampling campaigns analysis of thesoil organic carbon results and the overall analysis of the resultsand wrote the manuscript GG participated in the analysis of soil-quality-related properties the overall analysis of the results and thewriting of the manuscript LM participated in the overall analysisof the results and the writing of the manuscript CR and AT partici-pated in the analysis and interpretation of the stable isotope resultsRGR contributed to the design and the statistical analysis of soilorganic carbon (stock and fractions) and nitrogen data the overallanalysis of the results and the writing of the manuscript

Competing interests The authors declare that they have no con-flict of interest

Acknowledgements The authors would like to thankManuel Gonzaacutelez de Molina for providing valuable insightinto the management of the olive orchard in the decades before

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

180 J A Goacutemez et al Organic carbon across an olive orchard catena

rivers into aquatic ecosystems SOC is the most importantindicator of soil quality (Rajan et al 2010) and erosion-induced loss of SOC affects on-site soil fertility and off-site environment quality (Lal 2019) However the effects ofsoil erosion and the fate of the specific SOC fraction trans-ported by erosion in specific agricultural systems such asolive groves remain poorly understood therefore agroenvi-ronmental impacts of SOC dynamics and variability requiremore site- and crop-specific research

Olive trees one of the most important crops in theMediterranean region which account for approximately97 Mha (FAOSTAT 2019) have been linked to severe envi-ronmental impacts including the acceleration of erosion andsoil degradation (eg Beaufoy 2001 Scheidel and Kraus-mann 2011) In fact soil degradation is common in olive or-chards as they have been traditionally cultivated under rain-fed conditions on sloping land at relatively low tree densi-ties with limited canopy size due to pruning and under bare-soil management to optimize water use by the tree under thesemiarid conditions which characterize the Mediterraneanclimate (Goacutemez et al 2014) Indeed there are many stud-ies which have measured high erosion rates in olive orchardson sloping areas (eg Goacutemez et al 2014) although thesehigh erosion rates are not necessarily a direct consequence ofcurrent management In a study of historical erosion rates inseveral ancient olive orchards of Montefriacuteo (southern Spain)Vanwalleghem et al (2011) reported unsustainable erosionrates in the range of 23 to 68 Mg haminus1 yrminus1 during the 19thand early 20th centuries when these orchards were managedunder the same slope and rainfall conditions with bare soilalbeit based on animal plowing Vanwalleghem et al (2011)also reported a further increase in the erosion rates whenbare-soil management started to be implemented in these or-chards by mechanization and herbicides in the late 20th cen-tury In the last 5 decades (Ruiacutez de Castroviejo 1969) therehas been an attempt to control soil degradation while main-taining a favorable soilndashwater balance for the tree throughthe gradual development of temporary cover crops (grownduring the rainy season) (Goacutemez et al 2014) These higherosion rates have also been linked to the degradation of soilproperties observed in olive orchards For instance Goacutemezet al (2009b) measured in a 5-year experiment and in anolive grove a decrease in SOC aggregate stability and in-filtration rates with bare soil as compared to temporary covercrops Such scientific evidence which links changes in soilproperties to different erosion rates in olive orchards un-der controlled conditions is rarely reported in the literatureIndeed most of the studies aimed to connect soil proper-ties with different types of soil management in olive grovescome from surveys of soil properties in orchards with simi-lar soil types but with different soil management Examplesof these studies are those of Aacutelvarez et al (2010) and Sori-ano et al (2014) who found an improvement in soil prop-erties ndash particularly aggregate stability SOC and biologicalactivity ndash in organic olive orchards with cover crops com-

pared to those with bare soil In recent years these stud-ies have started to deepen our understanding of investigatingkey properties such as SOC For instance Vicente-Vicente etal (2017) evaluated the impact of cover crops in the distri-bution of unprotected and protected SOC in the top 15 cm ofthe soil Typically field studies take samples in a represen-tative area of the slope which is a common assumption inmany soil quality studies (eg Andrews and Carroll 2001)Although there is a limited number of experiments on thespatial variability of soil properties in olive orchards theysuggest the existence of significant in-field variability (egGargouri et al 2013 Huang et al 2017) Moreover Goacutemezet al (2012) suggested that regarding organic carbon part ofthis on-site variability of soil properties might be related toerosionndashdeposition processes

In-field variability associated with erosionndashdeposition pro-cesses is relatively well documented for organic carbon con-tent in field crops (eg De Gryze et al 2008 Mabit andBernard 1998 2010 Van Oost et al 2005) While thehuman-induced acceleration of soil erosion has depleted theSOC stock of agroecosystems the fates of SOC transportedacross the landscape and that deposited in depressional sitesare not fully understood despite the fact that these transfersmight explain a high proportion of the on-site variability ofsoil properties

Most of the erosion rates recorded or established in oliveorchards come from runoff plots or small catchment exper-iments (eg Goacutemez et al 2014) The use of the 137Cs ap-proach has demonstrated its potential in establishing long-term soil erosion rates with this specific land use An exam-ple of these studies is that of Mabit et al (2012) in whicherosion as well as deposition rates since the 1950s was de-termined in one ancient olive orchard in the municipality ofMontefriacuteo showing an average annual rate in the eroding partof the slope of 123 t haminus1 yrminus1 and an average depositionrate in the lower section of the hillslope much shorter thanthe eroding section of 131 t haminus1 yrminus1 This study involveda reference area for establishing precisely the initial 137Csinventory a natural undisturbed area located 200 m from theorchard As reported by Mabit et al (2012) based on 13 in-vestigated soil profiles the local reference 137Cs inventoryin this undisturbed area was evaluated at 1925plusmn250 Bq mminus2

(meanplusmn 2 standard error) with a coefficient of variation (CV)of 23

To complement andor to circumvent some limitation as-sociated with the use of this anthropogenic radioisotope (seeMabit et al 2008) and to maintain the capacity to deter-mine erosion and deposition rates without the need to use di-rect measurements other natural radioisotopes such as 210Pb(eg Mabit et al 2014 Matisoff 2014) or stable isotopessuch as δ15N or δ13C (eg Meusburger et al 2013) havebeen proposed

In this study we hypothesized that the contribution of thelong-term erosionndashdeposition processes to the in-field vari-ability of soil properties in olive orchards (or other woody

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 181

crops) at a medium-steep slope is relevant and should betaken into account when analyzing the effects of specificstrategies on SOC sequestration or on other soil propertiesIn addition we exploited the advantage provided by theunique location of an ancient olive orchard near an undis-turbed reference area and the previous information on thissite from studies on historical erosion rates (Vanwalleghemet al 2011 Mabit et al 2012) to fulfill the following objec-tives

1 to quantify the long-term variability in soil total organiccarbon and in their different fractions as well as soilquality indicators in relation to erosion and depositionareas in a historical olive orchard

2 to evaluate these differences in relation to the referencevalues found in an undisturbed natural area

3 to evaluate differences in stable isotope signatures (δ13Cand δ15N) and explore their potential for identifying de-graded areas within the olive orchard

2 Materials and methods

21 Description of the area

The study area is located in the municipality of Montefriacuteosouthwestern Spain (Fig 1) The municipality extension isaround 220 km2 of which 81 is cultivated mostly witholive trees The climate in the region is continental Mediter-ranean with a long-term (1960ndash2018) average annual pre-cipitation of 630 mm a mean annual evapotranspiration of750 mm and a yearly average temperature of 152 C It is amountainous area with elevation ranging between 800 and1600 m asl at the highest point (Sierra de Parapanda) Soilsampling took place in two areas around the archeologicalsite ldquoPentildea de los Gitanosrdquo where the soil is classified as Cal-cic Cambisol according to the FAO classification (Deckerset al 2004) The undisturbed reference area was inside anarcheological site (Fig 1) This undisturbed area is coveredby open Mediterranean forest interspersed with shrubs andannual grasses on limestone material (calcarenites) The sta-tus of this protected site guarantees that no anthropogenicactivities have impacted it for a long period of time approx-imately since the end of 16th century This area is coveredby natural vegetation typical of the Mediterranean regionmainly bushes like Pistacia lentiscus and Retama sphaero-carpa and herbaceous species such as Anthemis arvensisCalendula arvensis Borago officinalis Brachypodium sppBromus spp and Medicago spp The area studied was anolive orchard established in 1856 located within tens of me-ters of the reference area (Fig 1) Both areas were describedin detail in previous studies on historical erosion rates in theregion (Vanwalleghem et al 2011 Mabit et al 2012) Thisolive orchard is rainfed and soil management in the decadesbefore the sampling was based on bare soil with pruning

residues (trees pruned every 2 years) being chopped and lefton the soil surface Olive trees are fertilized annually with5 kg of 15 NPK spread below the tree canopy area

22 Soil sampling

The reference site adjacent to the olive orchard belongsto an archeologically protected site and is therefore a non-cropped area where neither erosion nor sedimentation pro-cesses take place (Fig 1) This site was sampled at 13 pointsacross a transect spaced at an average distance of 6 m withonly four of these sampling points used in this study (Fig 1)We use only four soil samples because not enough soil for theanalysis was collected in the other samples At each samplingpoint the excavation method was used Soil samples werecollected at 5 cm depth intervals until bedrock was reached(between 20 and 60 cm) The four samples used in this studyreached 40 cm depth or more Composite soil samples at10 cm intervals were prepared at the laboratory to performthe chemical analysis of the reference area as described be-low

In the olive orchard a hydraulic mechanical core samplerwas used It gently rotates and pushes an 8 cm diameter coreto sample eight points along a 452 m long catena (Fig 1) Tominimize soil disturbance soil sampling was made when soilwater content was between 40 and 80 of water-holdingcapacity Precautions were taken to assure that four rangesof soil depth (0ndash10 10ndash20 20ndash30 and 30ndash40 cm) were col-lected at each sampling point in the orchard soil

In a previous study soil erosion and deposition rates weredetermined at each sampling point comparing the 137Cs in-ventory among these points and that of the undisturbed refer-ence area (Mabit et al 2012) The positions of all samplingpoints were recorded by a real-time kinematic global posi-tioning system (RTK-GPS) at submeter resolution (Table 1)Overall 12 points were sampled 4 in the reference area and8 along the catena across the olive orchard with all of themreaching the bedrock below 40 cm depth

23 Physicochemical analysis

Soil samples were passed through a 2 mm sieve and homog-enized The fraction under 2 mm was determined and ex-pressed as a percentage of total soil on a dry mass basisSeparation of the four soil organic carbon (Corg) pools wasperformed by a combination of physical and chemical frac-tionation techniques through a three-step process developedby Six et al (2002) and modified by Stewart et al (2009)summarized here This three-step process isolates a total of12 fractions and it is based on the assumed link between theisolated fractions and the protection mechanisms involvedin the stabilization of organic C First a partial dispersionand physical fractionation of the soil are performed to ob-tain three size fractionsgt 250 mm (coarse nonprotected par-ticulate organic matter POM) 53ndash250 mm (microaggregate

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

182 J A Goacutemez et al Organic carbon across an olive orchard catena

Table 1 Location of the sampling points along the transect and associated soil redistribution rates derived from the 137Cs technique (adaptedfrom Mabit et al 2012) Negative values indicate net erosion and positive values net deposition

Point no Code Distance in transect (m) Elevation (m) Erosion deposition rate (t haminus1 yrminus1)

1 Cs1 0 1044 minus522 Cs3 664 10328 minus1783 Cs5 1250 10178 minus714 Cs7 1790 10068 minus1635 Cs12 3122 9868 minus1526 Cs13 3381 9848 597 Cs15 3888 9818 2478 Cs17 4295 9798 88

Figure 1 Site location and associated sampling scheme (a b c) Location map of the sampling area in Montefriacuteo southern Spain Referencesite delineated by the white line within a protected archeological site (yellow line) Yellow markers in (b) indicate the sampled transect withinthe olive orchard Numbering starts in the points at higher elevation see (d) for the elevation change in the transect in the orchard Yellowmarkers in (c) indicate sampling point in the reference area The map of Europe was designed by Freepik and the aerial images are takenfrom Google Earth (copy Google Earth 2018)

fraction) and lt 53 mm (easily dispersed silt and clay) Thisphysical fractionation is done on air-dried 2 mm soil sievedthrough a 250 mm sieve Material greater than 250 mm re-mained on the sieve Microaggregates were collected on a53 mm sieve that was subsequently wet-sieved to separate the

easily dispersed silt- and clay-sized fractions from the water-stable microaggregates The suspension was centrifuged at127 g for 7 min to separate the silt-sized fraction This su-pernatant was subsequently separated flocculated and cen-trifuged at 1730 g for 15 min to separate the clay-sized frac-

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 183

tion All fractions were dried in a 60 C oven and weighedAfterwards there was a second step involving a further frac-tionation of the microaggregate fraction isolated in the firststep A density flotation with sodium polytungstate was usedto isolate fine nonprotected POM (LF) after removing thefine nonprotected POM the heavy fraction was dispersedovernight by shaking and passed through a 53 mm sieveto separate the microaggregate-protected POM (gt 53 mm insize iPOM) from the microaggregate-derived silt- and clay-sized fractions The resulting suspension was centrifuged toseparate the microaggregate-derived silt-sized fraction fromthe clay-sized fraction as described above A final third stepinvolved the acid hydrolysis of each of the isolated silt- andclay-sized fractions The silt- and clay-sized fractions fromboth the density flotation and the initial dispersion and phys-ical fractionation were subjected to acid hydrolysis The un-protected pool includes the POM and LF fractions isolatedin the first and second fractionation steps respectively Thephysically protected SOC consists of the SOC measured inthe microaggregates It includes not only the iPOM but alsothe hydrolyzable and nonhydrolyzable SOC of the intermedi-ate fraction (53ndash250 microm) The chemically and biochemicallyprotected pools correspond to the hydrolyzable and nonhy-drolyzable SOC in the fine fraction (lt 53 microm) respectivelyIn all cases SOC fractions and in the bulk soil organic car-bon concentrations were determined by using the wet oxida-tion sulfuric acid and potassium dichromate method of An-derson and Ingram (1993)

Inorganic carbon was removed prior to stable isotope anal-ysis by acid fumigation following the method of Harris etal (2001) Moistened subsamples were exposed to the exha-lation of HCl in a desiccator overnight Afterwards the sam-ples were dried at 40 C before the stable isotope ratio wasmeasured The N measurements were done with unacidifiedsamples and the stable N isotope ratios and the C and N con-centrations were measured by isotope ratio mass spectrome-try (Isoprime 100 coupled with an Elementar Vario IsotopeSelect elemental analyzer both instruments supplied by El-ementar Langenselbold Germany) The instrumental stan-dard deviation for δ15N is 016 and 011 for δ13C Sta-ble isotopes are reported as delta values (permil) which are therelative differences between the isotope ratios of the samplesand the isotope ratio of a reference standard

In addition available phosphorus (Pavail) was determinedby the Olsen method (Olsen and Sommers 1982) and or-ganic nitrogen (Norg) was determined by the Kjeldahl method(Stevenson 1982) Water-stable aggregates (Wsagg) weremeasured using the method of Barthes and Roose (2002)Soil particle size distribution was determined using the hy-drometer method (Bouyoucos 1962) for the topsoil (0ndash10 cm) of the reference area and the olive orchard Two bulkdensity values were calculated for the whole profile (a) oneconsidering the mass of soil finer than 2 mm and (b) oneconsidering the mass of soil finer than 2 mm as well as thestone content which was determined from the excavation

and core sampling described above Additionally the bulkdensity of the topsoil (0ndash10 cm depth) was measured usinga manual soil core sample with a volume of 100 cm3 Soilcarbon stocks were calculated for the fine soil fraction af-ter discounting rock or stone fragments larger than 2 mm andconsidering bulk density They were then presented on equiv-alent soil mass as described by Wendt and Hauser (2013) Asproposed by Hassink and Whitmore (1997) theoretical val-ues of carbon saturation were calculated from the soil par-ticle analysis Finally the soil degradation index developedby Goacutemez et al (2009a) was calculated from the Corg Pavailand Wsagg

24 Statistical analysis

The overall effect of depth and area (reference site vs oliveorchard or eroded vs deposition area within the olive or-chard) was evaluated using a two-factor analysis of vari-ance (ANOVA) (p lt 005) Additionally for some compar-ison at similar soil depths values of soil properties betweentwo different areas were assessed using a one-way ANOVAtest (p lt 005) In both situations data were log-transformedwhen necessary to fulfill ANOVA requirements Exploratoryanalysis using principal component analysis (PCA) was per-formed in the olive orchard area using the variables and sam-pling depths showing significant differences in the ANOVAThis PCA was complemented by determining the linear cor-relation coefficient variables showing the highest load on thePCA and erosion deposition rates using the Pearson corre-lation test The statistical software package Stata SE141 wasused for these analyses

3 Results

31 Organic carbon concentration and distributionamong fractions

Table 2 shows the significance of the differences in bulk soilCorg and the various Corg fractions between reference andolive orchard plots at different soil depths and due to the in-teraction between both factors (Table 2a) It also shows theeffects of the erosion deposition ratio (Table 2b) Overallbulk soil Corg was always significantly higher in the referencearea as compared to the olive orchard (00001lt p lt 00013depending on depth) (Table 2a and Fig 2) and this was inde-pendent of the soil sampling depth Corg values on the refer-ence site were between 2 to 5 times higher than those of theolive orchard for a given depth with the greater differencesin the top 10 cm of the soil Soil depth has a significant ef-fect on bulk Corg and Corg fractions with values typically de-creasing with depth in both areas Corg concentrations in theunprotected and physically chemically and biochemicallyprotected fractions were significantly higher (p lt 00001)(Table 2a) in the reference site as compared to the olive or-chard and across the different depths (00001lt p lt 00007)

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

184 J A Goacutemez et al Organic carbon across an olive orchard catena

(Fig 3) Corg values were between 2 to 6 times higher for theunprotected and chemically protected fractions and between2 to 35 times higher for the physically and biochemicallyprotected fractions with differences tending to decrease withthe soil depth

Within the olive orchard there were statistically signifi-cant differences between the erosion and deposition areas forbulk Corg values (p = 00198) (Table 2b) Higher Corg val-ues (11 to 06 ) were observed in the deposition arealocated downslope whereas lower values (085 to 055 )were measured in the areas with net erosion in the upper andmid sections of the catena It is worth noting that these differ-ences between erosion and deposition areas are detected forthe overall analysis using a two-way ANOVA (Table 2b) al-though an individual analysis at each depth (Fig 4) does notdetect statistically significant differences probably due to themoderate number of replications Significant differences be-tween the deposition and eroding area were also found forthe unprotected (p = 004) and the physically (p = 00077)and chemically protected fractions (p = 00299) (Table 2b)However differences for the biochemically protected frac-tions (Table 2b Fig 4) were not significant

The percentage distribution of SOC among fractions wasalso significantly different between both areas (reference vsolive orchard) except for the biochemically protected fraction(p lt 00001) (Table 3a) and depths (p lt 0011) (Fig 5) Inthe reference area most of the Corg was unprotected (between50 and 65 approximately) with no significant trend withdepth (Table 3a Fig 5) followed in relative importance bythe chemically and physically protected fractions which con-tributed between 18 and 30 as well as 10 and 20 ofthe bulk soil Corg respectively The biochemically protectedfraction represents a very low percentage (between 4 and6 ) In the olive orchard Corg is stored predominantly inthe physically and chemically protected fractions which ac-counted for about 38 to 27 and 34 to 28 respec-tively followed by the pool of unprotected fractions (be-tween 22 to 32 ) (Fig 5) The biochemically protectedfraction represents between 4 and 11 of the organic car-bon stored in the olive orchard There are no clear differencesin the organic carbon distribution among the different frac-tions between the erosion and deposition areas in the oliveorchard with the exception of the physically protected frac-tions at 10ndash20 cm depth (p = 0024) (Fig 6)

32 Organic carbon stock

SOC stock in the reference area is approximately 160 t haminus1making it significantly higher at p lt 005 for an equivalentmass than that of the olive orchard (Fig 7) which stores be-tween 38 and 41 t haminus1 in the eroded and deposition areasrespectively There were no significant differences betweenthese two orchard areas Similar results were obtained acrossthe top 40 cm soil profile Clay silt and sand contents of thetopsoil (0ndash10 cm) along the catena in the olive orchard av-

eraged 41 37 and 22 respectively with similar av-erage values in the eroding and deposition areas Variabilitywas relatively low (average coefficient of variation of 17 )and there were no significant changes between the erosionand deposition areas In the reference area the soil has anaverage clay silt and sand content of 30 31 and 39 respectively also with a homogeneous distribution across thesampling area (coefficient of variation of 10 ) According tothe Hassink and Whitmore (1997) model the percentages oforganic carbon of maximum soil stable Corg are 324plusmn011 and 363plusmn019 in the reference site and olive orchard re-spectively So protected Corg in the reference and olive or-chard areas account for 498plusmn 115 and 2049plusmn 52 ofthe maximum soil stable Corg respectively in the topsoil

33 δ15N and δ13C isotopic signal

Figure 8 and Table 4a compare stable isotope delta values be-tween the reference site and the overall olive orchard accord-ing to depth There are significant statistical differences inδ15N δ 13C and the δ13C δ15N ratio between the two areas(p lt 0002 Table 4a) although in the case of δ 15N the dif-ferences at individual depths were not significant Soil depthalso had a significant effect (00026lt p lt 0029 Table 4a)When comparing differences between the erosion and depo-sition areas within the olive orchard we detected statisticallysignificant differences only in δ15N and the δ13C δ15N ra-tio (p = 001 Table 4b) and they were mostly marked in theupper 20 cm of the soil (Fig 8)

34 Soil quality of topsoil across the catena variabilitybetween eroded and deposition areas in the orchard

Figure 9 depicts the comparison of the Pavail Norg Wsaggand the soil degradation index (SDI Goacutemez et al 2009a)in the top 10 cm of the soil between the erosion and deposi-tion areas of the olive orchard and Table 5 shows a similarcomparison for Norg Pavail and bulk density at the differentsoil depths Pavail in the deposition area is significantly higherthan that of the erosion area in the topsoil (p = 0005 Fig 9)and for the whole profile (p = 001 Table 5) whereas no sig-nificant differences in individual soil layers were found forNorg and Wsagg The SDI which combines these three vari-ables was about 3 times higher in the eroded area than in thedeposition area

Table 6 shows the loads in the first three principal com-ponents (PCs) of the principal component analysis (PCA)for the variables used in this analysis More than 70 ofthe variance was explained by the first two PCs Soils ofthe eroded and deposition area were clearly separated in thespace defined by the two PCs (Fig 10) The variables withhigher contributions in PC 1 and 2 were related to Pavail con-centration in the 0ndash10 and 0ndash40 cm soil depths to δ15N andδ13C in the 10 to 30 cm soil depths and to Corg concentra-tion and distribution in some fractions also in soil depths be-

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 185

Figure 2 Comparison of average soil organic carbon concentration in bulk soil depending on soil depth distinguishing between the referencesite the whole olive orchard the eroded area of the olive orchard and the deposition area in this orchard The labels of the bars for eachdepth indicate the p value according to a one-way ANOVA for the reference area (dark blue) vs the whole olive orchard (dark green) andfor the eroding area (beige) vs the deposition area (light green lower label in italics)

Figure 3 Organic carbon concentration in the different soil organic carbon fractions at each depth comparing the reference site vs oliveorchard The labels of the bars indicate the p value according to a one-way ANOVA comparing treatments for the same soil depth and carbonfraction between the reference area and olive orchard

tween 10 and 30 cm (Table 6) The deposition area tendedto show higher values in PC 1 and PC 2 and there wasno clear tendency in the erosion area along the catena Thelinear correlation coefficient between the variables and theerosion deposition rate was rather significant (r gtplusmn0731)for most of the variables (Table 7) Interestingly Pavail was

highly positively correlated in the whole soil layer Amongthe two most robust correlations with erosion depositionrates were that with Pavail concentration across the 40 cm soildepth y = 03975x+ 98364 (r2

= 0907) and δ13C at 10ndash20 cm depth y =minus00094xminus 25318 (r2

= 0632)

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

186 J A Goacutemez et al Organic carbon across an olive orchard catena

Table 2 Results of the two-way ANOVA of soil organic carbon concentration Corg () in different fractions and in bulk soil In (a) arearefers to the reference site vs the olive orchard and in (b) area refers to eroded vs deposition areas in the olive orchard NS stands for notsignificant

(a) Model Bulk soil Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00000 00000 00000 00000 00000Depth (D) 00023 00022 00061 00190 NSAtimesD NS NS NS NS 00300

(b) Model Bulk soil Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00198 00400 00077 00299 NSDepth (D) 00081 00070 00058 00055 00847AtimesD NS NS NS NS NS

Figure 4 Organic carbon concentration in the different soil organic carbon fractions at each depth comparing the eroded vs the depositionarea within the olive orchard The labels of the bars indicate the p value according to a one-way ANOVA comparing treatments for the samesoil depth and carbon fraction between the reference area and olive orchard

4 Discussion

41 Organic carbon concentration and distributionamong fractions

After approximately 175 years of contrasted land use be-tween the undisturbed reference site and the olive orchardbulk soil organic carbon concentration and its fractions havebeen dramatically reduced in the olive orchard Current lev-els of Corg concentration in the soil profile are approximately20 to 25 of that found in the reference area covered bynatural vegetation in the area adjacent to the orchard This ra-

tio is similar albeit in the lower range to the comparison ofCorg in topsoil among olive orchards with different types ofmanagement and natural areas reported for the region (Mil-groom et al 2007) The increased soil disturbance the lowerannual rate of biomass returned to the soil and the highererosion rate in the olive orchard explain this difference Inboth areas the Corg is clearly stratified indicating that de-spite the different mechanisms involved there is a periodicinput of biomass from the olive trees (eg fall of senescenceleaves and tree pruning residues) plus the annual ground veg-etation Vicente-Vicente et al (2017) estimated this biomasscontribution in the range of 148 to 056 t haminus1 yrminus1 It is

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 187

Figure 5 Contribution () of the different fractions with respect to total soil organic carbon by depth comparing the reference site vs theolive orchard The labels of each fraction and depth are the p value according to a one-way ANOVA comparing the reference area vs theolive orchard for the same carbon fraction and soil depth

Figure 6 Fraction of total organic carbon stored in the different fractions by depth comparing the reference site vs the olive orchard Thelabels of each fraction and depth are the p value according to a one-way ANOVA comparing the reference area vs the olive orchard for thesame carbon fraction and soil depth

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

188 J A Goacutemez et al Organic carbon across an olive orchard catena

Table 3 Results of the two-way ANOVA of the distribution of the total soil organic carbon content in the soil among the different fractionsof soil organic carbon Corg In (a) area refers to the reference site vs the olive orchard and in (b) area refers to eroded vs deposition areasin the olive orchard NS stands for not significant

(a) Model Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00000 00000 00000 NSDepth (D) NS NS 00640 NSAtimesD NS 00059 NS NS

(b) Model Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 0091 00881 NS NSDepth (D) 0051 00214 NS 0033AtimesD NS NS NS NS

Figure 7 Cumulative soil organic carbon (SOC) stock across thesoil profile in terms of cumulative soil mass on the vertical axisDifferent letters for similar soil mass refer to statistically significantdifferences (KruskalndashWallis test at p lt 005) For this analysis cu-mulative soil organic carbons were interpolated linearly to the aver-age cumulative soil mass corresponding to all the points in the threeareas

worth noticing that the decrease in Corg as compared to thenatural area is much higher than the reported rates of increasein Corg in olive orchards using conservation agriculture (CA)techniques such as cover crops and incorporation of organicresidues from different sources In a meta-analysis Vicente-Vicente et al (2016) found a response ratio (the ratio of Corgunder CA management as compared to Corg in orchards withbare-soil management) of 11 to 19 suggesting that underCA management which combines cover crops and organicresidues Corg doubled as a maximum

Table 4 Results of the two-way ANOVA of the stable isotope sig-nal In (a) area refers to the reference site vs the olive orchard andin (b) area refers to eroded vs deposition areas in the olive orchardNS stands for not significant

(a) Model δ13C δ15N δ13C δ15N

Area (A) 00001 0002 0002Depth (D) 00026 00175 0029AtimesD NS NS NS

(b) Model δ13C δ15N δ13C δ15N

Area (A) NS 001 001Depth (D) NS NS NSAtimesD NS NS NS

Table 5 Results of the two-way ANOVA of some physical andchemical soil properties comparing eroded vs deposition areas inthe olive orchard NS stands for not significant

Model Norg Pavail Bulk density

Area (A) 00000 001 NSDepth (D) 00009 NS NSAtimesD NS NS NS

Combining all Corg data of the olive orchard the variabil-ity was about 35 which is similar to what has been re-ported so far in the few studies found on soil Corg variabilityin olive orchards For instance Gargouri et al (2013) indi-cated a 24 coefficient of variation (CV) in a 34 ha oliveorchard in Tunisia while Huang et al (2017) reported an av-erage CV of 41 in a 62 ha olive orchard in southern SpainNeither of these two studies reported clear trends in the distri-bution of Corg depending on topography Huang et al (2017)

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 189

Figure 8 13C and 15N isotopic signal of soil by depth distinguish-ing among reference site the whole olive orchard the eroded areaof the olive orchard and the deposition area in this orchard Thelabels of the bars for each depth indicate the p value according toa one-way ANOVA between the reference area (dark blue) vs thewhole olive orchard (dark green) and the eroding area (beige) vsthe deposition area (light green lower label in italics)

pointed out the additional difficulties in the determination ofCorg due to the topography heterogeneity although this wascompounded by the fact that within the orchards there weretwo areas with different planting dates for the trees Goacutemezet al (2012) reported a CV of 49 with higher Corg in ar-eas where there was a change in the slope gradient from thehillslope to a central channel draining into the catchment al-though they could not find a simple relationship between theincrease in content and the topographic indices Despite thefact that a lot of work has been done on the correlation be-tween erosion and deposition and the redistribution of soilCorg (eg Van Oost et al 2005) our study is to our knowl-edge the first attempt to quantify this in detail under the

Table 6 Loads of selected variables in the PCA for the first threeprincipal components (PCs) The values in brackets below PC 1 2and 3 indicate the percentage of variance explained by this PC Vari-ables in bold are those with a load higher than 90 of the variablewith the maximum load for this PC Conc refers to Corg concen-tration for this fraction and Frac means the relative contribution ofthis fraction to the total Corg for this soil depth

Variable at each depth interval (cm) PC 1 PC 2 PC 3(558) (176) (132)

Pavail 0ndash10 02298 008765 minus007278Pavail 0ndash40 02271 0101 minus008455δ13C 0ndash10 minus003385 02861 minus03302δ15N 0ndash10 01941 01677 01836Norg 0ndash10 02147 minus01375 minus00336δ13C δ15N 0ndash10 01594 0249 02211δ13C 10ndash20 minus02329 01461 004296δ15N 10ndash20 01856 007054 03121Norg 10ndash20 02317 minus008699 minus01404δ13C δ15N 10ndash20 01637 008512 03309δ13C 20ndash30 minus02316 006018 009789δ15N 20ndash30 01748 minus02656 01812Norg 20ndash30 02176 009979 minus007458δ13C δ15N 20ndash30 01684 minus02982 01774Corg conc 10ndash20 02421 minus002951 minus006407Corg unpr conc 10ndash20 02116 00331 00385Corg unpr Frac 10ndash20 minus009663 minus00509 0364Corg Phys Pro conc 10ndash20 02371 minus005728 minus01183Corg Phys Pro Frac 10ndash20 0123 0163 minus002953Corg Chem Pro conc 10ndash20 02072 003097 minus02335Corg Chem Pro Frac 10ndash20 minus004095 009461 minus04678Corg conc 20ndash30 02326 minus01108 minus005737Corg unpr conc 20ndash30 02002 minus02372 minus005071Corg unpr Rel 20ndash30 cm minus0132 minus03675 minus01379Corg Phys Pro conc 20ndash30 02257 005677 minus001036Corg Phys Pro Frac 20ndash30 009518 03528 01294Corg Chem Pro conc 20ndash30 02323 minus006802 minus01366Corg Chem Pro Frac 20ndash30 004565 04437 minus001278

agroenvironmental conditions of an olive orchard The vari-ability induced by the combined effects of water and tillageerosion in this olive orchard was similar to that described inother agroecosystems For instance Van Oost et al (2005)measured a clear correlation between the erosion and depo-sition rates and the topsoil Corg concentration which rangedbetween 08 of the erosion to 14 of the deposition sitesin the top 25 cm of the soil at two field crop sites under atemperate climate Besides this Bameri et al (2015) mea-sured a higher Corg in the lower part of a field crop site undersemiarid conditions where deposition of the eroded soil fromthe upper zones took place Overall in such landscapes culti-vated for a long time the cumulative effect of tillage and wa-ter erosion on the redistribution of soil across the slope hasbeen observed (Dlugoszlig et al 2012) These processes alsoproduce a vertical redistribution of Corg resulting in a rela-tively homogeneous profile in the tilled layer (top 15ndash20 cm)and a gradual decline below this depth as noted in our study

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

190 J A Goacutemez et al Organic carbon across an olive orchard catena

Figure 9 Soil available phosphorus (Pavail) organic nitrogen (Norg) aggregate stability (Wsagg) and soil degradation index (SDI) by depthcomparing eroded vs deposition areas within the olive orchard The labels of the bars indicate the p value according to a one-way ANOVAcomparing areas for the same soil depth

Figure 10 Scores on principal components 1 and 2 (PC 1 PC 2)for sampling points in the eroded and deposition areas in the oliveorchard

42 Organic carbon stock

The differences in soil organic carbon stock between the ref-erence site and the olive orchard are similar to those de-scribed previously when comparing cropland and forestedareas with the latter presenting a higher concentration ofCorg most of which is in the unprotected fraction whilethe cropland presented a higher fraction of the carbon inthe physically and chemically protected fractions (eg Poe-plau and Don 2013) This is likely due to the fact that un-der soil degradation processes such as water erosion and

low annual organic carbon input as is the case under oliveorchard land use most of the unprotected Corg decomposesrelatively quickly and a great proportion of the remaininglow SOC is protected In addition the mobilization of theunprotected Corg is expected to be reduced in the protectedforested area because of the canopy and the existing veg-etation on the ground that protects the soil against runoffand splash erosion processes In fact the protected Corg con-centration in the topsoil of the olive orchard in the erodedarea accounted for 186plusmn 39 of the upper limit of pro-tected Corg (364plusmn 023 ) according to the model of Has-sink and Whitmore (1997) Therefore the low unprotectedSOC concentration found in the olive orchard is an elementof concern in the increase in SOC stocks This is because pro-tected fractions are fueled from recently derived partially de-composed plant residues together with microbial and micro-meso- and macrofaunal debris (unprotected organic carbon)through processes like SOC aggregation into macro- andormicroaggregates (physically protected SOC) and complexSOC associations with clay and silt particles (chemicallyprotected SOC) which are disrupted in the cropland area incomparison with the reference area The distribution amongsoil Corg fractions in the orchard of this study was similarto the result obtained by Vicente-Vicente et al (2017) whomeasured Corg fraction distribution in olive oil orchards withtemporary cover crops with the exception of the unprotectedSOC which was much higher in soils under cover crops thanthat of our study under bare soil Also interesting is the dif-ficulty obtaining statistically significant differences in SOC

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 191

Tabl

e7

Pear

son

corr

elat

ion

coef

ficie

nts

oflin

ear

corr

elat

ion

betw

een

vari

able

sat

each

dept

hin

terv

al(c

m)

with

the

high

est

load

son

prin

cipa

lco

mpo

nent

s(s

eeTa

ble

6)lowast

Mea

nsst

atis

tical

lysi

gnifi

cant

atαlt

005

Ero

sde

pP a

vail

0ndash10

P ava

il0ndash

40δ

13C

10ndash2

0N

org

10ndash2

0δ

13C

20ndash3

0C

org

conc

C

org

Phys

C

org

conc

C

org

Phys

C

org

Che

m

Cor

gC

hem

10

ndash20

Pro

conc

20

ndash30

Pro

conc

Pr

oco

nc

Pro

Frac

10

ndash20

20ndash3

020

ndash30

20ndash3

0

Ero

sde

p1

P ava

il0ndash

100

746lowast

1P a

vail

0ndash40

095

3lowast0

857lowast

1δ

13C

10ndash2

0minus

082

3lowastminus

076

6lowast1

Nor

g10

ndash20

073

1lowast0

898lowast

083

9lowastminus

096

5lowast1

δ13

C20

ndash30

minus0

792lowast

minus0

893lowast

minus0

891lowast

088

8lowastminus

093

8lowast1

Cor

gco

nc1

0ndash20

075

7lowast0

843lowast

089

9lowastminus

092

9lowast0

920lowast

minus0

911lowast

1C

org

Phys

Pro

con

c10

ndash20

075

5lowast0

908lowast

088

2lowastminus

092

90

967lowast

minus0

985lowast

095

2lowast1

Cor

gco

nc2

0ndash30

078

6lowast0

799lowast

minus0

965lowast

092

7lowastminus

084

6lowast0

9453lowast

089

42lowast

1C

org

Phys

Pro

con

c20

ndash30

083

7lowast0

756lowast

minus0

846lowast

081

7lowastminus

071

4lowast0

8819lowast

079

28lowast

090

1lowast1

Cor

gC

hem

Pro

con

c20

ndash30

074

2lowast0

855lowast

085

9lowastminus

096

2lowast0

980lowast

minus0

896lowast

095

15lowast

094

34lowast

095

0lowast0

853lowast

1C

org

Che

mP

roF

rac

20ndash3

01

stock between the eroding and deposition areas (Fig 7) de-spite the apparent clear differences between the two areas insome other soil properties such as Pavail or δ15N and δ 13Cisotopic signatures in the subsoil (see below)

