Agriculture, Ecosystems and Environment 216 (2016) 247–257

Impact of conservation practices on soil aggregation and the carbonmanagement index after seven years of maize–wheat cropping systemin the Indian Himalayas

B.N. Ghosha,*, V.S. Meenab, N.M. Alama, Pradeep Dograa, Ranjan Bhattacharyyac,N.K. Sharmaa, P.K. Mishraa

a ICAR-Central Soil and Water Conservation Research and Training Institute, 218 Kaulagarh Road, Dehradun 248 195, Uttarakhand, IndiabCrop Production Division, ICAR-Vivekananda Institute of Hill Agriculture, Almora 263 601, IndiacCESCRA, NRL Building, ICAR-Indian Agricultural Research Institute, New Delhi 110 012, India

A R T I C L E I N F O

Article history:Received 17 February 2015Received in revised form 23 September 2015Accepted 28 September 2015Available online xxx

Keywords:Runoff and soil lossLabile C poolsSoil degradationMaize and wheat productivity

A B S T R A C T

The carbon management index (CMI) and labile organic carbon (LOC) pools are postulated as verysensitive indicators of changes in soil organic carbon (SOC) due to land degradation within a short time inresponse to management practices. To test this hypothesis, we investigated LOC and CMI under a fieldexperiment (2007–2013) in relation to runoff, soil loss, maize and wheat yields on a 2% (1.15�) land slopeof the Indian Himalayas. In this study, the impacts of several resource conservation practices, includingdifferent combinations of vegetative barriers (VB), minimum tillage (MT), different organic amendments(OA) and weed mulch, were evaluated. Results revealed that the plots under MT + OA with threeapplications of weed mulch had more SOC, macroaggregate-associated C concentrations and macro-aggregates than conventional tillage (CT) + NPK with chemical weed control. Carbon management indexvaried from 47 to 59 and 42 to 55% with different conservation practices at depths of 0–5 and 5–15 cmdepths, respectively. Incorporation of weed mulch along with application of OM, MT and VB (byPalmarosa) under MT improved CMI by 19.7 and 24.2% compared to CT plots with VB (by Panicum) andinorganic NPK at depths of 0–5 and 5–15 cm, respectively. Significant positive correlations were observedbetween CMI and maize yield (r = 0.948; n = 24; P < 0.01), CMI and wheat yield (r = 0.872; n = 24; P < 0.01)and CMI and wheat equivalent yield (r = 0.906; n = 24; P < 0.01). However, significant negativecorrelations were obtained for CMI and runoff (r = �0.701; n = 20; P < 0.01) and CMI and soil loss(r = �0.768; n = 20; P < 0.01). Results established that Palmarosa as VB along with OA plus weed-mulchunder MT was the best management practice for decreasing runoff and soil loss and increasing systemproductivity on a 2% slope in the region. The single value CMI was strongly positively correlated with cropproductivity and negatively correlated with soil loss. Hence, this single value CMI could potentially beused for assessment of soil degradation elsewhere.

ã 2015 Elsevier B.V. All rights reserved.

Contents lists available at ScienceDirect

Agriculture, Ecosystems and Environment

journal homepage: www.elsev ier .com/locate /agee

1. Introduction

Poor long-term management practices under intensive maize(Zea mays L.)–wheat (Triticum aestivum L.) cultivation on slopinglands have adverse impacts on soil–plant–environmental sustain-ability. Integrated fertilization, vegetative barriers (VB) and weedmulch with conservation tillage practices are methods to increase

* Corresponding author. Fax: +91 135275846.E-mail address: bnghosh62@gmail.com (B.N. Ghosh).

http://dx.doi.org/10.1016/j.agee.2015.09.0380167-8809/ã 2015 Elsevier B.V. All rights reserved.

soil organic carbon (SOC) and sustain crop productivity (Ghoshet al., 2012a; Bhattacharyya et al., 2012a; Das et al., 2014). TheConservation Technology Information Center (CTIC) definedconservation tillage as any tillage system that conserves soilwater, decreases soil erosion, and leaves �30% of the soil surfacecovered with residues after a main crop is planted. However, in thesub-temperate Indian Himalayas, where rainfall during summermonths is ample, conservation tillage practices designed for otherareas may be inappropriate (Bhattacharyya et al., 2012a,b, 2013a).Productivity of wheat-based cropping systems in the regiondecreased drastically with the advancing year of cultivation

248 B.N. Ghosh et al. / Agriculture, Ecosystems and Environment 216 (2016) 247–257

adopting no tillage (NT), despite improved moisture conservationand hydraulic conductivities (Bhattacharyya et al., 2009).

Integrated nutrient management (INM) can increase soilaggregate stability (Bhattacharyya et al., 2011). This tends toincrease the proportion of macro aggregates (>0.25 mm diameter)by binding microaggregates (0.053–0.25 mm diameter), therebydecreasing the proportion of microaggregates (Tisdall and Oades,1982). In addition, added organic matter (either direct addition orthrough the added root and residue inputs in the plots under INM)provides the substrate that leads to improved soil aggregation.Many researchers emphasize that C accumulation in soils is alteredby management practices (Six et al., 2002; Bhattacharyya et al.,2009; Das et al., 2014). This process is governed by the interplay ofseveral factors, including climate, soil type, substrate biochemistry,C inputs and associated microbial community structure(Majumder et al., 2007; Venkatesh et al., 2013). Managementpractices that ensure more residues are returned to soils areanticipated to increase SOC content. Identifying such systems orpractices is a priority for sustaining crop productivity.

Total SOC is composed of labile and recalcitrant pools. Labileorganic carbon (LOC) fraction comprises of the physical fraction(including particulate organic matter), chemical fraction (KMnO4-C) and biological fraction (including microbial biomass C, MBC).The recalcitrant pool (humus) resists decomposition (Mandal et al.,2013). The relative proportion of these fractions determines soilquality and is, therefore, a crucial factor in soil C dynamics.Bhattacharyya et al. (2009) reported that conventional cultivationhad the lowest total SOC content, whereas NPK + farmyard manure(FYM)-amended soils had the largest total SOC in the 0–45 cm soillayer after 30 years of soybean (Glycine max L.)–wheat cropping inthe Indian Himalayas. The highest proportion of the labile C wasobserved under NPK + FYM, whereas the proportion of non-labile Cfraction was more under NPK + FYM (Ghosh et al., 2006).

Changes in labile C pools occur within a short period (one to twoyears) and this pool (LOC) can be used to assess land management

Fig. 1. Experimental layout on a 2% (1.15�) slope, two vegetative barriers with 1 m widPalmarosa as vegetation strips, (c) Establishment and layout, gauging devices in a

barrier + NPK + conventional tillage; T2: VB + FYM + MT + 1 WM = vegetative barrier + favegetative barrier + farmyard manure + vermi-compost + minimum tillage + 2 weed-manure + vermi-compost + poultry manure + minimum tillage + 3 weed-mulch.

effects. Walkley Black C (WBC) or oxidizable soil organic C (asestimated following Walkley and Black, 1934) mostly representsthe entire labile C pool and some portion of long-lived C pools,which takes longer to change, due to land management effects (Sixet al., 1999). Hence, several workers reported that the KMnO4-oxidizable SOC or LOC is a more sensitive SOC indicator comparedwith total SOC or WBC (Moharana et al., 2012; Liu et al., 2014;Bhattacharyya et al., 2006). Thus, the carbon management index(CMI), developed from lability concepts, is considered the mosteffective tool for quantitative estimation of soil quality index (Blairet al., 1995). Several researchers (Ghosh et al., 2012b; Moharanaet al., 2012; Mandal et al., 2013; Six et al., 2000) have reported CMIas an early indicator of soil quality changes due to managementpractices and thus were able to assess best management practicesthat impede soil degradation.

