The soil organic carbon stabilization potential of old and ... · San Luis Obispo, California...

Biogeosciences 17 2971ndash2986 2020httpsdoiorg105194bg-17-2971-2020copy Author(s) 2020 This work is distributed underthe Creative Commons Attribution 40 License

The soil organic carbon stabilization potential of old andnew wheat cultivars a 13CO2-labeling studyMarijn Van de Broek1 Shiva Ghiasi2 Charlotte Decock13 Andreas Hund4 Samuel Abiven5 Cordula Friedli45Roland A Werner2 and Johan Six1

1Sustainable Agroecosystems Group Department of Environmental Systems Science Swiss Federal Institute of TechnologyETH Zuumlrich Zurich Switzerland2Grassland Sciences Group Department of Environmental Systems Science Swiss Federal Institute of TechnologyETH Zuumlrich Zurich Switzerland3Department of Natural Resources Management and Environmental Sciences California Polytechnic State UniversitySan Luis Obispo California 93407 USA4Group of Crop Science Department of Environmental Systems Science Swiss Federal Institute of TechnologyETH Zuumlrich Zurich Switzerland5Department of Geography University of Zuumlrich Zurich SwitzerlandThese authors contributed equally to this work

Correspondence Marijn Van de Broek (marijnvandebroekusysethzch)

Received 24 December 2019 ndash Discussion started 9 January 2020Revised 11 April 2020 ndash Accepted 29 April 2020 ndash Published 11 June 2020

Abstract Over the past decades average global wheat yieldshave increased by about 250 mainly due to the cultiva-tion of high-yielding wheat cultivars This selection processnot only affected aboveground parts of plants but in somecases also reduced root biomass with potentially large con-sequences for the amount of organic carbon (OC) transferredto the soil To study the effect of wheat breeding for high-yielding cultivars on subsoil OC dynamics two old and twonew wheat cultivars from the Swiss wheat breeding programwere grown for one growing season in 15 m deep lysime-ters and pulse labeled with 13CO2 to quantify the amountof assimilated carbon that was transferred belowground andcan potentially be stabilized in the soil The results show thatalthough the old wheat cultivars with higher root biomasstransferred more assimilated carbon belowground comparedto more recent cultivars no significant differences in net rhi-zodeposition were found between the different cultivars Asa consequence the long-term effect of wheat cultivar selec-tion on soil organic carbon (SOC) stocks will depend on theamount of root biomass that is stabilized in the soil Our re-sults suggest that the process of wheat selection for high-yielding cultivars resulted in lower amounts of belowgroundcarbon translocation with potentially important effects on

SOC stocks Further research is necessary to quantify thelong-term importance of this effect

1 Introduction

Soil management has a large influence on the size of the soilorganic carbon (SOC) stock in managed arable soils Thisis evident from the large decrease in SOC that is generallyobserved after soils under natural vegetation are convertedto arable land (Don et al 2011 Guo and Gifford 2002Poeplau et al 2011) As a consequence the mineralizationof SOC and the loss of forest caused by land use changehas contributed about 30 to the increase in atmosphericCO2 concentration since the onset of the industrial revolu-tion (Le Queacutereacute et al 2018) Current contributions of the agri-cultural sector to global warming have been estimated to beabout 11 but are mostly in the form of N2O and CH4 andnot anymore as CO2 (Tubiello et al 2015)

The rising awareness that there is potentially an opportu-nity to increase subsoil organic carbon (OC) stocks (Chen etal 2018) has led to the proposal that agricultural soils can bea sink of atmospheric CO2 by applying appropriate climate-

Published by Copernicus Publications on behalf of the European Geosciences Union

2972 M Van de Broek et al The soil organic carbon stabilization potential of old and new wheat cultivars

smart agricultural practices (Chenu et al 2018 Minasny etal 2017 Paustian et al 2016) Multiple management prac-tices have been shown to increase the OC content of cul-tivated soils including the application of organic amend-ments to soils (Sandeacuten et al 2018) increasing the amountof crop residues returned to the field (Lehtinen et al 2014)and planting of cover crops (Kong and Six 2010 Poeplauand Don 2015)

In addition growing crops with deeper roots andor higherroot biomass has been put forward as a strategy to increaseOC sequestration in arable soils (Kell 2011) while deeprooting can also decrease the effect of drought in climateswhere deep soil water is available during the main croppingseason (Wasson et al 2012) However a direct or marker-assisted selection for root traits is very rare in conven-tional breeding programs Accordingly we have very limitedknowledge on if and how breeders alter the root system andpotentially affect belowground carbon cycling One way toevaluate the effect of breederrsquos selection on root character-istics and subsoil carbon cycling is to compare old and newvarieties of the same breeding program For the Swiss wheatbreeding programs the selection process reduced the massand depth of roots under well-watered conditions (Friedli etal 2019) as has been found for other breeding programs(Aziz et al 2017) but modern genotypes enhanced root al-location to deep soil layers under drought However this pat-tern has not been observed consistently (Cholick et al 1977Feil 1992 Lupton et al 1974) To the best of our knowl-edge there is no information about the effect of breeding onchanges in subsoil OC dynamics and root respiration

One reason for the lack of quantitative data about the ef-fects of rooting depth on SOC sequestration is related todifficulties in measuring the amount of carbon transferredfrom roots to the soil (gross rhizodeposition) and the pro-portion of carbon that is eventually stabilized there (net rhi-zodeposition) after a portion of gross rhizodeposits are lostfrom the soil through microbial mineralization or leachingThe fact that rhizodeposition occurs below the soil surfacegreatly prevents direct observations of this ldquohidden half ofthe hidden halfrdquo of the SOC cycle (Pausch and Kuzyakov2018) First of all direct measurements of root exudationrates are hampered by the fact that rhizodeposits are usedby rhizosphere microorganisms within a couple of hours af-ter they are released resulting in very low concentrations ofroot carbon exudates in the soil (Kuzyakov 2006) Secondthe release of carbon exudates by agricultural crops is notequally divided throughout the growing season but mainlyoccurs in the first 1ndash2 months of the growing period and de-creases sharply thereafter (Gregory and Atwell 1991 Keithet al 1986 Kuzyakov and Domanski 2000 Pausch andKuzyakov 2018 Swinnen et al 1994) Third measurementsof the effects of rhizodeposits on changes in SOC stocks arefurther complicated by the priming effect ie their positiveeffect on the mineralization of native SOC (Fontaine et al2007 de Graaff et al 2009)

To overcome these difficulties rates of C rhizodepositioncan be measured by labeling plants with 13CO2 or 14CO2(Jones et al 2009 Kuzyakov and Domanski 2000) and sub-sequently tracing the amount of photosynthetically assimi-lated 13C or 14C label in the soil at the end of the growingseason (Kong and Six 2010 2012) However the continuousapplication of 13CO2 or 14CO2 during the course of an entiregrowing season to plants is often not feasible as this requiresthe setup of open-top chambers while continuously supply-ing the crops with the isotopic label which comes at a high fi-nancial cost Therefore plants are commonly labeled at fixedtime intervals during the growing season (repeated pulse la-beling) This results in reliable estimates of the partition-ing of assimilated carbon to different plant compartmentsas well as into the soil (Kong and Six 2010 Kuzyakov andDomanski 2000 Sun et al 2018)

In addition assessing the magnitude of the carbon transferfrom roots to the soil is not straightforward particularly un-der field conditions While carbon inputs from crops to thesoil are often derived from yield measurements (Keel et al2017 Kong et al 2005 Taghizadeh-Toosi et al 2016) thesequantities are often poorly related to root biomass or the mag-nitude of root exudates (Hirte et al 2018 Hu et al 2018)A better understanding of the factors controlling the rates ofcarbon rhizodeposition by different agricultural crops is thusnecessary to assess how different crops affect SOC cyclingand to provide SOC models with reliable rates of carbon in-puts to the soil

The present study addresses the following research ques-tion do wheat cultivars with shallow roots and lower rootbiomass lead to less net carbon rhizodeposition compared towheat cultivars with deeper roots and higher root biomassTo address this question four different bread wheat cultivarsfrom a century of Swiss wheat breeding (Fossati and Bra-bant 2003 Friedli et al 2019) were grown in large meso-cosms which allowed us to study the plantndashsoil system undercontrolled conditions that closely resemble a field situationWe hypothesized that wheat cultivars with shallow roots andlower root biomass would result in less net carbon rhizode-position over the course of a growing season compared tocultivars with deeper roots and higher root biomass

2 Materials and methods

21 Experimental setup

211 ETH mesocosm platform

To assess the effect of wheat root characteristics on netrhizodeposition in a realistic soil environment under con-trolled conditions an experiment was set up at the meso-cosm platform of the Sustainable Agroecosystems Group atthe Research Station for Plant Sciences Lindau (ETH ZuumlrichSwitzerland) The platform was located inside a greenhouse

Biogeosciences 17 2971ndash2986 2020 httpsdoiorg105194bg-17-2971-2020

M Van de Broek et al The soil organic carbon stabilization potential of old and new wheat cultivars 2973

and consisted of 12 cylindrical lysimeters with a diameterof 05 m and a height of 15 m constructed using 10 mmwide polyethylene (Fig S1 in the Supplement) The lysime-ters were equipped with probes installed at five differentdepths (0075 030 060 090 and 120 m below the surface)to measure the volumetric moisture content at a temporalresolution of 30 min (ECH2O EC-5 Decagon Devices US)and to sample soil pore water (Prenart Frederiksberg Den-mark) and soil pore air (Membrana Wuppertal Germany)The lysimeters were filled with mechanically homogenizedsoil collected from an agricultural field in Estavayer-le-LacSwitzerland The upper 015 m of the lysimeters was filledwith topsoil collected from the A horizon while the remain-der (015ndash135 m depth) was filled with subsoil The bottom015 m of the lysimeters (135ndash150 m depth) consisted ofa layer of gravel (Blaumlhton Erik Schweizer Switzerland) tofacilitate drainage of soil water through the bottom of thelysimeters The top and subsoil had a sandy clay loam tex-ture with 21 silt 21 clay and 58 sand and top- andsubsoil pH values were 78 and 75 respectively The OCconcentration of the top- and subsoil was 077plusmn 001 and040plusmn001 respectively with a C N ratio of 69 and 50respectively No carbonates were detected in the soil