43 δ15N and δ13C isotopic signal

Changes in vegetation types induced differences in δ13C be-tween the olive orchard and the reference area but as ex-pected no differences in δ13C were detected between the ero-sion and deposition areas in the olive orchard given the sameorigin of vegetation-derived organic matter ie C3 plantsInterestingly within the olive orchard significant differencesin δ15N were detected between the erosion and deposition ar-eas especially for the top 20 cm of soils (Fig 9) This sug-gests the potential of using δ15N as a variable for identifyingdegraded areas in olive growing fields as has been proposedfor other eroding regions in the world (eg Meusburger etal 2013) This might provide an alternative when other con-ventional or isotopic techniques are not available Neverthe-less further studies exploring this potential are necessary inorder to also consider the influence of the NndashP fertilizer mod-ifying the δ15N in relation to the reference area (eg Bate-man and Kelly 2007) The source of N in soil is multifariousand subject to a wide range of transformations that affect theδ15N signature therefore we can only speculate on the rea-sons for this difference in δ15N in a relatively homogeneousarea Bulk soil δ15N tended to be more positive (eg moreenriched in δ15N) as the N cycling rate increases soil micro-bial processes (eg N mineralization nitrification and deni-trification) resulting in products (eg nitrate N2O N2 andNH3) depleted in 15N while the substrate from which theywere formed becomes slightly enriched (Robison 2001) Thehigher δ15N signature of the soil at the deposition locationsuggests that the rates of processes involved in N cycling arehigher than in the erosion area and that it is in accordancewith the higher bulk Corg and Pavail contents and lower SDIof the deposition site The relatively lower soil δ15N signa-ture at the reference site could be partially due to the input oflitter N from the natural legumes and to the closed N cyclingwhich characterize natural forest ecosystems The trend in15N enrichment with soil depth as found at the referencesite is a common observation at forest and grassland sitesand has been related to different mechanisms including 15Nisotope discrimination during microbial N transformationsdifferential preservation of 15N-enriched soil organic mattercomponents during N decomposition and more recently tothe build-up of 15N-enriched microbial necromass (Huygenset al 2008) However there still remains the need for a care-ful calibration to an undisturbed reference site and a betterunderstanding of the influence of different vegetation in thereference and the studied area on the change in the δ15N sig-nal for its further use as an additional tool to determine soildegradation

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

192 J A Goacutemez et al Organic carbon across an olive orchard catena

44 Soil quality of topsoil across the catena variabilitybetween eroded and deposition areas in the orchard

This horizontal distribution due to tillage and water erosionalso simultaneously affected other soil properties and hasbeen described previously in other cultivated areas For in-stance De Gryze et al (2008) described how for a field croparea under conventional tillage in Belgium Pavail almost dou-bled (229 vs 122 mg kgminus1) in the depositional area as com-pared to the eroding upper part They also reported that inhalf of the field under conservation tillage these differencesin Pavail between the upper and lower areas of the field dis-appeared We have observed in our sampled orchard a pro-nounced increase in topsoil Pavail in the deposition area ofaround 400 as compared to the eroding part of the orchardwhich can be attributed to the deposition of enriched sedi-ment coming from the upslope area The cumulative effectsof the differences in Corg and Pavail and the trend towardshigher although nonsignificant Wsagg in the deposition arearesulted in two areas within the orchard with marked differ-ences in soil quality the eroded part which is within therange considered to be degraded in the region (Goacutemez et al2009a) and the depositional area representing 20 of theorchard transect length (Table 1) which is within the non-degraded range according to the same index Topographyand sediment redistribution by erosive processes introducea gradient of spatial variability that questions the conceptof representative area when it comes to describing a wholefield In fact several studies (eg Dell and Sharpley 2006)have suggested that the verification of compliance of envi-ronmental programs such as those related to Corg seques-tration should be based preferentially at least partially onmodeling approaches Our results raise the need for a care-ful delineation of subareas when analyzing soil quality indi-cators andor SOC carbon stock within the same field unitThey also warn about the difficulty of drawing hypothesesfor quantifying the differences between these delineated ar-eas in relation to specific soil properties For instance in ourstudy case erosionndashdeposition processes had a major impacton Pavail and soil quality but the impact on Corg concentra-tion and stock was moderate and extremely difficult to detectstatistically using a moderate number of samples

Our PCA and regression analyses confirmed the relativelyhigh variability of Corg and stock in relation to other soilquality indicators related to erosionndashdeposition processessuch as Pavail as discussed in the previous section While inthe catena studied in the current research the P cycle seemsto be mostly driven by sediment mobilization the Corg and Ncycles seem to be much more complex The moderate differ-ences in Corg and the homogeneity in δ15N and δ13C isotopicsignals between the eroding and deposition areas may be dueto several processes some of which were discussed abovesuch as spatial variability of carbon input due to biomassin the plot surface soil operations in the orchard and fer-tilization We found it both interesting and worth exploring

in future studies that significant correlations between erosionand deposition rates and Corg- δ15N- and δ 13C-related vari-ables were found for samples from the 10ndash20 and 20ndash30 cmlayers indicating that short-term disturbance by surface pro-cesses can mask experimental determination of the impact oferosion deposition processes in this olive orchard for thesevariables

5 Conclusions

1 The results indicate that erosion and deposition withinthe investigated old olive orchard have created a signifi-cant difference in soil properties along the catena whichis translated into different soil Corg Pavail and Norg con-tent δ15N and thus contrasting soil quality status

2 This variability was lower than that of the natural areawhich indicated a severe depletion of SOC as comparedto the natural area and a redistribution of available or-ganic carbon among the different SOC fractions

3 The results suggest that δ15N has the potential for beingused as an indicator of soil degradation although moreinvestigation in different agroecosystems would be re-quired for confirming this statement on a larger scale

4 This research highlights that proper understanding andmanagement of soil quality and Corg content in olive or-chards require consideration of the on-site spatial vari-ability induced by soil erosionndashdeposition processes

Data availability The basic data used in this paper are freelyavailable from the CSIC digital repository at httpsdoiorg1020350digitalCSIC12513 (Goacutemez Calero et al 2020)

Author contributions JAG coordinated and designed the over-all study conducted the field sampling campaigns analysis of thesoil organic carbon results and the overall analysis of the resultsand wrote the manuscript GG participated in the analysis of soil-quality-related properties the overall analysis of the results and thewriting of the manuscript LM participated in the overall analysisof the results and the writing of the manuscript CR and AT partici-pated in the analysis and interpretation of the stable isotope resultsRGR contributed to the design and the statistical analysis of soilorganic carbon (stock and fractions) and nitrogen data the overallanalysis of the results and the writing of the manuscript

Competing interests The authors declare that they have no con-flict of interest

Acknowledgements The authors would like to thankManuel Gonzaacutelez de Molina for providing valuable insightinto the management of the olive orchard in the decades before

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 181

crops) at a medium-steep slope is relevant and should betaken into account when analyzing the effects of specificstrategies on SOC sequestration or on other soil propertiesIn addition we exploited the advantage provided by theunique location of an ancient olive orchard near an undis-turbed reference area and the previous information on thissite from studies on historical erosion rates (Vanwalleghemet al 2011 Mabit et al 2012) to fulfill the following objec-tives

1 to quantify the long-term variability in soil total organiccarbon and in their different fractions as well as soilquality indicators in relation to erosion and depositionareas in a historical olive orchard

2 to evaluate these differences in relation to the referencevalues found in an undisturbed natural area

3 to evaluate differences in stable isotope signatures (δ13Cand δ15N) and explore their potential for identifying de-graded areas within the olive orchard

2 Materials and methods

21 Description of the area

The study area is located in the municipality of Montefriacuteosouthwestern Spain (Fig 1) The municipality extension isaround 220 km2 of which 81 is cultivated mostly witholive trees The climate in the region is continental Mediter-ranean with a long-term (1960ndash2018) average annual pre-cipitation of 630 mm a mean annual evapotranspiration of750 mm and a yearly average temperature of 152 C It is amountainous area with elevation ranging between 800 and1600 m asl at the highest point (Sierra de Parapanda) Soilsampling took place in two areas around the archeologicalsite ldquoPentildea de los Gitanosrdquo where the soil is classified as Cal-cic Cambisol according to the FAO classification (Deckerset al 2004) The undisturbed reference area was inside anarcheological site (Fig 1) This undisturbed area is coveredby open Mediterranean forest interspersed with shrubs andannual grasses on limestone material (calcarenites) The sta-tus of this protected site guarantees that no anthropogenicactivities have impacted it for a long period of time approx-imately since the end of 16th century This area is coveredby natural vegetation typical of the Mediterranean regionmainly bushes like Pistacia lentiscus and Retama sphaero-carpa and herbaceous species such as Anthemis arvensisCalendula arvensis Borago officinalis Brachypodium sppBromus spp and Medicago spp The area studied was anolive orchard established in 1856 located within tens of me-ters of the reference area (Fig 1) Both areas were describedin detail in previous studies on historical erosion rates in theregion (Vanwalleghem et al 2011 Mabit et al 2012) Thisolive orchard is rainfed and soil management in the decadesbefore the sampling was based on bare soil with pruning

residues (trees pruned every 2 years) being chopped and lefton the soil surface Olive trees are fertilized annually with5 kg of 15 NPK spread below the tree canopy area

22 Soil sampling

The reference site adjacent to the olive orchard belongsto an archeologically protected site and is therefore a non-cropped area where neither erosion nor sedimentation pro-cesses take place (Fig 1) This site was sampled at 13 pointsacross a transect spaced at an average distance of 6 m withonly four of these sampling points used in this study (Fig 1)We use only four soil samples because not enough soil for theanalysis was collected in the other samples At each samplingpoint the excavation method was used Soil samples werecollected at 5 cm depth intervals until bedrock was reached(between 20 and 60 cm) The four samples used in this studyreached 40 cm depth or more Composite soil samples at10 cm intervals were prepared at the laboratory to performthe chemical analysis of the reference area as described be-low

In the olive orchard a hydraulic mechanical core samplerwas used It gently rotates and pushes an 8 cm diameter coreto sample eight points along a 452 m long catena (Fig 1) Tominimize soil disturbance soil sampling was made when soilwater content was between 40 and 80 of water-holdingcapacity Precautions were taken to assure that four rangesof soil depth (0ndash10 10ndash20 20ndash30 and 30ndash40 cm) were col-lected at each sampling point in the orchard soil

In a previous study soil erosion and deposition rates weredetermined at each sampling point comparing the 137Cs in-ventory among these points and that of the undisturbed refer-ence area (Mabit et al 2012) The positions of all samplingpoints were recorded by a real-time kinematic global posi-tioning system (RTK-GPS) at submeter resolution (Table 1)Overall 12 points were sampled 4 in the reference area and8 along the catena across the olive orchard with all of themreaching the bedrock below 40 cm depth

23 Physicochemical analysis

Soil samples were passed through a 2 mm sieve and homog-enized The fraction under 2 mm was determined and ex-pressed as a percentage of total soil on a dry mass basisSeparation of the four soil organic carbon (Corg) pools wasperformed by a combination of physical and chemical frac-tionation techniques through a three-step process developedby Six et al (2002) and modified by Stewart et al (2009)summarized here This three-step process isolates a total of12 fractions and it is based on the assumed link between theisolated fractions and the protection mechanisms involvedin the stabilization of organic C First a partial dispersionand physical fractionation of the soil are performed to ob-tain three size fractionsgt 250 mm (coarse nonprotected par-ticulate organic matter POM) 53ndash250 mm (microaggregate

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

182 J A Goacutemez et al Organic carbon across an olive orchard catena

Table 1 Location of the sampling points along the transect and associated soil redistribution rates derived from the 137Cs technique (adaptedfrom Mabit et al 2012) Negative values indicate net erosion and positive values net deposition

Point no Code Distance in transect (m) Elevation (m) Erosion deposition rate (t haminus1 yrminus1)

1 Cs1 0 1044 minus522 Cs3 664 10328 minus1783 Cs5 1250 10178 minus714 Cs7 1790 10068 minus1635 Cs12 3122 9868 minus1526 Cs13 3381 9848 597 Cs15 3888 9818 2478 Cs17 4295 9798 88

Figure 1 Site location and associated sampling scheme (a b c) Location map of the sampling area in Montefriacuteo southern Spain Referencesite delineated by the white line within a protected archeological site (yellow line) Yellow markers in (b) indicate the sampled transect withinthe olive orchard Numbering starts in the points at higher elevation see (d) for the elevation change in the transect in the orchard Yellowmarkers in (c) indicate sampling point in the reference area The map of Europe was designed by Freepik and the aerial images are takenfrom Google Earth (copy Google Earth 2018)

fraction) and lt 53 mm (easily dispersed silt and clay) Thisphysical fractionation is done on air-dried 2 mm soil sievedthrough a 250 mm sieve Material greater than 250 mm re-mained on the sieve Microaggregates were collected on a53 mm sieve that was subsequently wet-sieved to separate the

easily dispersed silt- and clay-sized fractions from the water-stable microaggregates The suspension was centrifuged at127 g for 7 min to separate the silt-sized fraction This su-pernatant was subsequently separated flocculated and cen-trifuged at 1730 g for 15 min to separate the clay-sized frac-

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 183

tion All fractions were dried in a 60 C oven and weighedAfterwards there was a second step involving a further frac-tionation of the microaggregate fraction isolated in the firststep A density flotation with sodium polytungstate was usedto isolate fine nonprotected POM (LF) after removing thefine nonprotected POM the heavy fraction was dispersedovernight by shaking and passed through a 53 mm sieveto separate the microaggregate-protected POM (gt 53 mm insize iPOM) from the microaggregate-derived silt- and clay-sized fractions The resulting suspension was centrifuged toseparate the microaggregate-derived silt-sized fraction fromthe clay-sized fraction as described above A final third stepinvolved the acid hydrolysis of each of the isolated silt- andclay-sized fractions The silt- and clay-sized fractions fromboth the density flotation and the initial dispersion and phys-ical fractionation were subjected to acid hydrolysis The un-protected pool includes the POM and LF fractions isolatedin the first and second fractionation steps respectively Thephysically protected SOC consists of the SOC measured inthe microaggregates It includes not only the iPOM but alsothe hydrolyzable and nonhydrolyzable SOC of the intermedi-ate fraction (53ndash250 microm) The chemically and biochemicallyprotected pools correspond to the hydrolyzable and nonhy-drolyzable SOC in the fine fraction (lt 53 microm) respectivelyIn all cases SOC fractions and in the bulk soil organic car-bon concentrations were determined by using the wet oxida-tion sulfuric acid and potassium dichromate method of An-derson and Ingram (1993)

Inorganic carbon was removed prior to stable isotope anal-ysis by acid fumigation following the method of Harris etal (2001) Moistened subsamples were exposed to the exha-lation of HCl in a desiccator overnight Afterwards the sam-ples were dried at 40 C before the stable isotope ratio wasmeasured The N measurements were done with unacidifiedsamples and the stable N isotope ratios and the C and N con-centrations were measured by isotope ratio mass spectrome-try (Isoprime 100 coupled with an Elementar Vario IsotopeSelect elemental analyzer both instruments supplied by El-ementar Langenselbold Germany) The instrumental stan-dard deviation for δ15N is 016 and 011 for δ13C Sta-ble isotopes are reported as delta values (permil) which are therelative differences between the isotope ratios of the samplesand the isotope ratio of a reference standard

In addition available phosphorus (Pavail) was determinedby the Olsen method (Olsen and Sommers 1982) and or-ganic nitrogen (Norg) was determined by the Kjeldahl method(Stevenson 1982) Water-stable aggregates (Wsagg) weremeasured using the method of Barthes and Roose (2002)Soil particle size distribution was determined using the hy-drometer method (Bouyoucos 1962) for the topsoil (0ndash10 cm) of the reference area and the olive orchard Two bulkdensity values were calculated for the whole profile (a) oneconsidering the mass of soil finer than 2 mm and (b) oneconsidering the mass of soil finer than 2 mm as well as thestone content which was determined from the excavation

and core sampling described above Additionally the bulkdensity of the topsoil (0ndash10 cm depth) was measured usinga manual soil core sample with a volume of 100 cm3 Soilcarbon stocks were calculated for the fine soil fraction af-ter discounting rock or stone fragments larger than 2 mm andconsidering bulk density They were then presented on equiv-alent soil mass as described by Wendt and Hauser (2013) Asproposed by Hassink and Whitmore (1997) theoretical val-ues of carbon saturation were calculated from the soil par-ticle analysis Finally the soil degradation index developedby Goacutemez et al (2009a) was calculated from the Corg Pavailand Wsagg

24 Statistical analysis

The overall effect of depth and area (reference site vs oliveorchard or eroded vs deposition area within the olive or-chard) was evaluated using a two-factor analysis of vari-ance (ANOVA) (p lt 005) Additionally for some compar-ison at similar soil depths values of soil properties betweentwo different areas were assessed using a one-way ANOVAtest (p lt 005) In both situations data were log-transformedwhen necessary to fulfill ANOVA requirements Exploratoryanalysis using principal component analysis (PCA) was per-formed in the olive orchard area using the variables and sam-pling depths showing significant differences in the ANOVAThis PCA was complemented by determining the linear cor-relation coefficient variables showing the highest load on thePCA and erosion deposition rates using the Pearson corre-lation test The statistical software package Stata SE141 wasused for these analyses

3 Results

31 Organic carbon concentration and distributionamong fractions

Table 2 shows the significance of the differences in bulk soilCorg and the various Corg fractions between reference andolive orchard plots at different soil depths and due to the in-teraction between both factors (Table 2a) It also shows theeffects of the erosion deposition ratio (Table 2b) Overallbulk soil Corg was always significantly higher in the referencearea as compared to the olive orchard (00001lt p lt 00013depending on depth) (Table 2a and Fig 2) and this was inde-pendent of the soil sampling depth Corg values on the refer-ence site were between 2 to 5 times higher than those of theolive orchard for a given depth with the greater differencesin the top 10 cm of the soil Soil depth has a significant ef-fect on bulk Corg and Corg fractions with values typically de-creasing with depth in both areas Corg concentrations in theunprotected and physically chemically and biochemicallyprotected fractions were significantly higher (p lt 00001)(Table 2a) in the reference site as compared to the olive or-chard and across the different depths (00001lt p lt 00007)

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

184 J A Goacutemez et al Organic carbon across an olive orchard catena

(Fig 3) Corg values were between 2 to 6 times higher for theunprotected and chemically protected fractions and between2 to 35 times higher for the physically and biochemicallyprotected fractions with differences tending to decrease withthe soil depth

Within the olive orchard there were statistically signifi-cant differences between the erosion and deposition areas forbulk Corg values (p = 00198) (Table 2b) Higher Corg val-ues (11 to 06 ) were observed in the deposition arealocated downslope whereas lower values (085 to 055 )were measured in the areas with net erosion in the upper andmid sections of the catena It is worth noting that these differ-ences between erosion and deposition areas are detected forthe overall analysis using a two-way ANOVA (Table 2b) al-though an individual analysis at each depth (Fig 4) does notdetect statistically significant differences probably due to themoderate number of replications Significant differences be-tween the deposition and eroding area were also found forthe unprotected (p = 004) and the physically (p = 00077)and chemically protected fractions (p = 00299) (Table 2b)However differences for the biochemically protected frac-tions (Table 2b Fig 4) were not significant

The percentage distribution of SOC among fractions wasalso significantly different between both areas (reference vsolive orchard) except for the biochemically protected fraction(p lt 00001) (Table 3a) and depths (p lt 0011) (Fig 5) Inthe reference area most of the Corg was unprotected (between50 and 65 approximately) with no significant trend withdepth (Table 3a Fig 5) followed in relative importance bythe chemically and physically protected fractions which con-tributed between 18 and 30 as well as 10 and 20 ofthe bulk soil Corg respectively The biochemically protectedfraction represents a very low percentage (between 4 and6 ) In the olive orchard Corg is stored predominantly inthe physically and chemically protected fractions which ac-counted for about 38 to 27 and 34 to 28 respec-tively followed by the pool of unprotected fractions (be-tween 22 to 32 ) (Fig 5) The biochemically protectedfraction represents between 4 and 11 of the organic car-bon stored in the olive orchard There are no clear differencesin the organic carbon distribution among the different frac-tions between the erosion and deposition areas in the oliveorchard with the exception of the physically protected frac-tions at 10ndash20 cm depth (p = 0024) (Fig 6)

32 Organic carbon stock

SOC stock in the reference area is approximately 160 t haminus1making it significantly higher at p lt 005 for an equivalentmass than that of the olive orchard (Fig 7) which stores be-tween 38 and 41 t haminus1 in the eroded and deposition areasrespectively There were no significant differences betweenthese two orchard areas Similar results were obtained acrossthe top 40 cm soil profile Clay silt and sand contents of thetopsoil (0ndash10 cm) along the catena in the olive orchard av-

eraged 41 37 and 22 respectively with similar av-erage values in the eroding and deposition areas Variabilitywas relatively low (average coefficient of variation of 17 )and there were no significant changes between the erosionand deposition areas In the reference area the soil has anaverage clay silt and sand content of 30 31 and 39 respectively also with a homogeneous distribution across thesampling area (coefficient of variation of 10 ) According tothe Hassink and Whitmore (1997) model the percentages oforganic carbon of maximum soil stable Corg are 324plusmn011 and 363plusmn019 in the reference site and olive orchard re-spectively So protected Corg in the reference and olive or-chard areas account for 498plusmn 115 and 2049plusmn 52 ofthe maximum soil stable Corg respectively in the topsoil

33 δ15N and δ13C isotopic signal

Figure 8 and Table 4a compare stable isotope delta values be-tween the reference site and the overall olive orchard accord-ing to depth There are significant statistical differences inδ15N δ 13C and the δ13C δ15N ratio between the two areas(p lt 0002 Table 4a) although in the case of δ 15N the dif-ferences at individual depths were not significant Soil depthalso had a significant effect (00026lt p lt 0029 Table 4a)When comparing differences between the erosion and depo-sition areas within the olive orchard we detected statisticallysignificant differences only in δ15N and the δ13C δ15N ra-tio (p = 001 Table 4b) and they were mostly marked in theupper 20 cm of the soil (Fig 8)

34 Soil quality of topsoil across the catena variabilitybetween eroded and deposition areas in the orchard

Figure 9 depicts the comparison of the Pavail Norg Wsaggand the soil degradation index (SDI Goacutemez et al 2009a)in the top 10 cm of the soil between the erosion and deposi-tion areas of the olive orchard and Table 5 shows a similarcomparison for Norg Pavail and bulk density at the differentsoil depths Pavail in the deposition area is significantly higherthan that of the erosion area in the topsoil (p = 0005 Fig 9)and for the whole profile (p = 001 Table 5) whereas no sig-nificant differences in individual soil layers were found forNorg and Wsagg The SDI which combines these three vari-ables was about 3 times higher in the eroded area than in thedeposition area

Table 6 shows the loads in the first three principal com-ponents (PCs) of the principal component analysis (PCA)for the variables used in this analysis More than 70 ofthe variance was explained by the first two PCs Soils ofthe eroded and deposition area were clearly separated in thespace defined by the two PCs (Fig 10) The variables withhigher contributions in PC 1 and 2 were related to Pavail con-centration in the 0ndash10 and 0ndash40 cm soil depths to δ15N andδ13C in the 10 to 30 cm soil depths and to Corg concentra-tion and distribution in some fractions also in soil depths be-

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 185

Figure 2 Comparison of average soil organic carbon concentration in bulk soil depending on soil depth distinguishing between the referencesite the whole olive orchard the eroded area of the olive orchard and the deposition area in this orchard The labels of the bars for eachdepth indicate the p value according to a one-way ANOVA for the reference area (dark blue) vs the whole olive orchard (dark green) andfor the eroding area (beige) vs the deposition area (light green lower label in italics)

Figure 3 Organic carbon concentration in the different soil organic carbon fractions at each depth comparing the reference site vs oliveorchard The labels of the bars indicate the p value according to a one-way ANOVA comparing treatments for the same soil depth and carbonfraction between the reference area and olive orchard

tween 10 and 30 cm (Table 6) The deposition area tendedto show higher values in PC 1 and PC 2 and there wasno clear tendency in the erosion area along the catena Thelinear correlation coefficient between the variables and theerosion deposition rate was rather significant (r gtplusmn0731)for most of the variables (Table 7) Interestingly Pavail was

highly positively correlated in the whole soil layer Amongthe two most robust correlations with erosion depositionrates were that with Pavail concentration across the 40 cm soildepth y = 03975x+ 98364 (r2

= 0907) and δ13C at 10ndash20 cm depth y =minus00094xminus 25318 (r2

= 0632)

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

186 J A Goacutemez et al Organic carbon across an olive orchard catena

Table 2 Results of the two-way ANOVA of soil organic carbon concentration Corg () in different fractions and in bulk soil In (a) arearefers to the reference site vs the olive orchard and in (b) area refers to eroded vs deposition areas in the olive orchard NS stands for notsignificant

(a) Model Bulk soil Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00000 00000 00000 00000 00000Depth (D) 00023 00022 00061 00190 NSAtimesD NS NS NS NS 00300

(b) Model Bulk soil Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00198 00400 00077 00299 NSDepth (D) 00081 00070 00058 00055 00847AtimesD NS NS NS NS NS

Figure 4 Organic carbon concentration in the different soil organic carbon fractions at each depth comparing the eroded vs the depositionarea within the olive orchard The labels of the bars indicate the p value according to a one-way ANOVA comparing treatments for the samesoil depth and carbon fraction between the reference area and olive orchard

4 Discussion

41 Organic carbon concentration and distributionamong fractions

After approximately 175 years of contrasted land use be-tween the undisturbed reference site and the olive orchardbulk soil organic carbon concentration and its fractions havebeen dramatically reduced in the olive orchard Current lev-els of Corg concentration in the soil profile are approximately20 to 25 of that found in the reference area covered bynatural vegetation in the area adjacent to the orchard This ra-

tio is similar albeit in the lower range to the comparison ofCorg in topsoil among olive orchards with different types ofmanagement and natural areas reported for the region (Mil-groom et al 2007) The increased soil disturbance the lowerannual rate of biomass returned to the soil and the highererosion rate in the olive orchard explain this difference Inboth areas the Corg is clearly stratified indicating that de-spite the different mechanisms involved there is a periodicinput of biomass from the olive trees (eg fall of senescenceleaves and tree pruning residues) plus the annual ground veg-etation Vicente-Vicente et al (2017) estimated this biomasscontribution in the range of 148 to 056 t haminus1 yrminus1 It is

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 187

Figure 5 Contribution () of the different fractions with respect to total soil organic carbon by depth comparing the reference site vs theolive orchard The labels of each fraction and depth are the p value according to a one-way ANOVA comparing the reference area vs theolive orchard for the same carbon fraction and soil depth

Figure 6 Fraction of total organic carbon stored in the different fractions by depth comparing the reference site vs the olive orchard Thelabels of each fraction and depth are the p value according to a one-way ANOVA comparing the reference area vs the olive orchard for thesame carbon fraction and soil depth

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

188 J A Goacutemez et al Organic carbon across an olive orchard catena

Table 3 Results of the two-way ANOVA of the distribution of the total soil organic carbon content in the soil among the different fractionsof soil organic carbon Corg In (a) area refers to the reference site vs the olive orchard and in (b) area refers to eroded vs deposition areasin the olive orchard NS stands for not significant

(a) Model Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00000 00000 00000 NSDepth (D) NS NS 00640 NSAtimesD NS 00059 NS NS

(b) Model Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 0091 00881 NS NSDepth (D) 0051 00214 NS 0033AtimesD NS NS NS NS

Figure 7 Cumulative soil organic carbon (SOC) stock across thesoil profile in terms of cumulative soil mass on the vertical axisDifferent letters for similar soil mass refer to statistically significantdifferences (KruskalndashWallis test at p lt 005) For this analysis cu-mulative soil organic carbons were interpolated linearly to the aver-age cumulative soil mass corresponding to all the points in the threeareas

worth noticing that the decrease in Corg as compared to thenatural area is much higher than the reported rates of increasein Corg in olive orchards using conservation agriculture (CA)techniques such as cover crops and incorporation of organicresidues from different sources In a meta-analysis Vicente-Vicente et al (2016) found a response ratio (the ratio of Corgunder CA management as compared to Corg in orchards withbare-soil management) of 11 to 19 suggesting that underCA management which combines cover crops and organicresidues Corg doubled as a maximum

Table 4 Results of the two-way ANOVA of the stable isotope sig-nal In (a) area refers to the reference site vs the olive orchard andin (b) area refers to eroded vs deposition areas in the olive orchardNS stands for not significant

(a) Model δ13C δ15N δ13C δ15N

Area (A) 00001 0002 0002Depth (D) 00026 00175 0029AtimesD NS NS NS

(b) Model δ13C δ15N δ13C δ15N

Area (A) NS 001 001Depth (D) NS NS NSAtimesD NS NS NS

Table 5 Results of the two-way ANOVA of some physical andchemical soil properties comparing eroded vs deposition areas inthe olive orchard NS stands for not significant

Model Norg Pavail Bulk density

Area (A) 00000 001 NSDepth (D) 00009 NS NSAtimesD NS NS NS

Combining all Corg data of the olive orchard the variabil-ity was about 35 which is similar to what has been re-ported so far in the few studies found on soil Corg variabilityin olive orchards For instance Gargouri et al (2013) indi-cated a 24 coefficient of variation (CV) in a 34 ha oliveorchard in Tunisia while Huang et al (2017) reported an av-erage CV of 41 in a 62 ha olive orchard in southern SpainNeither of these two studies reported clear trends in the distri-bution of Corg depending on topography Huang et al (2017)

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 189

Figure 8 13C and 15N isotopic signal of soil by depth distinguish-ing among reference site the whole olive orchard the eroded areaof the olive orchard and the deposition area in this orchard Thelabels of the bars for each depth indicate the p value according toa one-way ANOVA between the reference area (dark blue) vs thewhole olive orchard (dark green) and the eroding area (beige) vsthe deposition area (light green lower label in italics)

pointed out the additional difficulties in the determination ofCorg due to the topography heterogeneity although this wascompounded by the fact that within the orchards there weretwo areas with different planting dates for the trees Goacutemezet al (2012) reported a CV of 49 with higher Corg in ar-eas where there was a change in the slope gradient from thehillslope to a central channel draining into the catchment al-though they could not find a simple relationship between theincrease in content and the topographic indices Despite thefact that a lot of work has been done on the correlation be-tween erosion and deposition and the redistribution of soilCorg (eg Van Oost et al 2005) our study is to our knowl-edge the first attempt to quantify this in detail under the

Table 6 Loads of selected variables in the PCA for the first threeprincipal components (PCs) The values in brackets below PC 1 2and 3 indicate the percentage of variance explained by this PC Vari-ables in bold are those with a load higher than 90 of the variablewith the maximum load for this PC Conc refers to Corg concen-tration for this fraction and Frac means the relative contribution ofthis fraction to the total Corg for this soil depth

Variable at each depth interval (cm) PC 1 PC 2 PC 3(558) (176) (132)

Pavail 0ndash10 02298 008765 minus007278Pavail 0ndash40 02271 0101 minus008455δ13C 0ndash10 minus003385 02861 minus03302δ15N 0ndash10 01941 01677 01836Norg 0ndash10 02147 minus01375 minus00336δ13C δ15N 0ndash10 01594 0249 02211δ13C 10ndash20 minus02329 01461 004296δ15N 10ndash20 01856 007054 03121Norg 10ndash20 02317 minus008699 minus01404δ13C δ15N 10ndash20 01637 008512 03309δ13C 20ndash30 minus02316 006018 009789δ15N 20ndash30 01748 minus02656 01812Norg 20ndash30 02176 009979 minus007458δ13C δ15N 20ndash30 01684 minus02982 01774Corg conc 10ndash20 02421 minus002951 minus006407Corg unpr conc 10ndash20 02116 00331 00385Corg unpr Frac 10ndash20 minus009663 minus00509 0364Corg Phys Pro conc 10ndash20 02371 minus005728 minus01183Corg Phys Pro Frac 10ndash20 0123 0163 minus002953Corg Chem Pro conc 10ndash20 02072 003097 minus02335Corg Chem Pro Frac 10ndash20 minus004095 009461 minus04678Corg conc 20ndash30 02326 minus01108 minus005737Corg unpr conc 20ndash30 02002 minus02372 minus005071Corg unpr Rel 20ndash30 cm minus0132 minus03675 minus01379Corg Phys Pro conc 20ndash30 02257 005677 minus001036Corg Phys Pro Frac 20ndash30 009518 03528 01294Corg Chem Pro conc 20ndash30 02323 minus006802 minus01366Corg Chem Pro Frac 20ndash30 004565 04437 minus001278

agroenvironmental conditions of an olive orchard The vari-ability induced by the combined effects of water and tillageerosion in this olive orchard was similar to that described inother agroecosystems For instance Van Oost et al (2005)measured a clear correlation between the erosion and depo-sition rates and the topsoil Corg concentration which rangedbetween 08 of the erosion to 14 of the deposition sitesin the top 25 cm of the soil at two field crop sites under atemperate climate Besides this Bameri et al (2015) mea-sured a higher Corg in the lower part of a field crop site undersemiarid conditions where deposition of the eroded soil fromthe upper zones took place Overall in such landscapes culti-vated for a long time the cumulative effect of tillage and wa-ter erosion on the redistribution of soil across the slope hasbeen observed (Dlugoszlig et al 2012) These processes alsoproduce a vertical redistribution of Corg resulting in a rela-tively homogeneous profile in the tilled layer (top 15ndash20 cm)and a gradual decline below this depth as noted in our study

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

190 J A Goacutemez et al Organic carbon across an olive orchard catena

Figure 9 Soil available phosphorus (Pavail) organic nitrogen (Norg) aggregate stability (Wsagg) and soil degradation index (SDI) by depthcomparing eroded vs deposition areas within the olive orchard The labels of the bars indicate the p value according to a one-way ANOVAcomparing areas for the same soil depth

Figure 10 Scores on principal components 1 and 2 (PC 1 PC 2)for sampling points in the eroded and deposition areas in the oliveorchard

42 Organic carbon stock

The differences in soil organic carbon stock between the ref-erence site and the olive orchard are similar to those de-scribed previously when comparing cropland and forestedareas with the latter presenting a higher concentration ofCorg most of which is in the unprotected fraction whilethe cropland presented a higher fraction of the carbon inthe physically and chemically protected fractions (eg Poe-plau and Don 2013) This is likely due to the fact that un-der soil degradation processes such as water erosion and

low annual organic carbon input as is the case under oliveorchard land use most of the unprotected Corg decomposesrelatively quickly and a great proportion of the remaininglow SOC is protected In addition the mobilization of theunprotected Corg is expected to be reduced in the protectedforested area because of the canopy and the existing veg-etation on the ground that protects the soil against runoffand splash erosion processes In fact the protected Corg con-centration in the topsoil of the olive orchard in the erodedarea accounted for 186plusmn 39 of the upper limit of pro-tected Corg (364plusmn 023 ) according to the model of Has-sink and Whitmore (1997) Therefore the low unprotectedSOC concentration found in the olive orchard is an elementof concern in the increase in SOC stocks This is because pro-tected fractions are fueled from recently derived partially de-composed plant residues together with microbial and micro-meso- and macrofaunal debris (unprotected organic carbon)through processes like SOC aggregation into macro- andormicroaggregates (physically protected SOC) and complexSOC associations with clay and silt particles (chemicallyprotected SOC) which are disrupted in the cropland area incomparison with the reference area The distribution amongsoil Corg fractions in the orchard of this study was similarto the result obtained by Vicente-Vicente et al (2017) whomeasured Corg fraction distribution in olive oil orchards withtemporary cover crops with the exception of the unprotectedSOC which was much higher in soils under cover crops thanthat of our study under bare soil Also interesting is the dif-ficulty obtaining statistically significant differences in SOC

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 191

Tabl

e7

Pear

son

corr

elat

ion

coef

ficie

nts

oflin

ear

corr

elat

ion

betw

een

vari

able

sat

each

dept

hin

terv

al(c

m)

with

the

high

est

load

son

prin

cipa

lco

mpo

nent

s(s

eeTa

ble

6)lowast

Mea

nsst

atis

tical

lysi

gnifi

cant

atαlt

005

Ero

sde

pP a

vail

0ndash10

P ava

il0ndash

40δ

13C

10ndash2

0N

org

10ndash2

0δ

13C

20ndash3

0C

org

conc

C

org

Phys

C

org

conc

C

org

Phys

C

org

Che

m

Cor

gC

hem

10

ndash20

Pro

conc

20

ndash30

Pro

conc

Pr

oco

nc

Pro

Frac

10

ndash20

20ndash3

020

ndash30

20ndash3

0

Ero

sde

p1

P ava

il0ndash

100

746lowast

1P a

vail

0ndash40

095

3lowast0

857lowast

1δ

13C

10ndash2

0minus

082

3lowastminus

076

6lowast1

Nor

g10

ndash20

073

1lowast0

898lowast

083

9lowastminus

096

5lowast1

δ13

C20

ndash30

minus0

792lowast

minus0

893lowast

minus0

891lowast

088

8lowastminus

093

8lowast1

Cor

gco

nc1

0ndash20

075

7lowast0

843lowast

089

9lowastminus

092

9lowast0

920lowast

minus0

911lowast

1C

org

Phys

Pro

con

c10

ndash20

075

5lowast0

908lowast

088

2lowastminus

092

90

967lowast

minus0

985lowast

095

2lowast1

Cor

gco

nc2

0ndash30

078

6lowast0

799lowast

minus0

965lowast

092

7lowastminus

084

6lowast0

9453lowast

089

42lowast

1C

org

Phys

Pro

con

c20

ndash30

083

7lowast0

756lowast

minus0

846lowast

081

7lowastminus

071

4lowast0

8819lowast

079

28lowast

090

1lowast1

Cor

gC

hem

Pro

con

c20

ndash30

074

2lowast0

855lowast

085

9lowastminus

096

2lowast0

980lowast

minus0

896lowast

095

15lowast

094

34lowast

095

0lowast0

853lowast

1C

org

Che

mP

roF

rac

20ndash3

01

stock between the eroding and deposition areas (Fig 7) de-spite the apparent clear differences between the two areas insome other soil properties such as Pavail or δ15N and δ 13Cisotopic signatures in the subsoil (see below)

