Effect of reduced tillage and mulching on soil health in Sesbania alley cropping based rainfed food - fodder systems

ABSTRACT An experiment was laid out in split-plot design with 3-Sesbania alley-based rainfed food-ufodder systems [Sesbania+(fodder sorghum–chickpea), Sesbania+(fodder sorghum–barley) and Sesbania+(fodder sorghum–mustard)] in main-plots and 6-reduced tillage and Sesbania mulching combinations in sub-plots to assess their impact on soil health. Minimum tillage (MT) during rainy season followed by zero tillage (ZT) during winter season along with mulch (MT-ZT+mulch) resulted in higher proportion of soil macro-aggregate, increased infiltration rate by 27.4%, hydraulic conductivity by 14.9% and water holding capacity by 5.9% (v/v) over conventional tillage (CT) during rainy season followed by conventional tillage during winter season without mulch (CT-CT without mulch). The MT-ZT+mulch had the highest soil organic carbon (7.3 g kg–1), microbial biomass carbon (226 mg kg–1 soil), SOC stock (14.9 Mg ha−1), carbon sequestration rate (1.72 Mg ha−1 year−1), soil quality index (SQI, 1.09) and enhanced the available soil N, P and K by 45.1, 47.2 and 22.2%, respectively, over CT-CT without mulching. The Sesbania alley + (fodder sorghum–mustard) cropping system had adverse effect on soil health while inclusion of chickpea in the system improved soil health and recorded the highest SQI (0.97).


Introduction
There has been increasing interest in conservation agriculture-based practices throughout the world due to their favourable effect on soil health, plant growth and productivity (Pradhan et al. 2018;Wang et al. 2020). Conservation agricultural practices have been found to provide several soil health benefits including the improvement of important soil quality parameters like soil organic carbon, available N, P and K, soil aggregation, and soil moisture dynamics, such as water holding capacity, infiltration and porosity (Parihar et al. 2016). Further, minimum soil disturbance associated with reduced or zero tillage improves soil biological activity, maintains optimum proportions of respiration gases in the rooting zone, moderates organic matter oxidation, increases porosity and limits reexposure of buried weed seeds to light, reducing their germination and emergence (Choudhary et al. 2018). Similarly, conservation agricultural practices are also reported to have favourable effect on environment through carbon sequestration and optimizing energy use (Parihar et al. 2020). Recycling of crop residues and plant biomass in the field is a viable option for replenishing soil fertility, improving soil physico-chemical and biological properties, rehabilitating degraded lands and sustaining crop yields (Ghosh et al. 2017). Cultivation of crops under zero tillage with adequate residue retention is cost effective, saves time, facilitates early sowing, conserves soil moisture and alleviates terminal drought stress. Therefore, in recent years, conservation agriculture is being widely promoted as a climate smart and alternative agricultural practice for sustainable intensification.
Despite the advantages conferred by conservation agriculture, however, yield variability with zero-tillage still remains a major concern among farmers ( Pittelkow et al. 2015). There are also tradeoffs among various uses of crop residues which limit its availability as mulch. About 30-40% of crop residue must be returned to soil for maintaining its health. However, most of the crop residues are used as livestock fodder and sometimes it is difficult to find surplus residues to be use as mulch especially under rainfed condition in arid and semi-arid tropics where livestock is the main source of livelihood (Ghosh et al. 2016). Therefore, alternative niches, such as, perennial Sesbania alley cropping, need to be explored as they have the potential to generate trimmings that can be used as mulch and animal feed. Sesbania sesban (L.) Merrill is the most productive multipurpose short-lived perennial deciduous shrub or tree that is widely distributed in the tropics and sub-tropics; and is usually planted by smallholder farmers. It is used for various purposes like green manuring, fuelwood, livestock feed, pasture improvement (through biological N fixation), weed control and have many medicinal properties. It can be grown as sole crop and in alley cropping with other food and fodder crops.
In India, conservation agriculture research is confined to irrigated conditions and rice-wheat cropping system (Pradhan et al. 2018). The possibility of double cropping is a challenge in rainfed regions of central India (Palsaniya et al. 2008) where droughts are frequent (Palsaniya et al. 2011) and average annual rainfall is showing a decreasing trend. The fields of farmers operating under rainfed conditions remain fallow during winter season after harvest of the rainy season crop (Palsaniya et al. 2009). Therefore, suitable and innovative rainfed cropping systems and conservation agriculture practices need to be identified to make double cropping a reality to increase productivity per unit land area while simultaneously improving soil health in such fragile rainfed agro-ecosystems. A second crop in the winter season is usually not possible in central India, without appropriate soil moisture management, because it is dry at this time of the year (Palsaniya et al. 2012). Therefore, conservation agricultural practices like zero tillage and minimum tillage (ZT and MT) and mulching using loppings from Sesbania alleys may provide a good opportunity for rainfed farmers to grow low water demanding winter crops such as chick pea, barley and mustard following rainy season fodder sorghum crop. We hypothesized that the potential productivity and soil health improving capacity of Sesbania alley-based rainfed food-fodder systems will significantly vary with tillage system, mulching and crop combinations used in the cropping system. The reduced tillage (MT-ZT) with Sesbania mulching would result improved soil physical, chemical and biological properties and soil quality index when compared to conventional tillage treatments under rainfed condition. Therefore, this study was undertaken to assess the effect of tillage system, mulching and cropping system on soil quality in Sesbania alley-based food-fodder systems.

