Early Survival and Growth of Six Afforestation Species on Abandoned Cropping Sites in Irrigated...

19
This article was downloaded by: [Aston University] On: 06 October 2014, At: 19:40 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK Arid Land Research and Management Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/uasr20 Early Survival and Growth of Six Afforestation Species on Abandoned Cropping Sites in Irrigated Drylands of the Aral Sea Basin Tilman Schachtsiek a , John P. A. Lamers a & Asia Khamzina a a ZEF (Center for Development Research), University of Bonn , Bonn , Germany Published online: 01 May 2014. To cite this article: Tilman Schachtsiek , John P. A. Lamers & Asia Khamzina (2014) Early Survival and Growth of Six Afforestation Species on Abandoned Cropping Sites in Irrigated Drylands of the Aral Sea Basin, Arid Land Research and Management, 28:4, 410-427, DOI: 10.1080/15324982.2013.855958 To link to this article: http://dx.doi.org/10.1080/15324982.2013.855958 PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms- and-conditions

Transcript of Early Survival and Growth of Six Afforestation Species on Abandoned Cropping Sites in Irrigated...

This article was downloaded by: [Aston University]On: 06 October 2014, At: 19:40Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Arid Land Research and ManagementPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/uasr20

Early Survival and Growth of SixAfforestation Species on AbandonedCropping Sites in Irrigated Drylands ofthe Aral Sea BasinTilman Schachtsiek a , John P. A. Lamers a & Asia Khamzina aa ZEF (Center for Development Research), University of Bonn ,Bonn , GermanyPublished online: 01 May 2014.

To cite this article: Tilman Schachtsiek , John P. A. Lamers & Asia Khamzina (2014) Early Survival andGrowth of Six Afforestation Species on Abandoned Cropping Sites in Irrigated Drylands of the Aral SeaBasin, Arid Land Research and Management, 28:4, 410-427, DOI: 10.1080/15324982.2013.855958

To link to this article: http://dx.doi.org/10.1080/15324982.2013.855958

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the“Content”) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoever orhowsoever caused arising directly or indirectly in connection with, in relation to or arisingout of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Early Survival and Growth of Six AfforestationSpecies on Abandoned Cropping Sites in Irrigated

Drylands of the Aral Sea Basin

Tilman Schachtsiek, John P. A. Lamers, and Asia Khamzina

ZEF (Center for Development Research), University of Bonn,Bonn, Germany

Afforestation is known as a low-input measure to rehabilitate marginalized irrigatedcroplands with a shallow, saline groundwater table. This study assessed the potentialof extending afforestation to sites long-term abandoned from cropping in the lowerreaches of the Amu Darya River, Central Asia. Tree survival and establishment weremonitored during two growing seasons following afforestation with six tree species attwo abandoned cropping sites. The sites, characterized by soil salinity and differentdepth to the groundwater table, received deficit irrigation of 154mm yr�1. Afforest-ation was feasible with Elaeagnus angustifolia, Ulmus pumila, Morus alba, andPopulus nivea x tremula given their survival rates of 75–91% on both sites aftertwo years. N2-fixing E. angustifolia was assessed as most promising among allspecies evidenced by highest survival and largest above-ground biomass increment(up to 904 kg ha�1 yr�1). The principal species of the native floodplain forest, Popu-lus euphratica and Salix nigra, failed to establish, showing survival rates <19%.Kaplan-Meier survival curves revealed differences in tree survival between the sites,indicating the necessity of a site-specific evaluation of the afforestation species. Bothcurrent plant-available water and soil salinity in the root zone significantly affectedtree survival, with distinct differences in the stress tolerance among the species.Afforesting long-term abandoned cropland is associated with higher risks and loweroutputs than tree planting in marginal agricultural areas, implying the necessity of asite-dependent evaluation including cost-benefit analyses of afforestation as opposedto natural re-vegetation of abandoned sites.

Keywords deficit irrigation, Elaeagnus angustifolia L., incidence rate ratios,kaplan-meier curves, salinity tolerance, shallow groundwater table

The Aral Sea Basin (ASB) is one of the largest irrigated areas in the world. About 75%of the irrigated area in the ASB is affected by land degradation caused by soil saliniza-tion (van Dijk et al., 1999) mainly as a result of elevated groundwater tables (GWT)(Pereira et al., 2009). Particularly in downstream areas, the soil salinity problem isaggravated by an insecure availability of irrigation water for land reclamation(Alcamo et al., 2003) by salt leaching and drainage through the constructed

Received 10 April 2013; accepted 12 October 2013.Address correspondence to Tilman Schachtsiek, ZEF (Center for Development

Research), University of Bonn, Walter-Flex Str. 3, 53113 Bonn, Germany. E-mail: [email protected]

Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/uasr.

Arid Land Research and Management, 28:410–427, 2014Copyright # Taylor & Francis Group, LLCISSN: 1532-4982 print=1532-4990 onlineDOI: 10.1080/15324982.2013.855958

410

Dow

nloa

ded

by [

Ast

on U

nive

rsity

] at

19:

40 0

6 O

ctob

er 2

014

collector-drainage systems (Tischbein et al., 2012). The growing water stress due toclimate change in the region (IPCC, 2007) and associated land degradation requireswater-saving options for land rehabilitation. In this view, an annual, water-demanding salt leaching to combat soil salinization might be unfeasible on the highlysaline croplands, which operate on the margin of economic viability or are unprofit-able for farmers (Djanibekov et al., 2012). Consequently, temporary, particularly indrought years (Dubovyk et al., 2013), and even long-term abandonment of degradedcropland parcels has become common in the ASB, leading to economic losses(Toderich et al., 2008).

In the ASB but also elsewhere in degraded drylands, afforestation has signifi-cantly contributed to ecological restoration by reducing soil erosion, lowering shallowsaline GWT, and replenishing soil nutrient stocks (Danso et al., 1992; Katyal andVlek, 2000; Cacho, 2001; Guo and Gifford, 2002; Heuperman et al., 2002; Toderichet al., 2008; Jiao et al., 2010). Furthermore, tree planting on degraded croplandpatches is suggested as a mitigating land-use strategy in response to global climatechange (FAO, 2000; Katyal and Vlek, 2000), and is therefore also incentivizedthrough the Clean Development Mechanism (CDM) of the Kyoto Protocol.