43 δ15N and δ13C isotopic signal

Changes in vegetation types induced differences in δ13C be-tween the olive orchard and the reference area but as ex-pected no differences in δ13C were detected between the ero-sion and deposition areas in the olive orchard given the sameorigin of vegetation-derived organic matter ie C3 plantsInterestingly within the olive orchard significant differencesin δ15N were detected between the erosion and deposition ar-eas especially for the top 20 cm of soils (Fig 9) This sug-gests the potential of using δ15N as a variable for identifyingdegraded areas in olive growing fields as has been proposedfor other eroding regions in the world (eg Meusburger etal 2013) This might provide an alternative when other con-ventional or isotopic techniques are not available Neverthe-less further studies exploring this potential are necessary inorder to also consider the influence of the NndashP fertilizer mod-ifying the δ15N in relation to the reference area (eg Bate-man and Kelly 2007) The source of N in soil is multifariousand subject to a wide range of transformations that affect theδ15N signature therefore we can only speculate on the rea-sons for this difference in δ15N in a relatively homogeneousarea Bulk soil δ15N tended to be more positive (eg moreenriched in δ15N) as the N cycling rate increases soil micro-bial processes (eg N mineralization nitrification and deni-trification) resulting in products (eg nitrate N2O N2 andNH3) depleted in 15N while the substrate from which theywere formed becomes slightly enriched (Robison 2001) Thehigher δ15N signature of the soil at the deposition locationsuggests that the rates of processes involved in N cycling arehigher than in the erosion area and that it is in accordancewith the higher bulk Corg and Pavail contents and lower SDIof the deposition site The relatively lower soil δ15N signa-ture at the reference site could be partially due to the input oflitter N from the natural legumes and to the closed N cyclingwhich characterize natural forest ecosystems The trend in15N enrichment with soil depth as found at the referencesite is a common observation at forest and grassland sitesand has been related to different mechanisms including 15Nisotope discrimination during microbial N transformationsdifferential preservation of 15N-enriched soil organic mattercomponents during N decomposition and more recently tothe build-up of 15N-enriched microbial necromass (Huygenset al 2008) However there still remains the need for a care-ful calibration to an undisturbed reference site and a betterunderstanding of the influence of different vegetation in thereference and the studied area on the change in the δ15N sig-nal for its further use as an additional tool to determine soildegradation

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

192 J A Goacutemez et al Organic carbon across an olive orchard catena

44 Soil quality of topsoil across the catena variabilitybetween eroded and deposition areas in the orchard

This horizontal distribution due to tillage and water erosionalso simultaneously affected other soil properties and hasbeen described previously in other cultivated areas For in-stance De Gryze et al (2008) described how for a field croparea under conventional tillage in Belgium Pavail almost dou-bled (229 vs 122 mg kgminus1) in the depositional area as com-pared to the eroding upper part They also reported that inhalf of the field under conservation tillage these differencesin Pavail between the upper and lower areas of the field dis-appeared We have observed in our sampled orchard a pro-nounced increase in topsoil Pavail in the deposition area ofaround 400 as compared to the eroding part of the orchardwhich can be attributed to the deposition of enriched sedi-ment coming from the upslope area The cumulative effectsof the differences in Corg and Pavail and the trend towardshigher although nonsignificant Wsagg in the deposition arearesulted in two areas within the orchard with marked differ-ences in soil quality the eroded part which is within therange considered to be degraded in the region (Goacutemez et al2009a) and the depositional area representing 20 of theorchard transect length (Table 1) which is within the non-degraded range according to the same index Topographyand sediment redistribution by erosive processes introducea gradient of spatial variability that questions the conceptof representative area when it comes to describing a wholefield In fact several studies (eg Dell and Sharpley 2006)have suggested that the verification of compliance of envi-ronmental programs such as those related to Corg seques-tration should be based preferentially at least partially onmodeling approaches Our results raise the need for a care-ful delineation of subareas when analyzing soil quality indi-cators andor SOC carbon stock within the same field unitThey also warn about the difficulty of drawing hypothesesfor quantifying the differences between these delineated ar-eas in relation to specific soil properties For instance in ourstudy case erosionndashdeposition processes had a major impacton Pavail and soil quality but the impact on Corg concentra-tion and stock was moderate and extremely difficult to detectstatistically using a moderate number of samples

Our PCA and regression analyses confirmed the relativelyhigh variability of Corg and stock in relation to other soilquality indicators related to erosionndashdeposition processessuch as Pavail as discussed in the previous section While inthe catena studied in the current research the P cycle seemsto be mostly driven by sediment mobilization the Corg and Ncycles seem to be much more complex The moderate differ-ences in Corg and the homogeneity in δ15N and δ13C isotopicsignals between the eroding and deposition areas may be dueto several processes some of which were discussed abovesuch as spatial variability of carbon input due to biomassin the plot surface soil operations in the orchard and fer-tilization We found it both interesting and worth exploring

in future studies that significant correlations between erosionand deposition rates and Corg- δ15N- and δ 13C-related vari-ables were found for samples from the 10ndash20 and 20ndash30 cmlayers indicating that short-term disturbance by surface pro-cesses can mask experimental determination of the impact oferosion deposition processes in this olive orchard for thesevariables

5 Conclusions

1 The results indicate that erosion and deposition withinthe investigated old olive orchard have created a signifi-cant difference in soil properties along the catena whichis translated into different soil Corg Pavail and Norg con-tent δ15N and thus contrasting soil quality status

2 This variability was lower than that of the natural areawhich indicated a severe depletion of SOC as comparedto the natural area and a redistribution of available or-ganic carbon among the different SOC fractions

3 The results suggest that δ15N has the potential for beingused as an indicator of soil degradation although moreinvestigation in different agroecosystems would be re-quired for confirming this statement on a larger scale

4 This research highlights that proper understanding andmanagement of soil quality and Corg content in olive or-chards require consideration of the on-site spatial vari-ability induced by soil erosionndashdeposition processes

Data availability The basic data used in this paper are freelyavailable from the CSIC digital repository at httpsdoiorg1020350digitalCSIC12513 (Goacutemez Calero et al 2020)

Author contributions JAG coordinated and designed the over-all study conducted the field sampling campaigns analysis of thesoil organic carbon results and the overall analysis of the resultsand wrote the manuscript GG participated in the analysis of soil-quality-related properties the overall analysis of the results and thewriting of the manuscript LM participated in the overall analysisof the results and the writing of the manuscript CR and AT partici-pated in the analysis and interpretation of the stable isotope resultsRGR contributed to the design and the statistical analysis of soilorganic carbon (stock and fractions) and nitrogen data the overallanalysis of the results and the writing of the manuscript

Competing interests The authors declare that they have no con-flict of interest

Acknowledgements The authors would like to thankManuel Gonzaacutelez de Molina for providing valuable insightinto the management of the olive orchard in the decades before

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

182 J A Goacutemez et al Organic carbon across an olive orchard catena

Table 1 Location of the sampling points along the transect and associated soil redistribution rates derived from the 137Cs technique (adaptedfrom Mabit et al 2012) Negative values indicate net erosion and positive values net deposition

Point no Code Distance in transect (m) Elevation (m) Erosion deposition rate (t haminus1 yrminus1)

1 Cs1 0 1044 minus522 Cs3 664 10328 minus1783 Cs5 1250 10178 minus714 Cs7 1790 10068 minus1635 Cs12 3122 9868 minus1526 Cs13 3381 9848 597 Cs15 3888 9818 2478 Cs17 4295 9798 88

Figure 1 Site location and associated sampling scheme (a b c) Location map of the sampling area in Montefriacuteo southern Spain Referencesite delineated by the white line within a protected archeological site (yellow line) Yellow markers in (b) indicate the sampled transect withinthe olive orchard Numbering starts in the points at higher elevation see (d) for the elevation change in the transect in the orchard Yellowmarkers in (c) indicate sampling point in the reference area The map of Europe was designed by Freepik and the aerial images are takenfrom Google Earth (copy Google Earth 2018)

fraction) and lt 53 mm (easily dispersed silt and clay) Thisphysical fractionation is done on air-dried 2 mm soil sievedthrough a 250 mm sieve Material greater than 250 mm re-mained on the sieve Microaggregates were collected on a53 mm sieve that was subsequently wet-sieved to separate the

easily dispersed silt- and clay-sized fractions from the water-stable microaggregates The suspension was centrifuged at127 g for 7 min to separate the silt-sized fraction This su-pernatant was subsequently separated flocculated and cen-trifuged at 1730 g for 15 min to separate the clay-sized frac-

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 183

tion All fractions were dried in a 60 C oven and weighedAfterwards there was a second step involving a further frac-tionation of the microaggregate fraction isolated in the firststep A density flotation with sodium polytungstate was usedto isolate fine nonprotected POM (LF) after removing thefine nonprotected POM the heavy fraction was dispersedovernight by shaking and passed through a 53 mm sieveto separate the microaggregate-protected POM (gt 53 mm insize iPOM) from the microaggregate-derived silt- and clay-sized fractions The resulting suspension was centrifuged toseparate the microaggregate-derived silt-sized fraction fromthe clay-sized fraction as described above A final third stepinvolved the acid hydrolysis of each of the isolated silt- andclay-sized fractions The silt- and clay-sized fractions fromboth the density flotation and the initial dispersion and phys-ical fractionation were subjected to acid hydrolysis The un-protected pool includes the POM and LF fractions isolatedin the first and second fractionation steps respectively Thephysically protected SOC consists of the SOC measured inthe microaggregates It includes not only the iPOM but alsothe hydrolyzable and nonhydrolyzable SOC of the intermedi-ate fraction (53ndash250 microm) The chemically and biochemicallyprotected pools correspond to the hydrolyzable and nonhy-drolyzable SOC in the fine fraction (lt 53 microm) respectivelyIn all cases SOC fractions and in the bulk soil organic car-bon concentrations were determined by using the wet oxida-tion sulfuric acid and potassium dichromate method of An-derson and Ingram (1993)

Inorganic carbon was removed prior to stable isotope anal-ysis by acid fumigation following the method of Harris etal (2001) Moistened subsamples were exposed to the exha-lation of HCl in a desiccator overnight Afterwards the sam-ples were dried at 40 C before the stable isotope ratio wasmeasured The N measurements were done with unacidifiedsamples and the stable N isotope ratios and the C and N con-centrations were measured by isotope ratio mass spectrome-try (Isoprime 100 coupled with an Elementar Vario IsotopeSelect elemental analyzer both instruments supplied by El-ementar Langenselbold Germany) The instrumental stan-dard deviation for δ15N is 016 and 011 for δ13C Sta-ble isotopes are reported as delta values (permil) which are therelative differences between the isotope ratios of the samplesand the isotope ratio of a reference standard

In addition available phosphorus (Pavail) was determinedby the Olsen method (Olsen and Sommers 1982) and or-ganic nitrogen (Norg) was determined by the Kjeldahl method(Stevenson 1982) Water-stable aggregates (Wsagg) weremeasured using the method of Barthes and Roose (2002)Soil particle size distribution was determined using the hy-drometer method (Bouyoucos 1962) for the topsoil (0ndash10 cm) of the reference area and the olive orchard Two bulkdensity values were calculated for the whole profile (a) oneconsidering the mass of soil finer than 2 mm and (b) oneconsidering the mass of soil finer than 2 mm as well as thestone content which was determined from the excavation

and core sampling described above Additionally the bulkdensity of the topsoil (0ndash10 cm depth) was measured usinga manual soil core sample with a volume of 100 cm3 Soilcarbon stocks were calculated for the fine soil fraction af-ter discounting rock or stone fragments larger than 2 mm andconsidering bulk density They were then presented on equiv-alent soil mass as described by Wendt and Hauser (2013) Asproposed by Hassink and Whitmore (1997) theoretical val-ues of carbon saturation were calculated from the soil par-ticle analysis Finally the soil degradation index developedby Goacutemez et al (2009a) was calculated from the Corg Pavailand Wsagg

24 Statistical analysis

The overall effect of depth and area (reference site vs oliveorchard or eroded vs deposition area within the olive or-chard) was evaluated using a two-factor analysis of vari-ance (ANOVA) (p lt 005) Additionally for some compar-ison at similar soil depths values of soil properties betweentwo different areas were assessed using a one-way ANOVAtest (p lt 005) In both situations data were log-transformedwhen necessary to fulfill ANOVA requirements Exploratoryanalysis using principal component analysis (PCA) was per-formed in the olive orchard area using the variables and sam-pling depths showing significant differences in the ANOVAThis PCA was complemented by determining the linear cor-relation coefficient variables showing the highest load on thePCA and erosion deposition rates using the Pearson corre-lation test The statistical software package Stata SE141 wasused for these analyses

3 Results

31 Organic carbon concentration and distributionamong fractions

Table 2 shows the significance of the differences in bulk soilCorg and the various Corg fractions between reference andolive orchard plots at different soil depths and due to the in-teraction between both factors (Table 2a) It also shows theeffects of the erosion deposition ratio (Table 2b) Overallbulk soil Corg was always significantly higher in the referencearea as compared to the olive orchard (00001lt p lt 00013depending on depth) (Table 2a and Fig 2) and this was inde-pendent of the soil sampling depth Corg values on the refer-ence site were between 2 to 5 times higher than those of theolive orchard for a given depth with the greater differencesin the top 10 cm of the soil Soil depth has a significant ef-fect on bulk Corg and Corg fractions with values typically de-creasing with depth in both areas Corg concentrations in theunprotected and physically chemically and biochemicallyprotected fractions were significantly higher (p lt 00001)(Table 2a) in the reference site as compared to the olive or-chard and across the different depths (00001lt p lt 00007)

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

184 J A Goacutemez et al Organic carbon across an olive orchard catena

(Fig 3) Corg values were between 2 to 6 times higher for theunprotected and chemically protected fractions and between2 to 35 times higher for the physically and biochemicallyprotected fractions with differences tending to decrease withthe soil depth

Within the olive orchard there were statistically signifi-cant differences between the erosion and deposition areas forbulk Corg values (p = 00198) (Table 2b) Higher Corg val-ues (11 to 06 ) were observed in the deposition arealocated downslope whereas lower values (085 to 055 )were measured in the areas with net erosion in the upper andmid sections of the catena It is worth noting that these differ-ences between erosion and deposition areas are detected forthe overall analysis using a two-way ANOVA (Table 2b) al-though an individual analysis at each depth (Fig 4) does notdetect statistically significant differences probably due to themoderate number of replications Significant differences be-tween the deposition and eroding area were also found forthe unprotected (p = 004) and the physically (p = 00077)and chemically protected fractions (p = 00299) (Table 2b)However differences for the biochemically protected frac-tions (Table 2b Fig 4) were not significant

The percentage distribution of SOC among fractions wasalso significantly different between both areas (reference vsolive orchard) except for the biochemically protected fraction(p lt 00001) (Table 3a) and depths (p lt 0011) (Fig 5) Inthe reference area most of the Corg was unprotected (between50 and 65 approximately) with no significant trend withdepth (Table 3a Fig 5) followed in relative importance bythe chemically and physically protected fractions which con-tributed between 18 and 30 as well as 10 and 20 ofthe bulk soil Corg respectively The biochemically protectedfraction represents a very low percentage (between 4 and6 ) In the olive orchard Corg is stored predominantly inthe physically and chemically protected fractions which ac-counted for about 38 to 27 and 34 to 28 respec-tively followed by the pool of unprotected fractions (be-tween 22 to 32 ) (Fig 5) The biochemically protectedfraction represents between 4 and 11 of the organic car-bon stored in the olive orchard There are no clear differencesin the organic carbon distribution among the different frac-tions between the erosion and deposition areas in the oliveorchard with the exception of the physically protected frac-tions at 10ndash20 cm depth (p = 0024) (Fig 6)

32 Organic carbon stock

SOC stock in the reference area is approximately 160 t haminus1making it significantly higher at p lt 005 for an equivalentmass than that of the olive orchard (Fig 7) which stores be-tween 38 and 41 t haminus1 in the eroded and deposition areasrespectively There were no significant differences betweenthese two orchard areas Similar results were obtained acrossthe top 40 cm soil profile Clay silt and sand contents of thetopsoil (0ndash10 cm) along the catena in the olive orchard av-

eraged 41 37 and 22 respectively with similar av-erage values in the eroding and deposition areas Variabilitywas relatively low (average coefficient of variation of 17 )and there were no significant changes between the erosionand deposition areas In the reference area the soil has anaverage clay silt and sand content of 30 31 and 39 respectively also with a homogeneous distribution across thesampling area (coefficient of variation of 10 ) According tothe Hassink and Whitmore (1997) model the percentages oforganic carbon of maximum soil stable Corg are 324plusmn011 and 363plusmn019 in the reference site and olive orchard re-spectively So protected Corg in the reference and olive or-chard areas account for 498plusmn 115 and 2049plusmn 52 ofthe maximum soil stable Corg respectively in the topsoil

33 δ15N and δ13C isotopic signal

Figure 8 and Table 4a compare stable isotope delta values be-tween the reference site and the overall olive orchard accord-ing to depth There are significant statistical differences inδ15N δ 13C and the δ13C δ15N ratio between the two areas(p lt 0002 Table 4a) although in the case of δ 15N the dif-ferences at individual depths were not significant Soil depthalso had a significant effect (00026lt p lt 0029 Table 4a)When comparing differences between the erosion and depo-sition areas within the olive orchard we detected statisticallysignificant differences only in δ15N and the δ13C δ15N ra-tio (p = 001 Table 4b) and they were mostly marked in theupper 20 cm of the soil (Fig 8)

34 Soil quality of topsoil across the catena variabilitybetween eroded and deposition areas in the orchard

Figure 9 depicts the comparison of the Pavail Norg Wsaggand the soil degradation index (SDI Goacutemez et al 2009a)in the top 10 cm of the soil between the erosion and deposi-tion areas of the olive orchard and Table 5 shows a similarcomparison for Norg Pavail and bulk density at the differentsoil depths Pavail in the deposition area is significantly higherthan that of the erosion area in the topsoil (p = 0005 Fig 9)and for the whole profile (p = 001 Table 5) whereas no sig-nificant differences in individual soil layers were found forNorg and Wsagg The SDI which combines these three vari-ables was about 3 times higher in the eroded area than in thedeposition area

Table 6 shows the loads in the first three principal com-ponents (PCs) of the principal component analysis (PCA)for the variables used in this analysis More than 70 ofthe variance was explained by the first two PCs Soils ofthe eroded and deposition area were clearly separated in thespace defined by the two PCs (Fig 10) The variables withhigher contributions in PC 1 and 2 were related to Pavail con-centration in the 0ndash10 and 0ndash40 cm soil depths to δ15N andδ13C in the 10 to 30 cm soil depths and to Corg concentra-tion and distribution in some fractions also in soil depths be-

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 185

Figure 2 Comparison of average soil organic carbon concentration in bulk soil depending on soil depth distinguishing between the referencesite the whole olive orchard the eroded area of the olive orchard and the deposition area in this orchard The labels of the bars for eachdepth indicate the p value according to a one-way ANOVA for the reference area (dark blue) vs the whole olive orchard (dark green) andfor the eroding area (beige) vs the deposition area (light green lower label in italics)

Figure 3 Organic carbon concentration in the different soil organic carbon fractions at each depth comparing the reference site vs oliveorchard The labels of the bars indicate the p value according to a one-way ANOVA comparing treatments for the same soil depth and carbonfraction between the reference area and olive orchard

tween 10 and 30 cm (Table 6) The deposition area tendedto show higher values in PC 1 and PC 2 and there wasno clear tendency in the erosion area along the catena Thelinear correlation coefficient between the variables and theerosion deposition rate was rather significant (r gtplusmn0731)for most of the variables (Table 7) Interestingly Pavail was

highly positively correlated in the whole soil layer Amongthe two most robust correlations with erosion depositionrates were that with Pavail concentration across the 40 cm soildepth y = 03975x+ 98364 (r2

= 0907) and δ13C at 10ndash20 cm depth y =minus00094xminus 25318 (r2

= 0632)

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

186 J A Goacutemez et al Organic carbon across an olive orchard catena

Table 2 Results of the two-way ANOVA of soil organic carbon concentration Corg () in different fractions and in bulk soil In (a) arearefers to the reference site vs the olive orchard and in (b) area refers to eroded vs deposition areas in the olive orchard NS stands for notsignificant

(a) Model Bulk soil Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00000 00000 00000 00000 00000Depth (D) 00023 00022 00061 00190 NSAtimesD NS NS NS NS 00300

(b) Model Bulk soil Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00198 00400 00077 00299 NSDepth (D) 00081 00070 00058 00055 00847AtimesD NS NS NS NS NS

Figure 4 Organic carbon concentration in the different soil organic carbon fractions at each depth comparing the eroded vs the depositionarea within the olive orchard The labels of the bars indicate the p value according to a one-way ANOVA comparing treatments for the samesoil depth and carbon fraction between the reference area and olive orchard

4 Discussion

41 Organic carbon concentration and distributionamong fractions

After approximately 175 years of contrasted land use be-tween the undisturbed reference site and the olive orchardbulk soil organic carbon concentration and its fractions havebeen dramatically reduced in the olive orchard Current lev-els of Corg concentration in the soil profile are approximately20 to 25 of that found in the reference area covered bynatural vegetation in the area adjacent to the orchard This ra-

tio is similar albeit in the lower range to the comparison ofCorg in topsoil among olive orchards with different types ofmanagement and natural areas reported for the region (Mil-groom et al 2007) The increased soil disturbance the lowerannual rate of biomass returned to the soil and the highererosion rate in the olive orchard explain this difference Inboth areas the Corg is clearly stratified indicating that de-spite the different mechanisms involved there is a periodicinput of biomass from the olive trees (eg fall of senescenceleaves and tree pruning residues) plus the annual ground veg-etation Vicente-Vicente et al (2017) estimated this biomasscontribution in the range of 148 to 056 t haminus1 yrminus1 It is

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 187

Figure 5 Contribution () of the different fractions with respect to total soil organic carbon by depth comparing the reference site vs theolive orchard The labels of each fraction and depth are the p value according to a one-way ANOVA comparing the reference area vs theolive orchard for the same carbon fraction and soil depth

Figure 6 Fraction of total organic carbon stored in the different fractions by depth comparing the reference site vs the olive orchard Thelabels of each fraction and depth are the p value according to a one-way ANOVA comparing the reference area vs the olive orchard for thesame carbon fraction and soil depth

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

188 J A Goacutemez et al Organic carbon across an olive orchard catena

Table 3 Results of the two-way ANOVA of the distribution of the total soil organic carbon content in the soil among the different fractionsof soil organic carbon Corg In (a) area refers to the reference site vs the olive orchard and in (b) area refers to eroded vs deposition areasin the olive orchard NS stands for not significant

(a) Model Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00000 00000 00000 NSDepth (D) NS NS 00640 NSAtimesD NS 00059 NS NS

(b) Model Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 0091 00881 NS NSDepth (D) 0051 00214 NS 0033AtimesD NS NS NS NS

Figure 7 Cumulative soil organic carbon (SOC) stock across thesoil profile in terms of cumulative soil mass on the vertical axisDifferent letters for similar soil mass refer to statistically significantdifferences (KruskalndashWallis test at p lt 005) For this analysis cu-mulative soil organic carbons were interpolated linearly to the aver-age cumulative soil mass corresponding to all the points in the threeareas

worth noticing that the decrease in Corg as compared to thenatural area is much higher than the reported rates of increasein Corg in olive orchards using conservation agriculture (CA)techniques such as cover crops and incorporation of organicresidues from different sources In a meta-analysis Vicente-Vicente et al (2016) found a response ratio (the ratio of Corgunder CA management as compared to Corg in orchards withbare-soil management) of 11 to 19 suggesting that underCA management which combines cover crops and organicresidues Corg doubled as a maximum

Table 4 Results of the two-way ANOVA of the stable isotope sig-nal In (a) area refers to the reference site vs the olive orchard andin (b) area refers to eroded vs deposition areas in the olive orchardNS stands for not significant

(a) Model δ13C δ15N δ13C δ15N

Area (A) 00001 0002 0002Depth (D) 00026 00175 0029AtimesD NS NS NS

(b) Model δ13C δ15N δ13C δ15N

Area (A) NS 001 001Depth (D) NS NS NSAtimesD NS NS NS

Table 5 Results of the two-way ANOVA of some physical andchemical soil properties comparing eroded vs deposition areas inthe olive orchard NS stands for not significant

Model Norg Pavail Bulk density

Area (A) 00000 001 NSDepth (D) 00009 NS NSAtimesD NS NS NS

Combining all Corg data of the olive orchard the variabil-ity was about 35 which is similar to what has been re-ported so far in the few studies found on soil Corg variabilityin olive orchards For instance Gargouri et al (2013) indi-cated a 24 coefficient of variation (CV) in a 34 ha oliveorchard in Tunisia while Huang et al (2017) reported an av-erage CV of 41 in a 62 ha olive orchard in southern SpainNeither of these two studies reported clear trends in the distri-bution of Corg depending on topography Huang et al (2017)

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 189

Figure 8 13C and 15N isotopic signal of soil by depth distinguish-ing among reference site the whole olive orchard the eroded areaof the olive orchard and the deposition area in this orchard Thelabels of the bars for each depth indicate the p value according toa one-way ANOVA between the reference area (dark blue) vs thewhole olive orchard (dark green) and the eroding area (beige) vsthe deposition area (light green lower label in italics)

pointed out the additional difficulties in the determination ofCorg due to the topography heterogeneity although this wascompounded by the fact that within the orchards there weretwo areas with different planting dates for the trees Goacutemezet al (2012) reported a CV of 49 with higher Corg in ar-eas where there was a change in the slope gradient from thehillslope to a central channel draining into the catchment al-though they could not find a simple relationship between theincrease in content and the topographic indices Despite thefact that a lot of work has been done on the correlation be-tween erosion and deposition and the redistribution of soilCorg (eg Van Oost et al 2005) our study is to our knowl-edge the first attempt to quantify this in detail under the

Table 6 Loads of selected variables in the PCA for the first threeprincipal components (PCs) The values in brackets below PC 1 2and 3 indicate the percentage of variance explained by this PC Vari-ables in bold are those with a load higher than 90 of the variablewith the maximum load for this PC Conc refers to Corg concen-tration for this fraction and Frac means the relative contribution ofthis fraction to the total Corg for this soil depth

Variable at each depth interval (cm) PC 1 PC 2 PC 3(558) (176) (132)

Pavail 0ndash10 02298 008765 minus007278Pavail 0ndash40 02271 0101 minus008455δ13C 0ndash10 minus003385 02861 minus03302δ15N 0ndash10 01941 01677 01836Norg 0ndash10 02147 minus01375 minus00336δ13C δ15N 0ndash10 01594 0249 02211δ13C 10ndash20 minus02329 01461 004296δ15N 10ndash20 01856 007054 03121Norg 10ndash20 02317 minus008699 minus01404δ13C δ15N 10ndash20 01637 008512 03309δ13C 20ndash30 minus02316 006018 009789δ15N 20ndash30 01748 minus02656 01812Norg 20ndash30 02176 009979 minus007458δ13C δ15N 20ndash30 01684 minus02982 01774Corg conc 10ndash20 02421 minus002951 minus006407Corg unpr conc 10ndash20 02116 00331 00385Corg unpr Frac 10ndash20 minus009663 minus00509 0364Corg Phys Pro conc 10ndash20 02371 minus005728 minus01183Corg Phys Pro Frac 10ndash20 0123 0163 minus002953Corg Chem Pro conc 10ndash20 02072 003097 minus02335Corg Chem Pro Frac 10ndash20 minus004095 009461 minus04678Corg conc 20ndash30 02326 minus01108 minus005737Corg unpr conc 20ndash30 02002 minus02372 minus005071Corg unpr Rel 20ndash30 cm minus0132 minus03675 minus01379Corg Phys Pro conc 20ndash30 02257 005677 minus001036Corg Phys Pro Frac 20ndash30 009518 03528 01294Corg Chem Pro conc 20ndash30 02323 minus006802 minus01366Corg Chem Pro Frac 20ndash30 004565 04437 minus001278

agroenvironmental conditions of an olive orchard The vari-ability induced by the combined effects of water and tillageerosion in this olive orchard was similar to that described inother agroecosystems For instance Van Oost et al (2005)measured a clear correlation between the erosion and depo-sition rates and the topsoil Corg concentration which rangedbetween 08 of the erosion to 14 of the deposition sitesin the top 25 cm of the soil at two field crop sites under atemperate climate Besides this Bameri et al (2015) mea-sured a higher Corg in the lower part of a field crop site undersemiarid conditions where deposition of the eroded soil fromthe upper zones took place Overall in such landscapes culti-vated for a long time the cumulative effect of tillage and wa-ter erosion on the redistribution of soil across the slope hasbeen observed (Dlugoszlig et al 2012) These processes alsoproduce a vertical redistribution of Corg resulting in a rela-tively homogeneous profile in the tilled layer (top 15ndash20 cm)and a gradual decline below this depth as noted in our study

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

190 J A Goacutemez et al Organic carbon across an olive orchard catena

Figure 9 Soil available phosphorus (Pavail) organic nitrogen (Norg) aggregate stability (Wsagg) and soil degradation index (SDI) by depthcomparing eroded vs deposition areas within the olive orchard The labels of the bars indicate the p value according to a one-way ANOVAcomparing areas for the same soil depth

Figure 10 Scores on principal components 1 and 2 (PC 1 PC 2)for sampling points in the eroded and deposition areas in the oliveorchard

42 Organic carbon stock

The differences in soil organic carbon stock between the ref-erence site and the olive orchard are similar to those de-scribed previously when comparing cropland and forestedareas with the latter presenting a higher concentration ofCorg most of which is in the unprotected fraction whilethe cropland presented a higher fraction of the carbon inthe physically and chemically protected fractions (eg Poe-plau and Don 2013) This is likely due to the fact that un-der soil degradation processes such as water erosion and

low annual organic carbon input as is the case under oliveorchard land use most of the unprotected Corg decomposesrelatively quickly and a great proportion of the remaininglow SOC is protected In addition the mobilization of theunprotected Corg is expected to be reduced in the protectedforested area because of the canopy and the existing veg-etation on the ground that protects the soil against runoffand splash erosion processes In fact the protected Corg con-centration in the topsoil of the olive orchard in the erodedarea accounted for 186plusmn 39 of the upper limit of pro-tected Corg (364plusmn 023 ) according to the model of Has-sink and Whitmore (1997) Therefore the low unprotectedSOC concentration found in the olive orchard is an elementof concern in the increase in SOC stocks This is because pro-tected fractions are fueled from recently derived partially de-composed plant residues together with microbial and micro-meso- and macrofaunal debris (unprotected organic carbon)through processes like SOC aggregation into macro- andormicroaggregates (physically protected SOC) and complexSOC associations with clay and silt particles (chemicallyprotected SOC) which are disrupted in the cropland area incomparison with the reference area The distribution amongsoil Corg fractions in the orchard of this study was similarto the result obtained by Vicente-Vicente et al (2017) whomeasured Corg fraction distribution in olive oil orchards withtemporary cover crops with the exception of the unprotectedSOC which was much higher in soils under cover crops thanthat of our study under bare soil Also interesting is the dif-ficulty obtaining statistically significant differences in SOC

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 191

Tabl

e7

Pear

son

corr

elat

ion

coef

ficie

nts

oflin

ear

corr

elat

ion

betw

een

vari

able

sat

each

dept

hin

terv

al(c

m)

with

the

high

est

load

son

prin

cipa

lco

mpo

nent

s(s

eeTa

ble

6)lowast

Mea

nsst

atis

tical

lysi

gnifi

cant

atαlt

005

Ero

sde

pP a

vail

0ndash10

P ava

il0ndash

40δ

13C

10ndash2

0N

org

10ndash2

0δ

13C

20ndash3

0C

org

conc

C

org

Phys

C

org

conc

C

org

Phys

C

org

Che

m

Cor

gC

hem

10

ndash20

Pro

conc

20

ndash30

Pro

conc

Pr

oco

nc

Pro

Frac

10

ndash20

20ndash3

020

ndash30

20ndash3

0

Ero

sde

p1

P ava

il0ndash

100

746lowast

1P a

vail

0ndash40

095

3lowast0

857lowast

1δ

13C

10ndash2

0minus

082

3lowastminus

076

6lowast1

Nor

g10

ndash20

073

1lowast0

898lowast

083

9lowastminus

096

5lowast1

δ13

C20

ndash30

minus0

792lowast

minus0

893lowast

minus0

891lowast

088

8lowastminus

093

8lowast1

Cor

gco

nc1

0ndash20

075

7lowast0

843lowast

089

9lowastminus

092

9lowast0

920lowast

minus0

911lowast

1C

org

Phys

Pro

con

c10

ndash20

075

5lowast0

908lowast

088

2lowastminus

092

90

967lowast

minus0

985lowast

095

2lowast1

Cor

gco

nc2

0ndash30

078

6lowast0

799lowast

minus0

965lowast

092

7lowastminus

084

6lowast0

9453lowast

089

42lowast

1C

org

Phys

Pro

con

c20

ndash30

083

7lowast0

756lowast

minus0

846lowast

081

7lowastminus

071

4lowast0

8819lowast

079

28lowast

090

1lowast1

Cor

gC

hem

Pro

con

c20

ndash30

074

2lowast0

855lowast

085

9lowastminus

096

2lowast0

980lowast

minus0

896lowast

095

15lowast

094

34lowast

095

0lowast0

853lowast

1C

org

Che

mP

roF

rac

20ndash3

01

stock between the eroding and deposition areas (Fig 7) de-spite the apparent clear differences between the two areas insome other soil properties such as Pavail or δ15N and δ 13Cisotopic signatures in the subsoil (see below)

43 δ15N and δ13C isotopic signal

Changes in vegetation types induced differences in δ13C be-tween the olive orchard and the reference area but as ex-pected no differences in δ13C were detected between the ero-sion and deposition areas in the olive orchard given the sameorigin of vegetation-derived organic matter ie C3 plantsInterestingly within the olive orchard significant differencesin δ15N were detected between the erosion and deposition ar-eas especially for the top 20 cm of soils (Fig 9) This sug-gests the potential of using δ15N as a variable for identifyingdegraded areas in olive growing fields as has been proposedfor other eroding regions in the world (eg Meusburger etal 2013) This might provide an alternative when other con-ventional or isotopic techniques are not available Neverthe-less further studies exploring this potential are necessary inorder to also consider the influence of the NndashP fertilizer mod-ifying the δ15N in relation to the reference area (eg Bate-man and Kelly 2007) The source of N in soil is multifariousand subject to a wide range of transformations that affect theδ15N signature therefore we can only speculate on the rea-sons for this difference in δ15N in a relatively homogeneousarea Bulk soil δ15N tended to be more positive (eg moreenriched in δ15N) as the N cycling rate increases soil micro-bial processes (eg N mineralization nitrification and deni-trification) resulting in products (eg nitrate N2O N2 andNH3) depleted in 15N while the substrate from which theywere formed becomes slightly enriched (Robison 2001) Thehigher δ15N signature of the soil at the deposition locationsuggests that the rates of processes involved in N cycling arehigher than in the erosion area and that it is in accordancewith the higher bulk Corg and Pavail contents and lower SDIof the deposition site The relatively lower soil δ15N signa-ture at the reference site could be partially due to the input oflitter N from the natural legumes and to the closed N cyclingwhich characterize natural forest ecosystems The trend in15N enrichment with soil depth as found at the referencesite is a common observation at forest and grassland sitesand has been related to different mechanisms including 15Nisotope discrimination during microbial N transformationsdifferential preservation of 15N-enriched soil organic mattercomponents during N decomposition and more recently tothe build-up of 15N-enriched microbial necromass (Huygenset al 2008) However there still remains the need for a care-ful calibration to an undisturbed reference site and a betterunderstanding of the influence of different vegetation in thereference and the studied area on the change in the δ15N sig-nal for its further use as an additional tool to determine soildegradation

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

192 J A Goacutemez et al Organic carbon across an olive orchard catena

44 Soil quality of topsoil across the catena variabilitybetween eroded and deposition areas in the orchard

This horizontal distribution due to tillage and water erosionalso simultaneously affected other soil properties and hasbeen described previously in other cultivated areas For in-stance De Gryze et al (2008) described how for a field croparea under conventional tillage in Belgium Pavail almost dou-bled (229 vs 122 mg kgminus1) in the depositional area as com-pared to the eroding upper part They also reported that inhalf of the field under conservation tillage these differencesin Pavail between the upper and lower areas of the field dis-appeared We have observed in our sampled orchard a pro-nounced increase in topsoil Pavail in the deposition area ofaround 400 as compared to the eroding part of the orchardwhich can be attributed to the deposition of enriched sedi-ment coming from the upslope area The cumulative effectsof the differences in Corg and Pavail and the trend towardshigher although nonsignificant Wsagg in the deposition arearesulted in two areas within the orchard with marked differ-ences in soil quality the eroded part which is within therange considered to be degraded in the region (Goacutemez et al2009a) and the depositional area representing 20 of theorchard transect length (Table 1) which is within the non-degraded range according to the same index Topographyand sediment redistribution by erosive processes introducea gradient of spatial variability that questions the conceptof representative area when it comes to describing a wholefield In fact several studies (eg Dell and Sharpley 2006)have suggested that the verification of compliance of envi-ronmental programs such as those related to Corg seques-tration should be based preferentially at least partially onmodeling approaches Our results raise the need for a care-ful delineation of subareas when analyzing soil quality indi-cators andor SOC carbon stock within the same field unitThey also warn about the difficulty of drawing hypothesesfor quantifying the differences between these delineated ar-eas in relation to specific soil properties For instance in ourstudy case erosionndashdeposition processes had a major impacton Pavail and soil quality but the impact on Corg concentra-tion and stock was moderate and extremely difficult to detectstatistically using a moderate number of samples

Our PCA and regression analyses confirmed the relativelyhigh variability of Corg and stock in relation to other soilquality indicators related to erosionndashdeposition processessuch as Pavail as discussed in the previous section While inthe catena studied in the current research the P cycle seemsto be mostly driven by sediment mobilization the Corg and Ncycles seem to be much more complex The moderate differ-ences in Corg and the homogeneity in δ15N and δ13C isotopicsignals between the eroding and deposition areas may be dueto several processes some of which were discussed abovesuch as spatial variability of carbon input due to biomassin the plot surface soil operations in the orchard and fer-tilization We found it both interesting and worth exploring

in future studies that significant correlations between erosionand deposition rates and Corg- δ15N- and δ 13C-related vari-ables were found for samples from the 10ndash20 and 20ndash30 cmlayers indicating that short-term disturbance by surface pro-cesses can mask experimental determination of the impact oferosion deposition processes in this olive orchard for thesevariables

5 Conclusions

1 The results indicate that erosion and deposition withinthe investigated old olive orchard have created a signifi-cant difference in soil properties along the catena whichis translated into different soil Corg Pavail and Norg con-tent δ15N and thus contrasting soil quality status

2 This variability was lower than that of the natural areawhich indicated a severe depletion of SOC as comparedto the natural area and a redistribution of available or-ganic carbon among the different SOC fractions