At the top of each lysimeter pneumatically activatedchambers were placed which were automatically closedwhen applying the 13CO2 label (see Sect 213) Thesechambers were made of stainless steel with fitted plexiglasspanes and covered a rectangular area of 05mtimes05m with aninitial height of 01 m Chamber heights were extended withincreasing plant height using one or two height extensionsof 05 m each (Fig S1)

212 Wheat cultivars and growth conditions

Four wheat (Triticum aestivum L) cultivars from the Swisswheat breeding program (Fossati and Brabant 2003 Friedliet al 2019) with different breeding ages were selectedMont-Calme 268 (introduced in 1926) Probus (1948) Zi-nal (2003) and CH Claro (2007) Generally more recent cul-tivars of this program on average have more shallow rootsand lower root biomass under well-watered conditions com-pared to the older cultivars (Friedli et al 2019) CH Clarowas selected as a modern variety with relatively deep root-ing

Before the wheat plants were transplanted to the lysime-ters wheat seeds were germinated in a greenhouse for 2ndash3 d on perforated antialgae foil laid over 2 mm moistenedfleece at a warm temperature (20 C during day and 18 Cduring night) and good light conditions Next the seedlingswere planted in containers filled with the same topsoil usedto fill the lysimeters and transferred to a climate chamberfor vernalization for 52 d (Baloch et al 2003) First theseedlings were kept 45 d at 4 C with 8 h of light per day anda light intensity of 10 klx During the 3 subsequent days day-light intensity was increased to 36 klx daytime temperature

was increased to 12 C and night temperature was increasedto 10 C During the last 4 d daytime temperature was in-creased to 16 C and night temperature to 12 C The rela-tive humidity was maintained at 60plusmn 10 during the entirevernalization period After vernalization 70 seedlings weretransplanted to every lysimeter corresponding to a plant den-sity of 387 plants mminus2 At the timing of transplanting theplants were at the onset of tillering

The experimental setup consisted of a randomized com-plete block design Each of the four wheat cultivars wasplanted in three lysimeters ie three replicates per cultivarresulting in a total of 12 lysimeters These were placed inthree blocks of four rows where each wheat cultivar wasplanted in one lysimeter in each block The plants weregrown in the greenhouse for about 5 months between 24 Au-gust 2015 and 1 February 2016 Despite uneven maturingof plants within and between the lysimeters all plants hadreached flowering stage at the time of harvest Fertilizer wasapplied to the soil lysimeters a first time on 5 October 2015at a rate of 84 kg N haminus1 36 kg P2O5 haminus1 48 kg K2O haminus1

and 9 kg Mg haminus1 and a second time on 4 December 2015 ata rate of 56 kg N haminus1 24 kg P2O5 haminus1 32 kg K2O haminus1 and6 kg Mg haminus1 The lysimeters were watered manually twiceper week with a similar amount of water to keep soil moistureclose to field capacity Differences in the amount of waterused by the different cultivars resulted in differences in thesoil water content between the cultivars (Fig S2) The tem-perature in the greenhouse was set to 20 C during the dayand 15 C during the night During the experiment the aver-age temperature in the greenhouse was 169 C with a mini-mum and maximum of 93 and 298 C respectively The av-erage humidity was 637 with a minimum and maximumof 353 and 864 respectively

213 Repeated 13C pulse labeling

In order to study carbon allocation within the atmospherendashplantndashsoil system a 13C-pulse-labeling approach was used99 at 13CO2 (Euriso-top Saint-Aubin France) was ap-plied once per week (1400 local time on Thursdays) by in-jecting 15 56 or 98 mL CO2 into each chamber depending onthe chamber extension used in order to yield a target 13CO2content of 58 at A weekly labeling frequency has beenshown to ensure a sufficient abundance of root-derived 13Cin the soil at the end of the experiment (Bromand et al 2001Kong and Six 2010) After chamber closure CO2 concen-tration in one chamber was monitored using a CO2 analyzer(Li-820 LICOR Lincoln US) Throughout the experimentCO2 concentrations were measured in the same chamber Af-ter the CO2 concentration dropped below 200 ppm another13CO2 pulse was injected to yield a postlabel CO2 concen-tration of 570 ppm in the chamber headspace The chamberlids were kept closed for 2 h after label injection to achievesufficient uptake and then reopened to avoid condensationOn the same day of pulse labeling all chambers were closed

httpsdoiorg105194bg-17-2971-2020 Biogeosciences 17 2971ndash2986 2020


overnight to recuperate 13C lost through night respiration andallowed to be taken up by the plants in the morning before re-opening the chambers

22 Measurements

221 Belowground CO2 concentration and δ13CO2

Soil gas sampling was performed once per week (Wednes-days) by attaching a pre-evacuated 110 mL crimp serum vialto a sampling port at each depth leaving it equilibratingovernight For each sample a 20 mL subsample was trans-ferred to a pre-evacuated Labco exetainer (12 mL) and usedto determine the CO2 concentration The CO2 concentra-tion of each sample was determined using a gas chromato-graph equipped with a thermal conductivity detector (Bruker456-GC Germany) In addition the δ13C value of CO2 wasmeasured for CO2 samples collected along the depth pro-files on the last sampling date using a GasBench II modifiedas described by Zeeman et al (2008) coupled to a Deltaplus

XP isotope ratio mass spectrometer (IRMS ThermoFisherGermany) The standard deviation of the measurements waslt 015 permil

222 Sampling and general soil analyses

At the end of the experiment the aboveground biomass of thewheat plants was harvested separately for each lysimeter andseparated into leaves ears and stems Soil from the lysime-ters was collected by destructive sampling to analyze bulkdensity root biomass and other soil properties The samplingwas done layer by layer After a soil layer had been sam-pled it was removed completely from the lysimeter and thenext layer was sampled From each depth increment (0ndash015015ndash045 045ndash075 075ndash105 105ndash135 m depth) fivesoil cores were collected per lysimeter using a soil core sam-pler (508 cm diameter Giddings Machine Company IncWindsor CO US) Three of the five cores per lysimeterand depth increment were used for the determination of rootbiomass based on a combination of buoyancy and sievingthrough a 530 microm sieve using a custom-built root washingstation The remaining two soil cores were sieved at 8 mmair-dried and stored for further analysis Prior to air-dryingthe fresh weight and volume for each core was determinedand a subsample was taken for the determination of gravimet-ric soil moisture content Bulk density was calculated basedon fresh weight gravimetric moisture content and core vol-ume Soil texture was measured using a particle size analyzer(LS 13 320 Beckman Coulter Indianapolis USA) Prior toanalysis 01 g of soil was shaken for 4 h with 4 mL of 10 sodium hexametaphosphate and sonicated for 1 min

223 Soil microbial biomass

Soil microbial biomass was extracted from soil samples thathad been frozen at minus20 C for 6 months immediately after

sampling Two subsamples of 40 g were taken from eachsieved soil sample One set was fumigated for 24 h usingchloroform Next total dissolved OC was extracted fromeach fumigated and nonfumigated subsample by shaking itin 200 mL 005 M K2SO4 for 1 h prior to filtering through aWhatman 42 filter paper Total OC concentrations in K2SO4extracts were determined using a CN analyzer (multi NC2100S analyzer Analytik Jena Germany) To determine mi-crobial biomass carbon per unit of dry soil the gravimetricsoil water content was determined by drying about 10 g ofeach soil sample at 105 C and subtracting the weights be-fore and after drying The carbon content of the soil micro-bial biomass was calculated according to Vance et al (1987)as

TOCMB =TOCFminusTOCNF

045 (1)

where TOCF and TOCNF are the total OC in fumigated andnonfumigated samples respectively The remainder of the fil-tered samples was freeze-dried in order to analyze the δ13Cvalue The δ13C value of soil microbial biomass was calcu-lated using mass balance according to Ruehr et al (2009)

δ13CMB =

(δ13CF middotCFminus δ

13CNF middotCNF)

CFminusCNF (2)

where CF and CNF represent total organic carbon content ofthe fumigated and nonfumigated samples respectively

224 Organic carbon concentration and isotopiccomposition of plant material soil organic carbonand microbes

The OC concentration and isotopic composition (δ13C) ofabove- and belowground plant material extracts from fumi-gated and nonfumigated soil and bulk soil were measuredby weighing 2 4 50 50 and 100 mg respectively of eachsample into Sn capsules (9mmtimes 5mm Saumlntis CH) for anal-ysis with a Flash EA 1112 series elemental analyzer (Ther-moFisher Germany) coupled to a Deltaplus XP IRMS via aConFlo III (Brooks et al 2003 Werner et al 1999 Wernerand Brand 2001) The measurement precision (SD) of thequality control standards tyrosine Tyr-Z1 and caffeine Caf-Z1 was 037 (permil) and 008 (permil) respectively for above- andbelowground plant material microbes and the soil samples

23 Data processing

231 Excess 13C calculations

The mass of 13C label that was recovered in (i) the above-ground vegetation (ii) roots of wheat plants and (iii) the soilwas calculated following Studer et al (2014)

mE(

13C)=

χE(13C

)middotm(C) middotM

(13C)

χ(

12C)middotM

(12C

)+χ

(13C

)middotM

(13C

) (3)



where mE(13C

)is the mass of recovered 13C label (g mminus2)

χE(13C

)is the excess atom fraction (unitless calculated fol-

lowing Coplen 2011) m(C) is the total mass (g mminus2) of CM

(12C)

and M(13C

)are the molar weight of 12C and 13C

(g molminus1) respectively and χ(12C

)and χ

(13C)

are the 12Cand 13C atom fraction (unitless) respectively

To calculate the excess atom fraction (χE(13C)) of thesoil compartment the isotopic composition of the soil atthe start of the experiment was used as the reference value(minus2645plusmn 004 permil for the topsoil minus2501plusmn 013 permil for thesubsoil) As all lysimeters were labeled with 13CO2 no con-trol treatment for the wheat plants was present Therefore aδ13C reference value of minus28 permil was assumed for the above-ground parts and roots of all wheat plants The calculation ofexcess 13C is very sensitive to variability in input parametervalues including the δ13C value of plant biomass and soilTherefore a sensitivity analysis was used to show that vary-ing the initial δ13C value of the wheat plants with plusmn3 permil atypical range over which δ13C values can vary in the field be-cause of eg precipitation (Kohn 2010) led to changes incalculated mE(13C) on the order of plusmn1 for abovegroundbiomass and plusmn 1 ndash5 for belowground biomass The ef-fect of the initial δ13C value of the biomass on the calculatedamount of recovered 13C label in the wheat plants was thuslimited Calculations of the effect of wheat cultivar on below-ground excess 13C were only performed for the upper 045 mof the lysimeters as missing data for deeper soil layers pre-vented including these layers in the statistical analyses andthe majority of root biomass was present in the upper 045 m(Fig 1)