Study site and soil
A field trial was carried at the Central Research Farm of ICAR-Indian Grassland and Fodder Research Institute Jhansi, India during 2013-2017. The site was located at 25° 27ʹ N latitude, 78° 33ʹ E longitude and 270 m altitude in the semi-arid continental monsoonal climatic zone. The study site received 1443, 617, 670 and 814 mm rainfall in 2013-2014, 2014-2015, 2015-2016 and 2016-2017, respectively, during crop period (June 01-May 31). The respective rainy days were 78, 46, 39 and 40. The monthly variation in rainfall during the study period is depicted in the Supplementary Figure 1. The study site experiences extreme temperature during summer (as high as 47°C) and winter (as low as 2°C). The site was occasionally cultivated for growing fodder cowpea during rainy season and oat during winter. The soil of the experimental plot was gravelly sandy clay loam (51.3% sand, 27.2% silt and 21.5% clay) and belonged to the hyper thermic family of Typic Haplustept (USDA 1999). The soil belonged to Cambisols (WRB 2015). The field capacity and permanent wilting point of the soil (0-30 cm) was 11.8 and 5.1% (w/w), respectively. On average, the experimental soil had pH 6.62 in 1:2.5 soil:water suspension, electrical conductivity 0.36 dS m −1 , bulk density 1.33 Mg m -3 , water holding capacity 56.2% (v/v), soil organic carbon 4.0 g kg −1 , KMnO 4 oxidizable N 110 kg ha −1 , 0.5 M NaHCO 3 extractable P 17.02 kg ha −1 and 1 N neutral NH 4 OAc extractable K 197.12 kg ha −1 in 0-30 cm soil layer.

Treatment details and crop husbandry
The detailed treatment description and protocols adopted (tillage and mulching) in the Sesbania alley-based food-fodder systems are described in Supplementary Table 1. The study was laid out in split-plot design with three replications. Three cropping systems namely Sesbania alley + (fodder sorghum -chickpea), Sesbania alley + (fodder sorghum -barley) and Sesbania alley + (fodder sorghum -mustard) were assigned to the main plots. The main-plot size was kept as 24 m � 6.5 m. Six combinations of conservation tillage and mulching practices, viz., conventional tillage during rainy season (CT) -conventional tillage during winter season (CT) without mulch, conventional tillage during rainy season (CT) -conventional tillage during winter (CT) with Sesbania mulch, minimum tillage during rainy season (MT) -zero tillage during winter season (ZT) without mulch, minimum tillage during rainy season (MT) -zero tillage during winter season (ZT) with Sesbania mulch, conventional tillage during rainy season (CT) -zero tillage during winter season (ZT) without mulch and conventional tillage during rainy season (CT) -zero tillage during winter season (ZT) with Sesbania mulch were laid out in sub-plots. The sub-plot size was 4 m � 6.5 m. The six months old seedlings of perennial Sesbania sesban were planted in well prepared pits at 3.25 meter row to row and 50 cm plant to plant spacing during rainy season in the first week of August, 2013 ( Figure 1). The detailed agronomic practices adopted for cultivation of all the annual crops under investigation are described in the Supplementary Table 2. Figure 1. Layout of experimental field. In between two hedge rows, sorghum during rainy season and barley, mustard and chickpea were grown during winter season as per treatment.