3 The results suggest that δ15N has the potential for beingused as an indicator of soil degradation although moreinvestigation in different agroecosystems would be re-quired for confirming this statement on a larger scale

4 This research highlights that proper understanding andmanagement of soil quality and Corg content in olive or-chards require consideration of the on-site spatial vari-ability induced by soil erosionndashdeposition processes

Data availability The basic data used in this paper are freelyavailable from the CSIC digital repository at httpsdoiorg1020350digitalCSIC12513 (Goacutemez Calero et al 2020)

Author contributions JAG coordinated and designed the over-all study conducted the field sampling campaigns analysis of thesoil organic carbon results and the overall analysis of the resultsand wrote the manuscript GG participated in the analysis of soil-quality-related properties the overall analysis of the results and thewriting of the manuscript LM participated in the overall analysisof the results and the writing of the manuscript CR and AT partici-pated in the analysis and interpretation of the stable isotope resultsRGR contributed to the design and the statistical analysis of soilorganic carbon (stock and fractions) and nitrogen data the overallanalysis of the results and the writing of the manuscript

Competing interests The authors declare that they have no con-flict of interest

Acknowledgements The authors would like to thankManuel Gonzaacutelez de Molina for providing valuable insightinto the management of the olive orchard in the decades before

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 183

tion All fractions were dried in a 60 C oven and weighedAfterwards there was a second step involving a further frac-tionation of the microaggregate fraction isolated in the firststep A density flotation with sodium polytungstate was usedto isolate fine nonprotected POM (LF) after removing thefine nonprotected POM the heavy fraction was dispersedovernight by shaking and passed through a 53 mm sieveto separate the microaggregate-protected POM (gt 53 mm insize iPOM) from the microaggregate-derived silt- and clay-sized fractions The resulting suspension was centrifuged toseparate the microaggregate-derived silt-sized fraction fromthe clay-sized fraction as described above A final third stepinvolved the acid hydrolysis of each of the isolated silt- andclay-sized fractions The silt- and clay-sized fractions fromboth the density flotation and the initial dispersion and phys-ical fractionation were subjected to acid hydrolysis The un-protected pool includes the POM and LF fractions isolatedin the first and second fractionation steps respectively Thephysically protected SOC consists of the SOC measured inthe microaggregates It includes not only the iPOM but alsothe hydrolyzable and nonhydrolyzable SOC of the intermedi-ate fraction (53ndash250 microm) The chemically and biochemicallyprotected pools correspond to the hydrolyzable and nonhy-drolyzable SOC in the fine fraction (lt 53 microm) respectivelyIn all cases SOC fractions and in the bulk soil organic car-bon concentrations were determined by using the wet oxida-tion sulfuric acid and potassium dichromate method of An-derson and Ingram (1993)

Inorganic carbon was removed prior to stable isotope anal-ysis by acid fumigation following the method of Harris etal (2001) Moistened subsamples were exposed to the exha-lation of HCl in a desiccator overnight Afterwards the sam-ples were dried at 40 C before the stable isotope ratio wasmeasured The N measurements were done with unacidifiedsamples and the stable N isotope ratios and the C and N con-centrations were measured by isotope ratio mass spectrome-try (Isoprime 100 coupled with an Elementar Vario IsotopeSelect elemental analyzer both instruments supplied by El-ementar Langenselbold Germany) The instrumental stan-dard deviation for δ15N is 016 and 011 for δ13C Sta-ble isotopes are reported as delta values (permil) which are therelative differences between the isotope ratios of the samplesand the isotope ratio of a reference standard

In addition available phosphorus (Pavail) was determinedby the Olsen method (Olsen and Sommers 1982) and or-ganic nitrogen (Norg) was determined by the Kjeldahl method(Stevenson 1982) Water-stable aggregates (Wsagg) weremeasured using the method of Barthes and Roose (2002)Soil particle size distribution was determined using the hy-drometer method (Bouyoucos 1962) for the topsoil (0ndash10 cm) of the reference area and the olive orchard Two bulkdensity values were calculated for the whole profile (a) oneconsidering the mass of soil finer than 2 mm and (b) oneconsidering the mass of soil finer than 2 mm as well as thestone content which was determined from the excavation

and core sampling described above Additionally the bulkdensity of the topsoil (0ndash10 cm depth) was measured usinga manual soil core sample with a volume of 100 cm3 Soilcarbon stocks were calculated for the fine soil fraction af-ter discounting rock or stone fragments larger than 2 mm andconsidering bulk density They were then presented on equiv-alent soil mass as described by Wendt and Hauser (2013) Asproposed by Hassink and Whitmore (1997) theoretical val-ues of carbon saturation were calculated from the soil par-ticle analysis Finally the soil degradation index developedby Goacutemez et al (2009a) was calculated from the Corg Pavailand Wsagg

24 Statistical analysis

The overall effect of depth and area (reference site vs oliveorchard or eroded vs deposition area within the olive or-chard) was evaluated using a two-factor analysis of vari-ance (ANOVA) (p lt 005) Additionally for some compar-ison at similar soil depths values of soil properties betweentwo different areas were assessed using a one-way ANOVAtest (p lt 005) In both situations data were log-transformedwhen necessary to fulfill ANOVA requirements Exploratoryanalysis using principal component analysis (PCA) was per-formed in the olive orchard area using the variables and sam-pling depths showing significant differences in the ANOVAThis PCA was complemented by determining the linear cor-relation coefficient variables showing the highest load on thePCA and erosion deposition rates using the Pearson corre-lation test The statistical software package Stata SE141 wasused for these analyses

3 Results

31 Organic carbon concentration and distributionamong fractions

Table 2 shows the significance of the differences in bulk soilCorg and the various Corg fractions between reference andolive orchard plots at different soil depths and due to the in-teraction between both factors (Table 2a) It also shows theeffects of the erosion deposition ratio (Table 2b) Overallbulk soil Corg was always significantly higher in the referencearea as compared to the olive orchard (00001lt p lt 00013depending on depth) (Table 2a and Fig 2) and this was inde-pendent of the soil sampling depth Corg values on the refer-ence site were between 2 to 5 times higher than those of theolive orchard for a given depth with the greater differencesin the top 10 cm of the soil Soil depth has a significant ef-fect on bulk Corg and Corg fractions with values typically de-creasing with depth in both areas Corg concentrations in theunprotected and physically chemically and biochemicallyprotected fractions were significantly higher (p lt 00001)(Table 2a) in the reference site as compared to the olive or-chard and across the different depths (00001lt p lt 00007)

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

184 J A Goacutemez et al Organic carbon across an olive orchard catena

(Fig 3) Corg values were between 2 to 6 times higher for theunprotected and chemically protected fractions and between2 to 35 times higher for the physically and biochemicallyprotected fractions with differences tending to decrease withthe soil depth

Within the olive orchard there were statistically signifi-cant differences between the erosion and deposition areas forbulk Corg values (p = 00198) (Table 2b) Higher Corg val-ues (11 to 06 ) were observed in the deposition arealocated downslope whereas lower values (085 to 055 )were measured in the areas with net erosion in the upper andmid sections of the catena It is worth noting that these differ-ences between erosion and deposition areas are detected forthe overall analysis using a two-way ANOVA (Table 2b) al-though an individual analysis at each depth (Fig 4) does notdetect statistically significant differences probably due to themoderate number of replications Significant differences be-tween the deposition and eroding area were also found forthe unprotected (p = 004) and the physically (p = 00077)and chemically protected fractions (p = 00299) (Table 2b)However differences for the biochemically protected frac-tions (Table 2b Fig 4) were not significant

The percentage distribution of SOC among fractions wasalso significantly different between both areas (reference vsolive orchard) except for the biochemically protected fraction(p lt 00001) (Table 3a) and depths (p lt 0011) (Fig 5) Inthe reference area most of the Corg was unprotected (between50 and 65 approximately) with no significant trend withdepth (Table 3a Fig 5) followed in relative importance bythe chemically and physically protected fractions which con-tributed between 18 and 30 as well as 10 and 20 ofthe bulk soil Corg respectively The biochemically protectedfraction represents a very low percentage (between 4 and6 ) In the olive orchard Corg is stored predominantly inthe physically and chemically protected fractions which ac-counted for about 38 to 27 and 34 to 28 respec-tively followed by the pool of unprotected fractions (be-tween 22 to 32 ) (Fig 5) The biochemically protectedfraction represents between 4 and 11 of the organic car-bon stored in the olive orchard There are no clear differencesin the organic carbon distribution among the different frac-tions between the erosion and deposition areas in the oliveorchard with the exception of the physically protected frac-tions at 10ndash20 cm depth (p = 0024) (Fig 6)

32 Organic carbon stock

SOC stock in the reference area is approximately 160 t haminus1making it significantly higher at p lt 005 for an equivalentmass than that of the olive orchard (Fig 7) which stores be-tween 38 and 41 t haminus1 in the eroded and deposition areasrespectively There were no significant differences betweenthese two orchard areas Similar results were obtained acrossthe top 40 cm soil profile Clay silt and sand contents of thetopsoil (0ndash10 cm) along the catena in the olive orchard av-

eraged 41 37 and 22 respectively with similar av-erage values in the eroding and deposition areas Variabilitywas relatively low (average coefficient of variation of 17 )and there were no significant changes between the erosionand deposition areas In the reference area the soil has anaverage clay silt and sand content of 30 31 and 39 respectively also with a homogeneous distribution across thesampling area (coefficient of variation of 10 ) According tothe Hassink and Whitmore (1997) model the percentages oforganic carbon of maximum soil stable Corg are 324plusmn011 and 363plusmn019 in the reference site and olive orchard re-spectively So protected Corg in the reference and olive or-chard areas account for 498plusmn 115 and 2049plusmn 52 ofthe maximum soil stable Corg respectively in the topsoil

33 δ15N and δ13C isotopic signal

Figure 8 and Table 4a compare stable isotope delta values be-tween the reference site and the overall olive orchard accord-ing to depth There are significant statistical differences inδ15N δ 13C and the δ13C δ15N ratio between the two areas(p lt 0002 Table 4a) although in the case of δ 15N the dif-ferences at individual depths were not significant Soil depthalso had a significant effect (00026lt p lt 0029 Table 4a)When comparing differences between the erosion and depo-sition areas within the olive orchard we detected statisticallysignificant differences only in δ15N and the δ13C δ15N ra-tio (p = 001 Table 4b) and they were mostly marked in theupper 20 cm of the soil (Fig 8)

34 Soil quality of topsoil across the catena variabilitybetween eroded and deposition areas in the orchard

Figure 9 depicts the comparison of the Pavail Norg Wsaggand the soil degradation index (SDI Goacutemez et al 2009a)in the top 10 cm of the soil between the erosion and deposi-tion areas of the olive orchard and Table 5 shows a similarcomparison for Norg Pavail and bulk density at the differentsoil depths Pavail in the deposition area is significantly higherthan that of the erosion area in the topsoil (p = 0005 Fig 9)and for the whole profile (p = 001 Table 5) whereas no sig-nificant differences in individual soil layers were found forNorg and Wsagg The SDI which combines these three vari-ables was about 3 times higher in the eroded area than in thedeposition area

Table 6 shows the loads in the first three principal com-ponents (PCs) of the principal component analysis (PCA)for the variables used in this analysis More than 70 ofthe variance was explained by the first two PCs Soils ofthe eroded and deposition area were clearly separated in thespace defined by the two PCs (Fig 10) The variables withhigher contributions in PC 1 and 2 were related to Pavail con-centration in the 0ndash10 and 0ndash40 cm soil depths to δ15N andδ13C in the 10 to 30 cm soil depths and to Corg concentra-tion and distribution in some fractions also in soil depths be-

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 185

Figure 2 Comparison of average soil organic carbon concentration in bulk soil depending on soil depth distinguishing between the referencesite the whole olive orchard the eroded area of the olive orchard and the deposition area in this orchard The labels of the bars for eachdepth indicate the p value according to a one-way ANOVA for the reference area (dark blue) vs the whole olive orchard (dark green) andfor the eroding area (beige) vs the deposition area (light green lower label in italics)

Figure 3 Organic carbon concentration in the different soil organic carbon fractions at each depth comparing the reference site vs oliveorchard The labels of the bars indicate the p value according to a one-way ANOVA comparing treatments for the same soil depth and carbonfraction between the reference area and olive orchard

tween 10 and 30 cm (Table 6) The deposition area tendedto show higher values in PC 1 and PC 2 and there wasno clear tendency in the erosion area along the catena Thelinear correlation coefficient between the variables and theerosion deposition rate was rather significant (r gtplusmn0731)for most of the variables (Table 7) Interestingly Pavail was

highly positively correlated in the whole soil layer Amongthe two most robust correlations with erosion depositionrates were that with Pavail concentration across the 40 cm soildepth y = 03975x+ 98364 (r2

= 0907) and δ13C at 10ndash20 cm depth y =minus00094xminus 25318 (r2

= 0632)

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

186 J A Goacutemez et al Organic carbon across an olive orchard catena

Table 2 Results of the two-way ANOVA of soil organic carbon concentration Corg () in different fractions and in bulk soil In (a) arearefers to the reference site vs the olive orchard and in (b) area refers to eroded vs deposition areas in the olive orchard NS stands for notsignificant

(a) Model Bulk soil Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00000 00000 00000 00000 00000Depth (D) 00023 00022 00061 00190 NSAtimesD NS NS NS NS 00300

(b) Model Bulk soil Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00198 00400 00077 00299 NSDepth (D) 00081 00070 00058 00055 00847AtimesD NS NS NS NS NS

Figure 4 Organic carbon concentration in the different soil organic carbon fractions at each depth comparing the eroded vs the depositionarea within the olive orchard The labels of the bars indicate the p value according to a one-way ANOVA comparing treatments for the samesoil depth and carbon fraction between the reference area and olive orchard

4 Discussion

41 Organic carbon concentration and distributionamong fractions

After approximately 175 years of contrasted land use be-tween the undisturbed reference site and the olive orchardbulk soil organic carbon concentration and its fractions havebeen dramatically reduced in the olive orchard Current lev-els of Corg concentration in the soil profile are approximately20 to 25 of that found in the reference area covered bynatural vegetation in the area adjacent to the orchard This ra-

tio is similar albeit in the lower range to the comparison ofCorg in topsoil among olive orchards with different types ofmanagement and natural areas reported for the region (Mil-groom et al 2007) The increased soil disturbance the lowerannual rate of biomass returned to the soil and the highererosion rate in the olive orchard explain this difference Inboth areas the Corg is clearly stratified indicating that de-spite the different mechanisms involved there is a periodicinput of biomass from the olive trees (eg fall of senescenceleaves and tree pruning residues) plus the annual ground veg-etation Vicente-Vicente et al (2017) estimated this biomasscontribution in the range of 148 to 056 t haminus1 yrminus1 It is

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 187

Figure 5 Contribution () of the different fractions with respect to total soil organic carbon by depth comparing the reference site vs theolive orchard The labels of each fraction and depth are the p value according to a one-way ANOVA comparing the reference area vs theolive orchard for the same carbon fraction and soil depth

Figure 6 Fraction of total organic carbon stored in the different fractions by depth comparing the reference site vs the olive orchard Thelabels of each fraction and depth are the p value according to a one-way ANOVA comparing the reference area vs the olive orchard for thesame carbon fraction and soil depth

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

188 J A Goacutemez et al Organic carbon across an olive orchard catena

Table 3 Results of the two-way ANOVA of the distribution of the total soil organic carbon content in the soil among the different fractionsof soil organic carbon Corg In (a) area refers to the reference site vs the olive orchard and in (b) area refers to eroded vs deposition areasin the olive orchard NS stands for not significant

(a) Model Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00000 00000 00000 NSDepth (D) NS NS 00640 NSAtimesD NS 00059 NS NS

(b) Model Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 0091 00881 NS NSDepth (D) 0051 00214 NS 0033AtimesD NS NS NS NS

Figure 7 Cumulative soil organic carbon (SOC) stock across thesoil profile in terms of cumulative soil mass on the vertical axisDifferent letters for similar soil mass refer to statistically significantdifferences (KruskalndashWallis test at p lt 005) For this analysis cu-mulative soil organic carbons were interpolated linearly to the aver-age cumulative soil mass corresponding to all the points in the threeareas

worth noticing that the decrease in Corg as compared to thenatural area is much higher than the reported rates of increasein Corg in olive orchards using conservation agriculture (CA)techniques such as cover crops and incorporation of organicresidues from different sources In a meta-analysis Vicente-Vicente et al (2016) found a response ratio (the ratio of Corgunder CA management as compared to Corg in orchards withbare-soil management) of 11 to 19 suggesting that underCA management which combines cover crops and organicresidues Corg doubled as a maximum

Table 4 Results of the two-way ANOVA of the stable isotope sig-nal In (a) area refers to the reference site vs the olive orchard andin (b) area refers to eroded vs deposition areas in the olive orchardNS stands for not significant

(a) Model δ13C δ15N δ13C δ15N

Area (A) 00001 0002 0002Depth (D) 00026 00175 0029AtimesD NS NS NS

(b) Model δ13C δ15N δ13C δ15N

Area (A) NS 001 001Depth (D) NS NS NSAtimesD NS NS NS

Table 5 Results of the two-way ANOVA of some physical andchemical soil properties comparing eroded vs deposition areas inthe olive orchard NS stands for not significant

Model Norg Pavail Bulk density

Area (A) 00000 001 NSDepth (D) 00009 NS NSAtimesD NS NS NS

Combining all Corg data of the olive orchard the variabil-ity was about 35 which is similar to what has been re-ported so far in the few studies found on soil Corg variabilityin olive orchards For instance Gargouri et al (2013) indi-cated a 24 coefficient of variation (CV) in a 34 ha oliveorchard in Tunisia while Huang et al (2017) reported an av-erage CV of 41 in a 62 ha olive orchard in southern SpainNeither of these two studies reported clear trends in the distri-bution of Corg depending on topography Huang et al (2017)

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 189

Figure 8 13C and 15N isotopic signal of soil by depth distinguish-ing among reference site the whole olive orchard the eroded areaof the olive orchard and the deposition area in this orchard Thelabels of the bars for each depth indicate the p value according toa one-way ANOVA between the reference area (dark blue) vs thewhole olive orchard (dark green) and the eroding area (beige) vsthe deposition area (light green lower label in italics)

pointed out the additional difficulties in the determination ofCorg due to the topography heterogeneity although this wascompounded by the fact that within the orchards there weretwo areas with different planting dates for the trees Goacutemezet al (2012) reported a CV of 49 with higher Corg in ar-eas where there was a change in the slope gradient from thehillslope to a central channel draining into the catchment al-though they could not find a simple relationship between theincrease in content and the topographic indices Despite thefact that a lot of work has been done on the correlation be-tween erosion and deposition and the redistribution of soilCorg (eg Van Oost et al 2005) our study is to our knowl-edge the first attempt to quantify this in detail under the

Table 6 Loads of selected variables in the PCA for the first threeprincipal components (PCs) The values in brackets below PC 1 2and 3 indicate the percentage of variance explained by this PC Vari-ables in bold are those with a load higher than 90 of the variablewith the maximum load for this PC Conc refers to Corg concen-tration for this fraction and Frac means the relative contribution ofthis fraction to the total Corg for this soil depth

Variable at each depth interval (cm) PC 1 PC 2 PC 3(558) (176) (132)

Pavail 0ndash10 02298 008765 minus007278Pavail 0ndash40 02271 0101 minus008455δ13C 0ndash10 minus003385 02861 minus03302δ15N 0ndash10 01941 01677 01836Norg 0ndash10 02147 minus01375 minus00336δ13C δ15N 0ndash10 01594 0249 02211δ13C 10ndash20 minus02329 01461 004296δ15N 10ndash20 01856 007054 03121Norg 10ndash20 02317 minus008699 minus01404δ13C δ15N 10ndash20 01637 008512 03309δ13C 20ndash30 minus02316 006018 009789δ15N 20ndash30 01748 minus02656 01812Norg 20ndash30 02176 009979 minus007458δ13C δ15N 20ndash30 01684 minus02982 01774Corg conc 10ndash20 02421 minus002951 minus006407Corg unpr conc 10ndash20 02116 00331 00385Corg unpr Frac 10ndash20 minus009663 minus00509 0364Corg Phys Pro conc 10ndash20 02371 minus005728 minus01183Corg Phys Pro Frac 10ndash20 0123 0163 minus002953Corg Chem Pro conc 10ndash20 02072 003097 minus02335Corg Chem Pro Frac 10ndash20 minus004095 009461 minus04678Corg conc 20ndash30 02326 minus01108 minus005737Corg unpr conc 20ndash30 02002 minus02372 minus005071Corg unpr Rel 20ndash30 cm minus0132 minus03675 minus01379Corg Phys Pro conc 20ndash30 02257 005677 minus001036Corg Phys Pro Frac 20ndash30 009518 03528 01294Corg Chem Pro conc 20ndash30 02323 minus006802 minus01366Corg Chem Pro Frac 20ndash30 004565 04437 minus001278

agroenvironmental conditions of an olive orchard The vari-ability induced by the combined effects of water and tillageerosion in this olive orchard was similar to that described inother agroecosystems For instance Van Oost et al (2005)measured a clear correlation between the erosion and depo-sition rates and the topsoil Corg concentration which rangedbetween 08 of the erosion to 14 of the deposition sitesin the top 25 cm of the soil at two field crop sites under atemperate climate Besides this Bameri et al (2015) mea-sured a higher Corg in the lower part of a field crop site undersemiarid conditions where deposition of the eroded soil fromthe upper zones took place Overall in such landscapes culti-vated for a long time the cumulative effect of tillage and wa-ter erosion on the redistribution of soil across the slope hasbeen observed (Dlugoszlig et al 2012) These processes alsoproduce a vertical redistribution of Corg resulting in a rela-tively homogeneous profile in the tilled layer (top 15ndash20 cm)and a gradual decline below this depth as noted in our study

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

190 J A Goacutemez et al Organic carbon across an olive orchard catena

Figure 9 Soil available phosphorus (Pavail) organic nitrogen (Norg) aggregate stability (Wsagg) and soil degradation index (SDI) by depthcomparing eroded vs deposition areas within the olive orchard The labels of the bars indicate the p value according to a one-way ANOVAcomparing areas for the same soil depth

Figure 10 Scores on principal components 1 and 2 (PC 1 PC 2)for sampling points in the eroded and deposition areas in the oliveorchard

42 Organic carbon stock

The differences in soil organic carbon stock between the ref-erence site and the olive orchard are similar to those de-scribed previously when comparing cropland and forestedareas with the latter presenting a higher concentration ofCorg most of which is in the unprotected fraction whilethe cropland presented a higher fraction of the carbon inthe physically and chemically protected fractions (eg Poe-plau and Don 2013) This is likely due to the fact that un-der soil degradation processes such as water erosion and

low annual organic carbon input as is the case under oliveorchard land use most of the unprotected Corg decomposesrelatively quickly and a great proportion of the remaininglow SOC is protected In addition the mobilization of theunprotected Corg is expected to be reduced in the protectedforested area because of the canopy and the existing veg-etation on the ground that protects the soil against runoffand splash erosion processes In fact the protected Corg con-centration in the topsoil of the olive orchard in the erodedarea accounted for 186plusmn 39 of the upper limit of pro-tected Corg (364plusmn 023 ) according to the model of Has-sink and Whitmore (1997) Therefore the low unprotectedSOC concentration found in the olive orchard is an elementof concern in the increase in SOC stocks This is because pro-tected fractions are fueled from recently derived partially de-composed plant residues together with microbial and micro-meso- and macrofaunal debris (unprotected organic carbon)through processes like SOC aggregation into macro- andormicroaggregates (physically protected SOC) and complexSOC associations with clay and silt particles (chemicallyprotected SOC) which are disrupted in the cropland area incomparison with the reference area The distribution amongsoil Corg fractions in the orchard of this study was similarto the result obtained by Vicente-Vicente et al (2017) whomeasured Corg fraction distribution in olive oil orchards withtemporary cover crops with the exception of the unprotectedSOC which was much higher in soils under cover crops thanthat of our study under bare soil Also interesting is the dif-ficulty obtaining statistically significant differences in SOC

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 191

Tabl

e7

Pear

son

corr

elat

ion

coef

ficie

nts

oflin

ear

corr

elat

ion

betw

een

vari

able

sat

each

dept

hin

terv

al(c

m)

with

the

high

est

load

son

prin

cipa

lco

mpo

nent

s(s

eeTa

ble

6)lowast

Mea

nsst

atis

tical

lysi

gnifi

cant

atαlt

005

Ero

sde

pP a

vail

0ndash10

P ava

il0ndash

40δ

13C

10ndash2

0N

org

10ndash2

0δ

13C

20ndash3

0C

org

conc

C

org

Phys

C

org

conc

C

org

Phys

C

org

Che

m

Cor

gC

hem

10

ndash20

Pro

conc

20

ndash30

Pro

conc

Pr

oco

nc

Pro

Frac

10

ndash20

20ndash3

020

ndash30

20ndash3

0

Ero

sde

p1

P ava

il0ndash

100

746lowast

1P a

vail

0ndash40

095

3lowast0

857lowast

1δ

13C

10ndash2

0minus

082

3lowastminus

076

6lowast1

Nor

g10

ndash20

073

1lowast0

898lowast

083

9lowastminus

096

5lowast1

δ13

C20

ndash30

minus0

792lowast

minus0

893lowast

minus0

891lowast

088

8lowastminus

093

8lowast1

Cor

gco

nc1

0ndash20

075

7lowast0

843lowast

089

9lowastminus

092

9lowast0

920lowast

minus0

911lowast

1C

org

Phys

Pro

con

c10

ndash20

075

5lowast0

908lowast

088

2lowastminus

092

90

967lowast

minus0

985lowast

095

2lowast1

Cor

gco

nc2

0ndash30

078

6lowast0

799lowast

minus0

965lowast

092

7lowastminus

084

6lowast0

9453lowast

089

42lowast

1C

org

Phys

Pro

con

c20

ndash30

083

7lowast0

756lowast

minus0

846lowast

081

7lowastminus

071

4lowast0

8819lowast

079

28lowast

090

1lowast1

Cor

gC

hem

Pro

con

c20

ndash30

074

2lowast0

855lowast

085

9lowastminus

096

2lowast0

980lowast

minus0

896lowast

095

15lowast

094

34lowast

095

0lowast0

853lowast

1C

org

Che

mP

roF

rac

20ndash3

01

stock between the eroding and deposition areas (Fig 7) de-spite the apparent clear differences between the two areas insome other soil properties such as Pavail or δ15N and δ 13Cisotopic signatures in the subsoil (see below)

43 δ15N and δ13C isotopic signal

Changes in vegetation types induced differences in δ13C be-tween the olive orchard and the reference area but as ex-pected no differences in δ13C were detected between the ero-sion and deposition areas in the olive orchard given the sameorigin of vegetation-derived organic matter ie C3 plantsInterestingly within the olive orchard significant differencesin δ15N were detected between the erosion and deposition ar-eas especially for the top 20 cm of soils (Fig 9) This sug-gests the potential of using δ15N as a variable for identifyingdegraded areas in olive growing fields as has been proposedfor other eroding regions in the world (eg Meusburger etal 2013) This might provide an alternative when other con-ventional or isotopic techniques are not available Neverthe-less further studies exploring this potential are necessary inorder to also consider the influence of the NndashP fertilizer mod-ifying the δ15N in relation to the reference area (eg Bate-man and Kelly 2007) The source of N in soil is multifariousand subject to a wide range of transformations that affect theδ15N signature therefore we can only speculate on the rea-sons for this difference in δ15N in a relatively homogeneousarea Bulk soil δ15N tended to be more positive (eg moreenriched in δ15N) as the N cycling rate increases soil micro-bial processes (eg N mineralization nitrification and deni-trification) resulting in products (eg nitrate N2O N2 andNH3) depleted in 15N while the substrate from which theywere formed becomes slightly enriched (Robison 2001) Thehigher δ15N signature of the soil at the deposition locationsuggests that the rates of processes involved in N cycling arehigher than in the erosion area and that it is in accordancewith the higher bulk Corg and Pavail contents and lower SDIof the deposition site The relatively lower soil δ15N signa-ture at the reference site could be partially due to the input oflitter N from the natural legumes and to the closed N cyclingwhich characterize natural forest ecosystems The trend in15N enrichment with soil depth as found at the referencesite is a common observation at forest and grassland sitesand has been related to different mechanisms including 15Nisotope discrimination during microbial N transformationsdifferential preservation of 15N-enriched soil organic mattercomponents during N decomposition and more recently tothe build-up of 15N-enriched microbial necromass (Huygenset al 2008) However there still remains the need for a care-ful calibration to an undisturbed reference site and a betterunderstanding of the influence of different vegetation in thereference and the studied area on the change in the δ15N sig-nal for its further use as an additional tool to determine soildegradation

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

192 J A Goacutemez et al Organic carbon across an olive orchard catena

44 Soil quality of topsoil across the catena variabilitybetween eroded and deposition areas in the orchard

This horizontal distribution due to tillage and water erosionalso simultaneously affected other soil properties and hasbeen described previously in other cultivated areas For in-stance De Gryze et al (2008) described how for a field croparea under conventional tillage in Belgium Pavail almost dou-bled (229 vs 122 mg kgminus1) in the depositional area as com-pared to the eroding upper part They also reported that inhalf of the field under conservation tillage these differencesin Pavail between the upper and lower areas of the field dis-appeared We have observed in our sampled orchard a pro-nounced increase in topsoil Pavail in the deposition area ofaround 400 as compared to the eroding part of the orchardwhich can be attributed to the deposition of enriched sedi-ment coming from the upslope area The cumulative effectsof the differences in Corg and Pavail and the trend towardshigher although nonsignificant Wsagg in the deposition arearesulted in two areas within the orchard with marked differ-ences in soil quality the eroded part which is within therange considered to be degraded in the region (Goacutemez et al2009a) and the depositional area representing 20 of theorchard transect length (Table 1) which is within the non-degraded range according to the same index Topographyand sediment redistribution by erosive processes introducea gradient of spatial variability that questions the conceptof representative area when it comes to describing a wholefield In fact several studies (eg Dell and Sharpley 2006)have suggested that the verification of compliance of envi-ronmental programs such as those related to Corg seques-tration should be based preferentially at least partially onmodeling approaches Our results raise the need for a care-ful delineation of subareas when analyzing soil quality indi-cators andor SOC carbon stock within the same field unitThey also warn about the difficulty of drawing hypothesesfor quantifying the differences between these delineated ar-eas in relation to specific soil properties For instance in ourstudy case erosionndashdeposition processes had a major impacton Pavail and soil quality but the impact on Corg concentra-tion and stock was moderate and extremely difficult to detectstatistically using a moderate number of samples

Our PCA and regression analyses confirmed the relativelyhigh variability of Corg and stock in relation to other soilquality indicators related to erosionndashdeposition processessuch as Pavail as discussed in the previous section While inthe catena studied in the current research the P cycle seemsto be mostly driven by sediment mobilization the Corg and Ncycles seem to be much more complex The moderate differ-ences in Corg and the homogeneity in δ15N and δ13C isotopicsignals between the eroding and deposition areas may be dueto several processes some of which were discussed abovesuch as spatial variability of carbon input due to biomassin the plot surface soil operations in the orchard and fer-tilization We found it both interesting and worth exploring

in future studies that significant correlations between erosionand deposition rates and Corg- δ15N- and δ 13C-related vari-ables were found for samples from the 10ndash20 and 20ndash30 cmlayers indicating that short-term disturbance by surface pro-cesses can mask experimental determination of the impact oferosion deposition processes in this olive orchard for thesevariables

5 Conclusions

1 The results indicate that erosion and deposition withinthe investigated old olive orchard have created a signifi-cant difference in soil properties along the catena whichis translated into different soil Corg Pavail and Norg con-tent δ15N and thus contrasting soil quality status

2 This variability was lower than that of the natural areawhich indicated a severe depletion of SOC as comparedto the natural area and a redistribution of available or-ganic carbon among the different SOC fractions

3 The results suggest that δ15N has the potential for beingused as an indicator of soil degradation although moreinvestigation in different agroecosystems would be re-quired for confirming this statement on a larger scale

4 This research highlights that proper understanding andmanagement of soil quality and Corg content in olive or-chards require consideration of the on-site spatial vari-ability induced by soil erosionndashdeposition processes

Data availability The basic data used in this paper are freelyavailable from the CSIC digital repository at httpsdoiorg1020350digitalCSIC12513 (Goacutemez Calero et al 2020)

Author contributions JAG coordinated and designed the over-all study conducted the field sampling campaigns analysis of thesoil organic carbon results and the overall analysis of the resultsand wrote the manuscript GG participated in the analysis of soil-quality-related properties the overall analysis of the results and thewriting of the manuscript LM participated in the overall analysisof the results and the writing of the manuscript CR and AT partici-pated in the analysis and interpretation of the stable isotope resultsRGR contributed to the design and the statistical analysis of soilorganic carbon (stock and fractions) and nitrogen data the overallanalysis of the results and the writing of the manuscript

Competing interests The authors declare that they have no con-flict of interest

Acknowledgements The authors would like to thankManuel Gonzaacutelez de Molina for providing valuable insightinto the management of the olive orchard in the decades before

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

184 J A Goacutemez et al Organic carbon across an olive orchard catena

(Fig 3) Corg values were between 2 to 6 times higher for theunprotected and chemically protected fractions and between2 to 35 times higher for the physically and biochemicallyprotected fractions with differences tending to decrease withthe soil depth

Within the olive orchard there were statistically signifi-cant differences between the erosion and deposition areas forbulk Corg values (p = 00198) (Table 2b) Higher Corg val-ues (11 to 06 ) were observed in the deposition arealocated downslope whereas lower values (085 to 055 )were measured in the areas with net erosion in the upper andmid sections of the catena It is worth noting that these differ-ences between erosion and deposition areas are detected forthe overall analysis using a two-way ANOVA (Table 2b) al-though an individual analysis at each depth (Fig 4) does notdetect statistically significant differences probably due to themoderate number of replications Significant differences be-tween the deposition and eroding area were also found forthe unprotected (p = 004) and the physically (p = 00077)and chemically protected fractions (p = 00299) (Table 2b)However differences for the biochemically protected frac-tions (Table 2b Fig 4) were not significant

The percentage distribution of SOC among fractions wasalso significantly different between both areas (reference vsolive orchard) except for the biochemically protected fraction(p lt 00001) (Table 3a) and depths (p lt 0011) (Fig 5) Inthe reference area most of the Corg was unprotected (between50 and 65 approximately) with no significant trend withdepth (Table 3a Fig 5) followed in relative importance bythe chemically and physically protected fractions which con-tributed between 18 and 30 as well as 10 and 20 ofthe bulk soil Corg respectively The biochemically protectedfraction represents a very low percentage (between 4 and6 ) In the olive orchard Corg is stored predominantly inthe physically and chemically protected fractions which ac-counted for about 38 to 27 and 34 to 28 respec-tively followed by the pool of unprotected fractions (be-tween 22 to 32 ) (Fig 5) The biochemically protectedfraction represents between 4 and 11 of the organic car-bon stored in the olive orchard There are no clear differencesin the organic carbon distribution among the different frac-tions between the erosion and deposition areas in the oliveorchard with the exception of the physically protected frac-tions at 10ndash20 cm depth (p = 0024) (Fig 6)

32 Organic carbon stock

SOC stock in the reference area is approximately 160 t haminus1making it significantly higher at p lt 005 for an equivalentmass than that of the olive orchard (Fig 7) which stores be-tween 38 and 41 t haminus1 in the eroded and deposition areasrespectively There were no significant differences betweenthese two orchard areas Similar results were obtained acrossthe top 40 cm soil profile Clay silt and sand contents of thetopsoil (0ndash10 cm) along the catena in the olive orchard av-

eraged 41 37 and 22 respectively with similar av-erage values in the eroding and deposition areas Variabilitywas relatively low (average coefficient of variation of 17 )and there were no significant changes between the erosionand deposition areas In the reference area the soil has anaverage clay silt and sand content of 30 31 and 39 respectively also with a homogeneous distribution across thesampling area (coefficient of variation of 10 ) According tothe Hassink and Whitmore (1997) model the percentages oforganic carbon of maximum soil stable Corg are 324plusmn011 and 363plusmn019 in the reference site and olive orchard re-spectively So protected Corg in the reference and olive or-chard areas account for 498plusmn 115 and 2049plusmn 52 ofthe maximum soil stable Corg respectively in the topsoil

33 δ15N and δ13C isotopic signal

Figure 8 and Table 4a compare stable isotope delta values be-tween the reference site and the overall olive orchard accord-ing to depth There are significant statistical differences inδ15N δ 13C and the δ13C δ15N ratio between the two areas(p lt 0002 Table 4a) although in the case of δ 15N the dif-ferences at individual depths were not significant Soil depthalso had a significant effect (00026lt p lt 0029 Table 4a)When comparing differences between the erosion and depo-sition areas within the olive orchard we detected statisticallysignificant differences only in δ15N and the δ13C δ15N ra-tio (p = 001 Table 4b) and they were mostly marked in theupper 20 cm of the soil (Fig 8)

34 Soil quality of topsoil across the catena variabilitybetween eroded and deposition areas in the orchard

Figure 9 depicts the comparison of the Pavail Norg Wsaggand the soil degradation index (SDI Goacutemez et al 2009a)in the top 10 cm of the soil between the erosion and deposi-tion areas of the olive orchard and Table 5 shows a similarcomparison for Norg Pavail and bulk density at the differentsoil depths Pavail in the deposition area is significantly higherthan that of the erosion area in the topsoil (p = 0005 Fig 9)and for the whole profile (p = 001 Table 5) whereas no sig-nificant differences in individual soil layers were found forNorg and Wsagg The SDI which combines these three vari-ables was about 3 times higher in the eroded area than in thedeposition area

Table 6 shows the loads in the first three principal com-ponents (PCs) of the principal component analysis (PCA)for the variables used in this analysis More than 70 ofthe variance was explained by the first two PCs Soils ofthe eroded and deposition area were clearly separated in thespace defined by the two PCs (Fig 10) The variables withhigher contributions in PC 1 and 2 were related to Pavail con-centration in the 0ndash10 and 0ndash40 cm soil depths to δ15N andδ13C in the 10 to 30 cm soil depths and to Corg concentra-tion and distribution in some fractions also in soil depths be-