232 Net carbon rhizodeposition

The absolute amount of carbon rhizodeposition for the dif-ferent depth segments in the lysimeters was calculated fol-lowing Janzen and Bruinsma (1989)

Rhizodeposition C=χE

(13C)

soil

χE(

13C)

root

middotCsoil (4)

where rhizodeposition C is expressed in grams per kilo-gram (g kgminus1) for the considered layer χE

(13C)

soil andχE

(13C)

root are the excess 13C atom fraction in the soil androots respectively calculated as described in Sect 231 andCsoil is the OC concentration of the considered soil layer(g kgminus1) This approach assumes that the isotopic enrichmentof rhizodeposits and roots is equal The absolute amount ofcarbon rhizodeposition for each soil layer was calculated bymultiplying rhizodeposition C (g kgminus1) with the carbon con-tent (kg) present in each of the respective layers We note thatdata on the C concentration and δ13C value of root biomasscould not be obtained from a number of soil layer below045 m depth for certain cultivars due to the limited rootbiomass that could be retrieved Therefore net C rhizodepo-sition was only calculated for the two uppermost soil layers

(0ndash045 m depth) as only for these layers all necessary datato calculate net C rhizodeposition were available for the threereplicates of every cultivar

233 Subsoil CO2 production

Depth profiles of subsoil CO2 production in the lysimeterswere calculated using the weekly measured depth profilesof CO2 concentration throughout the experiment To assessthe variability among the different lysimeters these calcu-lations were performed separately for every lysimeter andaverage CO2 production depth profiles were calculated foreach cultivar Measurements of CO2 concentration soil tem-perature and soil moisture content were performed at discretedepths (0075 030 060 090 and 120 m depth) Continu-ous depth profiles of these variables at a vertical resolution of005 m were obtained using linear interpolation Depth pro-files of CO2 production were calculated using the discretizedform of the mass balance equation of CO2 in a diffusive one-dimensional medium following Goffin et al (2014)

P(z)i =1(εi[CO2]i)

1t+Ftopi minusFboti

1z (5)

where P (z) is the CO2 production in layer i

(micromol CO2 mminus3 sminus1) over time span 1t t is the time(s) εi is the air-filled porosity in layer i (m3 mminus3) [CO2]i isthe CO2 concentration of layer i (micromol CO2 mminus3) Ftopi andFboti are the CO2 fluxes transported through the upper andlower boundaries of layer i (micromol CO2 mminus2 sminus1) during timespan 1t respectively and z is the depth (m) The verticalCO2 fluxes are calculated as (Goffin et al 2014)

Ftopi = minusDsiminus1i[CO2]iminus1minus [CO2]i

1z (6)

Fboti =minusDsii+1[CO2]i minus [CO2]i+1

1z (7)

where Dsij is the harmonic average of the effective diffu-sivity coefficient (Ds) between layers i and j and 1z is thelayer thickness The effective diffusivity coefficient is calcu-lated using a formula appropriate for repacked soils (Mol-drup et al 2000)

Dsi =D0tε25it

8i (8)

where D0 is the gas diffusion coefficient of CO2 in freeair over time span 1t (m2 sminus1) εi is the air-filled porosityof layer i over time span 1t (m3 mminus3) and 8i is the to-tal soil porosity of layer i (m3 mminus3) The total soil porositywas calculated as 8i = 1minus ρiρp where ρi is the soil bulkdensity (t mminus3) and ρp is the particle density (265 t mminus3)Due to the large vertical variability in measured bulk densitydepth profiles a constant bulk density profile was assumedfor the subsoil (below 015 m depth) calculated as the av-erage of the measured bulk density values for these layers



The air-filled porosity over time span 1t was calculated asthe difference between the total porosity (m3 mminus3) and theaverage measured water-filled pore space over time span 1t(m3 mminus3) The latter was measured throughout the experi-ment (see Sect 211) and corrected based on differences be-tween these measurements at the end of the experiment andthe measured volumetric water content of the sampled soil atthe end of the experiment For this purpose different correc-tion equations were used for (i) the upper soil layer (0ndash15 cm)and (ii) all deeper layers combined

The gas diffusion coefficient in free air was corrected forthe individual lysimeters for variations in temperature andsoil moisture throughout the experiment (Massman 1998)as

D0 =D0stpp0

p

(T

T0

)α (9)

where D0stp is the gas diffusion coefficient for CO2 in freeair under standard temperature (0 C) and pressure (1 atm)(1385times 10minus5 m2 sminus1 Massman 1998) and α is a coeffi-cient (181 Massman 1998) Semicontinuous measurementsof soil temperature in every lysimeter were used to calculateD0 values throughout the experiment while a constant at-mospheric pressure of 1 atm throughout the experiment wasassumed

To obtain depth profiles of the total amount of CO2 pro-duced by the different wheat cultivars during the experiment(expressed as g CO2 mminus2) the calculated CO2 productionrates between all measurement days (P(z)) were summedfor the time span of the experiment and converted to gramsof CO2 per square meter (g CO2 mminus2) using the molecularmass of CO2 (4401 g molminus1) We applied the boundary con-dition of the absence of a flux of CO2 at the bottom of thelysimeters It is noted that these calculations do not make adistinction between the source of CO2 production therebycombining both autotrophic and heterotrophic CO2 produc-tion (total soil respiration) For more information about thesemethods reference is made to Goffin et al (2014)

24 Statistics

To account for the three blocks in the randomized com-plete block design statistically significant differences be-tween aboveground characteristics of different cultivars werechecked using a two-way analysis of variance (ANOVA)without interactions (Dean et al 2015) followed by a Tukeytest based on the values obtained for the individual repli-cates (n= 3 for every cultivar) using a significance levelof 005 This was done after checking for homogeneity ofvariance (Levenersquos test) and normality (ShapirondashWilk test)using a confidence level of 005 These analyses were per-formed in MATLABreg The effects of cultivar and depth onsoil bulk density belowground biomass belowground C al-location and net C rhizodeposition were assessed using a lin-ear mixed-effects model with cultivar and depth being fixed

effects and blocks being treated as a random effect (lmerfunction in R R Core Team 2019) Next a pair-wise com-parison was used to check for statistically significant differ-ences between the cultivars (emmeans package in R) Below-ground biomass was log-transformed to increase normalityand homogeneity of variances for the latter analysis Uncer-tainties on reported variables are expressed as standard errors(n= 3)

3 Results

31 Aboveground biomass

The aboveground biomass produced at the end of the experi-ment was significantly different between Zinal and Probuswhile the aboveground biomass of CH Claro and Mont-Calme 268 was not significantly different from any othercultivar (Fig 1 Table 1) The biomass of the ears was sig-nificantly higher for Zinal compared to CH Claro Probusand Mont-Calme (Fig 1 Table S1 in the Supplement) It isnoted that these data should be interpreted with care sincenot all plants reached maturity at the time of harvest and arepotentially not representative for the biomass of the ears offull-grown plants No significant differences were found be-tween the δ13C values of aboveground biomass of the dif-ferent cultivars (Fig 2) The high δ13C values of the above-ground biomass of all wheat cultivars (266 permil on average)showed that a substantial amount of the 13CO2 tracer wasincorporated by all wheat plants (Fig 2)

32 Belowground biomass

The average root biomass was highest in the topsoil and sig-nificantly lower in the subsoil layers of all four wheat culti-vars (Fig 1b) Root biomass of Zinal was significantly lowercompared to the root biomass of Probus and Mont-Calme268 while the root biomass of CH Claro was not signif-icantly different from any of the other cultivars (Fig 1b)These differences were mostly present in the two uppermostsoil layers while root biomass was not significantly differ-ent between different cultivars at any depth except for Zi-nal and Mont-Calme 268 between 045 and 075 m depth(Fig 1) The root shoot ratio varied between 010plusmn002 and019plusmn 008 and was not significantly different between thedifferent cultivars (Table 1)

The depth profiles of the δ13C of root biomass were differ-ent between the old and more recent wheat cultivars (Fig 2)In the two uppermost soil layers no significant differenceswere detected between the δ13C values of root biomass of thedifferent cultivars These differences could not be checkedfor statistical significance in deeper soil layers due to a lackof sufficient recovered root biomass in each lysimeter Theδ13C values of the roots of the old wheat cultivars showedonly limited variation with depth with values between ca150 permil and 200 permil In contrast the δ13C values of the roots



Table 1 Characteristics (plusmn standard error n= 3) of the biomass of the different wheat cultivars at the end of the experiment Values thatshare a letter in the same column are not significantly different

Cultivar Aboveground biomass Root biomass Root shoot ratio

(year of release) Biomass (g mminus2) OC Biomass (g mminus2) OC

CH Claro (2007) 1064plusmn 207ab 405plusmn 03ab 107plusmn 28ab 387plusmn 19a 010plusmn 002a

Zinal (2003) 710plusmn 114b 400plusmn 014a 97plusmn 20b 391plusmn 11lowast 014plusmn 001a

Probus (1948) 1154plusmn 220a 419plusmn 02b 161plusmn 54a 381plusmn 05a 013plusmn 003a

Mont-Calme 268 (1926) 1119plusmn 174ab 408plusmn 04ab 205plusmn 67a 368plusmn 13a 019plusmn 008a

lowast Was excluded from statistical analysis due to missing data

Figure 1 Aboveground (a) and root (b) biomass of the differentwheat cultivars at the end of the experiment Bars represent the av-erage per wheat cultivar error bars show the standard error (n= 3)and circles show the individual data points The inset in (b) showsa detail of the subsoil root biomass If statistically significant differ-ences were present these are indicated with letters with variablessharing a letter not being significantly different

of the more recent wheat cultivars were highest in the two up-permost soil layers (0ndash45 cm) and showed an abrupt decreasewith depth in deeper soil layers