The vast body of research on tree species suitable for phytoremediation and landrehabilitation was largely summarized in the Forestry Compendium (CAB Inter-national, 2010). Tree species performance under water, salt, and nutrient stress con-ditions has been addressed also (e.g., Schonfield, 1992; Tomar et al., 2004; Miyamotoet al., 2004; Archibald et al., 2006), but a direct comparisons and an applicability ofthe findings are limited for the choice of suitable species for an afforestation of aban-doned croplands in the ASB. Previous studies in the lower Amu Darya River Basin ofthe ASB evaluated the performance of 15 tree species based on their physiological andsocio-economic characteristics, but under irrigated and slightly saline conditions(Khamzina et al., 2006; Lamers et al., 2006). Setting aside highly salinized, irrigatedcroplands for afforestation with salt-tolerant, nitrogen (N)-fixing tree species was alsoassessed as a low-input land rehabilitation measure in the ASB lowlands, where shal-low GWT enabled forestry practices with low irrigation input (Khamzina et al., 2008;2009; 2012; Djumaeva et al., 2013). These studies indicated that saved water and Nfertilizer resources would compensate for the agricultural production losses throughsustainably intensifying the land use in productive agricultural areas (Martius et al.,2004), while potential financial benefits could be offered from non-timber forestproducts and carbon (C) sequestration (Lamers et al., 2008; Djanibekov et al.,2012). In this context, afforestation with the tree species Elaeagnus angustifolia L.,Populus euphratica Oliv., and Ulmus pumila L. in areas with shallow, moderately sal-ine groundwater reduced irrigation to between 3 and 30% of the required volume forannual crop cultivation, due to an effective utilization of the groundwater by trees(Khamzina et al., 2008). Yet, the possibility of afforesting abandoned croplandshas not been addressed so far.

The depth to the GWT and associated salinity levels impact tree growth andin particular when groundwater is the main water source (Horton et al., 2001). Ashallow GWT usually prevails within the irrigated lowlands in the ASB (Tischbeinet al., 2012) but can be lower on sites long-term abandoned from cropping andirrigation (Dubovyk et al., 2012). It therefore should be of interest to farmers anddecision-makers alike whether or not afforestation can be expanded to croplandsthat have been abandoned for a longer period and where a rehabilitation of theland’s productive potential is particularly difficult due to an ongoing desertification

Afforestation Species on Abandoned Cropping Sites 411

Dow

nloa

ded

by [

Ast

on U

nive

rsity

] at

19:

40 0

6 O

ctob

er 2

014

(Genxu et al., 2004). Such abandoned cropping sites, characterized by nutrientdeficiency, high soil salinity, precarious water supplies due to tail-end locations inthe irrigation system, and by water-logging conditions or declined GWT, demandsite-specific assessments of tree species’ performances. The current study aims there-fore to determine if the previously observed performance of the tree species in mar-ginal agricultural areas that are continuously cropped despite low yields (Khamzinaet al., 2012) applies also to abandoned croplands, where degradation processes areno longer counterbalanced by agricultural inputs. It is hypothesized that other com-mercially important species also bear a potential for the afforestation of abandonedcropping sites if some irrigation inputs are provided to cope with water and salinitystress (Djanibekov et al., 2012). As insecure water availability is the main reasonfor cropland abandonment in the lower Amu Darya Basin (Tischbein et al., 2012),the feasibility of extending afforestation to such croplands needs to be assessedunder deficit irrigation while considering tree species’ tolerance to salt, water, andnutrient stress.

Initial survival of trees under sub-optimal growth conditions is of core impor-tance when evaluating a species’ suitability for afforestation purposes, and influencesthe amount of biomass and tradable goods from a plantation in the long run. Thus,this study used survival and early biomass increment and partitioning in responseto prevailing growth stress factors as indicators to evaluate the feasibility of multi-species afforestation on long-term abandoned cropping sites differing in depth tothe GWT.

Material and Methods

Study Sites

In 2010, two experimental tree plantations were established in the lower reachesof the Amu Darya River: (i) in the Khorezm region of Uzbekistan (G’oybu,41�30’44.90’’N, 60�34’05.60’’E) and (ii) in the southern part of the autonomousRepublic of Karakalpakstan (Beruniy, 41�43’03.10’’N, 60�49’18.58’’E; Figure 1).Belonging to the Central Asian semi-desert zone, the study area is characterizedby an arid, extreme continental climate, that is, dry and hot summers and cold win-ters with little precipitation. Annual precipitation averaging about 100mm occursmainly outside the growing season (Tischbein et al., 2012) and is exceeded at least

Figure 1. Location of the study region and the afforestation sites G’oybu and Beruniy.

412 T. Schachtsiek et al.

Dow

nloa

ded

by [

Ast

on U

nive

rsity

] at

19:

40 0

6 O

ctob

er 2

014

10-fold by the potential evapotranspiration (Ibrakhimov et al., 2007; Conrad et al.,2012). Year 2011 was characterized by a drought due to a reduced water intake fromthe Amu Darya for irrigation (Bekchanov et al., unpublished). Throughout the2-year study (2010–2011), air temperature averaged 13.4�C reaching its minimum(�18.8�C) in February and its maximum (42.4�C) in August, as recorded by a micro-meteorological stations installed next to the afforestation sites.

The sites had been selected due to their distinct differences in depths of GWT,with a significantly deeper water table in G’oybu compared to Beruniy. As initialguidance for the study site selection served the region-wide, long-term hydrologicalsurvey by Ibrakhimov et al. (2007). This was followed by on-site visits, measurements,and agreements with farmers and regional authorities prior to the afforestationexperiments. The soil in G’oybu had a silt-loamy texture (20% sand, 6% clay, 74%silt); whereas, lighter, sandy-loam soil was found in Beruniy (85% sand, 6% clay,9% silt). The available water content in the rooting zone (top 60 cm), defined as thedifference between soil moisture at field capacity and wilting point, amounted to157mm and 122mm in G’oybu and Beruniy, respectively. Both sites had been aban-doned for ten years from cotton and rice cropping due to low yields. The examinationof soil profiles at the onset of the experiment showed poor, to very poor concentra-tions of macronutrients in the crop rooting zone in both sites (Table 1). In 2010, fieldssurrounding the G’oybu site were cropped with water melon and corn, while inBeruniy, rice and cotton were the dominant crops. In the drought year 2011, lesswater-demanding crops were cropped on neighboring fields in Beruniy, whereas inG’oybu there were no cropping activities.

Tree Species

Six tree species were selected for the experiment based on the previously assessed keyphysiological characteristics and multipurpose potential (Khamzina et al., 2006). Allspecies were indigenous except for the introduced U. pumila and P. nivea x tremula,both widely planted in the study area for many decades. P. euphratica, Salix nigra,and E. angustifolia are natural phreatophytes and common species of the nativeriparian forest where the shallow GWT is essential for ecosystem functioning(Tupitsa, 2010). Earlier studies showed that E. angustifolia was capable of effectiveN2-fixation in saline soil conditions, thus significantly increasing soil N stocks(Khamzina et al., 2009). Leaves of several species were assessed suitable as fodderfor dairy livestock (Lamers and Khamzina, 2010), whereas leaves of Morus albaare traditionally used as silkworm feed (Kan et al., 2008). Additional tree productsinclude wicker from S. nigra, and firewood, which could be harvested from all speciesbut has the highest potential in species that tolerate frequent pruning and re-sproutreadily (Lamers and Khamzina, 2008).