A number of studies, reported in literature, provide importantinformation pertaining to several soil management practices, butfew studies integrate C pools and carbon lability into CMI as anapproach to appraise the capacity of conservation practices inarable soils to promote soil quality and decrease soil loss. Mulchingin between maize rows (90 � 20 cm; row to row � plant to plantdistance) is a common practice in this region (on gentle slopes), fordecreasing both raindrop impact and soil erosion (Ghosh, 2011).However, the availability of large quantities of mulching materials,from either crop residues or forest litter, is a constraint(Bhattacharyya et al., 2013b). Furthermore, weed management,either through weedicides or manual weeding, is a challenge,because of prolonged rains (22–26 days) in August. Hence, farmersof this region generally apply cut weeds between maize rows as acover-mulch, but the interval between cuttings varies amongfarmers. So weed management as ‘cut and mulch’ at one specifictime is called ‘one live mulch treatment.’ Farmers do practice up tothree live mulches during the rainy season.

To date, few systematic studies have been conducted on soildegradation assessment and mitigation options using the CMI. We

th at 50 m and 100 m of experimental plots (a) Panicum as vegetation strips, (b)maize–wheat cropping system. Treatment details: T1: VB + NPK + CT = vegetativermyard manure + minimum tillage + 1 weed-mulch; T3: VB + FYM + VC + 2 WM =mulch, and T4: VB + FYM + VC + PM + MT + 3 WM = vegetative barrier + farmyard

B.N. Ghosh et al. / Agriculture, Ecosystems and Environment 216 (2016) 247–257 249

hypothesized that the concept of CMI is applicable in sloping landswhere a significant portion of the non-labile C pool is lost throughsoil erosion (especially erosion of silt and clay fractions) thatcauses productivity decline over time. Information regardingassessment of soil aggregation/degradation based on CMI, andespecially its relationships with crop productivity and soil erosionrates, are limited. In this study, an attempt was made toquantitatively estimate CMI as affected by different soil conserva-tion measures and its relationship with maize–wheat systemproductivity on gently sloping (�2%) lands of the Indian Himalayas.Hence, the objectives of this study were:

(i) To determine the effects of long-term nutrient management,minimum tillage (MT) and weed mulch practices with VB onSOC pools and soil aggregation under a maize–wheat croppingsystem on a silty clay loam soil after seven years of cropping.

(ii) To assess the best management practice that mitigates soildegradation.

(iii) To establish relationships between CMI, mean (of seven years)runoff, soil loss and mean maize and wheat yields.

2. Materials and methods

2.1. Site details

The fixed plot experiment was initiated in June 2007 at theexperimental farm of the Central Soil and Water ConservationResearch and Training Institute (CSWCR & TI), Dehradun, India(30�20/40/N, 77�52/12//E; 516.5 m above mean sea-level on a 2%,1.15� slope) (Fig.1). The surface soil texture was silty clay loam (finemixed hyperthermic Typic Udorthent). The mean (1956–2011)annual precipitation is 1625 mm, with �80% falling during therainy season (June–September). Year-wise precipitation, numbersof rainy days and rainfall intensity during the maize crop growthperiod (15 June–15 September) are presented in Table 1. Theaverage daily maximum and minimum air temperatures rangedfrom 31.7 to 20.6 �C in June and 17.8 to 1.1 �C in January. Initialphysicochemical properties and fertility status of the experimentalsoil before commencing the study are reported in Table 2. Theexperiment was laid out in a randomized block design with thefollowing treatments:

T1: Panicum + 100:60:40 (N:P2O5:K2O) + conventional tillage(CT) + chemical method of weed control in both crops.

T2: Palmarosa + farmyard manure (FYM) at 5 t ha�1 + minimumtillage (MT) + 1-live mulch (at 20 days after sowing (DAS) of themaize crop).

T3: Palmarosa + FYM at 5 t ha�1 + vermi-compost (VC) at 1.0 tha�1 + MT + 2-live mulch (at 20 and 40 DAS of the maize crop).

T4: Palmarosa + FYM at 5 t ha�1 + VC at 1.0 t ha�1 + poultrymanure (PM) at 2.5 t ha�1 + MT + 3-live mulch (at 20, 40 and 60DAS of the maize crop).

The cropping system was maize (cultivar Kanchan)–wheat(cultivar PBW4). The growing period of maize was from June to the

Table 1Annual precipitation characteristics of the study area during the crop growth period.

Year Precipitation (mm) Number of rainy days Mea

2007 958.8 42 55.02008 1414.4 51 35.52009 961.6 41 58.02010 2070.0 48 56.22011 1336.8 54 28.02012 1304.4 46 54.42013 1820.0 52 40.1

first week of October (rainy season) and that for wheat, fromOctober to the first week of April (winter season or Rabi). Alltreatments were imposed in maize crops and wheat was grown onresidual fertility under rainfed conditions.

2.2. Tillage, manuring and mulching

Minimum tillage (a 50% tillage reduction) i.e. two tillageoperations, one before sowing of the maize crop (between late Juneand the first week of July) and one after harvesting (end ofSeptember) were conducted using a tractor drawn cultivator to�15 cm depth along with retention of 30% of maize stover residues.In conventional tillage, there were four tillage operations. The firsttillage was performed in the pre-monsoon season (April/May) andthe second one was performed in May/June, some 20–25 days afterthe first tillage. The objective of the second tillage was toincorporate FYM at maize sowing. The third tillage was conductedduring June/July for sowing of the maize crop and the fourth aftermaize harvest (September/October) at deeper depth (>15 cm)using a tractor drawn cultivator. FYM at 5 t ha�1, VC at 1.0 t ha�1 andPM at 2.5 t ha�1 were applied at the first tillage operation andthoroughly mixed with soil before maize sowing. Weeds were cutand mulched at 20 days after sowing (DAS) to make it a ‘1-livemuch treatment,’ at 20 DAS and 40 DAS to make it a ‘2-live mulchtreatment’ and at 20 DAS, 40 DAS and 60 DAS to make it a ‘3-livemulch treatment.’ Live mulch treatments were given only to themaize crops in all years. The fresh and dry biomasses added as livemulch were: 1.65, 0.52 for T2, 4.43, 1.47 for T3 and 6.99,2.18 t ha�1 yr�1 for T4, respectively. Maize was sown with theonset of monsoon rains (end of June to first week of July everyyear). Similar tillage operations were followed for the wheat crop.Wheat was sown before the third week of November. Fertilizerswere urea for N, single superphosphate for P and muriate of potashfor K. These were applied for maize at the recommended doses ofN, P, and K of 120, 60 and 60 kg ha�1, respectively. In all treatments,fertilizers were broadcast as basal doses. The wheat used theresidual fertility from the maize crop. Weeds were controlled usingattrazine and pendimethelen with active ingredient 1.5 mL L�1 ofwater in CT plots for maize and wheat crops, respectively. Onemanual weeding was performed in the wheat crop 30–35 DAS in allplots and no manure and fertilizer were applied to wheat. Maizegrain was harvested in September and wheat in April each year.Maize and wheat grain yields were expressed at 12% moisturebasis.

2.3. Soil sampling and processing

After wheat harvest (May 2014), two sets of triplicateundisturbed soil cores were collected from 0 to 5 and 5 to15 cm soil depths using a core sampler (7.5 cm diameter). Bulkdensity was determined using one sample set. Samples fromindividual plots (the second set) were thoroughly mixed, air-dried,and passed through a 4.75 mm sieve. We found no aggregates>4.75 mm diameter. Air-dried samples were placed in plastic bags

n precipitation intensity (mm h�1) Mean intensity of extreme events(mm h�1)

106.7 98.4 107.5 101.3 82.0 102.8

91.0

Table 2Physicochemical properties and fertility status of the experimental soil before commencing the study.