Soil sampling and analysis methodologies
The soil samples were taken in 2017 after harvest of the winter crops from fixed plots. The soil samples for physico-chemical properties (0-15 and 15-30 cm) and aggregate analysis (0-10 cm) were collected in triplicate from each treatment plot using a core sampler. The physico-chemical analysis results from the 0-15 cm soil layer are presented in the manuscript. The physico-chemical properties of soil in the 15-30 cm layer did not show significant variability. For aggregate analysis, only aggregates of 4-8 mm size were used. The soil samples were air dried in shade, ground and passed through a 2 mm mesh sieve. Bulk density (BD) of soil was determined by the core sampler method (Piper 1950) from three randomly selected points of each plot. Infiltration rate of soil was measured using double ring infiltrometer (Bouwer 1986) while saturated hydraulic conductivity (Ks) was determined using constant head method (Misra and Ahmed 1987). The soil aggregate analysis was done through wet sieving techniques using Yodders apparatus (Yodder 1936). The mean weight diameter (MWD) and geometric mean diameter (GMD) of aggregates were estimated as per Equation 1 and 2 given by Kemper and Rosenau (1986).
where, Wi, is the proportion of aggregates retained on the sieves in relation to the whole, Xi is the mean diameter of the class (mm), exp denote exponential, n is the number of size classes. Soil moisture content was measured during wet and dry season in 2016-17 cropping season using the gravimetric methods at soil depths of 0-15 and 15-30 cm. Conversion of gravimetric water content to volumetric water content was done using bulk density (Eq 3). Maximum water holding capacity of soil (WHC) was determined by equilibrating the soil with water through capillary action in a Keen Raczkowald (KR) box (Baruah and Barthakur 1999).
Soil moisture stockðm 3 ha À 1 Þ ¼ Soil moistureð% w w Þ � BDðMgm À 3 Þ � Soil depthðmÞ � 10000 100 : Soil pH was measured in aqueous soil extract in de-ionized water (1:2.5 soil: water) as described by Richards (1954). Soil salinity was measured as the electrical conductivity of the aqueous soil extract using conductivity bridge (Piper 1950). The soil organic carbon, available N, P and K were estimated using wet digestion method (Walkley and Black 1934), modified Kjeldahl method (Subbiah and Asija 1956), Olsen's method (Olsen et al. 1954) and Flame photometer method (Jackson 1973), respectively. The SOC stock for a layer of thickness was calculated using the methodology and Equation 4 described by Poeplau et al. (2017): Soil C sequestration rate (Mg C ha −1 year −1 ) was calculated as change in the SOC stock (Mg ha −1 ) divided by duration of experiment in years. For analysing biological properties, the soil samples were collected from 0-15 cm depth at the harvest of the last crop and sieved gently through a 4 mm mesh sieve to eliminate stones, plants roots and large organic substances. Then, the samples were passed through a 2-mm sieve and stored at 4°C until used for assaying of microbial count, microbial biomass carbon and soil enzyme activities (dehydrogenase, alkaline phosphatase and β-glucosidase). Total soil bacterial population was estimated by Waksman's (1952) method using nutrient agar medium at dilution of 10 −8 concentration. The inoculated petri-dishes were incubated at 30 ± 1°C. Total fungal population was determined using potato dextrose agar medium at dilution of 10 −3 concentration (Beuchat 1979). The soil microbial biomass carbon (SMBC) was estimated by the chloroform-fumigation incubation method (Nunan et al. 1998). Dehydrogenase and alkaline phosphatase activity in soil samples was estimated as per methods described by Casida et al. (1964) and Tabatabai and Bremner (1969), respectively. The β-glucosidase was estimated by determining the amount of p-nitrophenol released after 1 h of incubation with p-nitrophenyl-β -D-glucopyranoside (Eivazi and Tabatabai 1988).