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 185

Figure 2 Comparison of average soil organic carbon concentration in bulk soil depending on soil depth distinguishing between the referencesite the whole olive orchard the eroded area of the olive orchard and the deposition area in this orchard The labels of the bars for eachdepth indicate the p value according to a one-way ANOVA for the reference area (dark blue) vs the whole olive orchard (dark green) andfor the eroding area (beige) vs the deposition area (light green lower label in italics)

Figure 3 Organic carbon concentration in the different soil organic carbon fractions at each depth comparing the reference site vs oliveorchard The labels of the bars indicate the p value according to a one-way ANOVA comparing treatments for the same soil depth and carbonfraction between the reference area and olive orchard

tween 10 and 30 cm (Table 6) The deposition area tendedto show higher values in PC 1 and PC 2 and there wasno clear tendency in the erosion area along the catena Thelinear correlation coefficient between the variables and theerosion deposition rate was rather significant (r gtplusmn0731)for most of the variables (Table 7) Interestingly Pavail was

highly positively correlated in the whole soil layer Amongthe two most robust correlations with erosion depositionrates were that with Pavail concentration across the 40 cm soildepth y = 03975x+ 98364 (r2

= 0907) and δ13C at 10ndash20 cm depth y =minus00094xminus 25318 (r2

= 0632)

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

186 J A Goacutemez et al Organic carbon across an olive orchard catena

Table 2 Results of the two-way ANOVA of soil organic carbon concentration Corg () in different fractions and in bulk soil In (a) arearefers to the reference site vs the olive orchard and in (b) area refers to eroded vs deposition areas in the olive orchard NS stands for notsignificant

(a) Model Bulk soil Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00000 00000 00000 00000 00000Depth (D) 00023 00022 00061 00190 NSAtimesD NS NS NS NS 00300

(b) Model Bulk soil Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00198 00400 00077 00299 NSDepth (D) 00081 00070 00058 00055 00847AtimesD NS NS NS NS NS

Figure 4 Organic carbon concentration in the different soil organic carbon fractions at each depth comparing the eroded vs the depositionarea within the olive orchard The labels of the bars indicate the p value according to a one-way ANOVA comparing treatments for the samesoil depth and carbon fraction between the reference area and olive orchard

4 Discussion

41 Organic carbon concentration and distributionamong fractions

After approximately 175 years of contrasted land use be-tween the undisturbed reference site and the olive orchardbulk soil organic carbon concentration and its fractions havebeen dramatically reduced in the olive orchard Current lev-els of Corg concentration in the soil profile are approximately20 to 25 of that found in the reference area covered bynatural vegetation in the area adjacent to the orchard This ra-

tio is similar albeit in the lower range to the comparison ofCorg in topsoil among olive orchards with different types ofmanagement and natural areas reported for the region (Mil-groom et al 2007) The increased soil disturbance the lowerannual rate of biomass returned to the soil and the highererosion rate in the olive orchard explain this difference Inboth areas the Corg is clearly stratified indicating that de-spite the different mechanisms involved there is a periodicinput of biomass from the olive trees (eg fall of senescenceleaves and tree pruning residues) plus the annual ground veg-etation Vicente-Vicente et al (2017) estimated this biomasscontribution in the range of 148 to 056 t haminus1 yrminus1 It is

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 187

Figure 5 Contribution () of the different fractions with respect to total soil organic carbon by depth comparing the reference site vs theolive orchard The labels of each fraction and depth are the p value according to a one-way ANOVA comparing the reference area vs theolive orchard for the same carbon fraction and soil depth

Figure 6 Fraction of total organic carbon stored in the different fractions by depth comparing the reference site vs the olive orchard Thelabels of each fraction and depth are the p value according to a one-way ANOVA comparing the reference area vs the olive orchard for thesame carbon fraction and soil depth

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

188 J A Goacutemez et al Organic carbon across an olive orchard catena

Table 3 Results of the two-way ANOVA of the distribution of the total soil organic carbon content in the soil among the different fractionsof soil organic carbon Corg In (a) area refers to the reference site vs the olive orchard and in (b) area refers to eroded vs deposition areasin the olive orchard NS stands for not significant

(a) Model Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00000 00000 00000 NSDepth (D) NS NS 00640 NSAtimesD NS 00059 NS NS

(b) Model Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 0091 00881 NS NSDepth (D) 0051 00214 NS 0033AtimesD NS NS NS NS

Figure 7 Cumulative soil organic carbon (SOC) stock across thesoil profile in terms of cumulative soil mass on the vertical axisDifferent letters for similar soil mass refer to statistically significantdifferences (KruskalndashWallis test at p lt 005) For this analysis cu-mulative soil organic carbons were interpolated linearly to the aver-age cumulative soil mass corresponding to all the points in the threeareas

worth noticing that the decrease in Corg as compared to thenatural area is much higher than the reported rates of increasein Corg in olive orchards using conservation agriculture (CA)techniques such as cover crops and incorporation of organicresidues from different sources In a meta-analysis Vicente-Vicente et al (2016) found a response ratio (the ratio of Corgunder CA management as compared to Corg in orchards withbare-soil management) of 11 to 19 suggesting that underCA management which combines cover crops and organicresidues Corg doubled as a maximum

Table 4 Results of the two-way ANOVA of the stable isotope sig-nal In (a) area refers to the reference site vs the olive orchard andin (b) area refers to eroded vs deposition areas in the olive orchardNS stands for not significant

(a) Model δ13C δ15N δ13C δ15N

Area (A) 00001 0002 0002Depth (D) 00026 00175 0029AtimesD NS NS NS

(b) Model δ13C δ15N δ13C δ15N

Area (A) NS 001 001Depth (D) NS NS NSAtimesD NS NS NS

Table 5 Results of the two-way ANOVA of some physical andchemical soil properties comparing eroded vs deposition areas inthe olive orchard NS stands for not significant

Model Norg Pavail Bulk density

Area (A) 00000 001 NSDepth (D) 00009 NS NSAtimesD NS NS NS

Combining all Corg data of the olive orchard the variabil-ity was about 35 which is similar to what has been re-ported so far in the few studies found on soil Corg variabilityin olive orchards For instance Gargouri et al (2013) indi-cated a 24 coefficient of variation (CV) in a 34 ha oliveorchard in Tunisia while Huang et al (2017) reported an av-erage CV of 41 in a 62 ha olive orchard in southern SpainNeither of these two studies reported clear trends in the distri-bution of Corg depending on topography Huang et al (2017)

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 189

Figure 8 13C and 15N isotopic signal of soil by depth distinguish-ing among reference site the whole olive orchard the eroded areaof the olive orchard and the deposition area in this orchard Thelabels of the bars for each depth indicate the p value according toa one-way ANOVA between the reference area (dark blue) vs thewhole olive orchard (dark green) and the eroding area (beige) vsthe deposition area (light green lower label in italics)

pointed out the additional difficulties in the determination ofCorg due to the topography heterogeneity although this wascompounded by the fact that within the orchards there weretwo areas with different planting dates for the trees Goacutemezet al (2012) reported a CV of 49 with higher Corg in ar-eas where there was a change in the slope gradient from thehillslope to a central channel draining into the catchment al-though they could not find a simple relationship between theincrease in content and the topographic indices Despite thefact that a lot of work has been done on the correlation be-tween erosion and deposition and the redistribution of soilCorg (eg Van Oost et al 2005) our study is to our knowl-edge the first attempt to quantify this in detail under the

Table 6 Loads of selected variables in the PCA for the first threeprincipal components (PCs) The values in brackets below PC 1 2and 3 indicate the percentage of variance explained by this PC Vari-ables in bold are those with a load higher than 90 of the variablewith the maximum load for this PC Conc refers to Corg concen-tration for this fraction and Frac means the relative contribution ofthis fraction to the total Corg for this soil depth

Variable at each depth interval (cm) PC 1 PC 2 PC 3(558) (176) (132)

Pavail 0ndash10 02298 008765 minus007278Pavail 0ndash40 02271 0101 minus008455δ13C 0ndash10 minus003385 02861 minus03302δ15N 0ndash10 01941 01677 01836Norg 0ndash10 02147 minus01375 minus00336δ13C δ15N 0ndash10 01594 0249 02211δ13C 10ndash20 minus02329 01461 004296δ15N 10ndash20 01856 007054 03121Norg 10ndash20 02317 minus008699 minus01404δ13C δ15N 10ndash20 01637 008512 03309δ13C 20ndash30 minus02316 006018 009789δ15N 20ndash30 01748 minus02656 01812Norg 20ndash30 02176 009979 minus007458δ13C δ15N 20ndash30 01684 minus02982 01774Corg conc 10ndash20 02421 minus002951 minus006407Corg unpr conc 10ndash20 02116 00331 00385Corg unpr Frac 10ndash20 minus009663 minus00509 0364Corg Phys Pro conc 10ndash20 02371 minus005728 minus01183Corg Phys Pro Frac 10ndash20 0123 0163 minus002953Corg Chem Pro conc 10ndash20 02072 003097 minus02335Corg Chem Pro Frac 10ndash20 minus004095 009461 minus04678Corg conc 20ndash30 02326 minus01108 minus005737Corg unpr conc 20ndash30 02002 minus02372 minus005071Corg unpr Rel 20ndash30 cm minus0132 minus03675 minus01379Corg Phys Pro conc 20ndash30 02257 005677 minus001036Corg Phys Pro Frac 20ndash30 009518 03528 01294Corg Chem Pro conc 20ndash30 02323 minus006802 minus01366Corg Chem Pro Frac 20ndash30 004565 04437 minus001278

agroenvironmental conditions of an olive orchard The vari-ability induced by the combined effects of water and tillageerosion in this olive orchard was similar to that described inother agroecosystems For instance Van Oost et al (2005)measured a clear correlation between the erosion and depo-sition rates and the topsoil Corg concentration which rangedbetween 08 of the erosion to 14 of the deposition sitesin the top 25 cm of the soil at two field crop sites under atemperate climate Besides this Bameri et al (2015) mea-sured a higher Corg in the lower part of a field crop site undersemiarid conditions where deposition of the eroded soil fromthe upper zones took place Overall in such landscapes culti-vated for a long time the cumulative effect of tillage and wa-ter erosion on the redistribution of soil across the slope hasbeen observed (Dlugoszlig et al 2012) These processes alsoproduce a vertical redistribution of Corg resulting in a rela-tively homogeneous profile in the tilled layer (top 15ndash20 cm)and a gradual decline below this depth as noted in our study

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

190 J A Goacutemez et al Organic carbon across an olive orchard catena

Figure 9 Soil available phosphorus (Pavail) organic nitrogen (Norg) aggregate stability (Wsagg) and soil degradation index (SDI) by depthcomparing eroded vs deposition areas within the olive orchard The labels of the bars indicate the p value according to a one-way ANOVAcomparing areas for the same soil depth

Figure 10 Scores on principal components 1 and 2 (PC 1 PC 2)for sampling points in the eroded and deposition areas in the oliveorchard

42 Organic carbon stock

The differences in soil organic carbon stock between the ref-erence site and the olive orchard are similar to those de-scribed previously when comparing cropland and forestedareas with the latter presenting a higher concentration ofCorg most of which is in the unprotected fraction whilethe cropland presented a higher fraction of the carbon inthe physically and chemically protected fractions (eg Poe-plau and Don 2013) This is likely due to the fact that un-der soil degradation processes such as water erosion and

low annual organic carbon input as is the case under oliveorchard land use most of the unprotected Corg decomposesrelatively quickly and a great proportion of the remaininglow SOC is protected In addition the mobilization of theunprotected Corg is expected to be reduced in the protectedforested area because of the canopy and the existing veg-etation on the ground that protects the soil against runoffand splash erosion processes In fact the protected Corg con-centration in the topsoil of the olive orchard in the erodedarea accounted for 186plusmn 39 of the upper limit of pro-tected Corg (364plusmn 023 ) according to the model of Has-sink and Whitmore (1997) Therefore the low unprotectedSOC concentration found in the olive orchard is an elementof concern in the increase in SOC stocks This is because pro-tected fractions are fueled from recently derived partially de-composed plant residues together with microbial and micro-meso- and macrofaunal debris (unprotected organic carbon)through processes like SOC aggregation into macro- andormicroaggregates (physically protected SOC) and complexSOC associations with clay and silt particles (chemicallyprotected SOC) which are disrupted in the cropland area incomparison with the reference area The distribution amongsoil Corg fractions in the orchard of this study was similarto the result obtained by Vicente-Vicente et al (2017) whomeasured Corg fraction distribution in olive oil orchards withtemporary cover crops with the exception of the unprotectedSOC which was much higher in soils under cover crops thanthat of our study under bare soil Also interesting is the dif-ficulty obtaining statistically significant differences in SOC

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 191

Tabl

e7

Pear

son

corr

elat

ion

coef

ficie

nts

oflin

ear

corr

elat

ion

betw

een

vari

able

sat

each

dept

hin

terv

al(c

m)

with

the

high

est

load

son

prin

cipa

lco

mpo

nent

s(s

eeTa

ble

6)lowast

Mea

nsst

atis

tical

lysi

gnifi

cant

atαlt

005

Ero

sde

pP a

vail

0ndash10

P ava

il0ndash

40δ

13C

10ndash2

0N

org

10ndash2

0δ

13C

20ndash3

0C

org

conc

C

org

Phys

C

org

conc

C

org

Phys

C

org

Che

m

Cor

gC

hem

10

ndash20

Pro

conc

20

ndash30

Pro

conc

Pr

oco

nc

Pro

Frac

10

ndash20

20ndash3

020

ndash30

20ndash3

0

Ero

sde

p1

P ava

il0ndash

100

746lowast

1P a

vail

0ndash40

095

3lowast0

857lowast

1δ

13C

10ndash2

0minus

082

3lowastminus

076

6lowast1

Nor

g10

ndash20

073

1lowast0

898lowast

083

9lowastminus

096

5lowast1

δ13

C20

ndash30

minus0

792lowast

minus0

893lowast

minus0

891lowast

088

8lowastminus

093

8lowast1

Cor

gco

nc1

0ndash20

075

7lowast0

843lowast

089

9lowastminus

092

9lowast0

920lowast

minus0

911lowast

1C

org

Phys

Pro

con

c10

ndash20

075

5lowast0

908lowast

088

2lowastminus

092

90

967lowast

minus0

985lowast

095

2lowast1

Cor

gco

nc2

0ndash30

078

6lowast0

799lowast

minus0

965lowast

092

7lowastminus

084

6lowast0

9453lowast

089

42lowast

1C

org

Phys

Pro

con

c20

ndash30

083

7lowast0

756lowast

minus0

846lowast

081

7lowastminus

071

4lowast0

8819lowast

079

28lowast

090

1lowast1

Cor

gC

hem

Pro

con

c20

ndash30

074

2lowast0

855lowast

085

9lowastminus

096

2lowast0

980lowast

minus0

896lowast

095

15lowast

094

34lowast

095

0lowast0

853lowast

1C

org

Che

mP

roF

rac

20ndash3

01

stock between the eroding and deposition areas (Fig 7) de-spite the apparent clear differences between the two areas insome other soil properties such as Pavail or δ15N and δ 13Cisotopic signatures in the subsoil (see below)

43 δ15N and δ13C isotopic signal

Changes in vegetation types induced differences in δ13C be-tween the olive orchard and the reference area but as ex-pected no differences in δ13C were detected between the ero-sion and deposition areas in the olive orchard given the sameorigin of vegetation-derived organic matter ie C3 plantsInterestingly within the olive orchard significant differencesin δ15N were detected between the erosion and deposition ar-eas especially for the top 20 cm of soils (Fig 9) This sug-gests the potential of using δ15N as a variable for identifyingdegraded areas in olive growing fields as has been proposedfor other eroding regions in the world (eg Meusburger etal 2013) This might provide an alternative when other con-ventional or isotopic techniques are not available Neverthe-less further studies exploring this potential are necessary inorder to also consider the influence of the NndashP fertilizer mod-ifying the δ15N in relation to the reference area (eg Bate-man and Kelly 2007) The source of N in soil is multifariousand subject to a wide range of transformations that affect theδ15N signature therefore we can only speculate on the rea-sons for this difference in δ15N in a relatively homogeneousarea Bulk soil δ15N tended to be more positive (eg moreenriched in δ15N) as the N cycling rate increases soil micro-bial processes (eg N mineralization nitrification and deni-trification) resulting in products (eg nitrate N2O N2 andNH3) depleted in 15N while the substrate from which theywere formed becomes slightly enriched (Robison 2001) Thehigher δ15N signature of the soil at the deposition locationsuggests that the rates of processes involved in N cycling arehigher than in the erosion area and that it is in accordancewith the higher bulk Corg and Pavail contents and lower SDIof the deposition site The relatively lower soil δ15N signa-ture at the reference site could be partially due to the input oflitter N from the natural legumes and to the closed N cyclingwhich characterize natural forest ecosystems The trend in15N enrichment with soil depth as found at the referencesite is a common observation at forest and grassland sitesand has been related to different mechanisms including 15Nisotope discrimination during microbial N transformationsdifferential preservation of 15N-enriched soil organic mattercomponents during N decomposition and more recently tothe build-up of 15N-enriched microbial necromass (Huygenset al 2008) However there still remains the need for a care-ful calibration to an undisturbed reference site and a betterunderstanding of the influence of different vegetation in thereference and the studied area on the change in the δ15N sig-nal for its further use as an additional tool to determine soildegradation

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

192 J A Goacutemez et al Organic carbon across an olive orchard catena

44 Soil quality of topsoil across the catena variabilitybetween eroded and deposition areas in the orchard

This horizontal distribution due to tillage and water erosionalso simultaneously affected other soil properties and hasbeen described previously in other cultivated areas For in-stance De Gryze et al (2008) described how for a field croparea under conventional tillage in Belgium Pavail almost dou-bled (229 vs 122 mg kgminus1) in the depositional area as com-pared to the eroding upper part They also reported that inhalf of the field under conservation tillage these differencesin Pavail between the upper and lower areas of the field dis-appeared We have observed in our sampled orchard a pro-nounced increase in topsoil Pavail in the deposition area ofaround 400 as compared to the eroding part of the orchardwhich can be attributed to the deposition of enriched sedi-ment coming from the upslope area The cumulative effectsof the differences in Corg and Pavail and the trend towardshigher although nonsignificant Wsagg in the deposition arearesulted in two areas within the orchard with marked differ-ences in soil quality the eroded part which is within therange considered to be degraded in the region (Goacutemez et al2009a) and the depositional area representing 20 of theorchard transect length (Table 1) which is within the non-degraded range according to the same index Topographyand sediment redistribution by erosive processes introducea gradient of spatial variability that questions the conceptof representative area when it comes to describing a wholefield In fact several studies (eg Dell and Sharpley 2006)have suggested that the verification of compliance of envi-ronmental programs such as those related to Corg seques-tration should be based preferentially at least partially onmodeling approaches Our results raise the need for a care-ful delineation of subareas when analyzing soil quality indi-cators andor SOC carbon stock within the same field unitThey also warn about the difficulty of drawing hypothesesfor quantifying the differences between these delineated ar-eas in relation to specific soil properties For instance in ourstudy case erosionndashdeposition processes had a major impacton Pavail and soil quality but the impact on Corg concentra-tion and stock was moderate and extremely difficult to detectstatistically using a moderate number of samples

Our PCA and regression analyses confirmed the relativelyhigh variability of Corg and stock in relation to other soilquality indicators related to erosionndashdeposition processessuch as Pavail as discussed in the previous section While inthe catena studied in the current research the P cycle seemsto be mostly driven by sediment mobilization the Corg and Ncycles seem to be much more complex The moderate differ-ences in Corg and the homogeneity in δ15N and δ13C isotopicsignals between the eroding and deposition areas may be dueto several processes some of which were discussed abovesuch as spatial variability of carbon input due to biomassin the plot surface soil operations in the orchard and fer-tilization We found it both interesting and worth exploring

in future studies that significant correlations between erosionand deposition rates and Corg- δ15N- and δ 13C-related vari-ables were found for samples from the 10ndash20 and 20ndash30 cmlayers indicating that short-term disturbance by surface pro-cesses can mask experimental determination of the impact oferosion deposition processes in this olive orchard for thesevariables

5 Conclusions

1 The results indicate that erosion and deposition withinthe investigated old olive orchard have created a signifi-cant difference in soil properties along the catena whichis translated into different soil Corg Pavail and Norg con-tent δ15N and thus contrasting soil quality status

2 This variability was lower than that of the natural areawhich indicated a severe depletion of SOC as comparedto the natural area and a redistribution of available or-ganic carbon among the different SOC fractions

3 The results suggest that δ15N has the potential for beingused as an indicator of soil degradation although moreinvestigation in different agroecosystems would be re-quired for confirming this statement on a larger scale

4 This research highlights that proper understanding andmanagement of soil quality and Corg content in olive or-chards require consideration of the on-site spatial vari-ability induced by soil erosionndashdeposition processes

Data availability The basic data used in this paper are freelyavailable from the CSIC digital repository at httpsdoiorg1020350digitalCSIC12513 (Goacutemez Calero et al 2020)

Author contributions JAG coordinated and designed the over-all study conducted the field sampling campaigns analysis of thesoil organic carbon results and the overall analysis of the resultsand wrote the manuscript GG participated in the analysis of soil-quality-related properties the overall analysis of the results and thewriting of the manuscript LM participated in the overall analysisof the results and the writing of the manuscript CR and AT partici-pated in the analysis and interpretation of the stable isotope resultsRGR contributed to the design and the statistical analysis of soilorganic carbon (stock and fractions) and nitrogen data the overallanalysis of the results and the writing of the manuscript

Competing interests The authors declare that they have no con-flict of interest

Acknowledgements The authors would like to thankManuel Gonzaacutelez de Molina for providing valuable insightinto the management of the olive orchard in the decades before

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 185

Figure 2 Comparison of average soil organic carbon concentration in bulk soil depending on soil depth distinguishing between the referencesite the whole olive orchard the eroded area of the olive orchard and the deposition area in this orchard The labels of the bars for eachdepth indicate the p value according to a one-way ANOVA for the reference area (dark blue) vs the whole olive orchard (dark green) andfor the eroding area (beige) vs the deposition area (light green lower label in italics)

Figure 3 Organic carbon concentration in the different soil organic carbon fractions at each depth comparing the reference site vs oliveorchard The labels of the bars indicate the p value according to a one-way ANOVA comparing treatments for the same soil depth and carbonfraction between the reference area and olive orchard

tween 10 and 30 cm (Table 6) The deposition area tendedto show higher values in PC 1 and PC 2 and there wasno clear tendency in the erosion area along the catena Thelinear correlation coefficient between the variables and theerosion deposition rate was rather significant (r gtplusmn0731)for most of the variables (Table 7) Interestingly Pavail was

highly positively correlated in the whole soil layer Amongthe two most robust correlations with erosion depositionrates were that with Pavail concentration across the 40 cm soildepth y = 03975x+ 98364 (r2

= 0907) and δ13C at 10ndash20 cm depth y =minus00094xminus 25318 (r2

= 0632)

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

186 J A Goacutemez et al Organic carbon across an olive orchard catena

Table 2 Results of the two-way ANOVA of soil organic carbon concentration Corg () in different fractions and in bulk soil In (a) arearefers to the reference site vs the olive orchard and in (b) area refers to eroded vs deposition areas in the olive orchard NS stands for notsignificant

(a) Model Bulk soil Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00000 00000 00000 00000 00000Depth (D) 00023 00022 00061 00190 NSAtimesD NS NS NS NS 00300

(b) Model Bulk soil Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00198 00400 00077 00299 NSDepth (D) 00081 00070 00058 00055 00847AtimesD NS NS NS NS NS

Figure 4 Organic carbon concentration in the different soil organic carbon fractions at each depth comparing the eroded vs the depositionarea within the olive orchard The labels of the bars indicate the p value according to a one-way ANOVA comparing treatments for the samesoil depth and carbon fraction between the reference area and olive orchard

4 Discussion

41 Organic carbon concentration and distributionamong fractions

After approximately 175 years of contrasted land use be-tween the undisturbed reference site and the olive orchardbulk soil organic carbon concentration and its fractions havebeen dramatically reduced in the olive orchard Current lev-els of Corg concentration in the soil profile are approximately20 to 25 of that found in the reference area covered bynatural vegetation in the area adjacent to the orchard This ra-

tio is similar albeit in the lower range to the comparison ofCorg in topsoil among olive orchards with different types ofmanagement and natural areas reported for the region (Mil-groom et al 2007) The increased soil disturbance the lowerannual rate of biomass returned to the soil and the highererosion rate in the olive orchard explain this difference Inboth areas the Corg is clearly stratified indicating that de-spite the different mechanisms involved there is a periodicinput of biomass from the olive trees (eg fall of senescenceleaves and tree pruning residues) plus the annual ground veg-etation Vicente-Vicente et al (2017) estimated this biomasscontribution in the range of 148 to 056 t haminus1 yrminus1 It is

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 187

Figure 5 Contribution () of the different fractions with respect to total soil organic carbon by depth comparing the reference site vs theolive orchard The labels of each fraction and depth are the p value according to a one-way ANOVA comparing the reference area vs theolive orchard for the same carbon fraction and soil depth

Figure 6 Fraction of total organic carbon stored in the different fractions by depth comparing the reference site vs the olive orchard Thelabels of each fraction and depth are the p value according to a one-way ANOVA comparing the reference area vs the olive orchard for thesame carbon fraction and soil depth

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

188 J A Goacutemez et al Organic carbon across an olive orchard catena

Table 3 Results of the two-way ANOVA of the distribution of the total soil organic carbon content in the soil among the different fractionsof soil organic carbon Corg In (a) area refers to the reference site vs the olive orchard and in (b) area refers to eroded vs deposition areasin the olive orchard NS stands for not significant

(a) Model Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00000 00000 00000 NSDepth (D) NS NS 00640 NSAtimesD NS 00059 NS NS

(b) Model Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 0091 00881 NS NSDepth (D) 0051 00214 NS 0033AtimesD NS NS NS NS

Figure 7 Cumulative soil organic carbon (SOC) stock across thesoil profile in terms of cumulative soil mass on the vertical axisDifferent letters for similar soil mass refer to statistically significantdifferences (KruskalndashWallis test at p lt 005) For this analysis cu-mulative soil organic carbons were interpolated linearly to the aver-age cumulative soil mass corresponding to all the points in the threeareas

worth noticing that the decrease in Corg as compared to thenatural area is much higher than the reported rates of increasein Corg in olive orchards using conservation agriculture (CA)techniques such as cover crops and incorporation of organicresidues from different sources In a meta-analysis Vicente-Vicente et al (2016) found a response ratio (the ratio of Corgunder CA management as compared to Corg in orchards withbare-soil management) of 11 to 19 suggesting that underCA management which combines cover crops and organicresidues Corg doubled as a maximum

Table 4 Results of the two-way ANOVA of the stable isotope sig-nal In (a) area refers to the reference site vs the olive orchard andin (b) area refers to eroded vs deposition areas in the olive orchardNS stands for not significant

(a) Model δ13C δ15N δ13C δ15N

Area (A) 00001 0002 0002Depth (D) 00026 00175 0029AtimesD NS NS NS

(b) Model δ13C δ15N δ13C δ15N

Area (A) NS 001 001Depth (D) NS NS NSAtimesD NS NS NS

Table 5 Results of the two-way ANOVA of some physical andchemical soil properties comparing eroded vs deposition areas inthe olive orchard NS stands for not significant

Model Norg Pavail Bulk density

Area (A) 00000 001 NSDepth (D) 00009 NS NSAtimesD NS NS NS

Combining all Corg data of the olive orchard the variabil-ity was about 35 which is similar to what has been re-ported so far in the few studies found on soil Corg variabilityin olive orchards For instance Gargouri et al (2013) indi-cated a 24 coefficient of variation (CV) in a 34 ha oliveorchard in Tunisia while Huang et al (2017) reported an av-erage CV of 41 in a 62 ha olive orchard in southern SpainNeither of these two studies reported clear trends in the distri-bution of Corg depending on topography Huang et al (2017)

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 189

Figure 8 13C and 15N isotopic signal of soil by depth distinguish-ing among reference site the whole olive orchard the eroded areaof the olive orchard and the deposition area in this orchard Thelabels of the bars for each depth indicate the p value according toa one-way ANOVA between the reference area (dark blue) vs thewhole olive orchard (dark green) and the eroding area (beige) vsthe deposition area (light green lower label in italics)

pointed out the additional difficulties in the determination ofCorg due to the topography heterogeneity although this wascompounded by the fact that within the orchards there weretwo areas with different planting dates for the trees Goacutemezet al (2012) reported a CV of 49 with higher Corg in ar-eas where there was a change in the slope gradient from thehillslope to a central channel draining into the catchment al-though they could not find a simple relationship between theincrease in content and the topographic indices Despite thefact that a lot of work has been done on the correlation be-tween erosion and deposition and the redistribution of soilCorg (eg Van Oost et al 2005) our study is to our knowl-edge the first attempt to quantify this in detail under the

Table 6 Loads of selected variables in the PCA for the first threeprincipal components (PCs) The values in brackets below PC 1 2and 3 indicate the percentage of variance explained by this PC Vari-ables in bold are those with a load higher than 90 of the variablewith the maximum load for this PC Conc refers to Corg concen-tration for this fraction and Frac means the relative contribution ofthis fraction to the total Corg for this soil depth

Variable at each depth interval (cm) PC 1 PC 2 PC 3(558) (176) (132)

Pavail 0ndash10 02298 008765 minus007278Pavail 0ndash40 02271 0101 minus008455δ13C 0ndash10 minus003385 02861 minus03302δ15N 0ndash10 01941 01677 01836Norg 0ndash10 02147 minus01375 minus00336δ13C δ15N 0ndash10 01594 0249 02211δ13C 10ndash20 minus02329 01461 004296δ15N 10ndash20 01856 007054 03121Norg 10ndash20 02317 minus008699 minus01404δ13C δ15N 10ndash20 01637 008512 03309δ13C 20ndash30 minus02316 006018 009789δ15N 20ndash30 01748 minus02656 01812Norg 20ndash30 02176 009979 minus007458δ13C δ15N 20ndash30 01684 minus02982 01774Corg conc 10ndash20 02421 minus002951 minus006407Corg unpr conc 10ndash20 02116 00331 00385Corg unpr Frac 10ndash20 minus009663 minus00509 0364Corg Phys Pro conc 10ndash20 02371 minus005728 minus01183Corg Phys Pro Frac 10ndash20 0123 0163 minus002953Corg Chem Pro conc 10ndash20 02072 003097 minus02335Corg Chem Pro Frac 10ndash20 minus004095 009461 minus04678Corg conc 20ndash30 02326 minus01108 minus005737Corg unpr conc 20ndash30 02002 minus02372 minus005071Corg unpr Rel 20ndash30 cm minus0132 minus03675 minus01379Corg Phys Pro conc 20ndash30 02257 005677 minus001036Corg Phys Pro Frac 20ndash30 009518 03528 01294Corg Chem Pro conc 20ndash30 02323 minus006802 minus01366Corg Chem Pro Frac 20ndash30 004565 04437 minus001278

agroenvironmental conditions of an olive orchard The vari-ability induced by the combined effects of water and tillageerosion in this olive orchard was similar to that described inother agroecosystems For instance Van Oost et al (2005)measured a clear correlation between the erosion and depo-sition rates and the topsoil Corg concentration which rangedbetween 08 of the erosion to 14 of the deposition sitesin the top 25 cm of the soil at two field crop sites under atemperate climate Besides this Bameri et al (2015) mea-sured a higher Corg in the lower part of a field crop site undersemiarid conditions where deposition of the eroded soil fromthe upper zones took place Overall in such landscapes culti-vated for a long time the cumulative effect of tillage and wa-ter erosion on the redistribution of soil across the slope hasbeen observed (Dlugoszlig et al 2012) These processes alsoproduce a vertical redistribution of Corg resulting in a rela-tively homogeneous profile in the tilled layer (top 15ndash20 cm)and a gradual decline below this depth as noted in our study

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

190 J A Goacutemez et al Organic carbon across an olive orchard catena

Figure 9 Soil available phosphorus (Pavail) organic nitrogen (Norg) aggregate stability (Wsagg) and soil degradation index (SDI) by depthcomparing eroded vs deposition areas within the olive orchard The labels of the bars indicate the p value according to a one-way ANOVAcomparing areas for the same soil depth

Figure 10 Scores on principal components 1 and 2 (PC 1 PC 2)for sampling points in the eroded and deposition areas in the oliveorchard

42 Organic carbon stock

The differences in soil organic carbon stock between the ref-erence site and the olive orchard are similar to those de-scribed previously when comparing cropland and forestedareas with the latter presenting a higher concentration ofCorg most of which is in the unprotected fraction whilethe cropland presented a higher fraction of the carbon inthe physically and chemically protected fractions (eg Poe-plau and Don 2013) This is likely due to the fact that un-der soil degradation processes such as water erosion and

low annual organic carbon input as is the case under oliveorchard land use most of the unprotected Corg decomposesrelatively quickly and a great proportion of the remaininglow SOC is protected In addition the mobilization of theunprotected Corg is expected to be reduced in the protectedforested area because of the canopy and the existing veg-etation on the ground that protects the soil against runoffand splash erosion processes In fact the protected Corg con-centration in the topsoil of the olive orchard in the erodedarea accounted for 186plusmn 39 of the upper limit of pro-tected Corg (364plusmn 023 ) according to the model of Has-sink and Whitmore (1997) Therefore the low unprotectedSOC concentration found in the olive orchard is an elementof concern in the increase in SOC stocks This is because pro-tected fractions are fueled from recently derived partially de-composed plant residues together with microbial and micro-meso- and macrofaunal debris (unprotected organic carbon)through processes like SOC aggregation into macro- andormicroaggregates (physically protected SOC) and complexSOC associations with clay and silt particles (chemicallyprotected SOC) which are disrupted in the cropland area incomparison with the reference area The distribution amongsoil Corg fractions in the orchard of this study was similarto the result obtained by Vicente-Vicente et al (2017) whomeasured Corg fraction distribution in olive oil orchards withtemporary cover crops with the exception of the unprotectedSOC which was much higher in soils under cover crops thanthat of our study under bare soil Also interesting is the dif-ficulty obtaining statistically significant differences in SOC

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 191

Tabl

e7

Pear

son

corr

elat

ion

coef

ficie

nts

oflin

ear

corr

elat

ion

betw

een

vari

able

sat

each

dept

hin

terv

al(c

m)

with

the

high

est

load

son

prin

cipa

lco

mpo

nent

s(s

eeTa

ble

6)lowast

Mea

nsst

atis

tical

lysi

gnifi

cant

atαlt

005

Ero

sde

pP a

vail

0ndash10

P ava

il0ndash

40δ

13C

10ndash2

0N

org

10ndash2

0δ

13C

20ndash3

0C

org

conc

C

org

Phys

C

org

conc

C

org

Phys

C

org

Che

m

Cor

gC

hem

10

ndash20

Pro

conc

20

ndash30

Pro

conc

Pr

oco

nc

Pro

Frac

10

ndash20

20ndash3

020

ndash30

20ndash3

0

Ero

sde

p1

P ava

il0ndash

100

746lowast

1P a

vail

0ndash40

095

3lowast0

857lowast

1δ

13C

10ndash2

0minus

082

3lowastminus

076

6lowast1

Nor

g10

ndash20

073

1lowast0

898lowast

083

9lowastminus

096

5lowast1

δ13

C20

ndash30

minus0

792lowast

minus0

893lowast

minus0

891lowast

088

8lowastminus

093

8lowast1

Cor

gco

nc1

0ndash20

075

7lowast0

843lowast

089

9lowastminus

092

9lowast0

920lowast

minus0

911lowast

1C

org

Phys

Pro

con

c10

ndash20

075

5lowast0

908lowast

088

2lowastminus

092

90

967lowast

minus0

985lowast

095

2lowast1

Cor

gco

nc2

0ndash30

078

6lowast0

799lowast

minus0

965lowast

092

7lowastminus

084

6lowast0

9453lowast

089

42lowast

1C

org

Phys

Pro

con

c20

ndash30

083

7lowast0

756lowast

minus0

846lowast

081

7lowastminus

071

4lowast0

8819lowast

079

28lowast

090

1lowast1

Cor

gC

hem

Pro

con

c20

ndash30

074

2lowast0

855lowast

085

9lowastminus

096

2lowast0

980lowast

minus0

896lowast

095

15lowast

094

34lowast

095

0lowast0

853lowast

1C

org

Che

mP

roF

rac

20ndash3

01

stock between the eroding and deposition areas (Fig 7) de-spite the apparent clear differences between the two areas insome other soil properties such as Pavail or δ15N and δ 13Cisotopic signatures in the subsoil (see below)

43 δ15N and δ13C isotopic signal

Changes in vegetation types induced differences in δ13C be-tween the olive orchard and the reference area but as ex-pected no differences in δ13C were detected between the ero-sion and deposition areas in the olive orchard given the sameorigin of vegetation-derived organic matter ie C3 plantsInterestingly within the olive orchard significant differencesin δ15N were detected between the erosion and deposition ar-eas especially for the top 20 cm of soils (Fig 9) This sug-gests the potential of using δ15N as a variable for identifyingdegraded areas in olive growing fields as has been proposedfor other eroding regions in the world (eg Meusburger etal 2013) This might provide an alternative when other con-ventional or isotopic techniques are not available Neverthe-less further studies exploring this potential are necessary inorder to also consider the influence of the NndashP fertilizer mod-ifying the δ15N in relation to the reference area (eg Bate-man and Kelly 2007) The source of N in soil is multifariousand subject to a wide range of transformations that affect theδ15N signature therefore we can only speculate on the rea-sons for this difference in δ15N in a relatively homogeneousarea Bulk soil δ15N tended to be more positive (eg moreenriched in δ15N) as the N cycling rate increases soil micro-bial processes (eg N mineralization nitrification and deni-trification) resulting in products (eg nitrate N2O N2 andNH3) depleted in 15N while the substrate from which theywere formed becomes slightly enriched (Robison 2001) Thehigher δ15N signature of the soil at the deposition locationsuggests that the rates of processes involved in N cycling arehigher than in the erosion area and that it is in accordancewith the higher bulk Corg and Pavail contents and lower SDIof the deposition site The relatively lower soil δ15N signa-ture at the reference site could be partially due to the input oflitter N from the natural legumes and to the closed N cyclingwhich characterize natural forest ecosystems The trend in15N enrichment with soil depth as found at the referencesite is a common observation at forest and grassland sitesand has been related to different mechanisms including 15Nisotope discrimination during microbial N transformationsdifferential preservation of 15N-enriched soil organic mattercomponents during N decomposition and more recently tothe build-up of 15N-enriched microbial necromass (Huygenset al 2008) However there still remains the need for a care-ful calibration to an undisturbed reference site and a betterunderstanding of the influence of different vegetation in thereference and the studied area on the change in the δ15N sig-nal for its further use as an additional tool to determine soildegradation