Soil characteristics Mean (n = 3) � SD Method

Mechanical compositionSand (%)a 52.0 0.46 Bouyoucos (1962)Silt (%) 35.5 0.75Clay (%) 12.5 0.12pH (1:2.5 soil:water) 5.8 0.002 Jackson (1973)Bulk density (Mg m�3) 1.36 0.004 Veihmeyer and Hendrickson (1948)Oxidizable organic C (g kg�1) 6.0 0.004 Walkley and Black (1934)Available N (kg ha�1) 270 2.43 Subbiah and Asija (1956)Available P (kg ha�1) 24 1.27 Olsen et al. (1954)Available (kg ha�1) 110 3.24 Hanway and Heidel (1952)Infiltration rate (mm h�1)b 1.42 0.021 Devaum and Gifford (1984)

a Sand: 2000–50 mm; silt: 50–2 mm and clay <2 mm. �SD: standard deviation.b Infiltration rate of the soil profile.

250 B.N. Ghosh et al. / Agriculture, Ecosystems and Environment 216 (2016) 247–257

and stored at ambient laboratory temperature. A soil sub-samplewas taken from both depths and analysed for soil aggregation, totalSOC and labile and recalcitrant SOC pools, as reported below.

2.4. Separation of soil aggregates and fractionation of SOC

Aggregate-size separation was performed using a wet sievingmethod (Elliott, 1986). Soil samples (100-g air-dried <4.75 mm)were placed on top of a 2.0 mm sieve and submerged for 5 min indeionized water, to allow slaking (Kemper and Rosenau, 1986).Sieving was performed mechanically moving the sieve up anddown 3 cm, 50 times in 2 min using a modified Yoder’s apparatus. Aseries of three sieves (2000, 250 and 53 mm) was used to obtainfour aggregate fractions: (i) >2000 mm (large macroaggregates),(ii) 250–2000 mm (small macroaggregates), (iii) 53–250 mm(microaggregates) and (iv) <53 mm (silt-plus clay-size particles).

A subsample was taken from the collected soil suspension. Thesubsample was passed through the 53 mm sieve (‘silt clay’-sizedfraction). This subsample, along with soil aggregate fractionsretained on different sieves, was oven-dried at 50 �C, weighed andstored in glass jars for total SOC analysis. Different particle-sizefractions obtained from the aggregate analysis were measured andconverted into fractions. The fraction of aggregates <53 mm wascalculated by summing up the total mass of the material retainedon the >53 mm sieves and this was subtracted from the total weightof the soil taken for wet-sieving analysis. The mean weightdiameter (MWD) was calculated (van Bavel, 1949). Small and largemacroaggregates together constitute the macroaggregates.

The total SOC concentrations of bulk soil and in each aggregatefraction were determined following the dry combustion method(Nelson and Sommers, 1982) using a CHN analyzer. Since themeasured inorganic C (carbonates) contents of the samples werenil, the total soil C was equal to total SOC. Labile organic carbonpool was determined using 333 mM KMNO4 (Blair et al., 1995).Then the non-labile carbon (NLC) pool was estimated by deductingLOC from total SOC (total SOC � LOC = NLC).

Carbon management index (CMI) was calculated following Blairet al. (1995). Topsoil samples were also collected from the nearbyforest (30�20/40//N latitude, 77�52/12//E; 526.5 m above meansea). The forest consisted mainly of sal (Shorea robusta) trees andwas �100 years old. Undisturbed forest soils were considered asreference soil samples. The physicochemical properties of theforest soils indicated that sand ranged from 54.41 to 71.95%, clayfrom 7.51 to 21.14% and silt from 9.18 to 27.43%. The texture wasloamy sand, having porosity values ranging between 42.5 and50.6%. The surface soil was acidic (pH ranged from 5.34 to 6.46),having CEC of 18.4 cmol (p+) kg�1. The Doon sal forest soils werecharacterized by high exchangeable Ca2+, Mg2+ and SOM and werenot exposed to degradation processes (Mukesh et al., 2011).

CMI was calculated using the following equations:

Lability of CðLÞ ¼ C in fraction oxidized by KMnO4

C remaining unoxidized by KMnO4¼ CL

CNL

Liability IndexðLIÞ ¼ Lability of C in sample soilLability of C in reference soil

Carbon Pool IndexðCPIÞ ¼ Sample total CReference total C

¼ CT SampleCT Reference

Carbon Management Index ðCMIÞ ¼ CPI � LI � 100

2.5. Runoff and soil loss measurement

Runoff data were recorded at 0800 (local time) using a stagelevel recorder. Data were taken after each rainfall event from 15June to 15 September (rainy season) in all years (2007–2013) bymeasuring the hydrograph connected with a Coshocton wheel(Ghosh et al., 2012a). Daily precipitation was recorded (at 0800)using a rain gauge in all years. Runoff was collected and thoroughlystirred. About 1 L runoff water was taken from each tank todetermine accumulated sediment in the runoff tank for each plot.The resultant suspensions were filtered using Whatman 42 filterpaper (pore size 2.5 mm). Sediment on the filter paper was oven-dried for 24 h at 105 �C and weighed to obtain soil loss data (Ghoshet al., 2012a). Seasonal mean (of seven years) runoff and soil lossdata were calculated.

2.6. Grain yield and calculation of wheat equivalent yield

At harvest, grain yields of maize and wheat were determinedand there were three replicates per individual 16 m2 (2 � 8 m) plot.Mean (of seven years) grain yields of both crops were calculated.The maize grain yield was converted into wheat equivalent yield(WEY) based on year-wise market price of maize and wheat crops,following Bhattacharyya et al. (2009). The converted maize grainyield of a year in terms of WEY was then added to the actual wheatgrain yield in that year to obtain total system productivity.Therefore, WEY expresses total grain yield for the maize–wheatcropping system during 2007–08 to 2013–14.

2.7. Statistical analyses

All data were analysed using ANOVA for a randomized blockdesign. Duncan’s Multiple Range Test (DMRT) significant

B.N. Ghosh et al. / Agriculture, Ecosystems and Environment 216 (2016) 247–257 251

difference was used as a post hoc mean separation test (P < 0.05)using SAS 9.3 (SAS Institute, Cary, North Carolina, USA).

3. Results and discussion

3.1. Aggregate-size distribution

Aggregate size distribution in both soil depths were signifi-cantly impacted by management practices (Tables 2 and 3).Macroaggregates accounted for >51% of total aggregates. In topsoil(0–5 cm soil layer), these were the dominant water-stableaggregates (WSA). Significantly higher (60%) water-stable macro-aggregates was recorded in T4 plots compared with T1 in thetopsoil, with a concurrent decrease in microaggregates in the T2and T3 plots. A similar trend also was recorded in sub-surface soil(Table 3). Small macroaggregates were the greatest proportion ofthe whole soil, followed by aggregates <53 mm in topsoil. Plotsunder T4 had significantly more large and small microaggregatesthan T1 plots in both soil layers. The T4 plots had significantly morelarge macroaggregates, with a concomitant decrease in ‘silt + clay’sized aggregates compared with T1 plots in the topsoil. Conse-quently, T4 plots had significantly higher MWD and proportion ofmacroaggregates in topsoil.

In the sub-surface soil, size distributions of aggregates werealso significantly influenced by tillage, weed mulch and vegetativebarrier practices. Small macroaggregates comprised the greatestproportion of the whole soil, followed by aggregates <53 mm in the5–15 cm soil layer. Subsurface soil (5–15 cm depth) had 34% highermacroaggregates than microaggregates (Table 4). The fraction ofwater stable aggregates was significantly higher in T4 plots than inT3. Consequently, T3 plots had significantly higher MWD in the5–15 cm soil layer, followed by T4, T2 and T1 plots, in this order(Fig. 2). Aggregate data revealed that the macroaggregatesincreased by 39% and microaggregates decreased by 9% in T4 plots(with Palmarosa) compared with T1 plots (with Panicum). Decreasein microaggregates and increase in macroaggregates with appli-cation of conservation tillage (MT + OA + weed recycling throughmulch) with VB might have enhanced soil aggregation processes(Bhattacharyya et al., 2009, 2013b). Sediment deposition justbehind the grass barriers coupled with conservation tillageincreased the amount of macroaggregates in sloping lands ofthe Indian sub-Himalayas (Ghosh et al., 2012a).