33 Soil and soil organic carbon characteristics

The SOC concentration in the lysimeters was similar to theOC concentration of the initial soil (Fig 3a) A direct com-parison between the SOC concentration before and after theexperiment could not be made as no measurements of theOC concentration of the soil in the lysimeters before the startof the experiment could be made However the SOC concen-tration measured at the different depths in the lysimeters wassimilar to the OC concentration measured on the soil that wasused to fill the lysimeters (Fig 3a) No statistically signifi-cant differences in SOC concentration were found betweenthe different cultivars at any depth

The SOC in the two uppermost soil layers (0ndash45 cm) ofall wheat cultivars was enriched in 13C compared to the soilthat was used to fill the lysimeters (Fig 3b) Although theδ13C value of SOC was not significantly different at anydepth between any of the cultivars the largest increase inthe δ13C value of topsoil OC was observed for Probus andMont-Calme 268 (Fig 3b) indicating that the soil under theold cultivars incorporated more of the 13C label comparedto the more recent cultivars The limited difference between(i) the δ13C values of the soil used to fill the lysimetersand (ii) the measurements at the end of the experiment be-low a depth of 045 m indicates a lower amount of incor-porated 13C label in the subsoil Similarly the δ13C valueof topsoil microbial biomass was more positive compared todeeper soil layers for all cultivars indicating that microbesutilized more substrate enriched in 13C in the two upper-most soil layers compared to deeper soil layers (Fig 3c)Statistically significant differences were only detected in thelayer between 015 and 045 m depth where the δ13C valueof microbial biomass under Zinal was significantly lowercompared to Mont-Calme 268 However as the microbialbiomass under Zinal was substantially higher compared tounder Mont-Calme 268 in this layer (Fig S3) this does notnecessarily imply that microbes under Mont-Calme 268 in-corporated more excess 13C compared to under Zinal Depth



Figure 2 δ13C values of aboveground (a) and belowground (b)biomass for the different wheat cultivars at the end of the exper-iment Bars (a) and symbols (b) represent the average per wheatcultivar error bars show the standard error (n= 3) and symbolswithout error bars indicate samples for which no three replicateswere available Circles show the individual data points If statisti-cally significant differences were present for root biomass at thesame depth these are indicated with letters with variables sharinga letter not being significantly different and data points without errorbars being left out of the analyses

profiles of microbial biomass carbon were relatively constant(200ndash500 microg C g soilminus1) with no consistent differences be-tween different cultivars (Fig S3) The δ13C values of soilCO2 (δ13CO2) at the end of the experiment were similar forall wheat cultivars for the two uppermost layers (0ndash045 m)(Fig 3d) Deeper down the profile the CO2 under the oldwheat cultivars was more enriched in 13C compared to themore recent cultivars by an average of ca 30 permil The onlystatistically significant differences were detected in the low-ermost layer where the δ13C values of CO2 of Zinal andCH Claro were significantly lower compared to Mont-Calme268

There was no significant effect of cultivar on the bulk den-sity of the soil at the end of the experiment (F359 = 19p = 023) while there was a significant effect of depth onbulk density (F459 = 194 p lt 00005) The average bulkdensity of all lysimeters was highest in the topsoil (167plusmn012 Mg mminus3) and varied with depth (Fig S4a) The gravi-metric moisture content in the lysimeters at the end of theexperiment increased with depth for all cultivars from ca01 g gminus1 in the top layer to ca 015 g gminus1 in the bottomlayer (Fig S4b) and was only significantly different betweenMont-Calme 268 and Zinal in the uppermost soil layer Thesoil moisture content changed relatively little throughout theexperiment for all lysimeters after an initial phase of de-creasing soil moisture content at the onset of the experiment(Fig S2)

34 Excess 13C and carbon rhizodeposition

The total amount of 13C label that was present in the plantndashsoil system at the end of the experiment expressed as ex-cess 13C differed significantly between different wheat cul-tivars (Fig 4a) When accounting for excess 13C in above-ground biomass and in the soil and roots down to a depth of045 m the lowest amount of 13C label was found in the Zinallysimeters (119plusmn011 g mminus2) followed by CH Claro (164plusmn006 g mminus2) and the older wheat cultivars (205plusmn009 g mminus2

for Mont-Calme 268 and 201plusmn 019 g mminus2 for Probus)with the majority of 13C tracer in the aboveground biomass(Fig 4a) Despite these differences the relative distribu-tion of the assimilated 13C between aboveground biomassroots and soil was similar between the different wheat cul-tivars (Fig 4b) On average 807plusmn 17 of the assimilatedtracer ended up in aboveground biomass 84plusmn 15 in rootbiomass and 109plusmn 14 in the soil It is noted that root-respired 13C label is not included in this analysis which maylead to an underestimation of the fraction of 13C label thatwas allocated belowground

The total amount of net carbon rhizodeposition measuredat the end of the experiment down to 045 m decreasedwith depth for all wheat cultivars (Fig 4c) with this dif-ference only being statistically significant for CH ClaroThe highest amount of net carbon rhizodeposition was ob-served for Probus (108plusmn34 g C mminus2) followed by CH Claro(97plusmn24 g C mminus2) Mont-Calme (83plusmn29 g C mminus2) and Zinal(62plusmn 11 g C mminus2) There was thus no clear relationship be-tween the amount of net carbon rhizodeposition and year ofrelease of the wheat cultivars

35 CO2 concentration and production

Throughout the experiment the change in the CO2 concen-tration of the two uppermost soil layers was limited withaverage values for the topsoil between 470 and 761 ppmfor all cultivars (Fig 5) Deeper down the lysimeters rela-tively constant CO2 concentrations were observed during the



Figure 3 Depth profiles of organic carbon concentration (a) the δ13C value of organic carbon (b) the δ13C value of microbial biomass (c)and the δ13C value of soil CO2 (d) averaged per wheat cultivar at the end of the experiment Error bars represent the standard error (n= 3)lowast The initial soil indicates measurements performed on the soil that was used to fill the lysimeters prior to the experiments in (a) and (b) andmeasurements of the δ13CO2 depth profile at the beginning of the experiment in (d)

Figure 4 Absolute (a) and relative (b) distribution of excess 13C between aboveground biomass root biomass and soil for the differentwheat cultivars Soil compartments are calculated down to 045 m depth Panel (c) shows the total carbon rhizodeposition and root carbonfor the upper two soil layers for the different wheat cultivars (0ndash045 m depth) Error bars represent the standard error (n= 3) In (a) lettersindicate significant differences between the total amount of excess 13C of the different cultivars No significant differences in the amount ofexcess 13C in aboveground biomass root biomass or soil between the different cultivars were found In (c) letters are provided when thetotal belowground C allocation differed between the different depth layers which was only the case for CH Claro



Figure 5 Changes in the CO2 concentration (ppm) in the lysimetersthroughout the experiment for the four wheat cultivars The averageCO2 concentration of three replicates are shown (n= 3) Dots indi-cate the measured data points

first 3 weeks of the experiment ca 5000ndash10 000 ppm After3 weeks subsoil CO2 concentrations abruptly increased andremained high throughout the experiment These were sub-stantially larger for the older cultivars (with maximum valuesof ca 30 000 ppm) compared to the younger cultivars (withmaximum values ca 24 000 ppm)

Despite these high CO2 concentrations in the subsoil CO2production was mainly taking place in the topsoil with thehighest rates of CO2 production between 010 and 020 mdepth for all cultivars (Fig 6) For the young cultivars (Zinaland CH Claro) 95 of CO2 was produced above a depthof 03 m In contrast in older cultivars (Probus and Mont-Calme 268) 95 of CO2 was produced above a depth of055 and 06 m respectively Despite these observations nei-ther the calculated total amount of subsoil CO2 productionnor the depth above which 95 of CO2 was produced wassignificantly different between any of the cultivars

Figure 6 Depth profiles of calculated cumulative CO2 production(g CO2 mminus2 per 005 m depth layer) Dots show the calculated pro-duction rates thin lines show the calculated CO2 production for theindividual lysimeters and the thick lines show the average based onthree replicates (two for CH Claro)

4 Discussion

The aim of the present study was to assess differences in be-lowground carbon transfer and net rhizodeposition by wheatcultivars with different root biomass and rooting depth Ourresults show that although there are marked differences inboth the amount of carbon transferred belowground and thetiming of belowground carbon transfer there is no clear re-lationship between root characteristics and the amount of netrhizodeposition Therefore the fate of root biomass mightdetermine the total amount of subsoil carbon stabilization inthe long term

41 Plant biomass carbon dynamics and CO2production

No consistent differences in total aboveground biomass be-tween old and new wheat cultivars were observed Theaboveground biomass values were at the high end of reportedvalues for wheat plants in the field (Mathew et al 2017)while the lack of consistent differences in the biomass ofwheat cultivars released over a time span of multiple decadeshas generally been observed (Brancourt-Hulmel et al 2003Feil 1992 Lupton et al 1974 Wacker et al 2002)



The fraction of biomass in the grain-bearing ears washowever much larger for the modern wheat cultivars (on av-erage 9 and 47 of total aboveground biomass for CHClaro and Zinal respectively) compared to the old wheat cul-tivars (on average 1 and 2 for Mont-Calme 268 andProbus respectively) While an increase in the fraction ofbiomass allocated to grains is generally observed in old ver-sus modern wheat cultivars (Brancourt-Hulmel et al 2003Feil 1992 Shearman et al 2005) mostly as a consequenceof the introduction of reduced height genes (Tester and Lan-gridge 2010) the harvest index reported here for the old cul-tivars might have been underestimated because older culti-vars were not yet fully mature at plant harvest

The total root biomass of the older wheat cultivars wassubstantially larger compared to the more recent cultivarsalthough these differences were not consistently significantbetween all modern and old varieties (Table 1) These differ-ences were mostly apparent in the top 045 m of the lysime-ters (Fig 1) It is not clear if the lack of statistically sig-nificant differences in the root biomass within the deepersoil layers was due to (i) inability to collect all fine rootsfrom the soil or (ii) actual differences in root biomass Theseresults are in line with a recent study on the biomass ofroots of different wheat cultivars of the Swiss wheat breed-ing program including the cultivars used in our experiment(Friedli et al 2019) This study showed that under well-watered conditions older wheat cultivars had a substantiallyhigher root biomass compared to the more recently releasedwheat cultivars Similar results have been obtained for wheatcultivars released in eg Australia (Aziz et al 2017) andother countries around the world (Waines and Ehdaie 2007)The root shoot ratios of the wheat cultivars in our study(010plusmn 02ndash019plusmn 008 Table 1) were at the low end of re-ported values for wheat plants globally (Mathew et al 2017)but in line with reported values for wheat cultivars of theSwiss wheat breeding program including the cultivars usedin our study (an average value of 014 for all cultivars studiedby Friedli et al 2019)