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

192 J A Goacutemez et al Organic carbon across an olive orchard catena

44 Soil quality of topsoil across the catena variabilitybetween eroded and deposition areas in the orchard

This horizontal distribution due to tillage and water erosionalso simultaneously affected other soil properties and hasbeen described previously in other cultivated areas For in-stance De Gryze et al (2008) described how for a field croparea under conventional tillage in Belgium Pavail almost dou-bled (229 vs 122 mg kgminus1) in the depositional area as com-pared to the eroding upper part They also reported that inhalf of the field under conservation tillage these differencesin Pavail between the upper and lower areas of the field dis-appeared We have observed in our sampled orchard a pro-nounced increase in topsoil Pavail in the deposition area ofaround 400 as compared to the eroding part of the orchardwhich can be attributed to the deposition of enriched sedi-ment coming from the upslope area The cumulative effectsof the differences in Corg and Pavail and the trend towardshigher although nonsignificant Wsagg in the deposition arearesulted in two areas within the orchard with marked differ-ences in soil quality the eroded part which is within therange considered to be degraded in the region (Goacutemez et al2009a) and the depositional area representing 20 of theorchard transect length (Table 1) which is within the non-degraded range according to the same index Topographyand sediment redistribution by erosive processes introducea gradient of spatial variability that questions the conceptof representative area when it comes to describing a wholefield In fact several studies (eg Dell and Sharpley 2006)have suggested that the verification of compliance of envi-ronmental programs such as those related to Corg seques-tration should be based preferentially at least partially onmodeling approaches Our results raise the need for a care-ful delineation of subareas when analyzing soil quality indi-cators andor SOC carbon stock within the same field unitThey also warn about the difficulty of drawing hypothesesfor quantifying the differences between these delineated ar-eas in relation to specific soil properties For instance in ourstudy case erosionndashdeposition processes had a major impacton Pavail and soil quality but the impact on Corg concentra-tion and stock was moderate and extremely difficult to detectstatistically using a moderate number of samples

Our PCA and regression analyses confirmed the relativelyhigh variability of Corg and stock in relation to other soilquality indicators related to erosionndashdeposition processessuch as Pavail as discussed in the previous section While inthe catena studied in the current research the P cycle seemsto be mostly driven by sediment mobilization the Corg and Ncycles seem to be much more complex The moderate differ-ences in Corg and the homogeneity in δ15N and δ13C isotopicsignals between the eroding and deposition areas may be dueto several processes some of which were discussed abovesuch as spatial variability of carbon input due to biomassin the plot surface soil operations in the orchard and fer-tilization We found it both interesting and worth exploring

in future studies that significant correlations between erosionand deposition rates and Corg- δ15N- and δ 13C-related vari-ables were found for samples from the 10ndash20 and 20ndash30 cmlayers indicating that short-term disturbance by surface pro-cesses can mask experimental determination of the impact oferosion deposition processes in this olive orchard for thesevariables

5 Conclusions

1 The results indicate that erosion and deposition withinthe investigated old olive orchard have created a signifi-cant difference in soil properties along the catena whichis translated into different soil Corg Pavail and Norg con-tent δ15N and thus contrasting soil quality status

2 This variability was lower than that of the natural areawhich indicated a severe depletion of SOC as comparedto the natural area and a redistribution of available or-ganic carbon among the different SOC fractions

3 The results suggest that δ15N has the potential for beingused as an indicator of soil degradation although moreinvestigation in different agroecosystems would be re-quired for confirming this statement on a larger scale

4 This research highlights that proper understanding andmanagement of soil quality and Corg content in olive or-chards require consideration of the on-site spatial vari-ability induced by soil erosionndashdeposition processes

Data availability The basic data used in this paper are freelyavailable from the CSIC digital repository at httpsdoiorg1020350digitalCSIC12513 (Goacutemez Calero et al 2020)

Author contributions JAG coordinated and designed the over-all study conducted the field sampling campaigns analysis of thesoil organic carbon results and the overall analysis of the resultsand wrote the manuscript GG participated in the analysis of soil-quality-related properties the overall analysis of the results and thewriting of the manuscript LM participated in the overall analysisof the results and the writing of the manuscript CR and AT partici-pated in the analysis and interpretation of the stable isotope resultsRGR contributed to the design and the statistical analysis of soilorganic carbon (stock and fractions) and nitrogen data the overallanalysis of the results and the writing of the manuscript

Competing interests The authors declare that they have no con-flict of interest

Acknowledgements The authors would like to thankManuel Gonzaacutelez de Molina for providing valuable insightinto the management of the olive orchard in the decades before

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

186 J A Goacutemez et al Organic carbon across an olive orchard catena

Table 2 Results of the two-way ANOVA of soil organic carbon concentration Corg () in different fractions and in bulk soil In (a) arearefers to the reference site vs the olive orchard and in (b) area refers to eroded vs deposition areas in the olive orchard NS stands for notsignificant

(a) Model Bulk soil Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00000 00000 00000 00000 00000Depth (D) 00023 00022 00061 00190 NSAtimesD NS NS NS NS 00300

(b) Model Bulk soil Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00198 00400 00077 00299 NSDepth (D) 00081 00070 00058 00055 00847AtimesD NS NS NS NS NS

Figure 4 Organic carbon concentration in the different soil organic carbon fractions at each depth comparing the eroded vs the depositionarea within the olive orchard The labels of the bars indicate the p value according to a one-way ANOVA comparing treatments for the samesoil depth and carbon fraction between the reference area and olive orchard

4 Discussion

41 Organic carbon concentration and distributionamong fractions

After approximately 175 years of contrasted land use be-tween the undisturbed reference site and the olive orchardbulk soil organic carbon concentration and its fractions havebeen dramatically reduced in the olive orchard Current lev-els of Corg concentration in the soil profile are approximately20 to 25 of that found in the reference area covered bynatural vegetation in the area adjacent to the orchard This ra-

tio is similar albeit in the lower range to the comparison ofCorg in topsoil among olive orchards with different types ofmanagement and natural areas reported for the region (Mil-groom et al 2007) The increased soil disturbance the lowerannual rate of biomass returned to the soil and the highererosion rate in the olive orchard explain this difference Inboth areas the Corg is clearly stratified indicating that de-spite the different mechanisms involved there is a periodicinput of biomass from the olive trees (eg fall of senescenceleaves and tree pruning residues) plus the annual ground veg-etation Vicente-Vicente et al (2017) estimated this biomasscontribution in the range of 148 to 056 t haminus1 yrminus1 It is

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 187

Figure 5 Contribution () of the different fractions with respect to total soil organic carbon by depth comparing the reference site vs theolive orchard The labels of each fraction and depth are the p value according to a one-way ANOVA comparing the reference area vs theolive orchard for the same carbon fraction and soil depth

Figure 6 Fraction of total organic carbon stored in the different fractions by depth comparing the reference site vs the olive orchard Thelabels of each fraction and depth are the p value according to a one-way ANOVA comparing the reference area vs the olive orchard for thesame carbon fraction and soil depth

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

188 J A Goacutemez et al Organic carbon across an olive orchard catena

Table 3 Results of the two-way ANOVA of the distribution of the total soil organic carbon content in the soil among the different fractionsof soil organic carbon Corg In (a) area refers to the reference site vs the olive orchard and in (b) area refers to eroded vs deposition areasin the olive orchard NS stands for not significant

(a) Model Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00000 00000 00000 NSDepth (D) NS NS 00640 NSAtimesD NS 00059 NS NS

(b) Model Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 0091 00881 NS NSDepth (D) 0051 00214 NS 0033AtimesD NS NS NS NS

Figure 7 Cumulative soil organic carbon (SOC) stock across thesoil profile in terms of cumulative soil mass on the vertical axisDifferent letters for similar soil mass refer to statistically significantdifferences (KruskalndashWallis test at p lt 005) For this analysis cu-mulative soil organic carbons were interpolated linearly to the aver-age cumulative soil mass corresponding to all the points in the threeareas

worth noticing that the decrease in Corg as compared to thenatural area is much higher than the reported rates of increasein Corg in olive orchards using conservation agriculture (CA)techniques such as cover crops and incorporation of organicresidues from different sources In a meta-analysis Vicente-Vicente et al (2016) found a response ratio (the ratio of Corgunder CA management as compared to Corg in orchards withbare-soil management) of 11 to 19 suggesting that underCA management which combines cover crops and organicresidues Corg doubled as a maximum

Table 4 Results of the two-way ANOVA of the stable isotope sig-nal In (a) area refers to the reference site vs the olive orchard andin (b) area refers to eroded vs deposition areas in the olive orchardNS stands for not significant

(a) Model δ13C δ15N δ13C δ15N

Area (A) 00001 0002 0002Depth (D) 00026 00175 0029AtimesD NS NS NS

(b) Model δ13C δ15N δ13C δ15N

Area (A) NS 001 001Depth (D) NS NS NSAtimesD NS NS NS

Table 5 Results of the two-way ANOVA of some physical andchemical soil properties comparing eroded vs deposition areas inthe olive orchard NS stands for not significant

Model Norg Pavail Bulk density

Area (A) 00000 001 NSDepth (D) 00009 NS NSAtimesD NS NS NS

Combining all Corg data of the olive orchard the variabil-ity was about 35 which is similar to what has been re-ported so far in the few studies found on soil Corg variabilityin olive orchards For instance Gargouri et al (2013) indi-cated a 24 coefficient of variation (CV) in a 34 ha oliveorchard in Tunisia while Huang et al (2017) reported an av-erage CV of 41 in a 62 ha olive orchard in southern SpainNeither of these two studies reported clear trends in the distri-bution of Corg depending on topography Huang et al (2017)

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 189

Figure 8 13C and 15N isotopic signal of soil by depth distinguish-ing among reference site the whole olive orchard the eroded areaof the olive orchard and the deposition area in this orchard Thelabels of the bars for each depth indicate the p value according toa one-way ANOVA between the reference area (dark blue) vs thewhole olive orchard (dark green) and the eroding area (beige) vsthe deposition area (light green lower label in italics)

pointed out the additional difficulties in the determination ofCorg due to the topography heterogeneity although this wascompounded by the fact that within the orchards there weretwo areas with different planting dates for the trees Goacutemezet al (2012) reported a CV of 49 with higher Corg in ar-eas where there was a change in the slope gradient from thehillslope to a central channel draining into the catchment al-though they could not find a simple relationship between theincrease in content and the topographic indices Despite thefact that a lot of work has been done on the correlation be-tween erosion and deposition and the redistribution of soilCorg (eg Van Oost et al 2005) our study is to our knowl-edge the first attempt to quantify this in detail under the

Table 6 Loads of selected variables in the PCA for the first threeprincipal components (PCs) The values in brackets below PC 1 2and 3 indicate the percentage of variance explained by this PC Vari-ables in bold are those with a load higher than 90 of the variablewith the maximum load for this PC Conc refers to Corg concen-tration for this fraction and Frac means the relative contribution ofthis fraction to the total Corg for this soil depth

Variable at each depth interval (cm) PC 1 PC 2 PC 3(558) (176) (132)

Pavail 0ndash10 02298 008765 minus007278Pavail 0ndash40 02271 0101 minus008455δ13C 0ndash10 minus003385 02861 minus03302δ15N 0ndash10 01941 01677 01836Norg 0ndash10 02147 minus01375 minus00336δ13C δ15N 0ndash10 01594 0249 02211δ13C 10ndash20 minus02329 01461 004296δ15N 10ndash20 01856 007054 03121Norg 10ndash20 02317 minus008699 minus01404δ13C δ15N 10ndash20 01637 008512 03309δ13C 20ndash30 minus02316 006018 009789δ15N 20ndash30 01748 minus02656 01812Norg 20ndash30 02176 009979 minus007458δ13C δ15N 20ndash30 01684 minus02982 01774Corg conc 10ndash20 02421 minus002951 minus006407Corg unpr conc 10ndash20 02116 00331 00385Corg unpr Frac 10ndash20 minus009663 minus00509 0364Corg Phys Pro conc 10ndash20 02371 minus005728 minus01183Corg Phys Pro Frac 10ndash20 0123 0163 minus002953Corg Chem Pro conc 10ndash20 02072 003097 minus02335Corg Chem Pro Frac 10ndash20 minus004095 009461 minus04678Corg conc 20ndash30 02326 minus01108 minus005737Corg unpr conc 20ndash30 02002 minus02372 minus005071Corg unpr Rel 20ndash30 cm minus0132 minus03675 minus01379Corg Phys Pro conc 20ndash30 02257 005677 minus001036Corg Phys Pro Frac 20ndash30 009518 03528 01294Corg Chem Pro conc 20ndash30 02323 minus006802 minus01366Corg Chem Pro Frac 20ndash30 004565 04437 minus001278

agroenvironmental conditions of an olive orchard The vari-ability induced by the combined effects of water and tillageerosion in this olive orchard was similar to that described inother agroecosystems For instance Van Oost et al (2005)measured a clear correlation between the erosion and depo-sition rates and the topsoil Corg concentration which rangedbetween 08 of the erosion to 14 of the deposition sitesin the top 25 cm of the soil at two field crop sites under atemperate climate Besides this Bameri et al (2015) mea-sured a higher Corg in the lower part of a field crop site undersemiarid conditions where deposition of the eroded soil fromthe upper zones took place Overall in such landscapes culti-vated for a long time the cumulative effect of tillage and wa-ter erosion on the redistribution of soil across the slope hasbeen observed (Dlugoszlig et al 2012) These processes alsoproduce a vertical redistribution of Corg resulting in a rela-tively homogeneous profile in the tilled layer (top 15ndash20 cm)and a gradual decline below this depth as noted in our study

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

190 J A Goacutemez et al Organic carbon across an olive orchard catena

Figure 9 Soil available phosphorus (Pavail) organic nitrogen (Norg) aggregate stability (Wsagg) and soil degradation index (SDI) by depthcomparing eroded vs deposition areas within the olive orchard The labels of the bars indicate the p value according to a one-way ANOVAcomparing areas for the same soil depth

Figure 10 Scores on principal components 1 and 2 (PC 1 PC 2)for sampling points in the eroded and deposition areas in the oliveorchard

42 Organic carbon stock

The differences in soil organic carbon stock between the ref-erence site and the olive orchard are similar to those de-scribed previously when comparing cropland and forestedareas with the latter presenting a higher concentration ofCorg most of which is in the unprotected fraction whilethe cropland presented a higher fraction of the carbon inthe physically and chemically protected fractions (eg Poe-plau and Don 2013) This is likely due to the fact that un-der soil degradation processes such as water erosion and

low annual organic carbon input as is the case under oliveorchard land use most of the unprotected Corg decomposesrelatively quickly and a great proportion of the remaininglow SOC is protected In addition the mobilization of theunprotected Corg is expected to be reduced in the protectedforested area because of the canopy and the existing veg-etation on the ground that protects the soil against runoffand splash erosion processes In fact the protected Corg con-centration in the topsoil of the olive orchard in the erodedarea accounted for 186plusmn 39 of the upper limit of pro-tected Corg (364plusmn 023 ) according to the model of Has-sink and Whitmore (1997) Therefore the low unprotectedSOC concentration found in the olive orchard is an elementof concern in the increase in SOC stocks This is because pro-tected fractions are fueled from recently derived partially de-composed plant residues together with microbial and micro-meso- and macrofaunal debris (unprotected organic carbon)through processes like SOC aggregation into macro- andormicroaggregates (physically protected SOC) and complexSOC associations with clay and silt particles (chemicallyprotected SOC) which are disrupted in the cropland area incomparison with the reference area The distribution amongsoil Corg fractions in the orchard of this study was similarto the result obtained by Vicente-Vicente et al (2017) whomeasured Corg fraction distribution in olive oil orchards withtemporary cover crops with the exception of the unprotectedSOC which was much higher in soils under cover crops thanthat of our study under bare soil Also interesting is the dif-ficulty obtaining statistically significant differences in SOC

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 191

Tabl

e7

Pear

son

corr

elat

ion

coef

ficie

nts

oflin

ear

corr

elat

ion

betw

een

vari

able

sat

each

dept

hin

terv

al(c

m)

with

the

high

est

load

son

prin

cipa

lco

mpo

nent

s(s

eeTa

ble

6)lowast

Mea

nsst

atis

tical

lysi

gnifi

cant

atαlt

005

Ero

sde

pP a

vail

0ndash10

P ava

il0ndash

40δ

13C

10ndash2

0N

org

10ndash2

0δ

13C

20ndash3

0C

org

conc

C

org

Phys

C

org

conc

C

org

Phys

C

org

Che

m

Cor

gC

hem

10

ndash20

Pro

conc

20

ndash30

Pro

conc

Pr

oco

nc

Pro

Frac

10

ndash20

20ndash3

020

ndash30

20ndash3

0

Ero

sde

p1

P ava

il0ndash

100

746lowast

1P a

vail

0ndash40

095

3lowast0

857lowast

1δ

13C

10ndash2

0minus

082

3lowastminus

076

6lowast1

Nor

g10

ndash20

073

1lowast0

898lowast

083

9lowastminus

096

5lowast1

δ13

C20

ndash30

minus0

792lowast

minus0

893lowast

minus0

891lowast

088

8lowastminus

093

8lowast1

Cor

gco

nc1

0ndash20

075

7lowast0

843lowast

089

9lowastminus

092

9lowast0

920lowast

minus0

911lowast

1C

org

Phys

Pro

con

c10

ndash20

075

5lowast0

908lowast

088

2lowastminus

092

90

967lowast

minus0

985lowast

095

2lowast1

Cor

gco

nc2

0ndash30

078

6lowast0

799lowast

minus0

965lowast

092

7lowastminus

084

6lowast0

9453lowast

089

42lowast

1C

org

Phys

Pro

con

c20

ndash30

083

7lowast0

756lowast

minus0

846lowast

081

7lowastminus

071

4lowast0

8819lowast

079

28lowast

090

1lowast1

Cor

gC

hem

Pro

con

c20

ndash30

074

2lowast0

855lowast

085

9lowastminus

096

2lowast0

980lowast

minus0

896lowast

095

15lowast

094

34lowast

095

0lowast0

853lowast

1C

org

Che

mP

roF

rac

20ndash3

01

stock between the eroding and deposition areas (Fig 7) de-spite the apparent clear differences between the two areas insome other soil properties such as Pavail or δ15N and δ 13Cisotopic signatures in the subsoil (see below)

43 δ15N and δ13C isotopic signal

Changes in vegetation types induced differences in δ13C be-tween the olive orchard and the reference area but as ex-pected no differences in δ13C were detected between the ero-sion and deposition areas in the olive orchard given the sameorigin of vegetation-derived organic matter ie C3 plantsInterestingly within the olive orchard significant differencesin δ15N were detected between the erosion and deposition ar-eas especially for the top 20 cm of soils (Fig 9) This sug-gests the potential of using δ15N as a variable for identifyingdegraded areas in olive growing fields as has been proposedfor other eroding regions in the world (eg Meusburger etal 2013) This might provide an alternative when other con-ventional or isotopic techniques are not available Neverthe-less further studies exploring this potential are necessary inorder to also consider the influence of the NndashP fertilizer mod-ifying the δ15N in relation to the reference area (eg Bate-man and Kelly 2007) The source of N in soil is multifariousand subject to a wide range of transformations that affect theδ15N signature therefore we can only speculate on the rea-sons for this difference in δ15N in a relatively homogeneousarea Bulk soil δ15N tended to be more positive (eg moreenriched in δ15N) as the N cycling rate increases soil micro-bial processes (eg N mineralization nitrification and deni-trification) resulting in products (eg nitrate N2O N2 andNH3) depleted in 15N while the substrate from which theywere formed becomes slightly enriched (Robison 2001) Thehigher δ15N signature of the soil at the deposition locationsuggests that the rates of processes involved in N cycling arehigher than in the erosion area and that it is in accordancewith the higher bulk Corg and Pavail contents and lower SDIof the deposition site The relatively lower soil δ15N signa-ture at the reference site could be partially due to the input oflitter N from the natural legumes and to the closed N cyclingwhich characterize natural forest ecosystems The trend in15N enrichment with soil depth as found at the referencesite is a common observation at forest and grassland sitesand has been related to different mechanisms including 15Nisotope discrimination during microbial N transformationsdifferential preservation of 15N-enriched soil organic mattercomponents during N decomposition and more recently tothe build-up of 15N-enriched microbial necromass (Huygenset al 2008) However there still remains the need for a care-ful calibration to an undisturbed reference site and a betterunderstanding of the influence of different vegetation in thereference and the studied area on the change in the δ15N sig-nal for its further use as an additional tool to determine soildegradation

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

192 J A Goacutemez et al Organic carbon across an olive orchard catena

44 Soil quality of topsoil across the catena variabilitybetween eroded and deposition areas in the orchard

This horizontal distribution due to tillage and water erosionalso simultaneously affected other soil properties and hasbeen described previously in other cultivated areas For in-stance De Gryze et al (2008) described how for a field croparea under conventional tillage in Belgium Pavail almost dou-bled (229 vs 122 mg kgminus1) in the depositional area as com-pared to the eroding upper part They also reported that inhalf of the field under conservation tillage these differencesin Pavail between the upper and lower areas of the field dis-appeared We have observed in our sampled orchard a pro-nounced increase in topsoil Pavail in the deposition area ofaround 400 as compared to the eroding part of the orchardwhich can be attributed to the deposition of enriched sedi-ment coming from the upslope area The cumulative effectsof the differences in Corg and Pavail and the trend towardshigher although nonsignificant Wsagg in the deposition arearesulted in two areas within the orchard with marked differ-ences in soil quality the eroded part which is within therange considered to be degraded in the region (Goacutemez et al2009a) and the depositional area representing 20 of theorchard transect length (Table 1) which is within the non-degraded range according to the same index Topographyand sediment redistribution by erosive processes introducea gradient of spatial variability that questions the conceptof representative area when it comes to describing a wholefield In fact several studies (eg Dell and Sharpley 2006)have suggested that the verification of compliance of envi-ronmental programs such as those related to Corg seques-tration should be based preferentially at least partially onmodeling approaches Our results raise the need for a care-ful delineation of subareas when analyzing soil quality indi-cators andor SOC carbon stock within the same field unitThey also warn about the difficulty of drawing hypothesesfor quantifying the differences between these delineated ar-eas in relation to specific soil properties For instance in ourstudy case erosionndashdeposition processes had a major impacton Pavail and soil quality but the impact on Corg concentra-tion and stock was moderate and extremely difficult to detectstatistically using a moderate number of samples

Our PCA and regression analyses confirmed the relativelyhigh variability of Corg and stock in relation to other soilquality indicators related to erosionndashdeposition processessuch as Pavail as discussed in the previous section While inthe catena studied in the current research the P cycle seemsto be mostly driven by sediment mobilization the Corg and Ncycles seem to be much more complex The moderate differ-ences in Corg and the homogeneity in δ15N and δ13C isotopicsignals between the eroding and deposition areas may be dueto several processes some of which were discussed abovesuch as spatial variability of carbon input due to biomassin the plot surface soil operations in the orchard and fer-tilization We found it both interesting and worth exploring

in future studies that significant correlations between erosionand deposition rates and Corg- δ15N- and δ 13C-related vari-ables were found for samples from the 10ndash20 and 20ndash30 cmlayers indicating that short-term disturbance by surface pro-cesses can mask experimental determination of the impact oferosion deposition processes in this olive orchard for thesevariables

5 Conclusions

1 The results indicate that erosion and deposition withinthe investigated old olive orchard have created a signifi-cant difference in soil properties along the catena whichis translated into different soil Corg Pavail and Norg con-tent δ15N and thus contrasting soil quality status

2 This variability was lower than that of the natural areawhich indicated a severe depletion of SOC as comparedto the natural area and a redistribution of available or-ganic carbon among the different SOC fractions

3 The results suggest that δ15N has the potential for beingused as an indicator of soil degradation although moreinvestigation in different agroecosystems would be re-quired for confirming this statement on a larger scale

4 This research highlights that proper understanding andmanagement of soil quality and Corg content in olive or-chards require consideration of the on-site spatial vari-ability induced by soil erosionndashdeposition processes

Data availability The basic data used in this paper are freelyavailable from the CSIC digital repository at httpsdoiorg1020350digitalCSIC12513 (Goacutemez Calero et al 2020)

Author contributions JAG coordinated and designed the over-all study conducted the field sampling campaigns analysis of thesoil organic carbon results and the overall analysis of the resultsand wrote the manuscript GG participated in the analysis of soil-quality-related properties the overall analysis of the results and thewriting of the manuscript LM participated in the overall analysisof the results and the writing of the manuscript CR and AT partici-pated in the analysis and interpretation of the stable isotope resultsRGR contributed to the design and the statistical analysis of soilorganic carbon (stock and fractions) and nitrogen data the overallanalysis of the results and the writing of the manuscript

Competing interests The authors declare that they have no con-flict of interest

Acknowledgements The authors would like to thankManuel Gonzaacutelez de Molina for providing valuable insightinto the management of the olive orchard in the decades before

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 187

Figure 5 Contribution () of the different fractions with respect to total soil organic carbon by depth comparing the reference site vs theolive orchard The labels of each fraction and depth are the p value according to a one-way ANOVA comparing the reference area vs theolive orchard for the same carbon fraction and soil depth

Figure 6 Fraction of total organic carbon stored in the different fractions by depth comparing the reference site vs the olive orchard Thelabels of each fraction and depth are the p value according to a one-way ANOVA comparing the reference area vs the olive orchard for thesame carbon fraction and soil depth

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

188 J A Goacutemez et al Organic carbon across an olive orchard catena

Table 3 Results of the two-way ANOVA of the distribution of the total soil organic carbon content in the soil among the different fractionsof soil organic carbon Corg In (a) area refers to the reference site vs the olive orchard and in (b) area refers to eroded vs deposition areasin the olive orchard NS stands for not significant

(a) Model Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00000 00000 00000 NSDepth (D) NS NS 00640 NSAtimesD NS 00059 NS NS

(b) Model Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 0091 00881 NS NSDepth (D) 0051 00214 NS 0033AtimesD NS NS NS NS

Figure 7 Cumulative soil organic carbon (SOC) stock across thesoil profile in terms of cumulative soil mass on the vertical axisDifferent letters for similar soil mass refer to statistically significantdifferences (KruskalndashWallis test at p lt 005) For this analysis cu-mulative soil organic carbons were interpolated linearly to the aver-age cumulative soil mass corresponding to all the points in the threeareas

worth noticing that the decrease in Corg as compared to thenatural area is much higher than the reported rates of increasein Corg in olive orchards using conservation agriculture (CA)techniques such as cover crops and incorporation of organicresidues from different sources In a meta-analysis Vicente-Vicente et al (2016) found a response ratio (the ratio of Corgunder CA management as compared to Corg in orchards withbare-soil management) of 11 to 19 suggesting that underCA management which combines cover crops and organicresidues Corg doubled as a maximum

Table 4 Results of the two-way ANOVA of the stable isotope sig-nal In (a) area refers to the reference site vs the olive orchard andin (b) area refers to eroded vs deposition areas in the olive orchardNS stands for not significant

(a) Model δ13C δ15N δ13C δ15N

Area (A) 00001 0002 0002Depth (D) 00026 00175 0029AtimesD NS NS NS

(b) Model δ13C δ15N δ13C δ15N

Area (A) NS 001 001Depth (D) NS NS NSAtimesD NS NS NS

Table 5 Results of the two-way ANOVA of some physical andchemical soil properties comparing eroded vs deposition areas inthe olive orchard NS stands for not significant

Model Norg Pavail Bulk density

Area (A) 00000 001 NSDepth (D) 00009 NS NSAtimesD NS NS NS

Combining all Corg data of the olive orchard the variabil-ity was about 35 which is similar to what has been re-ported so far in the few studies found on soil Corg variabilityin olive orchards For instance Gargouri et al (2013) indi-cated a 24 coefficient of variation (CV) in a 34 ha oliveorchard in Tunisia while Huang et al (2017) reported an av-erage CV of 41 in a 62 ha olive orchard in southern SpainNeither of these two studies reported clear trends in the distri-bution of Corg depending on topography Huang et al (2017)

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 189

Figure 8 13C and 15N isotopic signal of soil by depth distinguish-ing among reference site the whole olive orchard the eroded areaof the olive orchard and the deposition area in this orchard Thelabels of the bars for each depth indicate the p value according toa one-way ANOVA between the reference area (dark blue) vs thewhole olive orchard (dark green) and the eroding area (beige) vsthe deposition area (light green lower label in italics)

pointed out the additional difficulties in the determination ofCorg due to the topography heterogeneity although this wascompounded by the fact that within the orchards there weretwo areas with different planting dates for the trees Goacutemezet al (2012) reported a CV of 49 with higher Corg in ar-eas where there was a change in the slope gradient from thehillslope to a central channel draining into the catchment al-though they could not find a simple relationship between theincrease in content and the topographic indices Despite thefact that a lot of work has been done on the correlation be-tween erosion and deposition and the redistribution of soilCorg (eg Van Oost et al 2005) our study is to our knowl-edge the first attempt to quantify this in detail under the

Table 6 Loads of selected variables in the PCA for the first threeprincipal components (PCs) The values in brackets below PC 1 2and 3 indicate the percentage of variance explained by this PC Vari-ables in bold are those with a load higher than 90 of the variablewith the maximum load for this PC Conc refers to Corg concen-tration for this fraction and Frac means the relative contribution ofthis fraction to the total Corg for this soil depth

Variable at each depth interval (cm) PC 1 PC 2 PC 3(558) (176) (132)

Pavail 0ndash10 02298 008765 minus007278Pavail 0ndash40 02271 0101 minus008455δ13C 0ndash10 minus003385 02861 minus03302δ15N 0ndash10 01941 01677 01836Norg 0ndash10 02147 minus01375 minus00336δ13C δ15N 0ndash10 01594 0249 02211δ13C 10ndash20 minus02329 01461 004296δ15N 10ndash20 01856 007054 03121Norg 10ndash20 02317 minus008699 minus01404δ13C δ15N 10ndash20 01637 008512 03309δ13C 20ndash30 minus02316 006018 009789δ15N 20ndash30 01748 minus02656 01812Norg 20ndash30 02176 009979 minus007458δ13C δ15N 20ndash30 01684 minus02982 01774Corg conc 10ndash20 02421 minus002951 minus006407Corg unpr conc 10ndash20 02116 00331 00385Corg unpr Frac 10ndash20 minus009663 minus00509 0364Corg Phys Pro conc 10ndash20 02371 minus005728 minus01183Corg Phys Pro Frac 10ndash20 0123 0163 minus002953Corg Chem Pro conc 10ndash20 02072 003097 minus02335Corg Chem Pro Frac 10ndash20 minus004095 009461 minus04678Corg conc 20ndash30 02326 minus01108 minus005737Corg unpr conc 20ndash30 02002 minus02372 minus005071Corg unpr Rel 20ndash30 cm minus0132 minus03675 minus01379Corg Phys Pro conc 20ndash30 02257 005677 minus001036Corg Phys Pro Frac 20ndash30 009518 03528 01294Corg Chem Pro conc 20ndash30 02323 minus006802 minus01366Corg Chem Pro Frac 20ndash30 004565 04437 minus001278

agroenvironmental conditions of an olive orchard The vari-ability induced by the combined effects of water and tillageerosion in this olive orchard was similar to that described inother agroecosystems For instance Van Oost et al (2005)measured a clear correlation between the erosion and depo-sition rates and the topsoil Corg concentration which rangedbetween 08 of the erosion to 14 of the deposition sitesin the top 25 cm of the soil at two field crop sites under atemperate climate Besides this Bameri et al (2015) mea-sured a higher Corg in the lower part of a field crop site undersemiarid conditions where deposition of the eroded soil fromthe upper zones took place Overall in such landscapes culti-vated for a long time the cumulative effect of tillage and wa-ter erosion on the redistribution of soil across the slope hasbeen observed (Dlugoszlig et al 2012) These processes alsoproduce a vertical redistribution of Corg resulting in a rela-tively homogeneous profile in the tilled layer (top 15ndash20 cm)and a gradual decline below this depth as noted in our study

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

190 J A Goacutemez et al Organic carbon across an olive orchard catena

Figure 9 Soil available phosphorus (Pavail) organic nitrogen (Norg) aggregate stability (Wsagg) and soil degradation index (SDI) by depthcomparing eroded vs deposition areas within the olive orchard The labels of the bars indicate the p value according to a one-way ANOVAcomparing areas for the same soil depth

Figure 10 Scores on principal components 1 and 2 (PC 1 PC 2)for sampling points in the eroded and deposition areas in the oliveorchard

42 Organic carbon stock

The differences in soil organic carbon stock between the ref-erence site and the olive orchard are similar to those de-scribed previously when comparing cropland and forestedareas with the latter presenting a higher concentration ofCorg most of which is in the unprotected fraction whilethe cropland presented a higher fraction of the carbon inthe physically and chemically protected fractions (eg Poe-plau and Don 2013) This is likely due to the fact that un-der soil degradation processes such as water erosion and

low annual organic carbon input as is the case under oliveorchard land use most of the unprotected Corg decomposesrelatively quickly and a great proportion of the remaininglow SOC is protected In addition the mobilization of theunprotected Corg is expected to be reduced in the protectedforested area because of the canopy and the existing veg-etation on the ground that protects the soil against runoffand splash erosion processes In fact the protected Corg con-centration in the topsoil of the olive orchard in the erodedarea accounted for 186plusmn 39 of the upper limit of pro-tected Corg (364plusmn 023 ) according to the model of Has-sink and Whitmore (1997) Therefore the low unprotectedSOC concentration found in the olive orchard is an elementof concern in the increase in SOC stocks This is because pro-tected fractions are fueled from recently derived partially de-composed plant residues together with microbial and micro-meso- and macrofaunal debris (unprotected organic carbon)through processes like SOC aggregation into macro- andormicroaggregates (physically protected SOC) and complexSOC associations with clay and silt particles (chemicallyprotected SOC) which are disrupted in the cropland area incomparison with the reference area The distribution amongsoil Corg fractions in the orchard of this study was similarto the result obtained by Vicente-Vicente et al (2017) whomeasured Corg fraction distribution in olive oil orchards withtemporary cover crops with the exception of the unprotectedSOC which was much higher in soils under cover crops thanthat of our study under bare soil Also interesting is the dif-ficulty obtaining statistically significant differences in SOC

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 191

Tabl

e7

Pear

son

corr

elat

ion

coef

ficie

nts

oflin

ear

corr

elat

ion

betw

een

vari

able

sat

each

dept

hin

terv

al(c

m)

with

the

high

est

load

son

prin

cipa

lco

mpo

nent

s(s

eeTa

ble

6)lowast

Mea

nsst

atis

tical

lysi

gnifi

cant

atαlt

005

Ero

sde

pP a

vail

0ndash10

P ava

il0ndash

40δ

13C

10ndash2

0N

org

10ndash2

0δ

13C

20ndash3

0C

org

conc

C

org

Phys

C

org

conc

C

org

Phys

C

org

Che

m

Cor

gC

hem

10

ndash20

Pro

conc

20

ndash30

Pro

conc

Pr

oco

nc

Pro

Frac

10

ndash20

20ndash3

020

ndash30

20ndash3

0

Ero

sde

p1

P ava

il0ndash

100

746lowast

1P a

vail

0ndash40

095

3lowast0

857lowast

1δ

13C

10ndash2

0minus

082

3lowastminus

076

6lowast1

Nor

g10

ndash20

073

1lowast0

898lowast

083

9lowastminus

096

5lowast1

δ13

C20

ndash30

minus0

792lowast

minus0

893lowast

minus0

891lowast

088

8lowastminus

093

8lowast1

Cor

gco

nc1

0ndash20

075

7lowast0

843lowast

089

9lowastminus

092

9lowast0

920lowast

minus0

911lowast

1C

org

Phys

Pro

con

c10

ndash20

075

5lowast0

908lowast

088

2lowastminus

092

90

967lowast

minus0

985lowast

095

2lowast1

Cor

gco

nc2

0ndash30

078

6lowast0

799lowast

minus0

965lowast

092

7lowastminus

084

6lowast0

9453lowast

089

42lowast

1C

org

Phys

Pro

con

c20

ndash30

083

7lowast0

756lowast

minus0

846lowast

081

7lowastminus

071

4lowast0

8819lowast

079

28lowast

090

1lowast1

Cor

gC

hem

Pro

con

c20

ndash30

074

2lowast0

855lowast

085

9lowastminus

096

2lowast0

980lowast

minus0

896lowast

095

15lowast

094

34lowast

095

0lowast0

853lowast

1C

org

Che

mP

roF

rac

20ndash3

01

stock between the eroding and deposition areas (Fig 7) de-spite the apparent clear differences between the two areas insome other soil properties such as Pavail or δ15N and δ 13Cisotopic signatures in the subsoil (see below)

43 δ15N and δ13C isotopic signal

Changes in vegetation types induced differences in δ13C be-tween the olive orchard and the reference area but as ex-pected no differences in δ13C were detected between the ero-sion and deposition areas in the olive orchard given the sameorigin of vegetation-derived organic matter ie C3 plantsInterestingly within the olive orchard significant differencesin δ15N were detected between the erosion and deposition ar-eas especially for the top 20 cm of soils (Fig 9) This sug-gests the potential of using δ15N as a variable for identifyingdegraded areas in olive growing fields as has been proposedfor other eroding regions in the world (eg Meusburger etal 2013) This might provide an alternative when other con-ventional or isotopic techniques are not available Neverthe-less further studies exploring this potential are necessary inorder to also consider the influence of the NndashP fertilizer mod-ifying the δ15N in relation to the reference area (eg Bate-man and Kelly 2007) The source of N in soil is multifariousand subject to a wide range of transformations that affect theδ15N signature therefore we can only speculate on the rea-sons for this difference in δ15N in a relatively homogeneousarea Bulk soil δ15N tended to be more positive (eg moreenriched in δ15N) as the N cycling rate increases soil micro-bial processes (eg N mineralization nitrification and deni-trification) resulting in products (eg nitrate N2O N2 andNH3) depleted in 15N while the substrate from which theywere formed becomes slightly enriched (Robison 2001) Thehigher δ15N signature of the soil at the deposition locationsuggests that the rates of processes involved in N cycling arehigher than in the erosion area and that it is in accordancewith the higher bulk Corg and Pavail contents and lower SDIof the deposition site The relatively lower soil δ15N signa-ture at the reference site could be partially due to the input oflitter N from the natural legumes and to the closed N cyclingwhich characterize natural forest ecosystems The trend in15N enrichment with soil depth as found at the referencesite is a common observation at forest and grassland sitesand has been related to different mechanisms including 15Nisotope discrimination during microbial N transformationsdifferential preservation of 15N-enriched soil organic mattercomponents during N decomposition and more recently tothe build-up of 15N-enriched microbial necromass (Huygenset al 2008) However there still remains the need for a care-ful calibration to an undisturbed reference site and a betterunderstanding of the influence of different vegetation in thereference and the studied area on the change in the δ15N sig-nal for its further use as an additional tool to determine soildegradation