3.2. Bulk soil organic carbon

Mineral fertilization, VB, weed mulch and MT practicessignificantly influenced total SOC concentrations after seven yearsin both soil layers (topsoil and 5–15 cm depth). Plots under T4 had23.2 and 13.9% higher bulk soil SOC in topsoil and 5–15 cm depth,respectively, compared with the farmers’ practice (T1). MaximumSOC content was found in T4 plots and the value was larger than theplots under T1, T2 and T3 in topsoil. Similar trends of bulk SOC were

Table 3Impact of long-term fertilization, vegetative barrier and tillage practices on soil aggreg

Treatments Size distribution of aggregates (mm)

>2000 250–2000 53–250

(g aggregate 100 g�1 dry soil)

T1: VB + NPK + CT 13.29b 30.62b 31.07a

T2: VB + FYM + MT + 1 WM 14.26b 38.58a 26.87b

T3: VB + FYM + VC + MT + 2 WM 22.06a 36.91a 22.45c

T4: VB + FYM + VC + PM + MT + 3 WM 15.75b 37.55a 21.69c

WSA: water-stable aggregates. Means followed by same letter within a column for a parDuncan’s Multiple Range Test (DMRT).

observed in the 5–15 cm soil layer (Fig. 3). This increase in total SOCwas probably due to significant increases in C inputs with organicamendments coupled with conservation tillage (Ghosh et al.,2012a; Das et al., 2013). Total SOC increased after FYM applicationin long-term experiments at Almora under a soybean–wheatcropping system (Bhattacharyya et al., 2011). In all managementpractices, LOC was higher in topsoil and decreased with depth. Thiswas mainly due to mineralizable and readily hydrolysable C inputs,resulting in more microbial activity in topsoil (Kaur et al., 2008).The SOC accumulation was greater in topsoil than at lower depths.This was mainly due to more addition of roots and plant biomass intopsoil and decreased nutrient concentration and biologicalactivity in the 5–15 cm soil layer (Ghosh et al., 2012b; Bhattachar-yya et al., 2013a).

3.3. Aggregate-associated carbon

Aggregate-associated C concentrations significantly increasedwith aggregate size (large and small macroaggregates vs micro-aggregates and silt + clay sized aggregates) and decreased with soildepth (Table 5). Plots under T4 had significantly higher aggregate-associated C in all the size classes compared with T1, T2 and T3 plotsin the 0–5 and 5–15 cm soil layers (Table 5). The 53–250 mm soilfractions did not significantly vary within the soil layers. The plotsunder T4 had higher total SOC, soil aggregation and macroaggre-gate-associated C concentrations compared with T1. The decreasein aggregate size with continuous CT could be due to macroaggre-gate disruption that might have exposed SOM previously protectedagainst oxidation (Pinheiro et al., 2004). In contrast, MT improvedmacroaggregation, particularly within topsoil (Bhattacharyyaet al., 2013b). Greater total SOC concentrations within macro-aggregates could be due to: (i) presence of decomposing roots andhyphae within macroaggregates, (ii) lower decomposable SOMassociated with these aggregates and (iii) direct contribution ofSOM to more microaggregates within macroaggregates (Pugetet al., 1995; Bhattacharyya et al., 2012a,b). This not only increasedSOC concentrations, but also contributed to their stabilization(Bhattacharyya et al., 2011). The significantly greater concen-trations of large and small macroaggregate-associated C indicatelong-lasting macroaggregate turnover rates due to less soildisturbance under MT (Six et al., 1999).

3.4. Carbon pools

Seven years of conservation practices under a maize–wheatcropping system also significantly influenced C pools (Tables 6 and7). Plots under T4 had significantly higher LOC and total SOC thanT3, T2 and T1 plots. But, in the case of C lability in the 0–5 and 5–15 cm soil layers, treatment impacts on this fraction were notsignificant (Table 7). The maize–wheat cropping system resulted ingreater accumulation of LOC and total SOC in topsoil. The

ation after seven years of maize–wheat cropping system in the 0–5 cm soil layer.

Macro-aggregates Micro-aggregates WSA

<53(g aggregate 100 g�1 dry soil) (%)

25.02a 43.91b 31.07a 43.00c

20.30ab 52.84a 26.87b 51.84b

18.58b 58.97a 22.45c 53.64b

25.01a 53.30a 21.69c 59.57a

ticular management practice are not significantly different at P < 0.05 according to

Table 4Impact of long-term fertilization, vegetative barrier and tillage practices on soil aggregation after seven years of maize–wheat cropping system in the 5–15 cm soil layer.

Treatments Size distribution of aggregates (mm) Macro-aggregates Micro-aggregates WSA

>2000 250–2000 53–250 <53(g aggregate 100 g�1 dry soil) (g aggregate 100 g�1 dry soil) (%)

T1: VB + NPK + CT 7.80c 30.55b 27.27bc 34.38a 38.34c 27.27bc 38.08b

T2: VB + FYM + MT +1 WM 9.47bc 35.28a 26.84c 28.40b 44.75b 26.84c 45.13a

T3: VB + FYM + VC + MT + 2 WM 13.24a 35.04a 30.64ab 21.09c 48.27a 30.64ab 46.92a

T4: VB + FYM + VC + PM + MT + 3 WM 11.29b 36.10a 32.76a 19.85c 47.40ab 32.76a 46.73a

WSA: water-stable aggregates. Mean followed by same letter within a column for a particular management practice are not significantly different at P < 0.05 according toDMRT.

Fig. 2. Impact of long-term fertilization, vegetative barriers, weed-mulch and tillage practices on mean weight diameter (mm) after seven years in the 0–5 cm and 5–15 cmsoil layers under a maize–wheat cropping system. Bars followed by the same upper-case letter between treatments within a soil layer are not significantly different at P < 0.05according to Duncan’s Multiple Range Test (DMRT). Error bars indicate standard deviations. T1: VB + NPK + CT = vegetative barrier + NPK + conventional tillage; T2:VB + FYM + MT + 1 WM = vegetative barrier + farmyard manure + minimum tillage + 1 weed-mulch; T3: VB + FYM + VC + 2 WM = vegetative barrier + farmyard manure + vermi-compost + minimum tillage + 2 weed-mulch, and T4: VB + FYM + VC + PM + MT + 3 WM = vegetative barrier + farmyard manure + vermi-compost + poultry manure + minimumtillage + 3 weed-mulch.

Fig. 3. Impacts of fertilization, vegetative barrier, weed-mulch and tillage practices on bulk soil organic C concentration (g kg–1 soil) after seven years in the 0–5 and 5–15 cmsoil layers under a maize–wheat cropping system. Bars followed by same upper-case letter between treatments within a soil layer are not significantly different at P < 0.05according to DMRT. Error bars indicate standard deviations. T1: VB + NPK + CT = vegetative barrier + NPK + conventional tillage; T2: VB + FYM + MT + 1 WM = vegetativebarrier + farmyard manure + minimum tillage + 1 weed mulch; T3: VB + FYM + VC + 2 WM = vegetative barrier + farmyard manure + vermi-compost + minimum tillage + 2 weedmulch, and T4: VB + FYM + VC + PM + MT + 3 WM = vegetative barrier + farmyard manure + vermi-compost + poultry manure + minimum tillage + 3 weed mulch.

252 B.N. Ghosh et al. / Agriculture, Ecosystems and Environment 216 (2016) 247–257

distribution of different carbon fractions followed the order:T4 > T3 > T2 > T1 in both soil layers.

It was conclusively established that tillage-nutrient-weedmanagement practices significantly affected different C poolsand the combined use of VB + FYM + VC + PM + MT + 3 WM (T4

treatment) had most impact. Among SOC fractions, there weremarginal increases in LOC contents in soils under INM treatments(T3 and T4) compared with farmers’ practice (T1). However, theincreases in non-labile C pools under T3 and T4 plots comparedwith T1 plots were more. While labile SOC pools augmented with

Table 5Impact of long-term fertilization, vegetative barrier and tillage practices on organic C contents within soil aggregate size classes in the 0–5 cm and 5–15 cm soil layers afterseven years of maize–wheat cropping.