The maximum rooting depth was similar between the oldand recent wheat cultivars (Fig 1b) This is in contrast withthe results from Friedli et al (2019) who found that the olderwheat cultivars had deeper roots (the depth above which 95 of roots were found (D95) was on average 101 cm) comparedto the more recent cultivars included in the present study (av-erage D95 of 85 cm) These differences might partly arisefrom the different setup used in both studies Both experi-ments were carried out in a controlled greenhouse environ-ment but Friedli et al (2019) used soil columns with a diam-eter of 011 m while in our study lysimeters with a diame-ter of 05 m were used Additional information about subsoilroot dynamics could be obtained from the measured depthprofiles of the CO2 concentration and 13CO2 with the lat-ter only being measured in the last phase of the experimentThe calculated depth profiles of CO2 production showed thatCO2 was being produced down to greater depths under the

old wheat cultivars (Fig 6) Combined with the higher δ13Cvalues of subsoil CO2 of the lysimeters under the old wheatcultivars at the end of the experiment (Fig 3d) this sug-gests that the roots of the old wheat cultivars respired CO2at greater depths compared to the recent wheat cultivars

The δ13C values of root biomass suggest that the temporalroot carbon dynamics of the old and recent wheat cultivarsdiffered substantially (Fig 2b) The root biomass of the oldwheat cultivars had a high δ13C value at all measured depthsindicating that the 13CO2 label was allocated to the roots atall depths throughout the experiment In contrast the rootbiomass of the recent wheat cultivars was greatly enrichedin 13C in the top 045 m while deeper roots were much lessenriched in 13C This suggests that both old and more re-cent wheat cultivars grew roots down to depths of gt 1 m inthe beginning of the experiment (when the total amount of13C assimilated by the plants was limited) while only theold cultivars kept on allocating carbon down to deep roots(gt 045 m) throughout the experiment (thus having assim-ilated more 13C over the period of root growth comparedto the more recent cultivars) The similar δ13C value of theaboveground biomass of all wheat cultivars (Fig 2a) suggeststhat the differences in δ13C values of the root biomass areunlikely to be caused by differences in the relative amount of13CO2 assimilated by the plants relative to unlabeled CO2Thus these results suggest that old wheat cultivars allocatephotosynthates down to their roots throughout a substantialpart of the plant growth phase while this is not the case formore recent cultivars

42 Carbon allocation by wheat plants

The partitioning of the 13C label was very similar betweenthe different wheat cultivars (Fig 4b) It is noted that theamount of rhizosphere-respired 13CO2 could not be includedin these calculations although this typically accounts forca 7 ndash14 of assimilated carbon in crops or 40 oftotal belowground C allocation (Kuzyakov and Domanski2000 Pausch and Kuzyakov 2018) The fraction of assim-ilated carbon that is transferred belowground reported hereis therefore underestimated The belowground transfer of ca20 of assimilated C for all cultivars is in line with previ-ous studies which have reported fractions of similar magni-tude for wheat plants when not accounting for rhizosphereCO2 respiration 18 ndash25 (Hirte et al 2018) 18 (as re-viewed by Kuzyakov and Domanski 2000) 15 (Keith etal 1986) 17 (Gregory and Atwell 1991) and 31 (Sunet al 2018) In contrast reported values of the partitioning ofbelowground translocated carbon by wheat plants to (i) rootsandor (ii) net rhizodeposition are much more variable withreported net rhizodeposition carbon as a percentage of to-tal belowground carbon (root carbon and net rhizodepositioncarbon combined) for wheat plants between 23 (as sum-marized by Kuzyakov and Domanski 2000) and 72 (Sunet al 2018) The results obtained here (68 ) are thus at



Table 2 Average belowground carbon allocation (net rhizodeposi-tion and root biomass combined) and net carbon rhizodeposition bythe different wheat cultivars calculated down to a depth of 045 m(variation is reported as the standard error n= 3) No statisticallysignificant differences in belowground C allocation or net C rhi-zodeposition between any of the cultivars were detected

Belowground Net CC allocation rhizodeposition

(g mminus2) (g mminus2)

CH Claro 131plusmn 26 97plusmn 24Zinal 97plusmn 14 62plusmn 11Probus 164plusmn 38 108plusmn 34Mont-Calme 268 154plusmn 39 83plusmn 29

the high end of reported values However they were similarto results from a field study in Switzerland which used twomodern Swiss wheat cultivars among which was CH Claro(58 Hirte et al 2018)

43 Rates of net carbon rhizodeposition

The total amount of carbon assimilated by the wheat cultivarsthat was transferred to roots and soil in the top 045 m at theend of the experiment ranged between 97plusmn14 g mminus2 (Zinal)and 164plusmn38 g mminus2 (Probus) (Table 2) It is noted that the to-tal amount of belowground carbon translocation by the wheatplants is underestimated as rhizosphere respiration could notbe included in our calculations These numbers are in therange of reported values for wheat plants of 94ndash295 g mminus2

(as summarized by Keith et al 1986) and the value reportedby Kuzyakov and Domanski (2000) (150 g mminus2) as well asthe reported amount for two recent wheat cultivars of theSwiss wheat breeding program (including CH Claro) of 110ndash134 g mminus2 (Hirte et al 2018)

In contrast to the total amount of carbon translocated be-lowground the amount of net carbon rhizodeposition was notconsistently different between the old and more recent wheatcultivars (62plusmn 11ndash108plusmn 34 g mminus2) (Fig 4 Table 2) Thesevalues are higher compared to values calculated by Pauschand Kuzyakov (2018) (18ndash34 g mminus2 depth unknown) andHirte et al (2018) (63ndash73 g mminus2 down to 075 m depth)

A large uncertainty associated with calculated valuesof subsoil carbon sequestration using isotopic labeling ap-proaches is related to the assumption that the isotopic en-richment of roots and rhizodeposits is similar (Eq 4) Thissimplification is made because of the difficulties in mea-suring quantitative characteristics of rhizodeposits in a soilmedium (Oburger and Jones 2018) but leads to erroneouscalculations of the amount of carbon rhizodeposition whenthis assumption is violated (Stevenel et al 2019) To as-sess the uncertainty of calculated values of subsoil carbonsequestration we calculated how these values differ whenthe value of root δ13C is varied with plusmn25 (Fig S5) This

results in calculated values of total carbon rhizodepositiondown to a depth of 045 m of 69ndash105 g mminus2 for Mont-Calme268 88ndash138 g mminus2 for Probus 51ndash78 g mminus2 for Zinal and81ndash121 g mminus2 for CH Claro or uncertainties in the amountof carbon rhizodeposits between minus18 and +28 Furtherresearch on the effect of the assumption of using root δ13Cvalues as a proxy for carbon rhizodeposits is thus necessaryto better quantify the effect on estimates of carbon sequestra-tion

44 The effect of old and recent wheat cultivars on netcarbon rhizodeposition

Our results indicate that the old wheat cultivars with deeperactive roots throughout the experiment and larger rootbiomass allocated more assimilated carbon belowgroundalthough the differences were not statistically significant(Fig 4c Table 2) However we found no evidence that wheatcultivars with larger root biomass lead to higher net carbonrhizodeposition (Table 2) Our hypothesis which stated thatwheat cultivars with larger root biomass and deeper rootswould lead to larger amounts of net carbon rhizodepositioncould therefore not be confirmed

The total amount of OC that will be stabilized in the soilby the studied wheat cultivars will therefore depend on thelong-term fate of the root biomass The root biomass washigher for the old wheat cultivars although these differenceswere mainly limited to the upper 045 m of the soil Due tothe destructive sampling of vegetation and soil at the end ofthe experiment the fate of root biomass after harvest couldnot be assessed Based on the results one could thereforehypothesize that the higher root biomass of old wheat cul-tivars would lead to larger rates of carbon sequestration inthe long term Similarly Mathew et al (2017) suggested thatgrowing grasses and maize plants would lead to larger SOCstocks because these plants have the highest total and rootbiomass compared to growing crops with a lower biomassHowever it is not straightforward to make predictions aboutthe amount of root biomass that will be stabilized in the soilin the long term as this depends on the efficiency with whichplant-derived biomass is incorporated in microbial biomass(Cotrufo et al 2013) and interactions between soil depththe microbial community composition and its substrate pref-erence (eg Kramer and Gleixner 2008) among other fac-tors During the past century there has been a continuingincrease in the importance of wheat cultivars with smallerroot biomass (Fossati and Brabant 2003 Friedli et al 2019Waines and Ehdaie 2007) This can have profound implica-tions for OC stocks of soils under wheat cultivation as rhi-zodeposition and root-derived carbon are the most importantinputs of OC to the soil (Kong and Six 2010) Testing thelong-term effect of the gradual change in wheat cultivars onOC inputs to the soil would thus require experiments that runover multiple growing seasons and allow the quantification



of the amount of root carbon that is eventually stabilized inthe soil

Correct knowledge on the amount of OC that is transferredbelowground by plants is necessary to reliably model SOCdynamics However this knowledge is currently limited andchanges in belowground carbon allocation due to the culti-vation of different cultivars are generally not considered inSOC models Moreover it has recently been shown that ac-counting for changes in belowground carbon allocation byrelating this to changes in aboveground biomass does not im-prove model results (Taghizadeh-Toosi et al 2016) Ratherit has been suggested that more reliable model results are ob-tained when crop-specific amounts of belowground carbonallocation are used independent of aboveground biomassproduction (Taghizadeh-Toosi et al 2016) Since model re-sults are very sensitive to the amount of carbon inputs (Keelet al 2017) and cereal crops are grown on ca 20 of crop-lands globally (Leff et al 2004) (covering ca 12 of globalland mass and storing ca 10 of global SOC in the uppermeter of soil Govers et al 2013) a correct assessment ofa potential decrease in belowground carbon inputs by wheatplants over the past century through the cultivation of differ-ent cultivars will have important implications for the simula-tion of changes in SOC on the global scale