Multivariate analysis and soil quality index (SQI)
The principal component analysis (PCA) was performed to screen minimum data set (MDS). We selected the PCs with eigen value >1 and those which described a minimum of 5% of variation. Within each PC, only highly weighted factors (having absolute values within 10% of the highest factor loading) were retained for the MDS. All the observations of each identified key MDS indicator were transformed using linear scoring technique by using more is better approach (Biswas et al. 2017). To remove redundancy in variables in each PC, correlation approach was used (Andrews and Carroll 2001). For development of SQI, the weightage of a variable was calculated as the ratio of percentage of the variation explained by the PC it belonged to, and the cumulative percent variation explained by all the four PCs having eigen value >1 (Eq5). After performing these steps, to obtain soil quality index (SQI), the weighted MDS indicator scores for each observation were summed up using the following function: where S = indicator score, W = PC weightage factor, i = variable number in MDS

Statistical analysis
All the data were subjected to analysis of variance (ANOVA) using the general linear model procedures of the Statistical Analysis System 9.3. Treatment means were compared by F-test (LSD) at α = 0.05. Since, the present trial was conducted in fixed plots at the same location, no homogeneity test was subjected. However, before performing ANOVA, all the data were subjected to Shapiro-Wilk normality test using the PROC UNIVARIATE procedure (P < 0.05) and data were found to be normal. Principal component analysis (PCA) was performed using the Genstat 18 th edition package (Genstat, Hempstead, UK).