Treatments Size distribution of aggregates (mm)in the 0–5 cm soil layer

Size distribution of aggregates (mm)in the 5–15 cm soil layer

>2000 250–2000 53–250 <53 >2000 250–2000 53–250 <53(g C kg�1 soil aggregate fraction)

T1: VB + NPK + CT 13.27c 11.03c 9.19a 5.93b 8.99c 8.04b 7.50a 5.67b

T2: VB + FYM + MT + 1 WM 13.31c 12.48b 8.66a 6.71b 9.68bc 9.27ab 7.93a 6.17b

T3: VB + FYM + VC + MT + 2 WM 15.29b 14.45a 8.65a 9.44a 10.59b 9.57a 8.00a 7.52a

T4: VB + FYM + VC + PM + MT + 3 WM 17.10a 14.06a 9.36a 9.40a 12.69a 9.90a 7.68a 7.77a

Mean followed by same letter within a column for a particular management practice is not significantly different at P < 0.05 according to DMRT.

Table 6Impact of long-term fertilization, vegetative barrier, weed mulch and tillage on carbon pools and carbon management index after seven years of maize–wheat croppingsystem in the 0–5 cm soil layer.

Treatments TOC LOC NLC LC LI CPI CMI(g C kg�1)

T1: VB + NPK + CT 9.00c 4.25a 4.75c 0.90a 1.92a 0.25c 47.23b

T2: VB + FYM + MT + 1 WM 9.75c 4.52b 5.23bc 0.87a 1.85a 0.27c 49.53b

T3: VB + FYM + VC + MT + 2 WM 10.83b 4.90c 5.92ab 0.83a 1.76a 0.30b 52.38ab

T4: VB + FYM + VC + PM + MT + 3 WM 11.79a 5.37d 6.42a 0.87a 1.85a 0.32a 58.82a

Control (forest soil) 36.37 11.64 24.73 0.47 1.00 1.00 100.00

Means followed by same letter within a column for a particular management practice are not significantly different at P < 0.05 according to DMRT. TOC: total organic carbon;LOC: labile organic carbon; NLC: non-labile C; CPI: carbon pool index; LC: lability of C; LI: lability index; CMI: carbon management index.

Table 7Impact of long-term fertilization, vegetative barrier, weed mulch and tillage on carbon pools and carbon management index after seven years of maize–wheat croppingsystem in the 5–15 cm soil layer.

Treatments TOC LOC NLC LC LI CPI CMI(g C kg�1)

T1: VB + NPK + CT 9.21c 3.98c 5.23a 0.77a 1.67ab 0.25c 41.89c

T2: VB + FYM + MT + 1 WM 9.82b 4.19c 5.64a 0.74a 1.62b 0.27b 43.45bc

T3: VB + FYM + VC + MT + 2 WM 10.16b 4.55b 5.60a 0.82a 1.79ab 0.28a 49.47ab

T4: VB + FYM + VC + PM + MT + 3 WM 10.66a 4.95a 5.71a 0.87a 1.91a 0.29a 55.25a

Control (forest soil) 36.67 11.52 25.15 0.46 1.00 1.00 100.00

Means followed by same letter within a column for a particular management practice are not significantly different at P < 0.05 according to DMRT. TOC: total organic carbon;LOC: labile organic carbon; NLC: non-labile C; CPI: carbon pool index; LC: lability of C; LI: lability index; CMI: carbon management index.

Fig. 4. Impacts of fertilization, vegetative barrier and tillage practices on mean runoff after seven years of maize–wheat cropping. Bars followed by the same upper-case letterbetween treatments within a soil layer are not significantly different at P < 0.05 according to DMRT. Error bars indicate standard deviations. T1: VB + NPK + CT = vegetativebarrier + NPK + conventional tillage; T2: VB + FYM + MT + 1 WM = vegetative barrier + farmyard manure + minimum tillage + 1 weed mulch; T3: VB + FYM + VC + 2 WM =vegetative barrier + farmyard manure + vermi-compost + minimum tillage + 2 weed mulch, and T4: VB + FYM + VC + PM + MT + 3 WM = vegetative barrier + farmyardmanure + vermi-compost + poultry manure + minimum tillage + 3 weed mulch.

B.N. Ghosh et al. / Agriculture, Ecosystems and Environment 216 (2016) 247–257 253

254 B.N. Ghosh et al. / Agriculture, Ecosystems and Environment 216 (2016) 247–257

the increase in the quantity of added manures through VC and PM,these manures were also very effective in boosting non-labile Cpools. The estimated biomass C inputs from root biomass,rhizodeposition, stubble, residues, organic amendments (FYM,VC and PM) and weed mulch were: 3.67, 4.44, 5.54 and6.66 t ha�1 yr�1 in T1, T2, T3 and T4, respectively. However, only0.07 t ha�1 (�2% of the estimated gross C input), 0.18 t ha�1 (�4%),0.28 t ha�1 (�5%) and 0.47 t ha�1 (�7%) C were retained within the<53 mm fraction of soil aggregates in plots under T1, T2, T3 and T4,respectively. Although large amounts of crop residues and weedmulch were received by the system, a very small portion wasretained because of loss of SOC through runoff and sediment(Ghosh et al., 2012a).

3.5. Carbon management index (CMI)

Carbon management index was significantly influenced byconservation management practices. Significantly higher CMI (8%)was recorded in topsoil compared with the sub-surface layer. In thecase of topsoil, it varied from 47 to 59%. However, in the 5–15 cmsoil layer, the CMI varied from 42 to 55% with differentmanagement practices (Tables 6 and 7). Maximum CMI wasrecorded in topsoil of plots under T4 and the CMI value was similarto that in T3 plots (Table 6). The substitution of FYM by PM and VCresulted in similar CMI values. Incorporation of weed mulch alongwith application of organic manures, MT and VB (T4) increased CMIby �19.7 and 24.2% over T1 plots in the 0–5 and 5–15 cm soil layers,respectively. Plots under T4 had significantly higher CMI than T1 inboth soil layers (Table 7).

Soil organic C pools directly affect soil physical, chemical andbiological properties (Blair and Crocker, 2000). This study indicatesthat CMI could be used as a more sensitive indicator comparedwith total SOC, for assessing changes due to management practicesin the Indian Himalayas under a maize–wheat system. Hence, thecombination of both SOC pool and CMI (Blair et al., 1995) couldresult in a useful parameter for assessing the ability of agriculturalmanagement practices to promote soil quality. Significantly higherCMI in plots under T4 compared with mineral fertilization is due tothe enhancement in annual C inputs and the difference in organicmatter quality, thus modifying the lability of C to an oxidized form(Bhattacharyya et al., 2011). Blair et al. (2006) also reported thatlong-term adoption of INM significantly increased CMI comparedto mineral fertilization.

Fig. 5. Impacts of fertilization, vegetative barrier and tillage practices on annual soil loss

of the Indian Himalayas. Bars followed by the same upper-case letter between treatmentsbars indicate standard deviations. T1: VB + NPK + CT = vegetative barrier + NPK + convenminimum tillage + 1 weed-mulch; T3: VB + FYM + VC + 2 WM = vegetative barrier +

VB + FYM + VC + PM + MT + 3 WM = vegetative barrier + farmyard manure + vermi-compos

3.6. Runoff and soil loss

Mean soil loss and runoff rates were significantly affected bynutrient management, tillage and vegetative barriers practices(Figs. 4 and 5). Highest mean annual soil loss and runoff rates wererecorded from T1 plots, while the lowest from T4 plots. Runoff wassignificantly influenced by management practices, as well as thedifferent intervals of weed mulch. Significantly lower runoff wasrecorded from T4 plots followed by T2 and T1 plots. However,maximum runoff loss was recorded from T1 plots and was �34%more than T4 plots. Soil loss varied from 3.14 to 5.14 t ha�1. Leastsignificant soil loss was recorded from T4 plots and was similar toT3plots. The plots under T1 had �39% higher soil loss than T4 plots.Incorporation of weed mulch along with application of organicmanure and MT practices decreased erosion rates loss (Fig. 5).Greater canopy cover of maize and ground cover by weed mulchwith T4 treatment decreased runoff compared with other treat-ments. Because of greater canopy cover and decreased disturbanceto soil by MT, soil loss was significantly less in T4 than in all othertreatments (Ghosh et al., 2012a, 2013). With mulch levels of 2 and4 t ha�1on a 10% slope, Lal (1975) observed that soil erosiondecreased by 97.0 and 99.6%, respectively, compared to soil erosionon non-mulched treatments.