Assessing the overall impact of the past evolution of wheatcultivars on SOC stocks also requires taking into account theamount of land needed to produce sufficient food For exam-ple if future research would show that more recent wheatcultivars lead to less SOC stabilization compared to oldercultivars this does not necessarily imply a net loss of SOC asa consequence of the historical shift to planting recently de-veloped wheat cultivars If the aim is to increase overall SOCstocks it might be more favorable to grow high-yield wheatcultivars that sequester less OC per unit area compared to alow-yielding cultivar if this results in a larger area of arableland that can be taken out of cultivation This land can be putunder native vegetation such as forest or grassland whichstores substantially more SOC compared to arable land (Job-baacutegy and Jackson 2000)

5 Conclusion

In this study four different wheat cultivars were grown inlysimeters and labeled with 13CO2 using repeated pulselabeling to quantify the effect of rooting depth and rootbiomass on net carbon rhizodeposition Our results show thatthere is no clear trend between the time of cultivar devel-opment and the amount of net carbon rhizodeposition withlarge variabilities being observed between replicates of thesame cultivars Based on these results the hypothesis thatwheat cultivars with a larger root biomass and deeper rootswould promote net carbon rhizodeposition was rejected Animportant remaining uncertainty is related to the fate of root

biomass after harvest which might contribute to the stabi-lized SOC pool over the long term

Data availability Additional figures and tables can be found in theSupplement The data associated with this paper are available in theSupplement and at httpsdoiorg101763275f349dpdr2 (Van deBroek and Ghiasi 2020)

Supplement The supplement related to this article is available on-line at httpsdoiorg105194bg-17-2971-2020-supplement

Author contributions CD SA AH CF and JS conceived the ideafor the study CD set up the lysimeter and labeling experiments andcollected the data SG CD SA and RAW performed lab analysesMVdB SG CD and JS analyzed and interpreted the data and per-formed the statistics MVdB and SG wrote the paper with contri-butions from CD AH SA JS CF and RAW MVdB and SG con-tributed equally to this paper

Competing interests The authors declare that they have no conflictof interest

Acknowledgements The authors are very grateful to Matti BarthelBenjamin Wild and Christopher Mikita for their help with setting upthe lysimeters data collection and interpretation The authors alsoappreciate the Grassland Sciences Group of ETH Zuumlrich for provid-ing the laboratory facility to perform part of the analyses performedin this study We thank Brigitta Herzog and Hansueli Zellweger fortheir help with greenhouse management and plant protection at theETH Research Station for Plant Sciences in Lindau Stefan Kar-lowsky and one anonymous reviewer are thanked for their compre-hensive feedback which improved the quality of this paper

Financial support This research has been supported by Plant Fel-lows a postdoctoral fellowship administered by the Zurich BaselPlant Science Center and funded under the European UnionrsquosSeventh Framework Programme for research technological devel-opment and demonstration (grant no GA-2010-267243 ndash PlantFellows) the Swiss National Science Foundation (project num-bers 205321_153545 ldquoCarINrdquo and 200021_160232) and ETH corestart-up funds provided to Johan Six

Review statement This paper was edited by Sara Vicca and re-viewed by Stefan Karlowsky and one anonymous referee

References

Aziz M M Palta J A Siddique K H M and SadrasV O Five decades of selection for yield reduced rootlength density and increased nitrogen uptake per unit root



length in Australian wheat varieties Plant Soil 413 181ndash192httpsdoiorg101007s11104-016-3059-y 2017

Baloch D M Karow R S Marx E Kling J G and Witt M DVernalization Studies with Pacific Northwest Wheat Agron J95 1201ndash1208 httpsdoiorg102134agronj20031201 2003

Brancourt-Hulmel M Doussinault G Lecomte C Beacuterard P LeBuanec B and Trottet M Genetic improvement of agronomictraits of winter wheat cultivars released in France from 1946 to1992 Crop Sci 43 37ndash45 2003

Bromand S Whalen J K Janzen H H Schjoerring JK and Ellert B H A pulse-labelling method to gener-ate 13C-enriched plant materials Plant Soil 235 253ndash257httpsdoiorg101023A1011922103323 2001

Brooks P D Geilmann H Werner R A and Brand W AImproved precision of coupled δ13C and δ15N measurementsfrom single samples using an elemental analyzerisotope ra-tio mass spectrometer combination with a post-column six-portvalve and selective CO2 trapping improved halide robustnessof the combustion Rapid Commun Mass Sp 17 1924ndash1926httpsdoiorg101002rcm1134 2003

Chen S Martin M P Saby N P A Walter C Angers D Aand Arrouays D Fine resolution map of top- and subsoil carbonsequestration potential in France Sci Total Environ 630 389ndash400 httpsdoiorg101016jscitotenv201802209 2018

Chenu C Angers D A Barreacute P Derrien D Arrouays Dand Balesdent J Increasing organic stocks in agricultural soilsKnowledge gaps and potential innovations Soil Till Res 18841ndash52 httpsdoiorg101016jstill201804011 2018

Cholick F A Welsh J R and Cole C V Rooting Patternsof Semi-dwarf and Tall Winter Wheat Cultivars under DrylandField Conditions Crop Sci 17 637ndash639 1977

Coplen T B Guidelines and recommended terms for ex-pression of stable-isotope-ratio and gas-ratio measure-ment results Rapid Commun Mass Sp 25 2538ndash2560httpsdoiorg101002rcm5129 2011

Cotrufo M F Wallenstein M D Boot C M Denef K andPaul E The Microbial Efficiency-Matrix Stabilization (MEMS)framework integrates plant litter decomposition with soil or-ganic matter stabilization Do labile plant inputs form sta-ble soil organic matter Glob Change Biol 19 988ndash995httpsdoiorg101111gcb12113 2013

Dean A Morris M Stufken J and Bingham D (Eds) Hand-book of design and analysis of experiments CRC Press BocaRaton FL 2015

de Graaff M-A Van Kessel C and Six J Rhizodeposition-induced decomposition increases N availability to wild and culti-vated wheat genotypes under elevated CO2 Soil Biol Biochem41 1094ndash1103 httpsdoiorg101016jsoilbio2009020152009

Don A Schu J and Freibauer A Impact of tropical land-usechange on soil organic carbon stocks ndash a meta-analysis GlobChange Biol 17 1658ndash1670 httpsdoiorg101111j1365-2486201002336x 2011

Feil B Breeding Progress in Small Grain Cereals ndash A Comparisonof Old and Modern Cultivars Plant Breed 108 1ndash11 1992

Fontaine S Barot S Barreacute P Bdioui N Mary B andRumpel C Stability of organic carbon in deep soil lay-ers controlled by fresh carbon supply Nature 450 277ndash280httpsdoiorg101038nature06275 2007

Fossati D and Brabant C La seacutelection du bleacute en Suisse Le pro-gramme des stations feacutedeacuterales Une progression impressionnanteRev Suisse Agric 35 169ndash180 2003

Friedli C N Abiven S Fossati D and Hund A Modern wheatsemi-dwarfs root deep on demand response of rooting depth todrought in a set of Swiss era wheats covering 100 years of breed-ing Euphytica 215 1ndash15 httpsdoiorg101007s10681-019-2404-7 2019

Goffin S Aubinet M Maier M Plain C Schack-Kirchner Hand Longdoz B Characterization of the soil CO2 productionand its carbon isotope composition in forest soil layers usingthe flux-gradient approach Agr Forest Meteorol 188 45ndash57httpsdoiorg101016jagrformet201311005 2014

Govers G Merckx R Van Oost K and van Wesemael B Man-aging Soil Organic Carbon for Global Benefits A STAP Tech-nical Report Global Environmental Facility Washington DC2013

Gregory P J and Atwell B J The fate of carbon in pulse-labelled crops of barley and wheat Plant Soil 136 205ndash213httpsdoiorg101007BF02150051 1991

Guo L B and Gifford R M Soil carbon stocks and landuse change a meta analysis Glob Change Biol 8 345ndash360httpsdoiorg101046j1354-1013200200486x 2002

Hirte J Leifeld J Abiven S Oberholzer H-R and Mayer JBelow ground carbon inputs to soil via root biomass and rhi-zodeposition of field-grown maize and wheat at harvest are inde-pendent of net primary productivity Agr Ecosyst Environ 265556ndash566 httpsdoiorg101016jagee201807010 2018

Hu T Soslashrensen P Wahlstroumlm E M Chirinda N SharifB Li X and Olesen J E Root biomass in cereals catchcrops and weeds can be reliably estimated without consider-ing aboveground biomass Agr Ecosyst Environ 251 141ndash148httpsdoiorg101016jagee201709024 2018

Janzen H H and Bruinsma Y Methodology for the quantifica-tion of root and rhizosphere nitrogen dynamics by exposure ofshoots to 15N-labelled ammonia Soil Biol Biochem 21 189ndash196 httpsdoiorg1010160038-0717(89)90094-1 1989

Jobbaacutegy E G and Jackson R B The vertical distribution ofsoil organic carbon and its relation to climate and vegeta-tion Ecol Appl 10 423ndash436 httpsdoiorg1018901051-0761(2000)010[0423TVDOSO]20CO2 2000

Jones D L Nguyen C and Finlay R D Carbon flow in therhizosphere carbon trading at the soilndashroot interface Plant Soil321 5ndash33 httpsdoiorg101007s11104-009-9925-0 2009

Keel S G Leifeld J Mayer J Taghizadeh-Toosi Aand Olesen J E Large uncertainty in soil carbon mod-elling related to method of calculation of plant carbon in-put in agricultural systems Eur J Soil Sci 68 953ndash963httpsdoiorg101111ejss12454 2017

Keith H Oades J M and Martin J K Input of carbon to soilfrom wheat plants Soil Biol Biochem 18 445ndash449 1986

Kell D B Breeding crop plants with deep roots their role in sus-tainable carbon nutrient and water sequestration Ann Bot 108407ndash418 httpsdoiorg101093aobmcr175 2011

Kohn M J Carbon isotope compositions of terrestrialC3 plants as indicators of (paleo)ecology and (pa-leo)climate P Natl Acad Sci USA 107 19691ndash19695httpsdoiorg101073pnas1004933107 2010



Kong A Y Y and Six J Tracing Root vs ResidueCarbon into Soils from Conventional and AlternativeCropping Systems Soil Sci Soc Am J 74 1201httpsdoiorg102136sssaj20090346 2010

Kong A Y Y and Six J Microbial community assimilation ofcover crop rhizodeposition within soil microenvironments in al-ternative and conventional cropping systems Plant Soil 356315ndash330 httpsdoiorg101007s11104-011-1120-4 2012