Experimental Set-Up and Measurements

The experimental sites were set up in a completely randomized block design. Prior totree planting in March 2010, the sites had been cleared of sparse natural vegetation,represented by shrubby and herbaceous halophytes (Tamarix spp., Cynodon dactylonL., Glycyrrhiza glabra L., Karelinia caspica Pall., Scrophularia leucoclada Bunge.),leveled and leached of salts. Trees of the six species were planted in pure-species plotsthat were replicated three times, amounting to 18 plots at each site. Planting density

Afforestation Species on Abandoned Cropping Sites 413

Dow

nloa

ded

by [

Ast

on U

nive

rsity

] at

19:

40 0

6 O

ctob

er 2

014

was 6,666 trees ha�1 within 10� 10m plots (1m� 1.5m spacing), with a 3.5-m bufferzone between the plots. The dense planting scheme was implemented to achievehigher biomass stock per unit of land area in the short-run and thus reduce thewaiting period before non-timber products can be harvested by farmers. After saltleaching, 12-month-old saplings were planted at a depth of about 30 cm. For S. nigra,cuttings of 20 cm were used, as saplings were not available.

From March until October in 2010 and 2011, the trees were irrigated weekly at arate corresponding to 10–15% of the evaporative demand, amounting to a total sea-sonal application of 154mm (about 3–5mm per week). Different methods were usedfor implementing the deficit irrigation because of differences in the GWT level henceexpected water contribution from the groundwater at the two study sites. In Beruniy,traditional furrow irrigation (Abdullaev, 2003) was applied as previously recom-mended for tree establishment at shallow GWT locations (Khamzina et al., 2008).The irrigation rate was controlled with Cipoletti weirs allowing the estimation ofthe water inflow from the upstream water depth (Forkutsa, 2006). In G’oybu, witha deeper GWT, irrigation was applied via a drip irrigation system consisting ofstabilized high-density polyethylene hoses, one pressure self-compensating dripper(C.N.L. type) per tree, a water storage tank, and a 140-mesh disk filter. The irrigationrate was controlled by automatic measuring valves shutting down after the application

Table 1. Soil total nitrogen (N), available phosphorus (P2O5), exchangeable potass-ium (K2O), and total organic carbon (SOC) at two study sites before afforestation inMarch 2010

G’oybu

LayerCm N %

P2O5 K2O SOC

mgkg�1 Evaluationa

mgkg�1 Evaluationa % Evaluationb

0–4 0.046 7.20 Very Low 163.70 Low 0.38 Poor4–15 0.033 5.00 Very Low 72.20 Very low 0.24 Poor15–30 0.018 4.70 Very Low 60.20 Very low 0.11 Very Poor30–60 0.024 4.40 Very Low 72.20 Very low 0.19 Very Poor60–81 0.020 7.20 Very Low 60.20 Very low 0.14 Very Poor81–101 0.017 5.00 Very Low 72.20 Very low 0.11 Very PoorBeruniy0–1 0.087 30.00 Low 207.10 Moderate 0.97 Rich1–4 0.075 20.70 Low 223.90 Moderate 0.90 Increased4–10 0.050 18.50 Low 207.10 Moderate 0.43 Poor10–20 0.027 10.40 Very Low 240.80 Moderate 0.21 Very Poor20–30 0.034 7.60 Very Low 257.60 Moderate 0.25 Poor30–45 0.046 6.80 Very Low 192.60 Low 0.36 Poor45–60 0.034 6.20 Very Low 163.70 Low 0.28 Poor60–80 0.021 6.20 Very Low 101.10 Low 0.12 Very Poor80–112 0.027 5.90 Very Low 149.30 Low 0.17 Very Poor

aMusaev, 2001;.bKrasnouhova et al., 1988.

414 T. Schachtsiek et al.

Dow

nloa

ded

by [

Ast

on U

nive

rsity

] at

19:

40 0

6 O

ctob

er 2

014

of the pre-set volume of water. As previous studies on irrigation methods showed thatthe irrigation methods do not directly influence tree survival (Khamzina et al., 2008),the effect of the different irrigation methods on tree survival was not further analyzedin the study.

For the determination of the initial sapling biomass (above- and below-ground),ten saplings of each species from the planting material used for the plantation set upwere randomly selected, separated into roots, stem, branches, and foliage fractions,dried and weighed with an electronic scale to the nearest gram. At the end of the grow-ing season in 2010 and 2011, that is, 7 and 19 months after planting (MaP), three treesper plot were harvested. The entire coarse root systems were excavated manually tomeasure the rooting depth and biomass. Fresh sub-samples of a known fresh massfrom each above-ground fraction were dried at 103�C in a forced air convection ovenuntil constant weight to allow the dry-weight estimation of the entire fraction. Thewhole coarse root systems were dried and weighed to avoid errors in weight estimatesdue to partial desiccation of roots during excavation. Biomass increment rates weredefined as the difference between the average dry weight of the initial plantingmaterial and that at 7 and 19 MaP.

Sapling survival was assessed at 3, 9, 12, and 19 MaP by counting the number ofliving trees per plot. Dead trees were replaced once in March 2011 to maintain ahomogeneous plot density in the experiment but were not analyzed.

The GWT levels were measured monthly in each plot through observation wellsinstalled to 200–220 cm depth. In addition, groundwater salinity was measured withan EC meter each time. Soil was sampled monthly with an auger for measuring sal-inity and moisture in each plot in 20 cm increments down to 100 cm depth. As a proxyof soil salinity, the electrical conductivity of the saturation paste (EC1:1) was mea-sured with a portable EC meter and converted into ECe using the relationshipECe ¼ EC�

1:1 3.6 established for soils in the Khorezm region (Shirokova et al.,2000). Soil moisture was measured gravimetrically and converted into volumetricunits using soil bulk density values as determined for the examined soil pits. Currentplant available water (CPAW) in the soil was calculated as the difference between themeasured volumetric soil moisture and the moisture at the permanent wiltingpoint (Goldhamer et al., 1999). For that purpose, pF curves were determined in thelaboratory with the pressure membrane method using samples collected from soil pitsat both study sites at one MaP.

Statistical Analyses

The effects of the species and study site factors on the above- and below-ground bio-mass increment were tested with the analysis of variance (ANOVA) at a significancelevel of p¼ 0.05. In case the ANOVA showed significant effects, this was followed bya Tukey post-hoc test to compare individual means.

The raw (unadjusted) survival data by species was analyzed using the nonpara-metric Kaplan-Meier estimator, a key method for analyzing censored survival data(Borgan, 2005). Differences between the survival curves of individual species (overalland by site) were tested with a log-rank test. A post-hoc Bonferroni adjusted test wascompleted to identify the significance of the species effect on each site.

Incidence rates and incidence rate ratios (IRR) adjusted for ties and dynamicsubject effects were estimated with a Poisson model with random effects usinggeneralized estimating equations and tested for their influence on the mortality rates.