The consistently smaller runoff in plots under T4 was possiblydue to the dense vegetation causing silt deposition (Sur andSandhu, 1994). Retention of weed mulch on plots under T4 alsocontributed to a significant decline in soil loss and runoff. Similarobservations were reported by Lal et al. (1996). Results showedthat weed cut and mulch with MT, along with Palmarosa as avegetation buffer strip, maintained soil in a non-disturbedcondition during high intensity rainfall events. Standing maizeprovided very good canopy cover, while vertical mulching resultedin greater water infiltration (4.24 mm h�1 in plots under T4) thanfields without residues (1.82 mm h�1 in plots under T1). Thisprotective mulching further decreased soil erosion rates (Gilleyet al.,1986). In addition to MT, the addition of organic amendmentson T4 plots might have also improved the soil structure, which ledto increased infiltration, thereby reducing runoff coefficients ineach rainfall event. These results support that Palmarosa alongwith weed derived mulch, organic amendments and MT couldensure soil and water conservation.

(mean of seven years) under the maize–wheat cropping on a 2% slope in the foothills within a soil layer are not significantly different at P < 0.05 according to DMRT. Errortional tillage; T2: VB + FYM + MT + 1 WM = vegetative barrier + farmyard manure +farmyard manure + vermi-compost + minimum tillage + 2 weed-mulch, and T4:t + poultry manure + minimum tillage + 3 weed-mulch.

Fig. 6. Impacts of fertilization, vegetative barrier and tillage practices on yield (mean of seven years) of maize, wheat and maize–wheat system productivity (in terms of wheatequivalent yield; WEY) after seven years of maize–wheat cropping. Bars followed by the same letter between treatments within a soil layer are not significantly different atP < 0.05 according to DMRT. Error bars indicate standard deviations. T1: VB + NPK + CT = vegetative barrier + NPK + conventional tillage; T2: VB + FYM + MT + 1 WM = vegetativebarrier + farmyard manure + minimum tillage + 1 weed mulch; T3: VB + FYM + VC + 2 WM = vegetative barrier + farmyard manure + vermi-compost + minimum tillage + 2 weedmulch, and T4: VB + FYM + VC + PM + MT + 3 WM = vegetative barrier + farmyard manure + vermi-compost + poultry manure + minimum tillage + 3 weed mulch.

B.N. Ghosh et al. / Agriculture, Ecosystems and Environment 216 (2016) 247–257 255

3.7. Maize and wheat yields and wheat equivalent yield

Maize grain yield was significantly higher with T4 than T1 plotsand lowest maize yield was recorded in plots under T2 (Fig. 6).However, significantly higher wheat grain yield was recorded inplots under T4 followed by T3, T2 and T1. Wheat equivalent yield(WEY) of T4 plots was significantly higher compared with T2 andWEY of T2 plots was similar to T3 and T1 (Fig. 6). Although maizeyield in plots under T4 was less during the initial years (data notpresented), mean (of seven years) maize yield of the sametreatment was highest. As wheat was grown on residual fertility,mean yield was highest in T4 plots and lowest with T1 treatment.The wheat yield did not show any definite trend in different yearsbecause of variable winter rainfall. But mean WEY was highest inplots under T4, which was 16% higher than T1.

During the initial years, the increase in maize yield in plotsunder T1 compared to T4 was probably due to applied mineralfertilizers that enhanced nutrient supplying capacity during theearly crop growth stages. In contrast, the organic nutrients appliedin the T4 treatment took some time to supply plant nutrients inavailable forms (Ghosh et al., 2006). Pandey et al. (2001) arguedthat crop–weed competition might decrease crop yield in theinitial years in plots where organic sources of nutrients wereapplied in combination with mineral fertilizers. Since 2011, T4remained the best treatment in respect of resource conservationand produced higher yields than T1. These findings highlight the

Table 8Correlation coefficients between parameters as influenced by long-term fertilization, vesystem (soil parameters were for 0–5 cm soil depth).

Parameters TOC LOC CMI Maize yield Whea

WBC 0.998** 0.716** 0.315 0.540* 0.566*

TOC 0.706** 0.529* 0.837** 0.868*

LOC 0.880** 0.904** 0.897*

CMI 0.948** 0.872*

Maize yield 0.419

Wheat yield

WEY

Soil loss

Runoff

WSA

* Correlation significant at P < 0.05, n = 24.** Correlation significant at P < 0.01, n = 24.

importance of the given management practice for facilitating thenutrient accumulation and moisture conservation (Bhattacharyyaet al., 2008). In contrast to the maize yield, plots under T4 producedthe highest wheat yield from the beginning (2007–08) and themean wheat yield in plots under T4 was �80% higher than T1.Higher grain yields of the residual wheat crops were probably dueto the accumulation of nutrients as recycled from weed biomassand organic amendments as well as soil moisture conservationpractices. Similar finding was reported earlier by Kundu et al.(2007) under comparable agro-climatic situations. The increase inWEY (Fig. 6) indicates the potential of the T4 soil managementsystem for enhancing grain yields. If adopted the net result couldbe sustainable crop production on sloping land in the region.

3.8. Relationships of grain yields and CMI with different parameters

Mean (of seven years) maize yield was positively correlatedwith CMI (r = 0.948**), LOC (r = 0.904**), total SOC (r = 0.837**) andWBC (r = 0.540*) in topsoil, while mean wheat yield wassignificantly correlated with CMI (r = 0.872**), LOC (r = 0.897**),total SOC (r = 0.868**) and WBC (r = 0.566*) (Table 8). Likewise,WEY was positively correlated with CMI (r = 0.906**), LOC(r = 0.917**), total SOC (r = 0.851**) and WBC (r = 0.580*). Positiverelationships between CMI with mean crop yields and SOC poolsimply that there appears to be a significant influence of CMI inincreasing crop yield and decreasing runoff and soil loss. SOC pools

getative barrier, weed mulch and tillage practices under a maize–wheat cropping

t yield WEY Soil loss Runoff WSA MWD

0.580* �0.710** �0.714** 0.726** 0.719*** 0.851** �0.713** �0.724** 0.717** 0.706*** 0.917** �0.717** �0.686** 0.734** 0.801*** 0.906** �0.768** �0.701** 0.508* 0.618**

0.890** �0.573** �0.358 0.173 0.3470.786** �0.558* �0.591** 0.702** 0.801**

�0.670** �0.540* 0.470* 0.638**

0.892** �0.533* �0.519*

�0.713** �0.611**

0.838**

256 B.N. Ghosh et al. / Agriculture, Ecosystems and Environment 216 (2016) 247–257

had positive associations with CMI, crop yields, WEY and negativeassociations between soil loss and runoff, indicating the role ofdifferent SOC pools in soil conservation. Moharana et al., (2012)also observed that changes in WBC, LBC and total SOC weresignificantly affected by CMI. The strongest positive correlationcoefficients were noted between CMI and total SOC (r = 0.998**)and that between CMI and maize yield (r = 0.948**). However,highest negative correlation coefficients were observed betweenCMI and soil loss (r = 768**) and between CMI and runoff (r = 701**)(Table 8). These values indicate that CMI was the single mostimportant parameter as affected by long-term adoption ofmanagement practices in a maize–wheat cropping system, as itwas most significantly correlated with both crop productivity andsoil erosion control (Table 8).