Kong A Y Y Six J Bryant D C Denison R F andvan Kessel C The Relationship between Carbon Input Ag-gregation and Soil Organic Carbon Stabilization in Sustain-able Cropping Systems Soil Sci Soc Am J 69 1078ndash1085httpsdoiorg102136sssaj20040215 2005

Kramer C and Gleixner G Soil organic matter in soil depthprofiles Distinct carbon preferences of microbial groups dur-ing carbon transformation Soil Biol Biochem 40 425ndash433httpsdoiorg101016jsoilbio200709016 2008

Kuzyakov Y Sources of CO2 efflux from soil and reviewof partitioning methods Soil Biol Biochem 38 425ndash448httpsdoiorg101016jsoilbio200508020 2006

Kuzyakov Y and Domanski G Carbon Input By Plants inthe Soil Review J Plant Nutr Soil Sci 163 421ndash431httpsdoiorg1010021522-2624(200008)1634lt421AID-JPLN421gt30CO2-R 2000

Leff B Ramankutty N and Foley J A Geographic distributionof major crops across the world Global Biogeochem Cy 181ndash27 httpsdoiorg1010292003GB002108 2004

Lehtinen T Schlatter N Baumgarten A Bechini L KruumlgerJ Grignani C Zavattaro L Costamagna C and Spiegel HEffect of crop residue incorporation on soil organic carbon andgreenhouse gas emissions in European agricultural soils SoilUse Manage 30 524ndash538 httpsdoiorg101111sum121512014

Le Queacutereacute C Andrew R M Friedlingstein P Sitch S PongratzJ Manning A C Korsbakken J I Peters G P CanadellJ G Jackson R B Boden T A Tans P P Andrews OD Arora V K Bakker D C E Barbero L Becker MBetts R A Bopp L Chevallier F Chini L P Ciais PCosca C E Cross J Currie K Gasser T Harris I HauckJ Haverd V Houghton R A Hunt C W Hurtt G Ily-ina T Jain A K Kato E Kautz M Keeling R F KleinGoldewijk K Koumlrtzinger A Landschuumltzer P Lefegravevre NLenton A Lienert S Lima I Lombardozzi D Metzl NMillero F Monteiro P M S Munro D R Nabel J E MS Nakaoka S Nojiri Y Padin X A Peregon A Pfeil BPierrot D Poulter B Rehder G Reimer J Roumldenbeck CSchwinger J Seacutefeacuterian R Skjelvan I Stocker B D Tian HTilbrook B Tubiello F N van der Laan-Luijkx I T van derWerf G R van Heuven S Viovy N Vuichard N WalkerA P Watson A J Wiltshire A J Zaehle S and Zhu DGlobal Carbon Budget 2017 Earth Syst Sci Data 10 405ndash448httpsdoiorg105194essd-10-405-2018 2018

Lupton F G H Oliver R H Ellis F B Barnes B T HowseK R Welbank P J and Taylor P J Root and shoot growth ofsemi-dwarf and taller winter wheats Ann Appl Biol 77 129ndash144 1974

Massman W J A review of the molecular diffusivities of H2OCO2 CH4 CO O3 SO2 NH3 N2O NO and NO2 in

air O2 and N2 near STP Atmos Environ 32 1111ndash1127httpsdoiorg101016S1352-2310(97)00391-9 1998

Mathew I Shimelis H Mutema M and Chaplot V Whatcrop type for atmospheric carbon sequestration Results froma global data analysis Agr Ecosyst Environ 243 34ndash46httpsdoiorg101016jagee201704008 2017

Minasny B Malone B P McBratney A B Angers DA Arrouays D Chambers A Chaplot V Chen Z-SCheng K Das B S Field D J Gimona A Hedley CB Hong S Y Mandal B Marchant B P Martin MMcConkey B G Mulder V L OrsquoRourke S Richer-de-Forges A C Odeh I Padarian J Paustian K Pan GPoggio L Savin I Stolbovoy V Stockmann U Sulae-man Y Tsui C-C Varinggen T-G van Wesemael B andWinowiecki L Soil carbon 4 per mille Geoderma 292 59ndash86httpsdoiorg101016jgeoderma201701002 2017

Moldrup P Olesen T Gamst J Schjoslashnning P YamaguchiT and Rolston D E Predicting the Gas Diffusion Coeffi-cient in Repacked Soil Soil Sci Soc Am J 64 1588ndash1594httpsdoiorg102136sssaj20006451588x 2000

Oburger E and Jones D L Sampling root exudatesndash Mission impossible Rhizosphere 6 116ndash133httpsdoiorg101016jrhisph201806004 2018

Pausch J and Kuzyakov Y Carbon input by roots intothe soil Quantification of rhizodeposition from rootto ecosystem scale Glob Change Biol 24 1ndash12httpsdoiorg101111gcb13850 2018

Paustian K Lehmann J Ogle S Reay D Robertson GP and Smith P Climate-smart soils Nature 532 49ndash57httpsdoiorg101038nature17174 2016

Poeplau C and Don A Carbon sequestration in agri-cultural soils via cultivation of cover crops ndash Ameta-analysis Agr Ecosyst Environ 200 33ndash41httpsdoiorg101016jagee201410024 2015

Poeplau C Don A Vesterdal L Le J van Wesemael BSchumacher J and Gensior A Temporal dynamics of soilorganic carbon after land-use change in the temperate zonendash carbon response functions as a model approach GlobChange Biol 17 2415ndash2427 httpsdoiorg101111j1365-2486201102408x 2011

R Core Team R A language and environment for statisticalcomputing R Foundation for Statistical Computing ViennaAustria available at httpswwwR-projectorg (last access28 May 2020) 2019

Ruehr N K Offermann C A Gessler A Winkler J B Fer-rio J P Buchmann N and Barnard R L Drought effects onallocation of recent carbon from beech leaves to soil CO2 ef-flux New Phytol 184 950ndash961 httpsdoiorg101111j1469-8137200903044x 2009

Sandeacuten T Spiegel H Stuumlger H-P Schlatter N Haslmayr H-PZavattaro L Grignani C Bechini L DrsquoHose T MolendijkL Pecio A Jarosz Z Guzmaacuten G Vanderlinden K GiraacuteldezJ V Mallast J and ten Berge H European long-term fieldexperiments knowledge gained about alternative managementpractices edited by Aitkenhead M Soil Use Manage 34 167ndash176 httpsdoiorg101111sum12421 2018

Shearman V J Sylvester-bradley R Scott R K and FoulkesM J Physiological Processes Associated with Wheat YieldProgress in the UK Crop Sci 45 175ndash185 2005



Stevenel P Frossard E Abiven S Rao I M Tamburini Fand Oberson A Using a Tri-Isotope (13C 15N 33P) LabellingMethod to Quantify Rhizodeposition in Methods in Rhizo-sphere Biology Research edited by Reinhardt D and SharmaA K 169ndash195 Springer Singapore 2019

Studer M S Siegwolf R T W and Abiven S Carbon transferpartitioning and residence time in the plant-soil system a com-parison of two 13CO2 labelling techniques Biogeosciences 111637ndash1648 httpsdoiorg105194bg-11-1637-2014 2014

Sun Z Chen Q Han X Bol R Qu B and Meng F Allo-cation of photosynthesized carbon in an intensively farmed win-ter wheatndashsoil system as revealed by 14CO2 pulse labelling SciRep-UK 8 1ndash10 httpsdoiorg101038s41598-018-21547-y2018

Swinnen J van Veen J A and Merckx R 14C pulse-labelingof field-grown spring wheat an evaluation of its use in rhizo-sphere carbon budget estimations Soil Biol Biochem 26 161ndash170 1994

Taghizadeh-Toosi A Christensen B T Glendining M and Ole-sen J E Consolidating soil carbon turnover models by im-proved estimates of belowground carbon input Sci Rep-UK6 1ndash7 httpsdoiorg101038srep32568 2016

Tester M and Langridge P Breeding technologies to increasecrop production in a changing world Science 327 818ndash822httpsdoiorg101126science1183700 2010

Tubiello F N Salvatore M Ferrara A F House J FedericiS Rossi S Biancalani R Condor Golec R D Jacobs HFlammini A Prosperi P Cardenas-Galindo P SchmidhuberJ Sanz Sanchez M J Srivastava N and Smith P The Con-tribution of Agriculture Forestry and other Land Use activitiesto Global Warming 1990ndash2012 Glob Change Biol 21 2655ndash2660 httpsdoiorg101111gcb12865 2015

Vance E D Brookes P C and Jenkinson D S An ex-traction method for measuring soil microbial biomass CSoil Biol Biochem 19 703ndash707 httpsdoiorg1010160038-0717(87)90052-6 1987

Van de Broek M and Ghiasi S Supplementary data for Vande Broek et al 2020 Biogeosciences Mendeley Data v2httpsdoiorg101763275f349dpdr2 2020

Wacker L Jacomet S and Koumlrner C Trends in Biomass Frac-tionation in Wheat and Barley from Wild Ancestors to Mod-ern Cultivars Plant Biol 4 258ndash265 httpsdoiorg101055s-2002-25735 2002

Waines J G and Ehdaie B Domestication and Crop Physiol-ogy Roots of Green-Revolution Wheat Ann Bot 100 991ndash998 httpsdoiorg101093aobmcm180 2007

Wasson A P Richards R A Chatrath R Misra S C PrasadS V S Rebetzke G J Kirkegaard J A Christopher J andWatt M Traits and selection strategies to improve root systemsand water uptake in water-limited wheat crops J Exp Bot 633485ndash3498 httpsdoiorg101093jxbers111 2012

Werner R A and Brand W A Referencing strategies and tech-niques in stable isotope ratio analysis Rapid Commun Mass Sp15 501ndash519 httpsdoiorg101002rcm258 2001

Werner R A Bruch B A and Brand W A ConFlo IIIndash an interface for high precision δ13C and δ15N analysiswith an extended dynamic range Rapid Commun MassSp 13 1237ndash1241 httpsdoiorg101002(SICI)1097-0231(19990715)1313lt1237AID-RCM633gt30CO2-C1999

Zeeman M J Werner R A Eugster W Siegwolf R T WWehrle G Mohn J and Buchmann N Optimization of auto-mated gas sample collection and isotope ratio mass spectromet-ric analysis of δ13C of CO2 in air Rapid Commun Mass Sp 223883ndash3892 httpsdoiorg101002rcm3772 2008