Soil physical properties
The dry bulk density was measured in 2017 for the three Sesbania-based alley cropping systems under six conservation tillage and mulching practices up to a depth of 15 cm (Table 1). At the end of eight cropping seasons of implementing different tillage and mulching treatment combinations in Sesbania-based alley cropping systems, bulk density was not significantly affected by tillage and cropping systems. The bulk density was only 2.3 and 1.5% lower in the Sesbania + (fodder sorghum -barley) alley cropping system than the Sesbania + (fodder sorghum -chickpea) and Sesbania + (fodder sorghum -mustard) system, respectively. Similarly, the bulk density was varied by only 0.8 to 3.1% due to mulching effect in all the tillage practices. The average IR was statistically non-significant among different Sesbania-based alley cropping systems (Table 1). Sesbania mulching significantly increased the IR irrespective of tillage practices. All the tillage practices with Sesbania mulching, viz., CT-CT + mulching, MT-ZT + mulching and CT-ZT + mulching recorded significantly higher IR than their respective treatments without mulching. Among the six tillage and mulching practices, the MT-ZT + mulch (22.3 mm h −1 ) and CT-ZT + mulch (20.7 mm h −1 ), being at par, recorded significantly higher IR than other tillage and mulching combinations. The hydraulic conductivity also did not vary significantly among all the Sesbania alley + fodder sorghum-based cropping systems. However, similar to IR, the Sesbania mulching significantly increased the hydraulic conductivity irrespective of tillage practices. The hydraulic conductivity in MT-ZT + mulch (17 mm h −1 ) and CT-ZT + mulch (16.4 mm h −1 ) were at par and recorded significantly higher values than other tillage and mulching treatments. The MT-ZT + mulch recorded 14.9% higher hydraulic conductivity than the CT-CT without mulch (14.8 mm h −1 ). The mean weight diameter (MWD) and geometric mean diameter (GMD) did not vary significantly among different Sesbania-based alley cropping systems (Table 1). However, significantly higher percentage of > 0.2 mm soil aggregates was observed under Sesbania + (fodder sorghumchickpea) alley cropping system compared to Sesbania + (fodder sorghum -barley) and Sesbania + (fodder sorghum -mustard) alley cropping systems. A reverse trend was observed in case of percentage of < 0.2 mm soil aggregates. Significantly higher MWD (0.84 mm), GMD (0.61 mm) and percentage of > 0.2 mm soil aggregates (49.2%) were observed under MT-ZT + mulch and CT-ZT + mulch over other treatments. Further, the above soil aggregation parameters were significantly higher under Sesbania mulching treatments in all the tillage practices over their respective treatments without mulching. However, percentage of <0.2 mm soil aggregates showed a reverse trend as observed in >0.2 mm soil aggregates. Water holding capacity (WHC) was not significantly affected by different Sesbania-based alley cropping systems. However, the MT-ZT + mulch (59.6%, v/v) and CT-ZT + mulch (58.6%, v/v), being at par, recorded significantly higher WHC than other tillage and mulching combinations. Further, the mulched treatments recorded significantly higher WHC as compared to un-mulched treatments irrespective of tillage practices. The WHC of soil was improved by 5.9% in MT-ZT + mulch over CT-CT without mulch at the end of 8th cropping cycle. There was no significant variation in soil moisture stock among different cropping systems during rainy season (Table 1). However, significantly higher soil moisture stock was found under Sesbania + (fodder sorghum -chickpea) (443 m 3 ha −1 ) cropping system compared to Sesbania + (fodder sorghum -mustard) (388 m 3 ha −1 ) and Sesbania + (fodder sorghum -barley) (331 m 3 ha −1 ) during winter season. The respective increase in soil moisture stock under Sesbania + (fodder sorghum -chickpea) system was 14.2% and 38.8% over Sesbania + (fodder sorghum -mustard) and Sesbania + (fodder sorghum -barley) cropping systems. The MT-ZT + mulch (591 m 3 ha −1 ) and CT-ZT + mulch (557 m 3 ha −1 ), being at par, recorded significantly higher soil moisture stock than other tillage and mulching treatments during rainy season. Moreover, the mulched treatments recorded significantly higher soil moisture stock as compared to un-mulched treatments irrespective of tillage practices. Similar trend was also observed during winter season. The soil moisture stock was improved by 30.1% in MT-ZT + mulch over CT-CT without mulch.