Afforestation Species on Abandoned Cropping Sites 415

Dow

nloa

ded

by [

Ast

on U

nive

rsity

] at

19:

40 0

6 O

ctob

er 2

014

The relationship between soil salinity and CPAW with tree mortality was consideredas target associations, adjusted for the eligible putative known confounders (site,plot, and species). Incidence rate ratios were used to analyze the proportional changein mortality rates when increasing independent variables by one unit (Long andFreese, 2006). Incidence rate ratios higher than one therefore indicate an increasingmortality with an increasing, independent variable, for example, salinity. Respect-ively, an IRR lower than one, indicates a tree mortality decrease with an increasing,independent variable. In the case of CPAW, exposition times were not included inthe calculation due to their high variability under irrigation within the measurementintervals. Therefore, incidence rates of the CPAW were not considered in the analy-sis, but IRR were used as indicators for the CPAW influence on tree survival. Allstatistical analyses were performed using SPSS 15.0 and the STATA 12.0 software.

Species Ranking

The species were ranked for their suitability for afforestation of abandoned croplandsin the study area based on the Kaplan-Meyer survival curves and the Tukey post-hoctests of the above- and below-ground biomass increments. The survival and growthrates significantly influence the amount of non-timber products and the early inde-pendence from irrigation due to the groundwater uptake by roots. Based on differingimportance of these criteria, they were respectively weighted 3, 2, and 1 in the ranking.

Results

Environmental Conditions of the Afforestation Sites

The GWT was significantly deeper in G’oybu than in Beruniy (p< 0.05) during bothgrowing seasons. During the 2010 season, the GWT fluctuated between 40 and 90 cmbelow the soil surface in Beruniy, and between 55 and 175 cm in G’oybu, droppingnotably at both sites by the end of August (Figure 2). In 2011 in Beruniy, theGWT declined slightly, fluctuating between 60 and 120 cm and averaging 93 cm. In

Figure 2. Annual course of the groundwater table at two afforestation sites measured with 18observation wells per site during the growing seasons in 2010 and 2011. The groundwater tablein G’oybu in 2011 was below the 200 cm reach of the observation wells. Vertical bars indicate95% confidence intervals.

416 T. Schachtsiek et al.

Dow

nloa

ded

by [

Ast

on U

nive

rsity

] at

19:

40 0

6 O

ctob

er 2

014

G’oybu, the level dropped below 220 cm, which was beyond the reach of theobservation wells (Figure 2). In Beruniy, groundwater salinity averaged 3.8 dS m�1

in both growing seasons thus exceeding the 2010 average measured in G’oybu(2.5 dS m�1).

Throughout the whole season in 2010, the volumetric soil moisture in the root-ing zone of 0–60 cm was significantly higher in Beruniy than in G’oybu (Figure 3).Nevertheless, no statistical difference in CPAW was observed between the sites, bothaveraging 11% in 2010 (p¼ 0.81) and 8.0% in 2011 (p¼ 0.48), which was the result ofdifferent water holding capacities associated with the different soil textures at thestudy sites. At both sites, soil moisture approached the wilting point in the midsea-son (July), thereafter recovering to the previous level and then dropping again inSeptember in Beruniy (Figure 3). That was in contrast to G’oybu, where soil waterremained at the low level till the end of the growing season.

Soil salinity varied depending on soil depth and time of the season (Figure 4).Generally, salinity levels in Beruniy, in conditions of the shallower GWT, exceededthose in G’oybu during the two-year observation period in all soil layers (p¼ 0.05).Within the soil profile, slightly higher salinity was observed in the topsoil (0–40 cm)on both study sites. During the first season 2010, Beruniy experienced higher salinityin the 0–20 cm and 20–40 cm layers than below the rooting depth in the 80–100 cmlayers. During the 2011 growing season, no statistically significant differences wereobserved in Beruniy (Figure 4). The soil salinity in G’oybu was significantly higherwithin the 20–40 cm layer (5.3 dS m�1) than in the deeper horizons of 60–100 cm in2010. In the following season, the top 40 cm layer (4.3 dS m�1) in G’oybu was signifi-cantly more saline than the deepest 80–100 cm layer (Figure 4).

On the annual average in Beruniy, soil salinity slightly declined from 6.9 dS m�1

to 6.5 dS m�1, which is within the medium salinity range (Abrol et al., 1988). InG’oybu, the inter-annual difference was also insignificant with annual averages of

Figure 3. Volumetric soil moisture in the rooting zone (0–60 cm) at the field capacity (FC), thepermanent wilting point (PWP), and as measured during the observation period on two studysites. Vertical bars indicate 95% confidence intervals.

Afforestation Species on Abandoned Cropping Sites 417

Dow

nloa

ded

by [

Ast

on U

nive

rsity

] at

19:

40 0

6 O

ctob

er 2

014

3.7 dS m�1 and 3.5 dS m�1 in 2010 and 2011, respectively, thus revealing a slightlysaline soil (Abrol et al., 1988).

Tree Survival

During the 19-month observation period, 2,984 trees were included in the analysis;during the same period 1,580 trees perished (53%). Comparison of the Kaplan-Meiercurves revealed that survival rates were higher in G’oybu than in Beruniy throughoutthe observation period (Figure 5). In G’oybu, E. angustifolia, M. alba, P. nivea xtremula, and U. pumila showed similar survival rates exceeding that of P. euphraticaand S. nigra. In contrast, in Beruniy, E. angustifolia outperformed all species whereasU. pumila and M. alba showed intermediate survival rates (Figure 5). Almost 60% ofthe P. euphratica trees on both sites perished within the first months after planting,and a complete mortality was observed after the second growing period (Figure 5).

Figure 4. Soil electrical conductivity (ECe) during the growing seasons 2010 and 2011 at twostudy sites according to 20 cm soil layers. Vertical bars indicate 95% confidence intervals.

Figure 5. Kaplan-Meier survival curves according to species and study sites. The table reportson the log-ranking tests of equality of the main factors. The vertical bars indicate 95% confi-dence intervals. Species with different superscripts differ significantly within one site.

418 T. Schachtsiek et al.

Dow

nloa

ded

by [

Ast

on U

nive

rsity

] at

19:

40 0

6 O

ctob

er 2

014

The CPAW and soil salinity in the rooting zone influenced species’ survival rateswith different intensity. In all species, increasing mortality was observed along withdecreasing CPAW, but this significantly affected only M. alba and U. pumila(Table 2) under the observed range of CPAW (>3% Vol.). The impact of CPAWon these species was similar, as evidenced by their similar IRRCPAW values(Table 2). All species except for S. nigra were significantly influenced by soil salinityin the observed range of up to 20 dS m�1(in single plot measurements) with signifi-cant differences in estimated IRRECe varying from 1.01 (U. pumila) to 1.12 (P. niveax tremula) (Table 2, Figure 6). Ulmus pumila was least impacted by soil salinity,followed by E. angustifolia (IRRECe¼ 1.01 and 1.07, respectively). The effect ofroot-zone soil salinity on M. alba and P. nivea x tremula was much higher, as sub-stantiated by the IRRECe values (Table 2).