Soil properties such as water stable aggregate (WSA) and MWDplay important roles in crop yield by virtue of improving soil waterretention characteristics, increasing plant available water capacityand decreased soil erodibility, which help reduce soil and nutrientlosses (Bhattacharyya et al., 2008). There were significantcorrelations between wheat yield as well as between maize yieldand MWD. However, there were no significant correlationsbetween maize yield and WSA and maize yield and MWD. Thismight be due to aggregate breakdown (especially in topsoil) due tohigh intensity precipitation during maize growing periods (rainyseasons). However, the total amount and intensity of precipitationwere much less during wheat growing periods than during rainyseasons. Hence, negligible breakdown of aggregates during wheatgrowth might have increased aggregate stability and influencedsignificant positive correlations with mean wheat yield (Table 8).

The results implied that changes in total SOC, CMI and labileSOC pools under different management practices were important.KMnO4 oxidation simulates microbial decomposition and, there-fore, KMnO4-C partly reflects the in situ enzymatic decompositionof labile SOM (Loginow et al., 1987). Therefore, we expected toobserve significant positive correlations among the labile SOCpools, as they have close inter-associations. These results agreewith those reported by Blair et al. (1995), Soon et al. (2007) andMoharana et al. (2012). Although LOC and WBC pools are smallcompared to total SOC, these pools are easily accessible. Thus, thelatter are more important as far as nutrient availability isconcerned during crop growth periods compared to total SOC,and are thus helpful in proposing sustainable management optionsfor arable soils.

4. Conclusions

Implementation of minimum tillage (MT) and addition oforganic matter in situ or ex situ enhanced soil aggregationprocesses and water stable aggregates and decreased long-termsoil erosion on a gentle slope (�2%) in the Indian Himalayas. Thecarbon management index (CMI) proved very effective in assessingthe best conservation management practice. It had significantpositive correlations with mean (of seven years) maize and wheatyields, and significant negative correlations with mean seasonal(rainy season) runoff and soil erosion rates. The estimation of CMIvalues of any conservation measures can quantitatively indicatesoil degradation status and thus provide valuable information forconservation planning and mitigating land degradation. Theresults indicated that soils of these regions have degraded by40–60% compared to native undisturbed forest soils. Significantpositive correlation of CMI with mean system productivity and itssignificant negative correlation with soil loss indicate the potentialof CMI for use as the single value index for prediction of soil loss inarable lands where gauging devices cannot be installed. However,this concept requires fine-tuning through further research towardsdeveloping a mathematical model in a particular agro-ecosystem.

The present study establishes that conservation measure withPalmarosa as vegetative barriers along with organic amendments(FYM + VC + PM) plus weed mulch application three times underMT is effective in decreasing runoff and soil erosion, increasingsystem productivity and improving soil quality, as assessedthrough CMI. Further research is required to extrapolate (fromthe field to regional scales) these management practices and theconcept (of soil erosion prediction by simple measurement of CMI).

Acknowledgements

Authors thank staff of the Central Laboratory, CSWCR&TI,Dehradun, India and Mr. Sanjay Kumar, Technical Officer, ICAR-Vivekananda Institute of Hill Agriculture, Almora and Mr. J.S.Deshwal, Technical Officer, CSWCR&TI, Dehradun for their kindassistance in analysing soil samples. The authors are grateful toProf. M.A. Fullen, Professor of Soil Technology, University ofWolverhampton, UK and Prof. S.K. Sanyal, former Vice Chancellor,Bidhan Chandra Krishi Viswavidyalaya, Mohanpur, Nadia, WestBengal, India for editing the manuscript.

References

Bhattacharyya, R., Ved-Prakash, R.B., Kundu, S., Gupta, H.S., 2006. Effects of tillageand crop rotations on pore size distribution and soil hydraulic conductivity insandy clay loam soil of the Indian Himalayas. Soil Tillage Res. 86, 129–140.

Bhattacharyya, R., Kundu, S., Ved-Prakash Gupta, H.S., 2008. Sustainability undercombined application of mineral and organic fertilizers in a rainfed soybean-wheat system of the Indian Himalayas. Eur. J. Agron. 28, 32–46.

Bhattacharyya, R., Ved-Prakash, R.B., Kundu, S., Srivastav, A.K., Gupta, H.S., 2009. Soilaggregation and organic matter in a sandy clay loam soil of the IndianHimalayas under different tillage and crop regimes. Agric. Ecosyst. Environ. 132,126–134.

Bhattacharyya, R., Kundu Srivastav, A.K., Gupta, H.S., Ved-Prakash Bhatt, J.C., 2011.Long term fertilization effects on soil organic carbon pools in a sandy loam soilof the Indian Himalayas. Plant Soil 341, 109–124.

Bhattacharyya, R., Tuti, M.D., Kundu, S., Bisht, J.K., Bhatt, J.C., 2012a. Conservationtillage impacts on soil aggregation and carbon pools in a sandy clay loam soil ofthe Indian Himalayas. Soil Sci. Soc. Am. J. 76, 617–627.

Bhattacharyya, R., Tuti, M.D., Bisht, J.K., Bhatt, J.C., Gupta, H.S., 2012b. Conservationtillage and fertilization impacts on soil aggregation and carbon pools in theIndian Himalayas under an irrigated rice–wheat rotation. Soil Sci. 177, 218–228.

Bhattacharyya, R., Pandey, S.C., Bisht, J.K., Bhatt, J.C., Gupta, H.S., Tuti, M.D., Mahanta,D., Mina, B.L., Singh, R.D., Chandra, S., Srivastva, A.K., Kundu, S., 2013a. Tillageand irrigation effects on soil aggregation and carbon pools in the Indian Sub-Himalayas. Agron. J. 105, 101–112.

Bhattacharyya, R., Das, T.K., Pramanik, P., Ganeshan, V., Saad, A.A., Sharma, A.R.,2013b. Impacts of conservation agriculture on soil aggregation and aggregateassociated N under an irrigated agroecosystem of the Indo-Gangetic Plains.Nutr. Cycl. Agroeosyst. 96, 185–202.

Blair, G.J., Lefroy, R.D.B., Lisle, L., 1995. Soil carbon fractions based on their degree ofoxidation, and the development of a carbon management index for agriculturalsystems. Aust. J. Agric. Res. 46, 1459–1466.

Blair, N., Crocker, G.J., 2000. Crop rotation effects on soil carbon and physical fertilityof two Australian soils. Aust. J. Soil Res. 38, 71–84.

Blair, N., Faulkner, R.D., Till, A.R., Poulton, P.R., 2006. Long-term managementimpactions on soil C, N and physical fertility. Part I: broadbalk experiment. SoilTillage Res. 91, 30–38.

Bouyoucos, G.J., 1962. Hydrometer method improved for making particle sizeanalysis of soil. Agron. J. 54, 464–465.

Das, B., Chakraborty, D., Singh, V.K., Aggarawal, P., Singh, R., Dwivedi, B.S., Mishra, R.P., 2014. Effect of integrated nutrient management practice on soil aggregateproperties: its stability and aggregate-associated carbon content in an intensiverice–wheat system. Soil Tillage Res. 136, 9–18.

Das, T.K., Bhattacharyya, R., Sharma, A.R., Das, S., Saad, A.A., Pathak, H., 2013. Impactsof conservation agriculture on total soil organic carbon retention potentialunder an irrigated agro-ecosystem of the western Indo-Gangetic Plains. Eur. J.Agron. 51, 34–42.

Devaum, M., Gifford, G.F., 1984. Variability of infiltration within large runoff plots onrangelands. J. Range Manag. 37, 523–528.

Elliott, E.T., 1986. Aggregate structure and carbon, nitrogen and phosphorus innative and cultivated soils. Soil Sci. Soc. Am. J. 50, 627–633.

Ghosh, B.N., Singh, R.D., Dadhwal, K.S., 2006. Available organic sources for IPNM inNorth-western Himalayan states for sustainable crop production. Ind. Farm. 56,10–12.