Soil chemical properties
The soil pH, electrical conductivity (EC), soil organic carbon (SOC) concentrations, available nutrients (N, P and K), SOC stock and carbon sequestration rate (CSR) did not vary significantly under different Sesbania-based alley cropping systems (Table 2). Similarly, tillage and Sesbania mulching also failed to bring significant effect on soil pH and EC. However, SOC concentrations, available nutrients (N, P and K), SOC stock and CSR showed significant variation under various tillage and Sesbania mulching treatments at the end of the 8 th cropping cycle. The MT-ZT + mulch and CT-ZT + mulch, being at par, registered significantly higher available soil nutrients (N, P and K) than other tillage and mulching combinations. Further, the mulched treatments in every tillage practice recorded significantly higher available soil N, P and K as compared to their respective un-mulched treatments. At the end of 4 th year, the available soil N, P and K increased by 45.1%, 47.2% and 22.2%, respectively, under MT-ZT + mulching as compared to CT-CT without mulching.
The SOC concentration was significantly higher under all the tillage treatments having mulching as compared to their un-mulched treatments. The MT-ZT + mulch, CT-ZT + mulch and CT-CT + mulch, being at par, recorded 65.9%, 56.8% and 43.2% higher SOC than CT-CT without mulching (4.4 g kg -1 ), respectively. Further, irrespective of mulching, the SOC and available nutrients (N, P and K) were found in decreasing order MT-ZT > CT-ZT > CT-CT. The SOC stock under MT-ZT + mulch (14.9 Mg ha −1 ) and CT-ZT + mulch (13.4 Mg ha −1 ) was at par and significantly higher over other tillage and mulching practices. The SOC stock was observed in order MT-ZT > CT-ZT > CT-CT. Further, the MT-ZT + mulch, CT-ZT + mulch and CT-CT + mulch recorded significantly higher SOC stock by 39.3%, 35.4% and 36.7% than their respective treatments without mulching. The carbon sequestration rate (CSR) followed similar trend to SOC concentrations and the MT-ZT + mulch, CT-ZT + mulch and CT-CT + mulch, being at par, recorded 588%, 436% and 336% higher CSR than CT-CT without mulching, respectively.
The reduced tillage practice (MT-ZT) + mulching recorded significantly the highest fungal population (16.8 cfu × 10 4 ) than the other tillage and mulching practices. However, the bacterial population was at par in MT-ZT + mulching (53.7 cfu × 10 7 ) and CT-ZT + mulching (54.5 cfu × 10 7 ), but significantly higher than other tillage and mulching practices. Moreover, the mulched plots recorded significantly higher fungal and bacterial population than the un-mulched plots irrespective of tillage. The SMBC ranged from 226 mg kg -1 soil in MT-ZT + mulching to 170 mg kg -1 soil in CT-CT without mulching. The MT-ZT + mulching and CT-ZT + mulching, being at par, recorded significantly higher SMBC compared to other tillage and mulching combinations. About 39.9% and 25.3% higher SMBC was observed under MT-ZT + mulching and CT-ZT + mulching, respectively, as compared to that under CT-CT without mulching. Further, the mulched plots recorded significantly higher SMBC than their un-mulched plots irrespective of tillage.
The Soil DHA, APA and β glucosidase were significantly higher in MT-ZT + mulching as compared to other tillage and mulching treatments (Figure 2). The DHA ranged from 29.8 to 14.3 μg TPF g −1 soil day −1 in different treatments. The CT-CT + mulch, MT-ZT + mulch and CT-ZT + mulch recorded 24.8%, 23.1% and 45.5% higher DHA than their respective un-mulched plots while the corresponding increase in APA was 17.7%, 26.1% and 16.2%. The MT-ZT + mulching treatment recorded 38.1% higher APA than the CT-CT without mulch treatment (the poorest performing treatment). The β glucosidase was significantly higher in MT-ZT + mulching treatment which recorded 32% higher value than CT-CT without mulch treatment. Further, mulching had significant effect on β glucosidase content and the CT-CT + mulch, MT-ZT + mulch and CT-ZT + mulch recorded 17.4%, 23% and 9.8% higher β glucosidase than their respective unmulched plots.

Soil quality index (SQI)
In the PCA of 15 variables, four principal components (PCs) were selected with eigen value of >1.0 which explained 71.98% of the variance in the data (Table 3). Available N, P, K and alkaline phosphatase activity were the high weight variables under PC1. In PC2, bacterial count and SOC and DHA were the high weight variables. The hydraulic conductivity was high weight variable in PC3. Similarly, geometric mean weigh and SOC were high weight variables in PC4. Soil quality index computed using the PCA, varied from 0.78 to 1.09 across the cropping systems and tillage management treatments (Figure 3). It was observed that tillage and mulching management practices played a significant role in influencing the soil quality index. Among the cropping systems, inclusion of chickpea (legume crop) in the cropping system recorded the highest SQI value of 0.97, while Sesbania alley + (sorghum -barley) and Sesbania alley + (sorghum -mustard) maintained SQI value of 0.93 and 0.90, respectively. Practice of MT-ZT + Sesbania leaves as mulch resulted in higher SQI of 1.09 followed by practice of CT-ZT + Sesbania mulch. It was interesting to note that Sesbania used as mulch maintained higher value of SQI in all the treatments irrespective of tillage.