Biomass Increment

In both growing seasons, the produced above- and below-ground biomass signifi-cantly varied according to tree species (Tables 3 and 4). Neither site nor interactionbetween site and species were statistically significant. At 7 MaP, E. angustifolia andP. nivea x tremula showed the highest above-ground biomass increment (ABI)among the species, of up to 473 and 245 kg ha�1 yr�1, respectively (Table 3). Inthe second growing season, biomass increments of E. angustifolia were higher thanthose of all other tree species on both study sites, with ABI ranging from 900 to1900 kg ha1 yr�1. Differences among the other species were statistically not signifi-cant (Table 3).

In terms of annual below-ground biomass increment (BBI) at 7 MaP, P. nivea xtremula overtook all other species (396–426 kg ha�1 yr�1; Table 4). The only excep-tion was E. angustifolia that showed the largest BBI with a borderline difference forP. nivea x tremula at 19 MaP.

Rooting Depth

During the whole observation period, rooting depths of all species did not exceed60 cm (Figure 7). Furthermore, no differences in rooting depths among species wereobserved, and site differences were negligible for both years at p¼ 0.05. Whereas inBeruniy rooting depths did not significantly increase over time, in G’oybu all speciesrooted significantly deeper during the second growing season (Figure 7).

Table 2. Incidence rate ratios (IRR) and corresponding p-values of the correlationof tree survival with the plant available water (CPAW) and soil electrical conduc-tivity (ECe) in the rooting zone

IRRCPAW p-value IRRECe p-value

Elaeagnus angustifolia 0.989 0.475 1.067 <0.001Morus alba 0.956 <0.001 1.096 <0.001Salix nigra 0.991 0.400 1.014 0.268Populus nivea x tremula 0.991 0.430 1.124 <0.001Ulmus pumila 0.957 0.013 1.014 0.003

Afforestation Species on Abandoned Cropping Sites 419

Dow

nloa

ded

by [

Ast

on U

nive

rsity

] at

19:

40 0

6 O

ctob

er 2

014

Species Ranking

Based on the weighted criteria of survival and biomass production rate, E. angusti-folia ranked first at the end of the observation period at both study sites whereas S.nigra was outperformed by all other species (Table 5). Otherwise, the ranking of thespecies suitability for afforestation differed between the two study sites. In G’oybu,

Table 3. Average above-ground biomass at planting and increments after the first (7MaP) and second (19 MaP) growing season according to tree species and study sites

Above-ground biomass increment(kg ha yr�1) (kg ha�1 yr�1)

Species

Total biomass(kg ha�1)at planting

G’oybu Beruniy

7 MaP 19 MaP 7 MaP 19 MaP

Elaeagnus angustifolia 195.1b 473.4a 1863.8a 377.2a 904.5a

Morus alba 7.0d 87.8b 75.6b 79.8a 62.0b

Populus nivea x tremula 251.9a 245.1ab 184.5b 314.8a 207.2b

Salix nigra 61.9c 14.2b 11.8b 28.3a 99.1b

Ulmus pumila 11.4d 174.0b 116.8b 93.8a 99.3b

ANOVA (p-values)

7 MaP 19 MaP

Species <0.001 <0.001Site 0.690 0.470Species � Site 0.829 0.188

Note: Values with different superscripts in one column differ significantly.

Figure 6. Incidence rate curves of the influence of average soil electrical conductivity inthe rooting zone (ECe) on the survival of five tree species. The vertical bars indicate 95%confidence intervals.

420 T. Schachtsiek et al.

Dow

nloa

ded

by [

Ast

on U

nive

rsity

] at

19:

40 0

6 O

ctob

er 2

014

Figure 7. Maximum rooting depth according to the tree species and average groundwatertable depth (GWT) during the 2010 and 2011 growing seasons on two study sites. Vertical barsindicate 95% confidence intervals.

Table 4. Average below-ground biomass at planting and increments after the first (7MaP) and second (19 MaP) growing season according to tree species and study sites

Below-ground biomass increment(kg ha yr�1) (kg ha�1 yr�1)

Species

Total biomass(kg ha�1)at planting

G’oybu Beruniy

7 MaP 19 MaP 7 MaP 19 MaP

Elaeagnus angustifolia 91.8b 214.4b 633.1a 301.54ab 376.6a

Morus alba 11.0c 87.4b 63.3b 87.7bc 63.5c

Populus nivea x tremula 193.9a 426.0a 223.0ab 396.4a 195.8ab

Salix nigra 0.0d 69.6b 80.5b 48.0c 63.1c

Ulmus pumila 17.6c 230.1b 147.7b 138.5bc 99.0bc

ANOVA (p-values)

7 MaP 19 MaP

Species <0.001 <0.001Site 0.735 0.216Species � Site 0.541 0.270

Note: Values with different superscripts in one column differ significantly.

Table 5. Weighted ranking of tree species suitability for afforest-ation of abandoned cropping sites in the Amu Darya lowlands

Species

Rank

G’oybu Beruniy

Elaeagnus angustifolia 1 1Morus alba 3 3Populus nivea x tremula 2 4Salix nigra 4 5Ulmus pumila 3 2

Afforestation Species on Abandoned Cropping Sites 421

Dow

nloa

ded

by [

Ast

on U

nive

rsity

] at

19:

40 0

6 O

ctob

er 2

014

P. nivea x tremula ranked second followed by M. alba and U. pumila. In Beruniy, U.pumila ranked second followed by M. alba. Populus euphratica was not included inthis ranking due to its early mortality.

Discussion

Environmental Conditions at Abandoned Cropping Sites

The annual GWT fluctuations at the study sites followed the regional pattern drivenby irrigation and leaching activities, with a shallow level observed during the irrigatedperiod and with a declined GWT in the fall-winter period, after the cessation of irri-gation (Ibrakhimov et al., 2007). The GWT subsidence in dry years, as evidenced in2000 and 2001 (Ibrakhimov et al., 2007) and, to a lesser extent, in 2008 and 2011 in thestudy region, was also revealed by our measurements in 2011. In G’oybu, the drop inthe GWT in 2011 as compared to the previous season was more severe due to theprevalence of uncultivated cropland in the surrounds, whereas in Beruniy, croppingand irrigation of the adjacent fields continued, although shifting towards lesswater-demanding crops. These differences in GWT between the study sites werereflected in their soil moisture dynamics. Soil moisture in G’oybu steadily declinedtowards the wilting point in the drought year (2011), due to the tree-water use anda drastic reduction in the GWT contribution to the soil water content. In Beruniy,the decline was likely compensated by the underground inflow from the neighboringirrigated fields.

The deficit irrigation applied weekly at both afforestation sites counteracted thetopsoil desiccation, thus supporting the survival and growth of the tree plantations.However, under conditions of declined GWT in G’oybu, larger amounts should havebeen applied to increase the moistened zone and stimulate root proliferation indepths below the observed 60 cm. Given the precarious availability of irrigationwater in dry years, particularly in abandoned cropland parcels with deeper GWT(Dubovyk et al., 2013), afforestation on such sites is associated with an increased riskof water stress.