Ghosh, B.N., 2011. Soil erosion control through use of grass vegetative filter strips inthe western Himalayan region. Eur. Soc. Soil Conserv. Newsl. 1, 26–32.

B.N. Ghosh et al. / Agriculture, Ecosystems and Environment 216 (2016) 247–257 257

Ghosh, B.N., Dogra, P., Bhattacharyya, R., Sharma, N.K., Dadhwal, K.S., 2012a. Effectsof grass vegetation strips on soil conservation and crop yield under rainfedconditions in the Indian sub-Himalayas. Soil Use Manag. 28, 635–646.

Ghosh, P.K., Venkatesh, M.S., Hazra, K.K., Kumar, N., 2012b. Long term effect of pulsesand nutrient management on soil organic carbon dynamics and sustainabilityon an Inceptisol of Indo-Gangetic plain of India. Exp. Agric. 48, 473–487.

Ghosh, B.N., Sharma, N.K., Dogra Pradeep, Mishra, P.K., 2013. Aromatic Grass BasedConservation Farming for Erosion Control and Maize–Wheat Productivity inNorth-western Himalayan Region. Tech. Brochur, CSWCRTI, Dehradun, pp. 1–8.

Gilley, J.E., Finkner, S.C., Spomer, R.G., Miclke, L.N., 1986. Runoff and erosion asaffected by crop residue. Part 1: total losses. Trans. ASAE 29, 157–160.

Hanway, J.J., Heidel, H., 1952. Soil analysis methods as used in Iowa state college soiltesting laboratory. Iowa State Coll. Bull. 57, 1–131.

Jackson, M.L., 1973. Soil Chemical Analysis, 2nd ed. Prentice Hall, New Delhi, India.Kaur, T., Brar, B.S., Dhillon, N.S., 2008. Soil organic matter dynamics as affected by

long-term use of organic and inorganic fertilizers under maize–wheat croppingsystem. Nutr. Cycl. Agroecosyst. 81, 59–69.

Kemper, W.D., Rosenau, R.C., 1986. Aggregate stability and size distribution, In:Klute, A. (Ed.), Methods of Soil Analysis. Part I. 2nd ed. Agron. Monogr. 9. ASAand SSSA, Madison, WI, pp. 425–442.

Kundu, S., Bhattacharyya, R., Ved-Prakash Ghosh, B.N., Gupta, H.S., 2007. Carbonsequestration and relationship between carbon addition and storage underrainfed soybean-wheat rotation in a sandy loam soil of the Indian Himalayas.Soil Tillage Res. 92, 87–95.

Lal, R., 1975. Role of mulching techniques in tropical soil and water management.Int. Inst. Trop. Agric. Tecnol. Bull. Ibadan, Nigeria 1, 38.

Lal, R.K., Sharma, J.R., Mishra, H.O.,1996. Varietal selection for high root biomass andoil yield in vetiver. Vetiver Newsl. 15, 16–18.

Loginow, W., Wisniewski, W., Gonet, S.S., Ciescinska, B., 1987. Fractionation oforganic carbon based on susceptibility to oxidation. Pol. J. Soil Sci. 20, 47–52.

Liu, M.Y., Chang, Q.R., Qi, Y.B., Liu, J., Chen, T., 2014. Aggregation and soil organiccarbon fractions under different land uses on the tableland of the Loess Plateauof China. Catena 115, 19–28.

Majumder, B., Mandal, B., Bandyopadhyay, P.K., Chaudhury, J., 2007. Soil organiccarbon pools and productivity relationships for a 34 year old rice–wheat–juteagro-ecosystem under different fertilizer treatments. Plant Soil 297, 53–67.

Mandal, N., Dwivedi, B.S., Meena, M.C., Singh, D., Datta, S.P., Tomar, R.K., Sharma, B.M., 2013. Effect of induced defoliation in pigeonpea, farmyard manure andsulphitation press mud on soil organic carbon fractions, mineral nitrogen andcrop yields in a pigeonpea–wheat cropping system. Field Crops Res. 154, 178–187.

Moharana, P.C., Sharma, B.M., Biswas, D.R., Dwivedi, B.S., Singh, R.V., 2012. Long-term effect of nutrient management on soil fertility and soil organic carbonpools under a 6-year-old pearl millet–wheat cropping system in an Inceptisol ofsubtropical India. Field Crops Res. 136, 32–41.

Mukesh, Manhas, R, K., Tripathi, A.K., Raina, A., K, Gupta, M, K., Kamboj, S.K., 2011.Sand and clay mineralogy of sal forest soils of the Doon Siwalik Himalayas. J.Earth Syst. Sci. 120 (1), 123–144.

Nelson, D.W., Sommers, L.E., 1982. otal carbon, organic carbon, and organic matter,2nd ed. Methods of Soil Analysis, 9(2. ASA Monograph, pp. 539–579.

Olsen, S.R., Cole, C.V., Watanabe, F.S., Dean, L., 1954. Estimation of availablephosphorus in soils by extraction with sodium bicarbonate. USDA Circ. 939. USGovt. Printing Office, Washington, D.C.

Pandey, A.K., Ved-Prakash Singh, R.D., Gupta, H.S., 2001. Contribution and impact ofproduction factors on growth yield attributes: yield and economics of rainfedwheat. Ind. J. Agron. 46, 674–681.

Pinheiro, E.F.M., Pereira, M.G., Anjos, L.H.C., 2004. Aggregate distribution and soilorganic matter under different tillage systems for vegetable crops in a RedLatosol from Brazil. Soil Tillage Res. 77, 79–84.

Puget, P., Chenu, C., Balesdent, J., 1995. Total and young organic matter distributionsin aggregates of silty cultivated soils. Eur. J. Soil Sci. 46, 449–459.

Six, J., Elliott, E.T., Paustian, K., 1999. Aggregate and soil organic matter dynamicsunder conventional and no-tillage systems. Soil Sci. Soc. Am. J. 63, 1350–1358.

Six, J., Elliott, E.T., Paustian, K., 2000. Soil macroaggregate turnover andmicroaggregate formation: a mechanism for C sequestration under no-tillageagriculture. Soil Biol. Biochem. 32, 2099–2103.

Six, J., Conant, R.T., Paul, E.A., Paustian, K., 2002. Stabilization mechanisms of soilorganic matter: implications for C-saturation of soils. Plant Soil 241, 155–176.

Soon, Y.K., Arshad, M.A., Haq, A., Lupwayi, N., 2007. The influence of 12 years oftillage and crop rotation on total and labile organic carbon in a sandy loam soil.Soil Tillage Res. 95, 38–46.

Subbiah, B.V., Asija, G.L.,1956. A rapid procedure for estimation of available nitrogenin soils. Curr. Sci. 25, 259–260.

Sur, H.S., Sandhu, I.S., 1994. Effect on different grass barriers on runoff, sediment lossand biomass production in the foothills of the Shiwaliks. Abstracts of the 8thISCO Conference, New Delhi, India, pp. 218–221.

Tisdall, J.M., Oades, J.M., 1982. Organic matter and water stable aggregates in soils. J.Soil Sci. 33, 141–163.

van Bavel, C.H.M., 1949. Mean weight diameter of soil aggregates as statistical indexof aggregation. Soil Sci. Soc. Am. J. Proc. 14, 20–23.

Veihmeyer, F.J., Hendrickson, A.H., 1948. Soil density and root penetration. Soil Sci.65, 487–494.

Venkatesh, M.S., Hazra, K.K., Ghosh, P.K., Praharaj, C.S., Kumar, N., 2013. Long-termeffect of pulses and nutrient management on soil carbon sequestration in Indo-Gangetic plains of India. Can. J. Soil Sci. 93, 127–136.

Walkley, A., Black, I.A., 1934. An examination of the Degtareff method fordetermining soil organic matter and a proposed modification of the chromicacid titration method. Soil Sci. 37, 29–38.