Soil physical properties
Lower but non-significant bulk density was recorded in the Sesbania + (fodder sorghumbarley) alley cropping system and in all the tillage practices with mulching. The bulk density of soil is usually not affected by short-term tillage and mulching practices. This study was carried out for 4 years. Within these four years, the mulching treatment was only applied for three years as the first year was used for the establishment of the Sesbania alleys. Other short-and medium-term studies also reported similar non-significant effect of reduced tillage and mulching on soil bulk density Dixit et al. 2019). Dam et al. (2005) observed that the soil bulk density was not affected significantly even in an eleven year experiment under different tillage and residue management practices. Bulk density is rather a complex soil property, usually not affected by short-term interventions and may also depend on native soil type, prevailing climate (temperature, rainfall, humidity) and other management practices like addition of organic manure, compost, green manuring, etc. Significantly higher IR and hydraulic conductivity under reduced tillage and Sesbania mulching may be attributed to the improved soil aggregation. Further, the minimum soil disturbance under reduced tillage and mulching maintained the pore continuum in soil that also resulted in higher IR ; SMBC-soil microbial biomass carbon (mg kg -1 ); DHA-dehydrogenase activity (µg TPF g -1 soil day -1 ); APA-alkaline phosphatases activity (μg ρ -NPP g -1 soil h -1 ); β glucosidase (μg PNP g −1 soil h -1 ). and hydraulic conductivity. The minimum soil disturbances under zero tillage results in large biopores, higher pore continuity and porosity (Saha et al. 2010). Contrary to this, the water movement within soil gets impeded due to continuous heavy tillage under CT-CT as it causes rapid soil structural decline through slaking and dispersion (Guzha 2004). The improved aggregate stability under Sesbania mulched plots was also resulted in higher IR and hydraulic conductivity (Table 1). Shukla et al. (2003) reported that infiltration rate and hydraulic conductivity of soil are largely governed by the soil aggregation, aggregate size distribution, geometry, continuity, and relative proportion of large aggregates.
Improved soil aggregation (with more, and stable aggregates with large macro-aggregates) was recorded under reduced tillage and Sesbania mulched plots. The results suggest that the higher SOC, microbial population and enzymatic activity under reduced tillage and Sesbania mulching treatment (Table 2 and Figure 2) improved soil aggregation and produced stable aggregates with large macroaggregates. On the other hand, the size of aggregates was reduced under CT due to mechanical abrasion of macro-aggregates and this further exposed SOM to agents of decomposition. The significant reduction in macro-aggregates (>0.25 mm) under CT was also reported by Mikha and Rice (2004) in a 10year study on silt loam soil in USA. Mulching and residues incorporation into soils counters the destructive action of tillage and increases soil aggregation (Jakhar et al. 2017).The significantly higher WHC and soil moisture stock under reduced tillage and mulching may be attributed to improved physico-chemical and biological properties of soil, moderation of soil temperature, reduced evaporation rate and lesser weed infestation. Libohova et al. (2018) reported higher soil WHC in reduced tillage due to the increased amount of soil organic matter. The higher soil water content in zero-tilled and mulched plots has been attributed to favourable soil temperatures and lower weed population (Bescansa et al. 2006). Comprehensive studies on conservation agricultural practices in rainfed regions (Pradhan et al. 2018) also revealed that these practices increased water storage, reduced water loss and wind erosion, improved water and energy use efficiency.

Soil chemical properties
The chemical properties, namely SOC concentrations, available nutrients (N, P and K), SOC stock and CSR significantly improved under various tillage and Sesbania mulching treatments. The soil organic carbon and available nutrient content are largely affected by the amount of nutrients mined, nutrients and crop residues replenished to the soil, farming systems adopted and agronomic management techniques followed Ghosh et al. 2017). Higher SOC and available nutrients (N, P and K) in mulched plots might be due to large amount of biomass (about 23.6 t ha -1 in three years) addition and more carbon sequestration (Supplementary Table 3 and Table 2) than unmulched treatments. Tillage and mulching practices may also influence the soil chemical properties by affecting the rate of mineralization and immobilization. Duiker and Lal (1999) reported that the conversion efficiency of the mulching carbon to SOC was lower in frequently tilled soil (8%) than in zero tillage (10%). Similarly, zero-tillage and mulching also reduce the decomposition rate of soil organic matter which leads to the building up of organic carbon content and available nutrients in soil with time (Gwenzi et al. 2009). The beneficial effect of mulching was evident on available N, P and K in present investigation as large amount (6.1, 8.5 and 9 t ha -1 during second, third and fourth year, respectively) of nutrient rich Sesbania mulch was applied over a period of six cropping seasons (Supplementary Table 3), that enhanced available nutrients in soil upon mineralization.