A shallow GWT significantly contributes to soil salinity (Forkutsa et al., 2009)when it exceeds a certain threshold level, above which it rises by capillarity towardthe soil surface (Hillel, 2000). This effect was pronounced in Beruniy, where moder-ately saline GWT during the entire observation period exceeded 1.5m, defined as thethreshold level in Khorezm (Rakhimbaev et al., 1992). This was 1.5 times more salinethan in G’oybu. In consequence, the Beruniy site was characterized by higher soilsalinity, at times above 8 dS m�1 hence beyond the tolerance of salt-sensitive plants(Kotuby-Amacher et al., 2000). This impacted tree survival and biomass production.

Tree Survival and Biomass Growth in Response to Prevalent Stress Factors

Among the tested species, only E. angustifolia performed soundly on both N-poorsites, given its ability for N2-fixation (Khamzina et al., 2009; Schachtsiek et al.,unpublished). Next to the nutrient status, the depth to the GWT and associatedmoisture and, particularly, salt regimes may explain the monitored differences in treesurvival between the sites, as observed regularly in shallow groundwater environ-ments (Horton et al., 2001).

In conditions of declined GWT, as in G’oybu, the deficit irrigation ensured amoisture content in the topsoil evidenced by the observed survival rates of over

422 T. Schachtsiek et al.

Dow

nloa

ded

by [

Ast

on U

nive

rsity

] at

19:

40 0

6 O

ctob

er 2

014

90% in most of the tested species, i.e., E. angustifolia, P. nivea x tremula, U. pumila,and M. alba. Nevertheless, survival rates of the latter two species depended on theamount of the soil moisture available in the rooting zone (as reflected in IRRCPAW).This indicates a higher water demand of these species, possibly to counterbalance theadverse impact of salinity, the most important stress factor for plant growth inirrigated croplands (Hillel, 2000).

Under the measured salinity range (0–20 dS m�1), all species except S. nigra wereaffected to a different extent. Ulmus pumila and E. angustifolia showed lowestIRRECe values illustrating a relatively high tolerance to soil salinity, as expressedin their best survival among the tested species at the more saline Beruniy site. Thisresult is in line with the salt tolerance ranking by Miyamoto et al. (2004) who statedthat E. angustifolia was able to tolerate salinity levels of 6-8 dS m�1, and so wasU. parvifolia, the only representative of Ulmus genus in this list. All other speciesincluded in the present experiment were classified by Miyamoto et al. (2004) as‘‘sensitive,’’ meaning that they start to show stress signs from 3 dS m�1 onward.Although the salinity tolerance of tree species depends on the genetic potential,tolerance is known to increase in older trees as well as under more favorable soilmoisture and nutrient conditions (Kozlowski, 1997).

In contrast to survival, the ANOVA revealed no significant site effect on biomassincrements. The tree biomass increments, above- and below-ground, significantly dif-fered only with respect to the species. The observed annual biomass increments for allspecies were up to five times lower compared to those in assessments of tree species’potential for afforestation of low-productive irrigated croplands in the same region(Khamzina et al., 2008; 2009; Djumaeva et al., 2013). The lower biomass productionin the present experiment was likely the result of the cumulative stress observed at theabandoned cropping sites, including water deficit, known to negatively affect treegrowth by inhibiting cell elongation and CO2 uptake (Chapman-King et al., 1986).In contrast to the previous afforestation trials, where tree roots established accessto the GWT moisture already during the first growing season, the root-groundwaterinterface was weaker in the G’oybu site with its relatively deep GWT. Although therooting systems in G’oybu deepened in response to the GWT decline in the droughtyear 2011, most of the water demand was satisfied by the applied deficit irrigation.Consequently, larger biomass increments can be expected with increased root contactwith the groundwater following a rise of the GWT in subsequent, non-drought years.

Species Ranking and Implications for Afforestation of Abandoned Cropland

The species ranking was generally in agreement with previous studies in the AmuDarya lowlands that emphasized the potential of E. angustifolia for the afforestationof low-productive irrigated croplands (Khamzina et al., 2006; 2008; Djumaeva et al.,2013). Nevertheless, the pronounced differences in species survival between the twostudy sites imply that a broad generalization and extrapolations of the resultsobtained in degraded croplands to long-term abandoned croplands should beavoided. In particular, the sensitivity of U. pumila andM. alba to soil water depletionmight require higher irrigation inputs when planting these species on abandonedcropping sites characterized by a lower GWT.

The declining survival of S. nigra was linked neither to moisture nor to salinityvariation in the root zone. The probable reason for the observed mortality is thatun-rooted cuttings were used as planting material. When planted as rooted saplings

Afforestation Species on Abandoned Cropping Sites 423

Dow

nloa

ded

by [

Ast

on U

nive

rsity

] at

19:

40 0

6 O

ctob

er 2

014

in other afforestation trials on sandy and silt-loamy, slightly saline hydromorphicsoils, S. nigra showed similar survival to that of E. angustifolia, U. pumila, andM. alba (Khamzina et al., 2006). Therefore, the impact of the used planting materialon the tree survival on degraded lands should be studied.

The low survival rate of P. euphratica (<50%) was previously observed in treeplantations on marginal, highly saline sites and also on productive farmland withover 0.9–2.0m deep GWT in Khorezm (Khamzina et al., 2006; 2008). This was alsogenerally reported with respect to artificial propagation of this species outside itsnatural habitat of riparian forest (Wang et al., 1996). Long-term monitoring onthe marginal site in Khorezm showed effective regeneration of P. euphratica via rootsuckering in consequent years (Khamzina et al., 2008). In contrast, trees in our studyexhibited continuous mortality over the 2-year observation period, resulting in nilsurvival in the second, drought-affected year 2011. Given such a high sensitivity ofthis species to water stress, the risk of which increases in abandoned cropping sites,P. euphratica is not recommended for afforestation in such locations.

All in all, the present afforestation experiments demonstrate the elevated risk ofafforesting croplands that had been long-term abandoned from cropping. Especiallyduring dry years, frequently occurring in the study region (Tischbein et al., 2012),both the accessibility of the GWT and availability of irrigation water are reducedand might not be sufficient for coping with salt stress during the early growth phase.This, consequently, limits the range of species to be included in afforestation ofthe abandoned cropping sites. Biomass production was likewise reduced, whichmight result in lower financial benefits than those estimated from afforestation ofmarginal croplands (Djanibekov et al., 2012). The observed biomass production inthe afforested plots is likely to be higher than on abandoned cropping sites that werenaturally re-vegetated by halophytic plants and showed sparse land cover (Dubovyket al., 2012). In this view, a cost-benefit analysis could be decisive before definingthe overall viability of afforestation versus the natural vegetation succession onabandoned croplands and their use for grazing.

Conclusions

The present findings generally support the feasibility of extending afforestationefforts to cropland parcels that have been abandoned for a longer period, butunderline the importance of a site-specific assessment given the increased risk andinput needs. A higher risk of the venture is particularly present during droughtperiods, which cause a decline in the GWT, thus requiring higher irrigation inputsto ensure early survival and growth of trees. On the other hand, tolerance to salinitystress was confirmed to be the factor determining tree survival under the observedmoisture and nutrient status of the soils. Tree growth rates are expected to increasedue to enhanced GWT in subsequent, non-drought years. The success of afforest-ation on abandoned croplands should therefore be evaluated over a longer obser-vation period.