Abstract

Introduction

Materials and methods

Experimental setup

ETH mesocosm platform

Wheat cultivars and growth conditions

Repeated 13C pulse labeling

Measurements

Belowground CO2 concentration and 13CO2

Sampling and general soil analyses

Soil microbial biomass

Organic carbon concentration and isotopic composition of plant material soil organic carbon and microbes

Data processing

Excess 13C calculations

Net carbon rhizodeposition

Subsoil CO2 production

Statistics

Results

Aboveground biomass

Belowground biomass

Soil and soil organic carbon characteristics

Excess 13C and carbon rhizodeposition

CO2 concentration and production

Discussion

Plant biomass carbon dynamics and CO2 production

Carbon allocation by wheat plants

Rates of net carbon rhizodeposition

The effect of old and recent wheat cultivars on net carbon rhizodeposition

Conclusion

Data availability

Supplement

Author contributions

Competing interests

Acknowledgements

Financial support

Review statement

References


smart agricultural practices (Chenu et al 2018 Minasny etal 2017 Paustian et al 2016) Multiple management prac-tices have been shown to increase the OC content of cul-tivated soils including the application of organic amend-ments to soils (Sandeacuten et al 2018) increasing the amountof crop residues returned to the field (Lehtinen et al 2014)and planting of cover crops (Kong and Six 2010 Poeplauand Don 2015)

In addition growing crops with deeper roots andor higherroot biomass has been put forward as a strategy to increaseOC sequestration in arable soils (Kell 2011) while deeprooting can also decrease the effect of drought in climateswhere deep soil water is available during the main croppingseason (Wasson et al 2012) However a direct or marker-assisted selection for root traits is very rare in conven-tional breeding programs Accordingly we have very limitedknowledge on if and how breeders alter the root system andpotentially affect belowground carbon cycling One way toevaluate the effect of breederrsquos selection on root character-istics and subsoil carbon cycling is to compare old and newvarieties of the same breeding program For the Swiss wheatbreeding programs the selection process reduced the massand depth of roots under well-watered conditions (Friedli etal 2019) as has been found for other breeding programs(Aziz et al 2017) but modern genotypes enhanced root al-location to deep soil layers under drought However this pat-tern has not been observed consistently (Cholick et al 1977Feil 1992 Lupton et al 1974) To the best of our knowl-edge there is no information about the effect of breeding onchanges in subsoil OC dynamics and root respiration

One reason for the lack of quantitative data about the ef-fects of rooting depth on SOC sequestration is related todifficulties in measuring the amount of carbon transferredfrom roots to the soil (gross rhizodeposition) and the pro-portion of carbon that is eventually stabilized there (net rhi-zodeposition) after a portion of gross rhizodeposits are lostfrom the soil through microbial mineralization or leachingThe fact that rhizodeposition occurs below the soil surfacegreatly prevents direct observations of this ldquohidden half ofthe hidden halfrdquo of the SOC cycle (Pausch and Kuzyakov2018) First of all direct measurements of root exudationrates are hampered by the fact that rhizodeposits are usedby rhizosphere microorganisms within a couple of hours af-ter they are released resulting in very low concentrations ofroot carbon exudates in the soil (Kuzyakov 2006) Secondthe release of carbon exudates by agricultural crops is notequally divided throughout the growing season but mainlyoccurs in the first 1ndash2 months of the growing period and de-creases sharply thereafter (Gregory and Atwell 1991 Keithet al 1986 Kuzyakov and Domanski 2000 Pausch andKuzyakov 2018 Swinnen et al 1994) Third measurementsof the effects of rhizodeposits on changes in SOC stocks arefurther complicated by the priming effect ie their positiveeffect on the mineralization of native SOC (Fontaine et al2007 de Graaff et al 2009)

To overcome these difficulties rates of C rhizodepositioncan be measured by labeling plants with 13CO2 or 14CO2(Jones et al 2009 Kuzyakov and Domanski 2000) and sub-sequently tracing the amount of photosynthetically assimi-lated 13C or 14C label in the soil at the end of the growingseason (Kong and Six 2010 2012) However the continuousapplication of 13CO2 or 14CO2 during the course of an entiregrowing season to plants is often not feasible as this requiresthe setup of open-top chambers while continuously supply-ing the crops with the isotopic label which comes at a high fi-nancial cost Therefore plants are commonly labeled at fixedtime intervals during the growing season (repeated pulse la-beling) This results in reliable estimates of the partition-ing of assimilated carbon to different plant compartmentsas well as into the soil (Kong and Six 2010 Kuzyakov andDomanski 2000 Sun et al 2018)

In addition assessing the magnitude of the carbon transferfrom roots to the soil is not straightforward particularly un-der field conditions While carbon inputs from crops to thesoil are often derived from yield measurements (Keel et al2017 Kong et al 2005 Taghizadeh-Toosi et al 2016) thesequantities are often poorly related to root biomass or the mag-nitude of root exudates (Hirte et al 2018 Hu et al 2018)A better understanding of the factors controlling the rates ofcarbon rhizodeposition by different agricultural crops is thusnecessary to assess how different crops affect SOC cyclingand to provide SOC models with reliable rates of carbon in-puts to the soil

The present study addresses the following research ques-tion do wheat cultivars with shallow roots and lower rootbiomass lead to less net carbon rhizodeposition compared towheat cultivars with deeper roots and higher root biomassTo address this question four different bread wheat cultivarsfrom a century of Swiss wheat breeding (Fossati and Bra-bant 2003 Friedli et al 2019) were grown in large meso-cosms which allowed us to study the plantndashsoil system undercontrolled conditions that closely resemble a field situationWe hypothesized that wheat cultivars with shallow roots andlower root biomass would result in less net carbon rhizode-position over the course of a growing season compared tocultivars with deeper roots and higher root biomass

2 Materials and methods

21 Experimental setup

211 ETH mesocosm platform

To assess the effect of wheat root characteristics on netrhizodeposition in a realistic soil environment under con-trolled conditions an experiment was set up at the meso-cosm platform of the Sustainable Agroecosystems Group atthe Research Station for Plant Sciences Lindau (ETH ZuumlrichSwitzerland) The platform was located inside a greenhouse
















22 Measurements










045 (1)


δ13CMB =


13CNF middotCNF)

CFminusCNF (2)




23 Data processing



mE(

13C)=

χE(13C

)middotm(C) middotM

(13C)

χ(

12C)middotM

(12C

)+χ

(13C

)middotM

(13C

) (3)



where mE(13C


χE(13C



(12C)

and M(13C



)and χ

(13C)






(13C)

soil

χE(

13C)

root

middotCsoil (4)


(13C)

soil andχE

(13C)





P(z)i =1(εi[CO2]i)

1t+Ftopi minusFboti

1z (5)




1z (6)


1z (7)


Dsi =D0tε25it

8i (8)






D0 =D0stpp0

p

(T

T0

)α (9)



24 Statistics



3 Results










































4 Discussion


































5 Conclusion










References














































































Abstract

Introduction


Experimental setup




Measurements





Data processing




Statistics

Results

Aboveground biomass

Belowground biomass




Discussion





Conclusion

Data availability

Supplement


Competing interests

Acknowledgements

Financial support

Review statement

References















22 Measurements










045 (1)


δ13CMB =


13CNF middotCNF)

CFminusCNF (2)




23 Data processing



mE(

13C)=

χE(13C

)middotm(C) middotM

(13C)

χ(

12C)middotM

(12C

)+χ

(13C

)middotM

(13C

) (3)



where mE(13C


χE(13C



(12C)

and M(13C



)and χ

(13C)






(13C)

soil

χE(

13C)

root

middotCsoil (4)


(13C)

soil andχE

(13C)





P(z)i =1(εi[CO2]i)

1t+Ftopi minusFboti

1z (5)




1z (6)


1z (7)


Dsi =D0tε25it

8i (8)






D0 =D0stpp0

p

(T

T0

)α (9)



24 Statistics



3 Results










































4 Discussion


































5 Conclusion










References














































































Abstract

Introduction


Experimental setup




Measurements





Data processing




Statistics

Results

Aboveground biomass

Belowground biomass




Discussion





Conclusion

Data availability

Supplement


Competing interests

Acknowledgements

Financial support

Review statement

References


where mE(13C


χE(13C



(12C)

and M(13C



)and χ

(13C)






(13C)

soil

χE(

13C)

root

middotCsoil (4)


(13C)

soil andχE

(13C)





P(z)i =1(εi[CO2]i)

1t+Ftopi minusFboti

1z (5)




1z (6)


1z (7)


Dsi =D0tε25it

8i (8)






D0 =D0stpp0

p

(T

T0

)α (9)



24 Statistics



3 Results










































4 Discussion


































5 Conclusion










References














































































Abstract

Introduction


Experimental setup




Measurements





Data processing




Statistics

Results

Aboveground biomass

Belowground biomass




Discussion





Conclusion

Data availability

Supplement


Competing interests

Acknowledgements

Financial support

Review statement

References




D0 =D0stpp0

p

(T

T0

)α (9)



24 Statistics



3 Results










































4 Discussion


































5 Conclusion










References














































































Abstract

Introduction


Experimental setup




Measurements





Data processing




Statistics

Results

Aboveground biomass

Belowground biomass




Discussion





Conclusion

Data availability

Supplement


Competing interests

Acknowledgements

Financial support

Review statement

References




































4 Discussion


































5 Conclusion










References














































































Abstract

Introduction


Experimental setup




Measurements





Data processing




Statistics

Results

Aboveground biomass

Belowground biomass




Discussion





Conclusion

Data availability

Supplement


Competing interests

Acknowledgements

Financial support

Review statement

References






























5 Conclusion










References














































































Abstract

Introduction


Experimental setup




Measurements





Data processing




Statistics

Results

Aboveground biomass

Belowground biomass




Discussion





Conclusion

Data availability

Supplement


Competing interests

Acknowledgements

Financial support

Review statement

References












































































Abstract

Introduction


Experimental setup




Measurements





Data processing




Statistics

Results

Aboveground biomass

Belowground biomass




Discussion





Conclusion

Data availability

Supplement


Competing interests

Acknowledgements

Financial support

Review statement

References

The soil organic carbon stabilization potential of old and ... · San Luis Obispo, California...

Documents

Transcript of The soil organic carbon stabilization potential of old and ... · San Luis Obispo, California...