Soil microbial and enzymatic activity
The soil bacterial population, dehydrogenase activity, alkaline phosphatase activity and β glucosidase were found the highest in Sesbania + (fodder sorghum -chickpea) system and the lowest in Sesbania + (fodder sorghum -mustard) cropping system. The higher and favourable microbial activity under chickpea-based system could be attributed to its biological nitrogen fixation ability ) while, the lowest soil biological properties under mustard might be due to its allelopathic effect. Mustard is known to contain glucosinolates, which on hydrolysis produces biocidal chemicals such as isothiocyanates, nitriles, and ionic thiocyanates (Fenwick et al. 1983). These compounds have been reported to have broad allelopathic effects and suppress a number of soil microorganisms (Mazzola et al. 2007). The microbial and enzymatic activities are largely governed by the availability of organic matter (the substrate for microbial growth) in the soil. Legume-based crop rotations, agroforestry, mulching, residue retention and organic inputs have favourable effect on the organic matter content in soil (Ghosh et al. 2017). Application of organic mulch and residues in the soil provides much needed organic substrate for microbial growth and stimulates intra and extracellular enzymatic activities. Choudhary et al. (2018)  organic carbon improved biological properties of soil due to residue retention under CA-based agricultural practices as evident from the significant positive correlation between SOC and other biological soil properties. The significantly higher correlation between enzymatic activity and soil organic C is probably due to the fact that increased organic matter supports greater microbial biomass and activity and provides a better environment for stabilizing and protecting extracellular enzymes (Balota et al. 2004). The microbial and enzymatic activities are highly correlated with ZT and negatively correlated with CT . The higher microbial population and enzymatic activities under reduced tillage practices may be attributed to less soil disturbance, better aeration, higher moisture and more organic matter content (Dixit et al. 2019). The increased water stable aggregates under conservation tillage may also have favourable effect in enhancing soil enzymes, SOC and MBC (Roldan et al. 2003).

Conclusions
The study highlighted that the inclusion of chickpea (legume crop) in the alley cropping system improved soil physico-chemical and biological properties and recorded the highest SQI (0.97). All the tillage practices, viz., CT-CT, MT-ZT and CT-ZT with Sesbania mulching, being at par, recorded significantly higher system productivity than their respective treatments without mulching. The MT-ZT + mulch and CT-ZT + mulch, being at par, recorded significantly higher IR, hydraulic conductivity, water holding capacity and stable soil aggregates than other tillage and mulching combinations. Sesbania mulching coupled with reduced tillage (MT-ZT and CT-ZT) had higher soil moisture stock besides improving soil organic carbon and available nutrients. It was also observed that MT-ZT + mulching led to significant improvement in soil microbial population, SMBC and soil enzymatic activities (DHA, APA and β glucosidase). The MT-ZT + Sesbania mulch resulted in higher SQI (1.09).
Our study suggests that MT-ZT with Sesbania mulching is effective in improving productivity, soil health and moisture regimes under fragile rainfed agro-ecosystems. However, a farmer needs to be choosy among rainfed winter crops as mustard was found to be soil depleting while chickpea as soil building. Further, the implication of the present investigation is that the innovative Sesbania alleysbased food-fodder systems can be a good solution for minimizing the trade-offs for use of crop residues and/or by-products as animal fodder or mulch material for the livestock keepers in rainfed areas of arid and semi-arid tropics. Such Sesbania alleys-based food-fodder systems along with reduced tillage and mulching need to be recommended and popularized all across the rainfed agroecosystems.

Disclosure statement
No potential conflict of interest was reported by the author(s).