In agreement with previous studies in the Amu Darya lowlands, the mostpromising afforestation species appeared to be E. angustifolia and, to a lesser degreeU. pumila, which performed with relative consistency on both sites. Populus nivea xtremula and, to a lesser degreeM. alba, should be considered for less saline sites only.The modest growth rates observed demand the assessment of the economic feasibilityof afforestation as opposed to leaving the land fallow.

424 T. Schachtsiek et al.

Dow

nloa

ded

by [

Ast

on U

nive

rsity

] at

19:

40 0

6 O

ctob

er 2

014

References

Abdullaev, U. 2003. Republic of Uzbekistan. Land Degradation Assessment in Dry Lands(LADA). State Design and Research Institute (Uzgip), Tashkent.

Abrol, I. P., J. S. P. Yadav, and F. I. Massoud. 1988. Salt-affected soils and their management.FAO Soils Bulletin 39 FAO, Rome: 1–131.

Alcamo, J., P. Doll, T. Henrichs, F. Kaspar, B. Lehner, T. Rosch, and S. Siebert. 2003. Devel-opment and testing of the WaterGAP 2 global model of water use and availability.Hydrological Science Journal 39: 317–338.

Archibald, R. D., R. J. Harper, J. E. D. Fox, and R. P. Silberstein. 2006. Tree performanceand root-zone salt accumulation in three dryland Australian plantations. AgroforestrySystems 66: 191–204.

Borgan, Ø. 2005. Kaplan–Meier Estimator, in P. Armitage and T. Colton, eds., Encyclopediaof Biostatistics. Wiley, Chichester and New York.

CAB International. 2010. Forestry Compendium. Wallingford, UK.Cacho, O. 2001. An analysis of externalities in agroforestry systems in the presence of land

degradation. Ecological Economics 39: 131–143.Chapman-King, R., T. M. Hinckley, and C. C. Grier. 1986. Growth responses of forest trees to

wastewater and sludge applications, p. 582. in D. W. Cole, C. L. Henry, and W. L. Nutter,eds., The Forest Alternative for Treatment and Utilization of Municipal and IndustrialWastes. University of Washington, University of Washington Press, Seattle.

Conrad, C., G. Schorcht, B. Tischbein, S. Davletov, M. Sultonov, and J. P. A. Lamers. 2012.Agro-meteorological trends of recent climate development in Khorezm and implicationsfor crop production, pp. 25–36, in C. Martius, I. Rudenko, J. P. A. Lamers and P. L. G.Vlek, eds., Cotton, Water, Salts And Soums - Economic and Ecological Restructuring inKhorezm, Uzbekistan. Dordrecht Heidelberg, London, New York.

Danso, S., G. Bowen, and N. Sanginga. 1992. Biological nitrogen fixation in trees in agroeco-systems. Plant and Soil 141(1–2): 177–196.

Djanibekov, U., A. Khamzina, N. Djanibekov, and J. P. A. Lamers. 2012. How attractive areshort-term CDM forestations in arid regions? The case of irrigated croplands in Uzbeki-stan. Forest Policy and Economics 21: 108–117.

Djumaeva, D., J. P. A. Lamers, A. Khamzina, and P. L. G. Vlek. 2013. The benefits ofphosphorus fertilization of trees grown on salinized croplands in the lower reaches ofAmu Darya, Uzbekistan. Agroforestry Systems 87: 555–569.

Dubovyk, O., G. Menz, C. Conrad, E. Kan, M. Machwitz, and A. Khamzina. 2013. Spatio-temporal analyses of cropland degradation in the irrigated lowlands of Uzbekistan usingremote sensing and logistic regression modeling. Environmental Monitoring and Assess-ment 185: 4775–4790.

Dubovyk, O., T. Schachtsiek, A. Khamzina, and G. Menz. 2012. Exploring the biophysicalpotential of dryland afforestation on degraded lands in the downstream of the AmuDarya River, Uzbekistan. Earth Bioresources and Life Quality 1.

FAO. 2000. Carbon Sequestration Options under the Clean Development Mechanism to AddressLand Degradation. World Soil Resources Reports, vol 92. Food and Agriculture Organi-zation of the United Nations, Rome.

Forkutsa, I. 2006. Modeling Water and Salt Dynamics under Irrigated Cotton with ShallowGroundwater in the Khorezm Region of Uzbekistan. Ecology and Development SeriesNo. 39. Cuvillier Verlag, Gottingen.

Forkutsa, I., R. Sommer, Y. I. Shirokova, J. P. A. Lamers, K. Kienzler, B. Tischbein, C.Martius,and P. L.G. Vlek. 2009.Modeling irrigated cotton with shallow groundwater in the Aral SeaBasin of Uzbekistan: II. Soil salinity dynamics. Irrigation Science 27(4): 319–330.

Genxu, W., M. Haiyan, Q. Ju, and C. Juan. 2004. Impact of land use changes on soil carbon,nitrogen and phosphorus and water pollution in an arid region of northwest China. SoilUse and Management 2004(20): 32–39.

Afforestation Species on Abandoned Cropping Sites 425

Dow

nloa

ded

by [

Ast

on U

nive

rsity

] at

19:

40 0

6 O

ctob

er 2

014

Goldhamer, D. A., E. Fereres, M. M. Girona, J. Girona, and M. Cohen. 1999. Sensitivity ofcontinuous and discrete plant and soil water status monitoring in peach trees subjectedto deficit irrigation. Journal of the American Society for Horticultural Science 124(4):437–444.

Guo, L. B., and R. M. Gifford. 2002. Soil Carbon stocks and land use change: A meta analysis.Global Change Biology 8: 345–360.

Heuperman, A., A. Kapoor, and H. Denecke. 2002. Biodrainage - Principles, Experiences andApplications. Food and Agriculture Organization of the United Nations, Rome.

Hillel, D. 2000. Salinity Management for Sustainable Agriculture: Integrating Science, Environ-ment, and Economics. The World Bank, Washington, DC.

Horton, L., T. E. Kolb, and S. C. Hart. 2001. Physiological response to groundwater depthvaries among species and with river flow regulation. Ecological Applications 11(4):1046–1059.

Ibrakhimov, M., A. Khamzina, I. Forkutsa, G. Paluasheva, J. P. A. Lamers, B. Tischbein, P.L. G. Vlek, and C. Martius. 2007. Groundwater table and salinity: Spatial and temporaldistribution and influence on soil salinization in Khorezm region (Uzbekistan, Aral SeaBasin). Irrigation and Drainage Systems 21: 219–236.

IPCC 2007. 4th Assessment Report Climate Change 2007: Mitigation of Climate Change.Cambridge University Press, Cambridge, United Kingdom and New York, NY.