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

192 J A Goacutemez et al Organic carbon across an olive orchard catena

44 Soil quality of topsoil across the catena variabilitybetween eroded and deposition areas in the orchard

This horizontal distribution due to tillage and water erosionalso simultaneously affected other soil properties and hasbeen described previously in other cultivated areas For in-stance De Gryze et al (2008) described how for a field croparea under conventional tillage in Belgium Pavail almost dou-bled (229 vs 122 mg kgminus1) in the depositional area as com-pared to the eroding upper part They also reported that inhalf of the field under conservation tillage these differencesin Pavail between the upper and lower areas of the field dis-appeared We have observed in our sampled orchard a pro-nounced increase in topsoil Pavail in the deposition area ofaround 400 as compared to the eroding part of the orchardwhich can be attributed to the deposition of enriched sedi-ment coming from the upslope area The cumulative effectsof the differences in Corg and Pavail and the trend towardshigher although nonsignificant Wsagg in the deposition arearesulted in two areas within the orchard with marked differ-ences in soil quality the eroded part which is within therange considered to be degraded in the region (Goacutemez et al2009a) and the depositional area representing 20 of theorchard transect length (Table 1) which is within the non-degraded range according to the same index Topographyand sediment redistribution by erosive processes introducea gradient of spatial variability that questions the conceptof representative area when it comes to describing a wholefield In fact several studies (eg Dell and Sharpley 2006)have suggested that the verification of compliance of envi-ronmental programs such as those related to Corg seques-tration should be based preferentially at least partially onmodeling approaches Our results raise the need for a care-ful delineation of subareas when analyzing soil quality indi-cators andor SOC carbon stock within the same field unitThey also warn about the difficulty of drawing hypothesesfor quantifying the differences between these delineated ar-eas in relation to specific soil properties For instance in ourstudy case erosionndashdeposition processes had a major impacton Pavail and soil quality but the impact on Corg concentra-tion and stock was moderate and extremely difficult to detectstatistically using a moderate number of samples

Our PCA and regression analyses confirmed the relativelyhigh variability of Corg and stock in relation to other soilquality indicators related to erosionndashdeposition processessuch as Pavail as discussed in the previous section While inthe catena studied in the current research the P cycle seemsto be mostly driven by sediment mobilization the Corg and Ncycles seem to be much more complex The moderate differ-ences in Corg and the homogeneity in δ15N and δ13C isotopicsignals between the eroding and deposition areas may be dueto several processes some of which were discussed abovesuch as spatial variability of carbon input due to biomassin the plot surface soil operations in the orchard and fer-tilization We found it both interesting and worth exploring

in future studies that significant correlations between erosionand deposition rates and Corg- δ15N- and δ 13C-related vari-ables were found for samples from the 10ndash20 and 20ndash30 cmlayers indicating that short-term disturbance by surface pro-cesses can mask experimental determination of the impact oferosion deposition processes in this olive orchard for thesevariables

5 Conclusions

1 The results indicate that erosion and deposition withinthe investigated old olive orchard have created a signifi-cant difference in soil properties along the catena whichis translated into different soil Corg Pavail and Norg con-tent δ15N and thus contrasting soil quality status

2 This variability was lower than that of the natural areawhich indicated a severe depletion of SOC as comparedto the natural area and a redistribution of available or-ganic carbon among the different SOC fractions

3 The results suggest that δ15N has the potential for beingused as an indicator of soil degradation although moreinvestigation in different agroecosystems would be re-quired for confirming this statement on a larger scale

4 This research highlights that proper understanding andmanagement of soil quality and Corg content in olive or-chards require consideration of the on-site spatial vari-ability induced by soil erosionndashdeposition processes

Data availability The basic data used in this paper are freelyavailable from the CSIC digital repository at httpsdoiorg1020350digitalCSIC12513 (Goacutemez Calero et al 2020)

Author contributions JAG coordinated and designed the over-all study conducted the field sampling campaigns analysis of thesoil organic carbon results and the overall analysis of the resultsand wrote the manuscript GG participated in the analysis of soil-quality-related properties the overall analysis of the results and thewriting of the manuscript LM participated in the overall analysisof the results and the writing of the manuscript CR and AT partici-pated in the analysis and interpretation of the stable isotope resultsRGR contributed to the design and the statistical analysis of soilorganic carbon (stock and fractions) and nitrogen data the overallanalysis of the results and the writing of the manuscript

Competing interests The authors declare that they have no con-flict of interest

Acknowledgements The authors would like to thankManuel Gonzaacutelez de Molina for providing valuable insightinto the management of the olive orchard in the decades before

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

188 J A Goacutemez et al Organic carbon across an olive orchard catena

Table 3 Results of the two-way ANOVA of the distribution of the total soil organic carbon content in the soil among the different fractionsof soil organic carbon Corg In (a) area refers to the reference site vs the olive orchard and in (b) area refers to eroded vs deposition areasin the olive orchard NS stands for not significant

(a) Model Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 00000 00000 00000 NSDepth (D) NS NS 00640 NSAtimesD NS 00059 NS NS

(b) Model Corg fraction

Not protected Physically Chemically Biochemicallyprotected protected protected

Area (A) 0091 00881 NS NSDepth (D) 0051 00214 NS 0033AtimesD NS NS NS NS

Figure 7 Cumulative soil organic carbon (SOC) stock across thesoil profile in terms of cumulative soil mass on the vertical axisDifferent letters for similar soil mass refer to statistically significantdifferences (KruskalndashWallis test at p lt 005) For this analysis cu-mulative soil organic carbons were interpolated linearly to the aver-age cumulative soil mass corresponding to all the points in the threeareas

worth noticing that the decrease in Corg as compared to thenatural area is much higher than the reported rates of increasein Corg in olive orchards using conservation agriculture (CA)techniques such as cover crops and incorporation of organicresidues from different sources In a meta-analysis Vicente-Vicente et al (2016) found a response ratio (the ratio of Corgunder CA management as compared to Corg in orchards withbare-soil management) of 11 to 19 suggesting that underCA management which combines cover crops and organicresidues Corg doubled as a maximum

Table 4 Results of the two-way ANOVA of the stable isotope sig-nal In (a) area refers to the reference site vs the olive orchard andin (b) area refers to eroded vs deposition areas in the olive orchardNS stands for not significant

(a) Model δ13C δ15N δ13C δ15N

Area (A) 00001 0002 0002Depth (D) 00026 00175 0029AtimesD NS NS NS

(b) Model δ13C δ15N δ13C δ15N

Area (A) NS 001 001Depth (D) NS NS NSAtimesD NS NS NS

Table 5 Results of the two-way ANOVA of some physical andchemical soil properties comparing eroded vs deposition areas inthe olive orchard NS stands for not significant

Model Norg Pavail Bulk density

Area (A) 00000 001 NSDepth (D) 00009 NS NSAtimesD NS NS NS

Combining all Corg data of the olive orchard the variabil-ity was about 35 which is similar to what has been re-ported so far in the few studies found on soil Corg variabilityin olive orchards For instance Gargouri et al (2013) indi-cated a 24 coefficient of variation (CV) in a 34 ha oliveorchard in Tunisia while Huang et al (2017) reported an av-erage CV of 41 in a 62 ha olive orchard in southern SpainNeither of these two studies reported clear trends in the distri-bution of Corg depending on topography Huang et al (2017)

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 189

Figure 8 13C and 15N isotopic signal of soil by depth distinguish-ing among reference site the whole olive orchard the eroded areaof the olive orchard and the deposition area in this orchard Thelabels of the bars for each depth indicate the p value according toa one-way ANOVA between the reference area (dark blue) vs thewhole olive orchard (dark green) and the eroding area (beige) vsthe deposition area (light green lower label in italics)

pointed out the additional difficulties in the determination ofCorg due to the topography heterogeneity although this wascompounded by the fact that within the orchards there weretwo areas with different planting dates for the trees Goacutemezet al (2012) reported a CV of 49 with higher Corg in ar-eas where there was a change in the slope gradient from thehillslope to a central channel draining into the catchment al-though they could not find a simple relationship between theincrease in content and the topographic indices Despite thefact that a lot of work has been done on the correlation be-tween erosion and deposition and the redistribution of soilCorg (eg Van Oost et al 2005) our study is to our knowl-edge the first attempt to quantify this in detail under the

Table 6 Loads of selected variables in the PCA for the first threeprincipal components (PCs) The values in brackets below PC 1 2and 3 indicate the percentage of variance explained by this PC Vari-ables in bold are those with a load higher than 90 of the variablewith the maximum load for this PC Conc refers to Corg concen-tration for this fraction and Frac means the relative contribution ofthis fraction to the total Corg for this soil depth

Variable at each depth interval (cm) PC 1 PC 2 PC 3(558) (176) (132)

Pavail 0ndash10 02298 008765 minus007278Pavail 0ndash40 02271 0101 minus008455δ13C 0ndash10 minus003385 02861 minus03302δ15N 0ndash10 01941 01677 01836Norg 0ndash10 02147 minus01375 minus00336δ13C δ15N 0ndash10 01594 0249 02211δ13C 10ndash20 minus02329 01461 004296δ15N 10ndash20 01856 007054 03121Norg 10ndash20 02317 minus008699 minus01404δ13C δ15N 10ndash20 01637 008512 03309δ13C 20ndash30 minus02316 006018 009789δ15N 20ndash30 01748 minus02656 01812Norg 20ndash30 02176 009979 minus007458δ13C δ15N 20ndash30 01684 minus02982 01774Corg conc 10ndash20 02421 minus002951 minus006407Corg unpr conc 10ndash20 02116 00331 00385Corg unpr Frac 10ndash20 minus009663 minus00509 0364Corg Phys Pro conc 10ndash20 02371 minus005728 minus01183Corg Phys Pro Frac 10ndash20 0123 0163 minus002953Corg Chem Pro conc 10ndash20 02072 003097 minus02335Corg Chem Pro Frac 10ndash20 minus004095 009461 minus04678Corg conc 20ndash30 02326 minus01108 minus005737Corg unpr conc 20ndash30 02002 minus02372 minus005071Corg unpr Rel 20ndash30 cm minus0132 minus03675 minus01379Corg Phys Pro conc 20ndash30 02257 005677 minus001036Corg Phys Pro Frac 20ndash30 009518 03528 01294Corg Chem Pro conc 20ndash30 02323 minus006802 minus01366Corg Chem Pro Frac 20ndash30 004565 04437 minus001278

agroenvironmental conditions of an olive orchard The vari-ability induced by the combined effects of water and tillageerosion in this olive orchard was similar to that described inother agroecosystems For instance Van Oost et al (2005)measured a clear correlation between the erosion and depo-sition rates and the topsoil Corg concentration which rangedbetween 08 of the erosion to 14 of the deposition sitesin the top 25 cm of the soil at two field crop sites under atemperate climate Besides this Bameri et al (2015) mea-sured a higher Corg in the lower part of a field crop site undersemiarid conditions where deposition of the eroded soil fromthe upper zones took place Overall in such landscapes culti-vated for a long time the cumulative effect of tillage and wa-ter erosion on the redistribution of soil across the slope hasbeen observed (Dlugoszlig et al 2012) These processes alsoproduce a vertical redistribution of Corg resulting in a rela-tively homogeneous profile in the tilled layer (top 15ndash20 cm)and a gradual decline below this depth as noted in our study

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

190 J A Goacutemez et al Organic carbon across an olive orchard catena

Figure 9 Soil available phosphorus (Pavail) organic nitrogen (Norg) aggregate stability (Wsagg) and soil degradation index (SDI) by depthcomparing eroded vs deposition areas within the olive orchard The labels of the bars indicate the p value according to a one-way ANOVAcomparing areas for the same soil depth

Figure 10 Scores on principal components 1 and 2 (PC 1 PC 2)for sampling points in the eroded and deposition areas in the oliveorchard

42 Organic carbon stock

The differences in soil organic carbon stock between the ref-erence site and the olive orchard are similar to those de-scribed previously when comparing cropland and forestedareas with the latter presenting a higher concentration ofCorg most of which is in the unprotected fraction whilethe cropland presented a higher fraction of the carbon inthe physically and chemically protected fractions (eg Poe-plau and Don 2013) This is likely due to the fact that un-der soil degradation processes such as water erosion and

low annual organic carbon input as is the case under oliveorchard land use most of the unprotected Corg decomposesrelatively quickly and a great proportion of the remaininglow SOC is protected In addition the mobilization of theunprotected Corg is expected to be reduced in the protectedforested area because of the canopy and the existing veg-etation on the ground that protects the soil against runoffand splash erosion processes In fact the protected Corg con-centration in the topsoil of the olive orchard in the erodedarea accounted for 186plusmn 39 of the upper limit of pro-tected Corg (364plusmn 023 ) according to the model of Has-sink and Whitmore (1997) Therefore the low unprotectedSOC concentration found in the olive orchard is an elementof concern in the increase in SOC stocks This is because pro-tected fractions are fueled from recently derived partially de-composed plant residues together with microbial and micro-meso- and macrofaunal debris (unprotected organic carbon)through processes like SOC aggregation into macro- andormicroaggregates (physically protected SOC) and complexSOC associations with clay and silt particles (chemicallyprotected SOC) which are disrupted in the cropland area incomparison with the reference area The distribution amongsoil Corg fractions in the orchard of this study was similarto the result obtained by Vicente-Vicente et al (2017) whomeasured Corg fraction distribution in olive oil orchards withtemporary cover crops with the exception of the unprotectedSOC which was much higher in soils under cover crops thanthat of our study under bare soil Also interesting is the dif-ficulty obtaining statistically significant differences in SOC

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 191

Tabl

e7

Pear

son

corr

elat

ion

coef

ficie

nts

oflin

ear

corr

elat

ion

betw

een

vari

able

sat

each

dept

hin

terv

al(c

m)

with

the

high

est

load

son

prin

cipa

lco

mpo

nent

s(s

eeTa

ble

6)lowast

Mea

nsst

atis

tical

lysi

gnifi

cant

atαlt

005

Ero

sde

pP a

vail

0ndash10

P ava

il0ndash

40δ

13C

10ndash2

0N

org

10ndash2

0δ

13C

20ndash3

0C

org

conc

C

org

Phys

C

org

conc

C

org

Phys

C

org

Che

m

Cor

gC

hem

10

ndash20

Pro

conc

20

ndash30

Pro

conc

Pr

oco

nc

Pro

Frac

10

ndash20

20ndash3

020

ndash30

20ndash3

0

Ero

sde

p1

P ava

il0ndash

100

746lowast

1P a

vail

0ndash40

095

3lowast0

857lowast

1δ

13C

10ndash2

0minus

082

3lowastminus

076

6lowast1

Nor

g10

ndash20

073

1lowast0

898lowast

083

9lowastminus

096

5lowast1

δ13

C20

ndash30

minus0

792lowast

minus0

893lowast

minus0

891lowast

088

8lowastminus

093

8lowast1

Cor

gco

nc1

0ndash20

075

7lowast0

843lowast

089

9lowastminus

092

9lowast0

920lowast

minus0

911lowast

1C

org

Phys

Pro

con

c10

ndash20

075

5lowast0

908lowast

088

2lowastminus

092

90

967lowast

minus0

985lowast

095

2lowast1

Cor

gco

nc2

0ndash30

078

6lowast0

799lowast

minus0

965lowast

092

7lowastminus

084

6lowast0

9453lowast

089

42lowast

1C

org

Phys

Pro

con

c20

ndash30

083

7lowast0

756lowast

minus0

846lowast

081

7lowastminus

071

4lowast0

8819lowast

079

28lowast

090

1lowast1

Cor

gC

hem

Pro

con

c20

ndash30

074

2lowast0

855lowast

085

9lowastminus

096

2lowast0

980lowast

minus0

896lowast

095

15lowast

094

34lowast

095

0lowast0

853lowast

1C

org

Che

mP

roF

rac

20ndash3

01

stock between the eroding and deposition areas (Fig 7) de-spite the apparent clear differences between the two areas insome other soil properties such as Pavail or δ15N and δ 13Cisotopic signatures in the subsoil (see below)

43 δ15N and δ13C isotopic signal

Changes in vegetation types induced differences in δ13C be-tween the olive orchard and the reference area but as ex-pected no differences in δ13C were detected between the ero-sion and deposition areas in the olive orchard given the sameorigin of vegetation-derived organic matter ie C3 plantsInterestingly within the olive orchard significant differencesin δ15N were detected between the erosion and deposition ar-eas especially for the top 20 cm of soils (Fig 9) This sug-gests the potential of using δ15N as a variable for identifyingdegraded areas in olive growing fields as has been proposedfor other eroding regions in the world (eg Meusburger etal 2013) This might provide an alternative when other con-ventional or isotopic techniques are not available Neverthe-less further studies exploring this potential are necessary inorder to also consider the influence of the NndashP fertilizer mod-ifying the δ15N in relation to the reference area (eg Bate-man and Kelly 2007) The source of N in soil is multifariousand subject to a wide range of transformations that affect theδ15N signature therefore we can only speculate on the rea-sons for this difference in δ15N in a relatively homogeneousarea Bulk soil δ15N tended to be more positive (eg moreenriched in δ15N) as the N cycling rate increases soil micro-bial processes (eg N mineralization nitrification and deni-trification) resulting in products (eg nitrate N2O N2 andNH3) depleted in 15N while the substrate from which theywere formed becomes slightly enriched (Robison 2001) Thehigher δ15N signature of the soil at the deposition locationsuggests that the rates of processes involved in N cycling arehigher than in the erosion area and that it is in accordancewith the higher bulk Corg and Pavail contents and lower SDIof the deposition site The relatively lower soil δ15N signa-ture at the reference site could be partially due to the input oflitter N from the natural legumes and to the closed N cyclingwhich characterize natural forest ecosystems The trend in15N enrichment with soil depth as found at the referencesite is a common observation at forest and grassland sitesand has been related to different mechanisms including 15Nisotope discrimination during microbial N transformationsdifferential preservation of 15N-enriched soil organic mattercomponents during N decomposition and more recently tothe build-up of 15N-enriched microbial necromass (Huygenset al 2008) However there still remains the need for a care-ful calibration to an undisturbed reference site and a betterunderstanding of the influence of different vegetation in thereference and the studied area on the change in the δ15N sig-nal for its further use as an additional tool to determine soildegradation

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

192 J A Goacutemez et al Organic carbon across an olive orchard catena

44 Soil quality of topsoil across the catena variabilitybetween eroded and deposition areas in the orchard

This horizontal distribution due to tillage and water erosionalso simultaneously affected other soil properties and hasbeen described previously in other cultivated areas For in-stance De Gryze et al (2008) described how for a field croparea under conventional tillage in Belgium Pavail almost dou-bled (229 vs 122 mg kgminus1) in the depositional area as com-pared to the eroding upper part They also reported that inhalf of the field under conservation tillage these differencesin Pavail between the upper and lower areas of the field dis-appeared We have observed in our sampled orchard a pro-nounced increase in topsoil Pavail in the deposition area ofaround 400 as compared to the eroding part of the orchardwhich can be attributed to the deposition of enriched sedi-ment coming from the upslope area The cumulative effectsof the differences in Corg and Pavail and the trend towardshigher although nonsignificant Wsagg in the deposition arearesulted in two areas within the orchard with marked differ-ences in soil quality the eroded part which is within therange considered to be degraded in the region (Goacutemez et al2009a) and the depositional area representing 20 of theorchard transect length (Table 1) which is within the non-degraded range according to the same index Topographyand sediment redistribution by erosive processes introducea gradient of spatial variability that questions the conceptof representative area when it comes to describing a wholefield In fact several studies (eg Dell and Sharpley 2006)have suggested that the verification of compliance of envi-ronmental programs such as those related to Corg seques-tration should be based preferentially at least partially onmodeling approaches Our results raise the need for a care-ful delineation of subareas when analyzing soil quality indi-cators andor SOC carbon stock within the same field unitThey also warn about the difficulty of drawing hypothesesfor quantifying the differences between these delineated ar-eas in relation to specific soil properties For instance in ourstudy case erosionndashdeposition processes had a major impacton Pavail and soil quality but the impact on Corg concentra-tion and stock was moderate and extremely difficult to detectstatistically using a moderate number of samples

Our PCA and regression analyses confirmed the relativelyhigh variability of Corg and stock in relation to other soilquality indicators related to erosionndashdeposition processessuch as Pavail as discussed in the previous section While inthe catena studied in the current research the P cycle seemsto be mostly driven by sediment mobilization the Corg and Ncycles seem to be much more complex The moderate differ-ences in Corg and the homogeneity in δ15N and δ13C isotopicsignals between the eroding and deposition areas may be dueto several processes some of which were discussed abovesuch as spatial variability of carbon input due to biomassin the plot surface soil operations in the orchard and fer-tilization We found it both interesting and worth exploring

in future studies that significant correlations between erosionand deposition rates and Corg- δ15N- and δ 13C-related vari-ables were found for samples from the 10ndash20 and 20ndash30 cmlayers indicating that short-term disturbance by surface pro-cesses can mask experimental determination of the impact oferosion deposition processes in this olive orchard for thesevariables

5 Conclusions

1 The results indicate that erosion and deposition withinthe investigated old olive orchard have created a signifi-cant difference in soil properties along the catena whichis translated into different soil Corg Pavail and Norg con-tent δ15N and thus contrasting soil quality status

2 This variability was lower than that of the natural areawhich indicated a severe depletion of SOC as comparedto the natural area and a redistribution of available or-ganic carbon among the different SOC fractions

3 The results suggest that δ15N has the potential for beingused as an indicator of soil degradation although moreinvestigation in different agroecosystems would be re-quired for confirming this statement on a larger scale

4 This research highlights that proper understanding andmanagement of soil quality and Corg content in olive or-chards require consideration of the on-site spatial vari-ability induced by soil erosionndashdeposition processes

Data availability The basic data used in this paper are freelyavailable from the CSIC digital repository at httpsdoiorg1020350digitalCSIC12513 (Goacutemez Calero et al 2020)

Author contributions JAG coordinated and designed the over-all study conducted the field sampling campaigns analysis of thesoil organic carbon results and the overall analysis of the resultsand wrote the manuscript GG participated in the analysis of soil-quality-related properties the overall analysis of the results and thewriting of the manuscript LM participated in the overall analysisof the results and the writing of the manuscript CR and AT partici-pated in the analysis and interpretation of the stable isotope resultsRGR contributed to the design and the statistical analysis of soilorganic carbon (stock and fractions) and nitrogen data the overallanalysis of the results and the writing of the manuscript

Competing interests The authors declare that they have no con-flict of interest

Acknowledgements The authors would like to thankManuel Gonzaacutelez de Molina for providing valuable insightinto the management of the olive orchard in the decades before

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 189

Figure 8 13C and 15N isotopic signal of soil by depth distinguish-ing among reference site the whole olive orchard the eroded areaof the olive orchard and the deposition area in this orchard Thelabels of the bars for each depth indicate the p value according toa one-way ANOVA between the reference area (dark blue) vs thewhole olive orchard (dark green) and the eroding area (beige) vsthe deposition area (light green lower label in italics)

pointed out the additional difficulties in the determination ofCorg due to the topography heterogeneity although this wascompounded by the fact that within the orchards there weretwo areas with different planting dates for the trees Goacutemezet al (2012) reported a CV of 49 with higher Corg in ar-eas where there was a change in the slope gradient from thehillslope to a central channel draining into the catchment al-though they could not find a simple relationship between theincrease in content and the topographic indices Despite thefact that a lot of work has been done on the correlation be-tween erosion and deposition and the redistribution of soilCorg (eg Van Oost et al 2005) our study is to our knowl-edge the first attempt to quantify this in detail under the

Table 6 Loads of selected variables in the PCA for the first threeprincipal components (PCs) The values in brackets below PC 1 2and 3 indicate the percentage of variance explained by this PC Vari-ables in bold are those with a load higher than 90 of the variablewith the maximum load for this PC Conc refers to Corg concen-tration for this fraction and Frac means the relative contribution ofthis fraction to the total Corg for this soil depth

Variable at each depth interval (cm) PC 1 PC 2 PC 3(558) (176) (132)

Pavail 0ndash10 02298 008765 minus007278Pavail 0ndash40 02271 0101 minus008455δ13C 0ndash10 minus003385 02861 minus03302δ15N 0ndash10 01941 01677 01836Norg 0ndash10 02147 minus01375 minus00336δ13C δ15N 0ndash10 01594 0249 02211δ13C 10ndash20 minus02329 01461 004296δ15N 10ndash20 01856 007054 03121Norg 10ndash20 02317 minus008699 minus01404δ13C δ15N 10ndash20 01637 008512 03309δ13C 20ndash30 minus02316 006018 009789δ15N 20ndash30 01748 minus02656 01812Norg 20ndash30 02176 009979 minus007458δ13C δ15N 20ndash30 01684 minus02982 01774Corg conc 10ndash20 02421 minus002951 minus006407Corg unpr conc 10ndash20 02116 00331 00385Corg unpr Frac 10ndash20 minus009663 minus00509 0364Corg Phys Pro conc 10ndash20 02371 minus005728 minus01183Corg Phys Pro Frac 10ndash20 0123 0163 minus002953Corg Chem Pro conc 10ndash20 02072 003097 minus02335Corg Chem Pro Frac 10ndash20 minus004095 009461 minus04678Corg conc 20ndash30 02326 minus01108 minus005737Corg unpr conc 20ndash30 02002 minus02372 minus005071Corg unpr Rel 20ndash30 cm minus0132 minus03675 minus01379Corg Phys Pro conc 20ndash30 02257 005677 minus001036Corg Phys Pro Frac 20ndash30 009518 03528 01294Corg Chem Pro conc 20ndash30 02323 minus006802 minus01366Corg Chem Pro Frac 20ndash30 004565 04437 minus001278

agroenvironmental conditions of an olive orchard The vari-ability induced by the combined effects of water and tillageerosion in this olive orchard was similar to that described inother agroecosystems For instance Van Oost et al (2005)measured a clear correlation between the erosion and depo-sition rates and the topsoil Corg concentration which rangedbetween 08 of the erosion to 14 of the deposition sitesin the top 25 cm of the soil at two field crop sites under atemperate climate Besides this Bameri et al (2015) mea-sured a higher Corg in the lower part of a field crop site undersemiarid conditions where deposition of the eroded soil fromthe upper zones took place Overall in such landscapes culti-vated for a long time the cumulative effect of tillage and wa-ter erosion on the redistribution of soil across the slope hasbeen observed (Dlugoszlig et al 2012) These processes alsoproduce a vertical redistribution of Corg resulting in a rela-tively homogeneous profile in the tilled layer (top 15ndash20 cm)and a gradual decline below this depth as noted in our study

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

190 J A Goacutemez et al Organic carbon across an olive orchard catena

Figure 9 Soil available phosphorus (Pavail) organic nitrogen (Norg) aggregate stability (Wsagg) and soil degradation index (SDI) by depthcomparing eroded vs deposition areas within the olive orchard The labels of the bars indicate the p value according to a one-way ANOVAcomparing areas for the same soil depth

Figure 10 Scores on principal components 1 and 2 (PC 1 PC 2)for sampling points in the eroded and deposition areas in the oliveorchard

42 Organic carbon stock

The differences in soil organic carbon stock between the ref-erence site and the olive orchard are similar to those de-scribed previously when comparing cropland and forestedareas with the latter presenting a higher concentration ofCorg most of which is in the unprotected fraction whilethe cropland presented a higher fraction of the carbon inthe physically and chemically protected fractions (eg Poe-plau and Don 2013) This is likely due to the fact that un-der soil degradation processes such as water erosion and

low annual organic carbon input as is the case under oliveorchard land use most of the unprotected Corg decomposesrelatively quickly and a great proportion of the remaininglow SOC is protected In addition the mobilization of theunprotected Corg is expected to be reduced in the protectedforested area because of the canopy and the existing veg-etation on the ground that protects the soil against runoffand splash erosion processes In fact the protected Corg con-centration in the topsoil of the olive orchard in the erodedarea accounted for 186plusmn 39 of the upper limit of pro-tected Corg (364plusmn 023 ) according to the model of Has-sink and Whitmore (1997) Therefore the low unprotectedSOC concentration found in the olive orchard is an elementof concern in the increase in SOC stocks This is because pro-tected fractions are fueled from recently derived partially de-composed plant residues together with microbial and micro-meso- and macrofaunal debris (unprotected organic carbon)through processes like SOC aggregation into macro- andormicroaggregates (physically protected SOC) and complexSOC associations with clay and silt particles (chemicallyprotected SOC) which are disrupted in the cropland area incomparison with the reference area The distribution amongsoil Corg fractions in the orchard of this study was similarto the result obtained by Vicente-Vicente et al (2017) whomeasured Corg fraction distribution in olive oil orchards withtemporary cover crops with the exception of the unprotectedSOC which was much higher in soils under cover crops thanthat of our study under bare soil Also interesting is the dif-ficulty obtaining statistically significant differences in SOC

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 191

Tabl

e7

Pear

son

corr

elat

ion

coef

ficie

nts

oflin

ear

corr

elat

ion

betw

een

vari

able

sat

each

dept

hin

terv

al(c

m)

with

the

high

est

load

son

prin

cipa

lco

mpo

nent

s(s

eeTa

ble

6)lowast

Mea

nsst

atis

tical

lysi

gnifi

cant

atαlt

005

Ero

sde

pP a

vail

0ndash10

P ava

il0ndash

40δ

13C

10ndash2

0N

org

10ndash2

0δ

13C

20ndash3

0C

org

conc

C

org

Phys

C

org

conc

C

org

Phys

C

org

Che

m

Cor

gC

hem

10

ndash20

Pro

conc

20

ndash30

Pro

conc

Pr

oco

nc

Pro

Frac

10

ndash20

20ndash3

020

ndash30

20ndash3

0

Ero

sde

p1

P ava

il0ndash

100

746lowast

1P a

vail

0ndash40

095

3lowast0

857lowast

1δ

13C

10ndash2

0minus

082

3lowastminus

076

6lowast1

Nor

g10

ndash20

073

1lowast0

898lowast

083

9lowastminus

096

5lowast1

δ13

C20

ndash30

minus0

792lowast

minus0

893lowast

minus0

891lowast

088

8lowastminus

093

8lowast1

Cor

gco

nc1

0ndash20

075

7lowast0

843lowast

089

9lowastminus

092

9lowast0

920lowast

minus0

911lowast

1C

org

Phys

Pro

con

c10

ndash20

075

5lowast0

908lowast

088

2lowastminus

092

90

967lowast

minus0

985lowast

095

2lowast1

Cor

gco

nc2

0ndash30

078

6lowast0

799lowast

minus0

965lowast

092

7lowastminus

084

6lowast0

9453lowast

089

42lowast

1C

org

Phys

Pro

con

c20

ndash30

083

7lowast0

756lowast

minus0

846lowast

081

7lowastminus

071

4lowast0

8819lowast

079

28lowast

090

1lowast1

Cor

gC

hem

Pro

con

c20

ndash30

074

2lowast0

855lowast

085

9lowastminus

096

2lowast0

980lowast

minus0

896lowast

095

15lowast

094

34lowast

095

0lowast0

853lowast

1C

org

Che

mP

roF

rac

20ndash3

01

stock between the eroding and deposition areas (Fig 7) de-spite the apparent clear differences between the two areas insome other soil properties such as Pavail or δ15N and δ 13Cisotopic signatures in the subsoil (see below)

43 δ15N and δ13C isotopic signal

Changes in vegetation types induced differences in δ13C be-tween the olive orchard and the reference area but as ex-pected no differences in δ13C were detected between the ero-sion and deposition areas in the olive orchard given the sameorigin of vegetation-derived organic matter ie C3 plantsInterestingly within the olive orchard significant differencesin δ15N were detected between the erosion and deposition ar-eas especially for the top 20 cm of soils (Fig 9) This sug-gests the potential of using δ15N as a variable for identifyingdegraded areas in olive growing fields as has been proposedfor other eroding regions in the world (eg Meusburger etal 2013) This might provide an alternative when other con-ventional or isotopic techniques are not available Neverthe-less further studies exploring this potential are necessary inorder to also consider the influence of the NndashP fertilizer mod-ifying the δ15N in relation to the reference area (eg Bate-man and Kelly 2007) The source of N in soil is multifariousand subject to a wide range of transformations that affect theδ15N signature therefore we can only speculate on the rea-sons for this difference in δ15N in a relatively homogeneousarea Bulk soil δ15N tended to be more positive (eg moreenriched in δ15N) as the N cycling rate increases soil micro-bial processes (eg N mineralization nitrification and deni-trification) resulting in products (eg nitrate N2O N2 andNH3) depleted in 15N while the substrate from which theywere formed becomes slightly enriched (Robison 2001) Thehigher δ15N signature of the soil at the deposition locationsuggests that the rates of processes involved in N cycling arehigher than in the erosion area and that it is in accordancewith the higher bulk Corg and Pavail contents and lower SDIof the deposition site The relatively lower soil δ15N signa-ture at the reference site could be partially due to the input oflitter N from the natural legumes and to the closed N cyclingwhich characterize natural forest ecosystems The trend in15N enrichment with soil depth as found at the referencesite is a common observation at forest and grassland sitesand has been related to different mechanisms including 15Nisotope discrimination during microbial N transformationsdifferential preservation of 15N-enriched soil organic mattercomponents during N decomposition and more recently tothe build-up of 15N-enriched microbial necromass (Huygenset al 2008) However there still remains the need for a care-ful calibration to an undisturbed reference site and a betterunderstanding of the influence of different vegetation in thereference and the studied area on the change in the δ15N sig-nal for its further use as an additional tool to determine soildegradation

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

192 J A Goacutemez et al Organic carbon across an olive orchard catena

44 Soil quality of topsoil across the catena variabilitybetween eroded and deposition areas in the orchard

This horizontal distribution due to tillage and water erosionalso simultaneously affected other soil properties and hasbeen described previously in other cultivated areas For in-stance De Gryze et al (2008) described how for a field croparea under conventional tillage in Belgium Pavail almost dou-bled (229 vs 122 mg kgminus1) in the depositional area as com-pared to the eroding upper part They also reported that inhalf of the field under conservation tillage these differencesin Pavail between the upper and lower areas of the field dis-appeared We have observed in our sampled orchard a pro-nounced increase in topsoil Pavail in the deposition area ofaround 400 as compared to the eroding part of the orchardwhich can be attributed to the deposition of enriched sedi-ment coming from the upslope area The cumulative effectsof the differences in Corg and Pavail and the trend towardshigher although nonsignificant Wsagg in the deposition arearesulted in two areas within the orchard with marked differ-ences in soil quality the eroded part which is within therange considered to be degraded in the region (Goacutemez et al2009a) and the depositional area representing 20 of theorchard transect length (Table 1) which is within the non-degraded range according to the same index Topographyand sediment redistribution by erosive processes introducea gradient of spatial variability that questions the conceptof representative area when it comes to describing a wholefield In fact several studies (eg Dell and Sharpley 2006)have suggested that the verification of compliance of envi-ronmental programs such as those related to Corg seques-tration should be based preferentially at least partially onmodeling approaches Our results raise the need for a care-ful delineation of subareas when analyzing soil quality indi-cators andor SOC carbon stock within the same field unitThey also warn about the difficulty of drawing hypothesesfor quantifying the differences between these delineated ar-eas in relation to specific soil properties For instance in ourstudy case erosionndashdeposition processes had a major impacton Pavail and soil quality but the impact on Corg concentra-tion and stock was moderate and extremely difficult to detectstatistically using a moderate number of samples

Our PCA and regression analyses confirmed the relativelyhigh variability of Corg and stock in relation to other soilquality indicators related to erosionndashdeposition processessuch as Pavail as discussed in the previous section While inthe catena studied in the current research the P cycle seemsto be mostly driven by sediment mobilization the Corg and Ncycles seem to be much more complex The moderate differ-ences in Corg and the homogeneity in δ15N and δ13C isotopicsignals between the eroding and deposition areas may be dueto several processes some of which were discussed abovesuch as spatial variability of carbon input due to biomassin the plot surface soil operations in the orchard and fer-tilization We found it both interesting and worth exploring

in future studies that significant correlations between erosionand deposition rates and Corg- δ15N- and δ 13C-related vari-ables were found for samples from the 10ndash20 and 20ndash30 cmlayers indicating that short-term disturbance by surface pro-cesses can mask experimental determination of the impact oferosion deposition processes in this olive orchard for thesevariables

5 Conclusions

1 The results indicate that erosion and deposition withinthe investigated old olive orchard have created a signifi-cant difference in soil properties along the catena whichis translated into different soil Corg Pavail and Norg con-tent δ15N and thus contrasting soil quality status

2 This variability was lower than that of the natural areawhich indicated a severe depletion of SOC as comparedto the natural area and a redistribution of available or-ganic carbon among the different SOC fractions

3 The results suggest that δ15N has the potential for beingused as an indicator of soil degradation although moreinvestigation in different agroecosystems would be re-quired for confirming this statement on a larger scale

4 This research highlights that proper understanding andmanagement of soil quality and Corg content in olive or-chards require consideration of the on-site spatial vari-ability induced by soil erosionndashdeposition processes

Data availability The basic data used in this paper are freelyavailable from the CSIC digital repository at httpsdoiorg1020350digitalCSIC12513 (Goacutemez Calero et al 2020)