Jiao, J., Z. Zhang, W. Bai, W. Jia, and N. Wang. 2010. Assessing the ecological success ofrestoration by afforestation on the Chinese loess plateau. Restoration Ecology 20(2):240–249.

Kan, E., J. P. A. Lamers, R. Eshchanov, and A. Khamzina. 2008. Small-scale farmers’ percep-tions and knowledge of tree intercropping systems in the Khorezm Region of Uzbekistan.Forests, Trees and Livelihoods 18(4): 355–372.

Katyal, J. C., and P. L. G. Vlek. 2000. Desertification - Concept, causes and amelioration.ZEF-Discussion Papers on Development Policy 33. Center for Development Research(ZEF), Bonn.

Khamzina, A., J. P. A. Lamers, and P. L. G. Vlek. 2008. Tree establishment under deficitirrigation on degraded agricultural land in the lower Amu Darya River region, AralSea Basin. Forest Ecology and Management 255(1): 168–178.

Khamzina, A., J. P. A. Lamers, and P. L. G. Vlek. 2009. Nitrogen fixation by Elaeagnus angu-stifolia L. in the reclamation of degraded croplands of Central Asia. Tree Physiology29(6): 799–808.

Khamzina, A., J. P. A. Lamers, and P. L. G. Vlek. 2012. Conversion of degraded cropland totree plantations for ecosystem and livelihood benefits, pp. 235–248, in C. Martius, I.Rudenko, J. P. A. Lamers and P. L. G. Vlek, eds., Cotton, Water, Salts and Soums -Economic and Ecological Restructuring in Khorezm, Uzbekistan. Dordrecht Heidelberg,London, New York.

Khamzina, A., J. P. A. Lamers, M. Worbes, E. Botman, and P. L. G. Vlek. 2006. Assessingthe potential of trees for afforestation of degraded landscapes in the Aral Sea Basin ofUzbekistan. Agroforestry Systems 66(2): 129–141.

Kotuby-Amacher, J., R. Koenig, and B. Kitchen. 2000. Salinity and Plant Tolerance. UtahState University Cooperative Extension Publication AG-SO-03. Logan.

Kozlowski, T. T. 1997. Responses of Woody Plants to Flooding and Salinity. Tree PhysiologyMonograph No 1.

Krasnouhova, R. A., E. G. Bessonov. 1988. Methodical instructions on humus mapping forsoils of Uzbek SSR (in Russian). Gosagroprom UzSSR, Tashkent.

Lamers, J. P. A., and A. Khamzina. 2008. Fuelwood production in the degraded agriculturalareas of the Aral Sea Basin, Uzbekistan. Bois et Forets des Tropiques 297(3): 43–53.

Lamers, J. P. A., and A. Khamzina. 2010. Seasonal quality profile and production of foliagefrom trees grown on degraded cropland in arid Uzbekistan, Central Asia. Journal ofAnimal Physiology and Animal Nutrition 94(5), e77–e85.

426 T. Schachtsiek et al.

Dow

nloa

ded

by [

Ast

on U

nive

rsity

] at

19:

40 0

6 O

ctob

er 2

014

Lamers, J. P. A., A. Khamzina, and M. Worbes. 2006. The analyses of physiological andmorphological attributes of 10 tree species for early determination of their suitabilityto afforest degraded landscapes in the Aral Sea Basin of Uzbekistan. Forest Ecologyand Management 221(1–3): 249–259.

Long, J. S., and J. Freese. 2006. Regression for Categorical Dependent Variables UsingSTATA. 2nd ed. Stata Press, College Station, TX.

Martius, C., J. P. A. Lamers, P. Wehrheim, A. Schoeller-Schletter, R. Eshchanov, A. Tupitsa,A. Khamzina, A. Akrahamov, and P. L. G. Vlek. 2004. Developing Sustainable Land andWater Management for the Aral Sea Basin through an Interdisciplinary Research. Waterin Agriculture. ACIAR, Australian Center for International Agricultural Research,Canberra.

Miyamoto, S., I. Martinez, M. Padilla, A. Portillo, and D. Ornelas. 2004. Landscape PlantLists for Salt Tolerance Assessment. Texas A&M University Research Center and El PasoWater Utilities.

Musaev, B. S. 2001. Agrochemistry (in Uzbek). Shark, Tashkent.Pereira, L. S., P. Paredes, E. D. Eholpankulov, O. P. Inchenkova, P. R. Teodoro, and M. G.

Horst. 2009. Irrigation scheduling strategies for cotton to cope with water scarcity in theFergana Valley, Central Asia. Agricultural Water Management 96: 723–735.

Rakhimbaev, F. M., N. F. Bezpalov, M. K. Khamidov, K. T. Isabaev, and D. Alieva. 1992.Peculiarities of Crop Irrigation in Lower Amu Darya River Areas (in Russian). FAN,Tashkent, Uzbekistan.

Schonfield, N. J. 1992. Tree planting for dryland salinity control in Australia. AgroforestrySystems 20: 1–23.

Shirokova, Y. I., N. Sharafutdinova, and I. Forkutsa. 2000. Use of electrical conductivityinstead of soluble salts for soil salinity monitoring in Central Asia. Irrigation and DrainageSystems 14: 199–205.

Tischbein, B., U. K. Awan, I. Abdullaev, I. Bobojonov, C. Conrad, I. Forkutsa, M.Ibrakhimov, and G. Poluasheva. 2012. Water management in Khorezm: CurrentSituation and Options for Improvement (Hydrological Perspective), pp. 69–92, inC. Martius, I. Rudenko, J. P. A. Lamers and P. L. G. Vlek, eds., Cotton, Water, Saltsand Soums - Economic and Ecological Restructuring in Khorezm, Uzbekistan. DordrechtHeidelberg, London, New York.

Toderich, K., T. Tsukatani, I. Shoaib, I. Massino, M. Wilhelm, S. Yusupov, T. Kuliev, and S.Ruziev. 2008. Extent of salt affected land in Central Asia: Biosaline agriculture andutilization of the salt-affected resources. Discussion Paper No. 648. Kyoto Institute ofEconomic Research Kyoto University, Japan.

Tomar, O. S., P. S. Minhas, V. K. Sharma, Y. P. Singh, and R. K. Gupta. 2004. Performanceof 31 tree species and soil conditions in a plantation established with saline irrigation.Forest Ecology and Management 177: 333–346.

Tupitsa, A. 2010. Photogrammetric techniques for the functional assessment of tree and forestresources in Khorezm, Uzbekistan. Ecology and Development Series No. 71. CuvillierVerlag Gottingen.

van Dijk, A., R. den Haan, P. van den Hoven, and R. Bos. 1999. Can WARMAP save theAral Sea? Remote Sensing Nieuwsbrief 2000 91: 31–33.

Wang, S., B. Chen, and H. Li. 1996. Euphrates poplar forest. China Environmental SciencePress, Beijing. pp. 17–151.

Afforestation Species on Abandoned Cropping Sites 427

Dow

nloa

ded

by [

Ast

on U

nive

rsity

] at

19:

40 0

6 O

ctob

er 2

014