Author contributions JAG coordinated and designed the over-all study conducted the field sampling campaigns analysis of thesoil organic carbon results and the overall analysis of the resultsand wrote the manuscript GG participated in the analysis of soil-quality-related properties the overall analysis of the results and thewriting of the manuscript LM participated in the overall analysisof the results and the writing of the manuscript CR and AT partici-pated in the analysis and interpretation of the stable isotope resultsRGR contributed to the design and the statistical analysis of soilorganic carbon (stock and fractions) and nitrogen data the overallanalysis of the results and the writing of the manuscript

Competing interests The authors declare that they have no con-flict of interest

Acknowledgements The authors would like to thankManuel Gonzaacutelez de Molina for providing valuable insightinto the management of the olive orchard in the decades before

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

190 J A Goacutemez et al Organic carbon across an olive orchard catena

Figure 9 Soil available phosphorus (Pavail) organic nitrogen (Norg) aggregate stability (Wsagg) and soil degradation index (SDI) by depthcomparing eroded vs deposition areas within the olive orchard The labels of the bars indicate the p value according to a one-way ANOVAcomparing areas for the same soil depth

Figure 10 Scores on principal components 1 and 2 (PC 1 PC 2)for sampling points in the eroded and deposition areas in the oliveorchard

42 Organic carbon stock

The differences in soil organic carbon stock between the ref-erence site and the olive orchard are similar to those de-scribed previously when comparing cropland and forestedareas with the latter presenting a higher concentration ofCorg most of which is in the unprotected fraction whilethe cropland presented a higher fraction of the carbon inthe physically and chemically protected fractions (eg Poe-plau and Don 2013) This is likely due to the fact that un-der soil degradation processes such as water erosion and

low annual organic carbon input as is the case under oliveorchard land use most of the unprotected Corg decomposesrelatively quickly and a great proportion of the remaininglow SOC is protected In addition the mobilization of theunprotected Corg is expected to be reduced in the protectedforested area because of the canopy and the existing veg-etation on the ground that protects the soil against runoffand splash erosion processes In fact the protected Corg con-centration in the topsoil of the olive orchard in the erodedarea accounted for 186plusmn 39 of the upper limit of pro-tected Corg (364plusmn 023 ) according to the model of Has-sink and Whitmore (1997) Therefore the low unprotectedSOC concentration found in the olive orchard is an elementof concern in the increase in SOC stocks This is because pro-tected fractions are fueled from recently derived partially de-composed plant residues together with microbial and micro-meso- and macrofaunal debris (unprotected organic carbon)through processes like SOC aggregation into macro- andormicroaggregates (physically protected SOC) and complexSOC associations with clay and silt particles (chemicallyprotected SOC) which are disrupted in the cropland area incomparison with the reference area The distribution amongsoil Corg fractions in the orchard of this study was similarto the result obtained by Vicente-Vicente et al (2017) whomeasured Corg fraction distribution in olive oil orchards withtemporary cover crops with the exception of the unprotectedSOC which was much higher in soils under cover crops thanthat of our study under bare soil Also interesting is the dif-ficulty obtaining statistically significant differences in SOC

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 191

Tabl

e7

Pear

son

corr

elat

ion

coef

ficie

nts

oflin

ear

corr

elat

ion

betw

een

vari

able

sat

each

dept

hin

terv

al(c

m)

with

the

high

est

load

son

prin

cipa

lco

mpo

nent

s(s

eeTa

ble

6)lowast

Mea

nsst

atis

tical

lysi

gnifi

cant

atαlt

005

Ero

sde

pP a

vail

0ndash10

P ava

il0ndash

40δ

13C

10ndash2

0N

org

10ndash2

0δ

13C

20ndash3

0C

org

conc

C

org

Phys

C

org

conc

C

org

Phys

C

org

Che

m

Cor

gC

hem

10

ndash20

Pro

conc

20

ndash30

Pro

conc

Pr

oco

nc

Pro

Frac

10

ndash20

20ndash3

020

ndash30

20ndash3

0

Ero

sde

p1

P ava

il0ndash

100

746lowast

1P a

vail

0ndash40

095

3lowast0

857lowast

1δ

13C

10ndash2

0minus

082

3lowastminus

076

6lowast1

Nor

g10

ndash20

073

1lowast0

898lowast

083

9lowastminus

096

5lowast1

δ13

C20

ndash30

minus0

792lowast

minus0

893lowast

minus0

891lowast

088

8lowastminus

093

8lowast1

Cor

gco

nc1

0ndash20

075

7lowast0

843lowast

089

9lowastminus

092

9lowast0

920lowast

minus0

911lowast

1C

org

Phys

Pro

con

c10

ndash20

075

5lowast0

908lowast

088

2lowastminus

092

90

967lowast

minus0

985lowast

095

2lowast1

Cor

gco

nc2

0ndash30

078

6lowast0

799lowast

minus0

965lowast

092

7lowastminus

084

6lowast0

9453lowast

089

42lowast

1C

org

Phys

Pro

con

c20

ndash30

083

7lowast0

756lowast

minus0

846lowast

081

7lowastminus

071

4lowast0

8819lowast

079

28lowast

090

1lowast1

Cor

gC

hem

Pro

con

c20

ndash30

074

2lowast0

855lowast

085

9lowastminus

096

2lowast0

980lowast

minus0

896lowast

095

15lowast

094

34lowast

095

0lowast0

853lowast

1C

org

Che

mP

roF

rac

20ndash3

01

stock between the eroding and deposition areas (Fig 7) de-spite the apparent clear differences between the two areas insome other soil properties such as Pavail or δ15N and δ 13Cisotopic signatures in the subsoil (see below)

43 δ15N and δ13C isotopic signal

Changes in vegetation types induced differences in δ13C be-tween the olive orchard and the reference area but as ex-pected no differences in δ13C were detected between the ero-sion and deposition areas in the olive orchard given the sameorigin of vegetation-derived organic matter ie C3 plantsInterestingly within the olive orchard significant differencesin δ15N were detected between the erosion and deposition ar-eas especially for the top 20 cm of soils (Fig 9) This sug-gests the potential of using δ15N as a variable for identifyingdegraded areas in olive growing fields as has been proposedfor other eroding regions in the world (eg Meusburger etal 2013) This might provide an alternative when other con-ventional or isotopic techniques are not available Neverthe-less further studies exploring this potential are necessary inorder to also consider the influence of the NndashP fertilizer mod-ifying the δ15N in relation to the reference area (eg Bate-man and Kelly 2007) The source of N in soil is multifariousand subject to a wide range of transformations that affect theδ15N signature therefore we can only speculate on the rea-sons for this difference in δ15N in a relatively homogeneousarea Bulk soil δ15N tended to be more positive (eg moreenriched in δ15N) as the N cycling rate increases soil micro-bial processes (eg N mineralization nitrification and deni-trification) resulting in products (eg nitrate N2O N2 andNH3) depleted in 15N while the substrate from which theywere formed becomes slightly enriched (Robison 2001) Thehigher δ15N signature of the soil at the deposition locationsuggests that the rates of processes involved in N cycling arehigher than in the erosion area and that it is in accordancewith the higher bulk Corg and Pavail contents and lower SDIof the deposition site The relatively lower soil δ15N signa-ture at the reference site could be partially due to the input oflitter N from the natural legumes and to the closed N cyclingwhich characterize natural forest ecosystems The trend in15N enrichment with soil depth as found at the referencesite is a common observation at forest and grassland sitesand has been related to different mechanisms including 15Nisotope discrimination during microbial N transformationsdifferential preservation of 15N-enriched soil organic mattercomponents during N decomposition and more recently tothe build-up of 15N-enriched microbial necromass (Huygenset al 2008) However there still remains the need for a care-ful calibration to an undisturbed reference site and a betterunderstanding of the influence of different vegetation in thereference and the studied area on the change in the δ15N sig-nal for its further use as an additional tool to determine soildegradation

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

192 J A Goacutemez et al Organic carbon across an olive orchard catena

44 Soil quality of topsoil across the catena variabilitybetween eroded and deposition areas in the orchard

This horizontal distribution due to tillage and water erosionalso simultaneously affected other soil properties and hasbeen described previously in other cultivated areas For in-stance De Gryze et al (2008) described how for a field croparea under conventional tillage in Belgium Pavail almost dou-bled (229 vs 122 mg kgminus1) in the depositional area as com-pared to the eroding upper part They also reported that inhalf of the field under conservation tillage these differencesin Pavail between the upper and lower areas of the field dis-appeared We have observed in our sampled orchard a pro-nounced increase in topsoil Pavail in the deposition area ofaround 400 as compared to the eroding part of the orchardwhich can be attributed to the deposition of enriched sedi-ment coming from the upslope area The cumulative effectsof the differences in Corg and Pavail and the trend towardshigher although nonsignificant Wsagg in the deposition arearesulted in two areas within the orchard with marked differ-ences in soil quality the eroded part which is within therange considered to be degraded in the region (Goacutemez et al2009a) and the depositional area representing 20 of theorchard transect length (Table 1) which is within the non-degraded range according to the same index Topographyand sediment redistribution by erosive processes introducea gradient of spatial variability that questions the conceptof representative area when it comes to describing a wholefield In fact several studies (eg Dell and Sharpley 2006)have suggested that the verification of compliance of envi-ronmental programs such as those related to Corg seques-tration should be based preferentially at least partially onmodeling approaches Our results raise the need for a care-ful delineation of subareas when analyzing soil quality indi-cators andor SOC carbon stock within the same field unitThey also warn about the difficulty of drawing hypothesesfor quantifying the differences between these delineated ar-eas in relation to specific soil properties For instance in ourstudy case erosionndashdeposition processes had a major impacton Pavail and soil quality but the impact on Corg concentra-tion and stock was moderate and extremely difficult to detectstatistically using a moderate number of samples

Our PCA and regression analyses confirmed the relativelyhigh variability of Corg and stock in relation to other soilquality indicators related to erosionndashdeposition processessuch as Pavail as discussed in the previous section While inthe catena studied in the current research the P cycle seemsto be mostly driven by sediment mobilization the Corg and Ncycles seem to be much more complex The moderate differ-ences in Corg and the homogeneity in δ15N and δ13C isotopicsignals between the eroding and deposition areas may be dueto several processes some of which were discussed abovesuch as spatial variability of carbon input due to biomassin the plot surface soil operations in the orchard and fer-tilization We found it both interesting and worth exploring

in future studies that significant correlations between erosionand deposition rates and Corg- δ15N- and δ 13C-related vari-ables were found for samples from the 10ndash20 and 20ndash30 cmlayers indicating that short-term disturbance by surface pro-cesses can mask experimental determination of the impact oferosion deposition processes in this olive orchard for thesevariables

5 Conclusions

1 The results indicate that erosion and deposition withinthe investigated old olive orchard have created a signifi-cant difference in soil properties along the catena whichis translated into different soil Corg Pavail and Norg con-tent δ15N and thus contrasting soil quality status

2 This variability was lower than that of the natural areawhich indicated a severe depletion of SOC as comparedto the natural area and a redistribution of available or-ganic carbon among the different SOC fractions

3 The results suggest that δ15N has the potential for beingused as an indicator of soil degradation although moreinvestigation in different agroecosystems would be re-quired for confirming this statement on a larger scale

4 This research highlights that proper understanding andmanagement of soil quality and Corg content in olive or-chards require consideration of the on-site spatial vari-ability induced by soil erosionndashdeposition processes

Data availability The basic data used in this paper are freelyavailable from the CSIC digital repository at httpsdoiorg1020350digitalCSIC12513 (Goacutemez Calero et al 2020)

Author contributions JAG coordinated and designed the over-all study conducted the field sampling campaigns analysis of thesoil organic carbon results and the overall analysis of the resultsand wrote the manuscript GG participated in the analysis of soil-quality-related properties the overall analysis of the results and thewriting of the manuscript LM participated in the overall analysisof the results and the writing of the manuscript CR and AT partici-pated in the analysis and interpretation of the stable isotope resultsRGR contributed to the design and the statistical analysis of soilorganic carbon (stock and fractions) and nitrogen data the overallanalysis of the results and the writing of the manuscript

Competing interests The authors declare that they have no con-flict of interest

Acknowledgements The authors would like to thankManuel Gonzaacutelez de Molina for providing valuable insightinto the management of the olive orchard in the decades before

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 191

Tabl

e7

Pear

son

corr

elat

ion

coef

ficie

nts

oflin

ear

corr

elat

ion

betw

een

vari

able

sat

each

dept

hin

terv

al(c

m)

with

the

high

est

load

son

prin

cipa

lco

mpo

nent

s(s

eeTa

ble

6)lowast

Mea

nsst

atis

tical

lysi

gnifi

cant

atαlt

005

Ero

sde

pP a

vail

0ndash10

P ava

il0ndash

40δ

13C

10ndash2

0N

org

10ndash2

0δ

13C

20ndash3

0C

org

conc

C

org

Phys

C

org

conc

C

org

Phys

C

org

Che

m

Cor

gC

hem

10

ndash20

Pro

conc

20

ndash30

Pro

conc

Pr

oco

nc

Pro

Frac

10

ndash20

20ndash3

020

ndash30

20ndash3

0

Ero

sde

p1

P ava

il0ndash

100

746lowast

1P a

vail

0ndash40

095

3lowast0

857lowast

1δ

13C

10ndash2

0minus

082

3lowastminus

076

6lowast1

Nor

g10

ndash20

073

1lowast0

898lowast

083

9lowastminus

096

5lowast1

δ13

C20

ndash30

minus0

792lowast

minus0

893lowast

minus0

891lowast

088

8lowastminus

093

8lowast1

Cor

gco

nc1

0ndash20

075

7lowast0

843lowast

089

9lowastminus

092

9lowast0

920lowast

minus0

911lowast

1C

org

Phys

Pro

con

c10

ndash20

075

5lowast0

908lowast

088

2lowastminus

092

90

967lowast

minus0

985lowast

095

2lowast1

Cor

gco

nc2

0ndash30

078

6lowast0

799lowast

minus0

965lowast

092

7lowastminus

084

6lowast0

9453lowast

089

42lowast

1C

org

Phys

Pro

con

c20

ndash30

083

7lowast0

756lowast

minus0

846lowast

081

7lowastminus

071

4lowast0

8819lowast

079

28lowast

090

1lowast1

Cor

gC

hem

Pro

con

c20

ndash30

074

2lowast0

855lowast

085

9lowastminus

096

2lowast0

980lowast

minus0

896lowast

095

15lowast

094

34lowast

095

0lowast0

853lowast

1C

org

Che

mP

roF

rac

20ndash3

01

stock between the eroding and deposition areas (Fig 7) de-spite the apparent clear differences between the two areas insome other soil properties such as Pavail or δ15N and δ 13Cisotopic signatures in the subsoil (see below)

43 δ15N and δ13C isotopic signal

Changes in vegetation types induced differences in δ13C be-tween the olive orchard and the reference area but as ex-pected no differences in δ13C were detected between the ero-sion and deposition areas in the olive orchard given the sameorigin of vegetation-derived organic matter ie C3 plantsInterestingly within the olive orchard significant differencesin δ15N were detected between the erosion and deposition ar-eas especially for the top 20 cm of soils (Fig 9) This sug-gests the potential of using δ15N as a variable for identifyingdegraded areas in olive growing fields as has been proposedfor other eroding regions in the world (eg Meusburger etal 2013) This might provide an alternative when other con-ventional or isotopic techniques are not available Neverthe-less further studies exploring this potential are necessary inorder to also consider the influence of the NndashP fertilizer mod-ifying the δ15N in relation to the reference area (eg Bate-man and Kelly 2007) The source of N in soil is multifariousand subject to a wide range of transformations that affect theδ15N signature therefore we can only speculate on the rea-sons for this difference in δ15N in a relatively homogeneousarea Bulk soil δ15N tended to be more positive (eg moreenriched in δ15N) as the N cycling rate increases soil micro-bial processes (eg N mineralization nitrification and deni-trification) resulting in products (eg nitrate N2O N2 andNH3) depleted in 15N while the substrate from which theywere formed becomes slightly enriched (Robison 2001) Thehigher δ15N signature of the soil at the deposition locationsuggests that the rates of processes involved in N cycling arehigher than in the erosion area and that it is in accordancewith the higher bulk Corg and Pavail contents and lower SDIof the deposition site The relatively lower soil δ15N signa-ture at the reference site could be partially due to the input oflitter N from the natural legumes and to the closed N cyclingwhich characterize natural forest ecosystems The trend in15N enrichment with soil depth as found at the referencesite is a common observation at forest and grassland sitesand has been related to different mechanisms including 15Nisotope discrimination during microbial N transformationsdifferential preservation of 15N-enriched soil organic mattercomponents during N decomposition and more recently tothe build-up of 15N-enriched microbial necromass (Huygenset al 2008) However there still remains the need for a care-ful calibration to an undisturbed reference site and a betterunderstanding of the influence of different vegetation in thereference and the studied area on the change in the δ15N sig-nal for its further use as an additional tool to determine soildegradation

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

192 J A Goacutemez et al Organic carbon across an olive orchard catena

44 Soil quality of topsoil across the catena variabilitybetween eroded and deposition areas in the orchard

This horizontal distribution due to tillage and water erosionalso simultaneously affected other soil properties and hasbeen described previously in other cultivated areas For in-stance De Gryze et al (2008) described how for a field croparea under conventional tillage in Belgium Pavail almost dou-bled (229 vs 122 mg kgminus1) in the depositional area as com-pared to the eroding upper part They also reported that inhalf of the field under conservation tillage these differencesin Pavail between the upper and lower areas of the field dis-appeared We have observed in our sampled orchard a pro-nounced increase in topsoil Pavail in the deposition area ofaround 400 as compared to the eroding part of the orchardwhich can be attributed to the deposition of enriched sedi-ment coming from the upslope area The cumulative effectsof the differences in Corg and Pavail and the trend towardshigher although nonsignificant Wsagg in the deposition arearesulted in two areas within the orchard with marked differ-ences in soil quality the eroded part which is within therange considered to be degraded in the region (Goacutemez et al2009a) and the depositional area representing 20 of theorchard transect length (Table 1) which is within the non-degraded range according to the same index Topographyand sediment redistribution by erosive processes introducea gradient of spatial variability that questions the conceptof representative area when it comes to describing a wholefield In fact several studies (eg Dell and Sharpley 2006)have suggested that the verification of compliance of envi-ronmental programs such as those related to Corg seques-tration should be based preferentially at least partially onmodeling approaches Our results raise the need for a care-ful delineation of subareas when analyzing soil quality indi-cators andor SOC carbon stock within the same field unitThey also warn about the difficulty of drawing hypothesesfor quantifying the differences between these delineated ar-eas in relation to specific soil properties For instance in ourstudy case erosionndashdeposition processes had a major impacton Pavail and soil quality but the impact on Corg concentra-tion and stock was moderate and extremely difficult to detectstatistically using a moderate number of samples

Our PCA and regression analyses confirmed the relativelyhigh variability of Corg and stock in relation to other soilquality indicators related to erosionndashdeposition processessuch as Pavail as discussed in the previous section While inthe catena studied in the current research the P cycle seemsto be mostly driven by sediment mobilization the Corg and Ncycles seem to be much more complex The moderate differ-ences in Corg and the homogeneity in δ15N and δ13C isotopicsignals between the eroding and deposition areas may be dueto several processes some of which were discussed abovesuch as spatial variability of carbon input due to biomassin the plot surface soil operations in the orchard and fer-tilization We found it both interesting and worth exploring

in future studies that significant correlations between erosionand deposition rates and Corg- δ15N- and δ 13C-related vari-ables were found for samples from the 10ndash20 and 20ndash30 cmlayers indicating that short-term disturbance by surface pro-cesses can mask experimental determination of the impact oferosion deposition processes in this olive orchard for thesevariables

5 Conclusions

1 The results indicate that erosion and deposition withinthe investigated old olive orchard have created a signifi-cant difference in soil properties along the catena whichis translated into different soil Corg Pavail and Norg con-tent δ15N and thus contrasting soil quality status

2 This variability was lower than that of the natural areawhich indicated a severe depletion of SOC as comparedto the natural area and a redistribution of available or-ganic carbon among the different SOC fractions

3 The results suggest that δ15N has the potential for beingused as an indicator of soil degradation although moreinvestigation in different agroecosystems would be re-quired for confirming this statement on a larger scale

4 This research highlights that proper understanding andmanagement of soil quality and Corg content in olive or-chards require consideration of the on-site spatial vari-ability induced by soil erosionndashdeposition processes

Data availability The basic data used in this paper are freelyavailable from the CSIC digital repository at httpsdoiorg1020350digitalCSIC12513 (Goacutemez Calero et al 2020)

Author contributions JAG coordinated and designed the over-all study conducted the field sampling campaigns analysis of thesoil organic carbon results and the overall analysis of the resultsand wrote the manuscript GG participated in the analysis of soil-quality-related properties the overall analysis of the results and thewriting of the manuscript LM participated in the overall analysisof the results and the writing of the manuscript CR and AT partici-pated in the analysis and interpretation of the stable isotope resultsRGR contributed to the design and the statistical analysis of soilorganic carbon (stock and fractions) and nitrogen data the overallanalysis of the results and the writing of the manuscript

Competing interests The authors declare that they have no con-flict of interest

Acknowledgements The authors would like to thankManuel Gonzaacutelez de Molina for providing valuable insightinto the management of the olive orchard in the decades before

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

192 J A Goacutemez et al Organic carbon across an olive orchard catena

44 Soil quality of topsoil across the catena variabilitybetween eroded and deposition areas in the orchard

This horizontal distribution due to tillage and water erosionalso simultaneously affected other soil properties and hasbeen described previously in other cultivated areas For in-stance De Gryze et al (2008) described how for a field croparea under conventional tillage in Belgium Pavail almost dou-bled (229 vs 122 mg kgminus1) in the depositional area as com-pared to the eroding upper part They also reported that inhalf of the field under conservation tillage these differencesin Pavail between the upper and lower areas of the field dis-appeared We have observed in our sampled orchard a pro-nounced increase in topsoil Pavail in the deposition area ofaround 400 as compared to the eroding part of the orchardwhich can be attributed to the deposition of enriched sedi-ment coming from the upslope area The cumulative effectsof the differences in Corg and Pavail and the trend towardshigher although nonsignificant Wsagg in the deposition arearesulted in two areas within the orchard with marked differ-ences in soil quality the eroded part which is within therange considered to be degraded in the region (Goacutemez et al2009a) and the depositional area representing 20 of theorchard transect length (Table 1) which is within the non-degraded range according to the same index Topographyand sediment redistribution by erosive processes introducea gradient of spatial variability that questions the conceptof representative area when it comes to describing a wholefield In fact several studies (eg Dell and Sharpley 2006)have suggested that the verification of compliance of envi-ronmental programs such as those related to Corg seques-tration should be based preferentially at least partially onmodeling approaches Our results raise the need for a care-ful delineation of subareas when analyzing soil quality indi-cators andor SOC carbon stock within the same field unitThey also warn about the difficulty of drawing hypothesesfor quantifying the differences between these delineated ar-eas in relation to specific soil properties For instance in ourstudy case erosionndashdeposition processes had a major impacton Pavail and soil quality but the impact on Corg concentra-tion and stock was moderate and extremely difficult to detectstatistically using a moderate number of samples

Our PCA and regression analyses confirmed the relativelyhigh variability of Corg and stock in relation to other soilquality indicators related to erosionndashdeposition processessuch as Pavail as discussed in the previous section While inthe catena studied in the current research the P cycle seemsto be mostly driven by sediment mobilization the Corg and Ncycles seem to be much more complex The moderate differ-ences in Corg and the homogeneity in δ15N and δ13C isotopicsignals between the eroding and deposition areas may be dueto several processes some of which were discussed abovesuch as spatial variability of carbon input due to biomassin the plot surface soil operations in the orchard and fer-tilization We found it both interesting and worth exploring

in future studies that significant correlations between erosionand deposition rates and Corg- δ15N- and δ 13C-related vari-ables were found for samples from the 10ndash20 and 20ndash30 cmlayers indicating that short-term disturbance by surface pro-cesses can mask experimental determination of the impact oferosion deposition processes in this olive orchard for thesevariables

5 Conclusions

1 The results indicate that erosion and deposition withinthe investigated old olive orchard have created a signifi-cant difference in soil properties along the catena whichis translated into different soil Corg Pavail and Norg con-tent δ15N and thus contrasting soil quality status

2 This variability was lower than that of the natural areawhich indicated a severe depletion of SOC as comparedto the natural area and a redistribution of available or-ganic carbon among the different SOC fractions

3 The results suggest that δ15N has the potential for beingused as an indicator of soil degradation although moreinvestigation in different agroecosystems would be re-quired for confirming this statement on a larger scale

4 This research highlights that proper understanding andmanagement of soil quality and Corg content in olive or-chards require consideration of the on-site spatial vari-ability induced by soil erosionndashdeposition processes

Data availability The basic data used in this paper are freelyavailable from the CSIC digital repository at httpsdoiorg1020350digitalCSIC12513 (Goacutemez Calero et al 2020)

Author contributions JAG coordinated and designed the over-all study conducted the field sampling campaigns analysis of thesoil organic carbon results and the overall analysis of the resultsand wrote the manuscript GG participated in the analysis of soil-quality-related properties the overall analysis of the results and thewriting of the manuscript LM participated in the overall analysisof the results and the writing of the manuscript CR and AT partici-pated in the analysis and interpretation of the stable isotope resultsRGR contributed to the design and the statistical analysis of soilorganic carbon (stock and fractions) and nitrogen data the overallanalysis of the results and the writing of the manuscript

Competing interests The authors declare that they have no con-flict of interest

Acknowledgements The authors would like to thankManuel Gonzaacutelez de Molina for providing valuable insightinto the management of the olive orchard in the decades before

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020

J A Goacutemez et al Organic carbon across an olive orchard catena 193

sampling Furthermore the authors are grateful to the three anony-mous reviewers of the manuscript version in Soil Discussion fortheir care and attention to the manuscript All their comments andindications have greatly helped to increase the quality of our workThis study was supported by the projects AGL2015-40128-C03-01of the Spanish Government SHui (GA 773903) of the EuropeanCommission and EU-FEDER funds

Financial support This research has been supported by theSpanish Ministry of Economy and Competitiveness (grant noAGL2015-40128-C03-01) and the European Commission (SHuigrant no 773903)

Review statement This paper was edited by Olivier Evrard andreviewed by three anonymous referees

References

Aacutelvarez S Soriano M A Landa B B and Goacutemez J ASoil properties in organic olive groves compared with that innatural areas in a mountainous landscape in southern SpainSoil Use Manage 23 404ndash416 httpsdoiorg101111j1475-2743200700104x 2010

Anderson J M and Ingram J S I (Eds) Tropical soil biology andfertility A handbook of methods CAB International Walling-ford UK 1993

Andrews S S and Carroll C R Designing a soil qual-ity assessment tool for sustainable agroecosystem manage-ment Ecol Appl 11 1573ndash1585 httpsdoiorg1018901051-0761(2001)011[1573DASQAT]20CO2 2001

Bameri A Khormali F Amir K and Dehghani A Spatial vari-ability of soil organic carbon in different hillslope positions inToshan area Golestan Province Iran Geostatistical approachesJ Mt Sci 12 1422ndash1433 httpsdoiorg101007s11629-014-3213-z 2015

Barthes B and Roose E Aggregate stability as an indicatorof soil susceptibility to runoff and erosion validation at sev-eral levels Catena 47 133ndash149 httpsdoiorg101016S0341-8162(01)00180-1 2002

Bateman A S and Kelly S D Fertilizer nitrogen iso-tope signatures Isot Environ Healt S 43 237ndash247httpsdoiorg10108010256010701550732 2007

Beaufoy G EU Policies for olive farming-Unsustainable on allCounts Brussels WWF amp BirdLife Joint Report 20 pp 2001

Bouyoucos G J Hydrometer method improvement for mak-ing particle size analysis of soils Agron J 54 179ndash186httpsdoiorg102134agronj196200021962005400050028x1962

Deckers J Spaargaren O and Dondeyne S Soil survey as a basisfor land evaluation in Land Use Land Cover and Soil SciencesVolume II Land Evaluation Encyclopedia of Life Support Sys-tems edited by Verheyen W EOLSS-UNESCO Publ OxfordUK 29ndash58 2004

De Gryze S Six J Bossuyt H Van Oost K and Merckx RThe relationship between landform and the distribution of soil CN and P under conventional and minimum tillage Geoderma

144 180ndash188 httpsdoiorg101016jgeoderma2007110132008

Dell C J and Sharpley A N Spatial variation of soil organiccarbon in a Northeastern US watershed J Soil Water Conserv61 129ndash136 httpsdoiorg101007s11676-014-0533-3 2006

Dlugoszlig V Fiener P Van Oost K and Schneider K Modelbased analysis of lateral and vertical soil carbon fluxesinduced by soil redistribution processes in a small agri-cultural catchment Earth Surf Proc Land 37 193ndash208httpsdoiorg101002esp2246 2012

FAOSTAT Food and Agriculture Statistics of FAO Publicationsavailable at httpwwwfaoorgfaostatenhome last accessJanuary 2019

Gargouri K Rigane H Arous I and Touil FEvolution of soil organic carbon in an olive or-chard under arid climate Sci Hortic 152 102ndash108httpsdoiorg101016jscienta201211025 2013

Goacutemez J A Aacutelvarez S and Soriano M A Devel-opment of a soil degradation assessment tool for or-ganic olive groves in southern Spain Catena 79 9ndash17httpsdoiorg101016jcatena200905002 2009a

Goacutemez J A Sobrinho T A Giraacuteldez J V and Fereres ESoil management effects on runoff erosion and soil propertiesin an olive grove of Southern Spain Soil Till Res 102 5ndash13httpsdoiorg101016jstill200805005 2009b

Goacutemez J A Burguet M Taguas M E Perez R Ayuso J LVanwalleghem T Giraldez J V and Vanderlinden K Distri-bution of soil organic carbon in two small agricultural Mediter-ranean catchments Geophys Res Abstr EGU2012-4232 EGUGeneral Assembly 2012 Vienna Austria 2012

Goacutemez J A Infante-Amate J Gonzaacutelez de Molina M Van-walleghem T Taguas E V and Lorite I Review Olivecultivation its impact on soil erosion and its progression intoyield impacts in Southern Spain in the past as a key to a fu-ture of increasing climate uncertainty Agriculture 4 170ndash200httpsdoiorg103390agriculture4020170 2014

Goacutemez Calero J A Guzmaacuten G Toloza A Resch C Garciacutea-Ruiz R and Mabit L Basic data underlying research inmanuscript ldquoVariation of soil organic carbon stable isotopesand soil quality indicators across an erosionndashdeposition catenain a historical Spanish olive orchardrdquo CSIC digital repositoryhttpsdoiorg1020350digitalCSIC12513 2020

Harris D Horwath W and Kessel C Acid fumigation of soilsto remove carbonates prior to total organic carbon or Carbon-13 isotopic analysis Soil Sci Soc Am J 65 1853ndash1856httpsdoiorg102136sssaj20011853 2001

Hassink J and Whitmore A P A model of the physical protectionof organic matter in soils Soil Sci Soc Am J 61 131ndash139httpsdoiorg102136sssaj199703615995006100010020x1997

Huang J Pedrera-Parrilla A Vanderlinden K Taguas E VGoacutemez J A and Triantafilis J Potential to map depth-specificsoil organic matter content across an olive grove using quasi-2dand quasi-3d inversion of DUALEM-21 data Catena 152 207ndash217 httpsdoiorg101016jcatena201701017 2017

Huygens D Denef K Vandeweyer R Godoy R Van Cleem-put O and Boeckx P Do Nitrogen isotope patterns reflect mi-crobial colonization of soil organic matter fractions Biol Fert

wwwsoil-journalnet61792020 SOIL 6 179ndash194 2020

194 J A Goacutemez et al Organic carbon across an olive orchard catena

Soils 44 955ndash964 httpsdoiorg101007s00374-008-0280-82008

Lal R Soil erosion and the global carbon budget Environ Int29 437ndash450 httpsdoiorg101016S0160-4120(02)00192-72003

Lal R Cover cropping and the ldquo4 per Thou-sandrdquo proposal J Soil Water Conserv 70 141Ahttpsdoiorg102489jswc706141A 2015

Lal R Accelerated soil erosion as a source of at-mospheric CO2 Soil Till Res 188 35ndash40httpsdoiorg101016jstill201802001 2019

Lugato E Paustian K Panagos P Jones A and Borrelli PQuantifying the erosion effect on current carbon budget of Eu-ropean agricultural soils at high spatial resolution Glob ChangeBiol 22 1976ndash1984 httpsdoiorg101111gcb13198 2016

Mabit L and Bernard C Relationship between soil in-ventories and chemical properties in a small intensivelycropped watershed CR Acad Sci II A 327 527ndash532httpsdoiorg101016S1251-8050(99)80034-2 1998

Mabit L and Bernard C Spatial distribution and content of soilorganic matter in an agricultural field in Eastern Canada as esti-mated from geostatistical tools Earth Surf Proc Land 35 278ndash283 httpsdoiorg101002esp1907 2010

Mabit L Benmansour M and Walling D E Com-parative advantages and limitations of Fallout radionu-clides (137Cs 210Pb and 7Be) to assess soil erosion andsedimentation J Environ Radioactiv 99 1799ndash1807httpsdoiorg101016jjenvrad200808009 2008

Mabit L Chhem-Kieth S Toloza A Vanwalleghem TBernard C Infante-Amate J Gonzaacutelez de Molina Mand Goacutemez J A Radioisotopic and physicochemical back-ground indicators to assess soil degradation affecting olive or-chards in southern Spain Agr Ecosyst Environ 159 70ndash80httpsdoiorg101016jagee201206014 2012

Mabit L Benmansour M Abril J M Walling D E Meus-burger K Iurian A R Bernard C Tarjaacuten S OwensP N Blake W H and Alewell C Fallout 210Pb asa soil and sediment tracer in catchment sediment bud-get investigations A review Earth-Sci Rev 138 335ndash351httpsdoiorg101016jearscirev201406007 2014

Matisoff G 210Pb as a tracer of soil erosion sedimentsource area identification and particle transport in the ter-restrial environment J Environ Radioactiv 138 343ndash354httpsdoiorg101016jjenvrad201403008 2014

Meusburger K Mabit L Park J-H Sandor T and AlewellC Combined use of stable isotopes and fallout radionuclides assoil erosion indicators in a forested mountain site South KoreaBiogeosciences 10 5627ndash5638 httpsdoiorg105194bg-10-5627-2013 2013

Milgroom J Soriano M A Garrido J M Goacutemez J Aand Fereres E The influence of a shift from conventionalto organic olive farming on soil management and erosionrisk in southern Spain Renew Agr Food Syst 22 1ndash10httpsdoiorg101017S1742170507001500 2007

Olsen S R and Sommers L E Phosphorus in Methods of soilanalysis Part 2 Chemical and microbiological properties editedby Page A L Miller R H and Keeney D ASA amp SSSA403ndash430 Madison WI USA 1982

Poeplau C and Don A Sensitivity of soil organiccarbon stocks and fractions to different land-usechanges across Europe Geoderma 192 189ndash201httpsdoiorg101016jgeoderma201208003 2013

Rajan K Natarajan A Kumar K S A Badrinath M S andGowda R C Soil organic carbon ndash the most reliable indicatorfor monitoring land degradation by soil erosion Curr Sci 99823ndash827 2010

Robison D δ15N as an integrator of the nitrogen cycleTrends Ecol Evol 16 153ndash162 httpsdoiorg101016S0169-5347(00)02098-X 2001

Ruiacutez de Castroviejo J Explotacioacuten de olivares en asociacioacuten contreacutebol subterraacuteneo Agricultura 443 135ndash139 1969

Scheidel A and Krausmann F Diet trade and landuse A socio-ecological analysis of the transformationof the olive oil system Land Use Policy 28 47ndash56httpsdoiorg101016jlandusepol201004008 2011

Six J Callewaert P Lenders S De Gryz S Morris SJ Gregorich E G and Paustian K Measuring and un-derstanding carbon storage in afforested soils by physi-cal fractionation Soil Sci Soc Am J 66 1981ndash1987httpsdoiorg102136sssaj20021981 2002

Soriano M A Aacutelvarez S Landa B B and GoacutemezJ A Soil properties in organic olive orchards follow-ing different weed management in a rolling landscape ofAndalusia Spain Renew Agr Food Syst 29 83ndash91httpsdoiorg101017S1742170512000361 2014

Stevenson F J Nitrogen-Organic forms in Methods of soil anal-ysis Part 2 Chemical and microbiological properties edited byPage A L Miller R H and Keeney D ASA amp SSSA 625ndash641 Madison WI USA 1982

Stewart C E Paustian K Conant R T Plante A F and SixJ Soil carbon saturation Implications for measurable carbonpool dynamics in long-term incubations Soil Biol Biochem 41357ndash366 httpsdoiorg101016jsoilbio200811011 2009

Van Oost K Govers G Quine T A Heckrath G Olesen JE De Gryze S and Merckx R Landscape-scale modelingof carbon cycling under the impact of soil redistribution Therole of tillage erosion Global Biogeochem Cy 19 GB4014httpsdoiorg1010292005GB002471 2005

Vanwalleghem T Infante-Amate J Gonzaacutelez de Molina MSoto Fernaacutendez D and Goacutemez J Quantifying the effect ofhistorical soil management on soil erosion rates in Mediter-ranean olive Orchards Agr Ecosyst Environ 142 341ndash351httpsdoiorg101016jagee201106003 2011

Vicente-Vicente J L Garciacutea-Ruiz R Francaviglia R Aguil-era E and Smith P Soil carbon sequestration rates underMediterranean woody crops using recommended managementpractices A meta-analysis Agr Ecosyst Environ 235 204ndash214 httpsdoiorg101016jagee201610024 2016

Vicente-Vicente J L Goacutemez-Muntildeoz B Hinojosa-CentenoM B Smith P and Garciacutea-Ruiz R Carbon satura-tion and assessment of soil organic carbon fractions inMediterranean rainfed olive orchards under plant covermanagement Agr Ecosyst Environ 245 135ndash146httpsdoiorg101016jagee201705020 2017

Wendt J W and Hauser S An equivalent soil mass procedure formonitoring soil organic carbon in multiple soil layers Eur J SoilSci 64 58ndash65 httpsdoiorg101111ejss12002 2013

SOIL 6 179ndash194 2020 wwwsoil-journalnet61792020