The high production costs in dairy farming were attributed to feed purchased from external markets

They added that flock sizes were determined by resource availability, which caused annual variability, as competition for land resources for grazing in the region was increasing.Goat rearing was a largely seasonal activity, and it was more predominant in the summer or taken up based on the income needs of individual HHs.Herd sizes ranged from 10 to 55 goats per household, depending on whether goat rearing was a primary or supplementary source of income.Focus-group participants stated that goats were grazed mainly on lands with tree and shrub cover.Large goat flocks would require large tracts of grazing land.The scarcity of such land has therefore reduced goat keeping.Sheep-rearing HHs rarely depended on off-farm labor, whereas those rearing goats frequently depended on off-farm labor or agricultural wage work.HHs also depended on markets or the public food distribution system to meet their food needs, although dependence was greater for goat-rearing HHs.The least prevalent category was the CWDL system.Most HHs in this category were medium farmers , followed by small and marginal farmers.Only 4% were large farmers.Crop production in this system veered towards food crops, using seasonally available water resources, and limited external inputs.In this case, livestock keeping was integrated and intended to support crop production.According to the participants in the focus groups, this had been the most prevalent system in the 1990s.Diverse livestock species were reared in this system.Crop and livestock products were consumed predominantly at home, and only surplus production, if available, was sold.The primary source of income for these HHs consisted of remittances from family members working in cities.Land and herd sizes are presented in Table 2.Land and herd sizes differed across systems.HHs in the CSR system had the highest herd size in comparison to all systems.The CD, CSR, and CWDL systems had comparable land sizes.

The LWL system was not comparable to any system in both land and herd sizes.In this section, we present the results of the economic-performance study of only the CWL, CD, and CSR systems ,ebb flow table as they provided consistent income from agriculture.The revenue, costs of production, and total GM per household are displayed in Table 3 for all three systems under study.For HHs in the CD and CSR systems, GM comprised income from crop and livestock production.For those in the CWL system, it consisted of crop production and off-farm activities.The economic performance of crop production is explained by the various crop management and input requirements across systems.As indicated in the focus-group discussions, the CWL and CD systems limited crop production to the monsoon agricultural season each year, while the CSR system managed crop production for both monsoon and winter seasons each year.In terms of inputs, the CWL system incurred the highest crop production costs, followed by the CSR system and then the CD system.The differences were due to the types of crops grown, with cash crops having higher costs, associated inputs, and the availability of livestock manure, which replaced expenses for inorganic inputs.Detailed information about the production costs for livestock rearing is presented in Table 4.The costs for dairy farming were substantially higher than those for small ruminants.The CD system also exhibited the greatest variation in the GM.Some of the HHs in this system had negative GMs in the summer due to low milk production combined with high feed requirements by the cattle.The CSR system was the most profitable farming system, due to low feed costs and high market price for meat.The highest costs per annum in the CSR system were for animal health care and for leasing grazing lands.These factors nevertheless did not seem to impair economic performance.The CSR system managed to obtain a high GM in the summer, the most unproductive season in dryland regions due to high temperatures and water shortages.Another factor addressed in the survey was loan access and repayment.The findings revealed that HHs took loans from multiple sources to continue farming.Among the three systems, CD HHs had the most loans from cooperative banks , local pawnbrokers , and selfhelp groups simultaneously.The loan values were also higher in comparison to those of HHs in the other two systems.In the CWL system, HHs accessed government crop loans , self-help group loans , and government schemes to manage crop production.In contrast, only 50% of CSR HHs took loans, and only from cooperative banks.

Focus group discussions indicated that loans from self-help groups were entirely availed by women, however, loans from banks were from both genders, as women also had access to banks.Despite formal credit sources, informal credit sources are still being accessed particularly by the CD HHs.This situation can be related to the high investment and production costs in dairying farming, where formal credit options work due to pending loan repayment.The GM of the HHs in the CSR system was statistically higher than that of those obtained by HHs in the CD and CWL systems.There were no statistical differences between the CD and CSR groups.The linear regression analysis revealed that factors explaining the GM were dependent on the farming system.First, caste and family type were not significant in any of the farming systems.For the CWL system, land size was the only statistically significant variable clarifying the GM.For the CSR system, both herd size and land size were significantly and positively correlated with GM.For the CD system, however, none of these variables was statistically significant.In Fig.2 below, we further illustrate the relationship between herd size and GM, which helps to explain why herd size is an explanatory variable for the CSR system, but not for the CD system.For the CSR system, GM increased along with herd size, as indicated by the significant regression line.In contrast, the CD system exhibited high variation, as CD farms with low herd size obtained both negative and positive GMs, while those with large herd sizes obtained only moderate to low GMs.The CWL system had the lowest gains and no livestock.The characterization of farming systems in the study region revealed that the CWL, CD, and CSR systems were variants of intensive, specialized, and market-oriented farming systems, while the LWL and CWDL were variants of subsistence farming systems.The majority of the HHs in the region fell into two farm categories: CWL and CD.The CSR system, although lucrative, was dominated by the BC communities in the region, given that sheep rearing has been their traditional occupation for generations.

For LWL and CWDL systems, livestock rearing was a need-based livelihood activity, and it usually involved poultry and seasonal goat rearing.Although the CWDL system was the most prevalent in the past, the majority of HHs have now transitioned away from this system.Further analysis of the three systems revealed that the CWL system is a medium-input/low-output system, the CD system is a high-input/highoutput system, and the CSR system is a medium-input/medium-output system.In terms of economic performance, the CSR system showed the best performance, as explained by the low water requirements and low feed production costs.The profitability of this system was further enhanced by growing market demand and the current market price for small ruminant meat.The system also adapted to the dynamic context by adjusting herd sizes to the decreasing availability of common property resources.All these factors make the CSR system suited to the dryland context.Despite having the highest revenues, the CD system was less profitable, due to high production costs.This system exhibited high variability in GMs from moderate to substantially negative records across HHs.This variability might have been due to the influence of other factors not included in this study.In addition, the consistent income obtained from dairy farming came at the expense of crop production in the winter season, as scarce water resources were diverted for dairying.This strategy resulted in the loss of additional income for CD HHs, in contrast to those in the CSR system, which cultivated crops for two seasons each year, in addition to rearing small ruminants.These findings thus suggest that engaging in dairy production may not be a resilient option for HHs in semi-arid regions.The CWL system consistently exhibited low economic performance, with low revenues attributable primarily to higher production costs for cash crops and market volatility.In line with other studies such as, Sallu et al.; Ten Napel et al.; Ayeb-Karlsson et al.; Kuchimanchi et al.we find that the trend of intensification and specialization in farming, particularly in the CWL and CD systems, has increased generic risks and decreased flexibility for coping with disturbances and shocks.For example, the CWL system reflected the absence of crop diversity and livestock and was dependent on off-farm employment, which was not regularly available.

The lack of crop livestock integration at the farm level increases dependence on inorganic fertilizers , which reduces soil carbon levels, subsequently affecting soil fertility, crop productivity, and revenue in the long term.These factors make HHs in this system more reliant on external inputs and market conditions to continue farming, leading to higher risks in the long term.The CD system was the most desired by HHs in the region, as it provided consistent income throughout the year.However, this system had compromised GMs and can be seen as entailing high risk,hydroponic grow table as dairy farming is heavily dependent on external markets for feed resources, scarce water resources and milk collection.Small landholdings limit feed production and increase the amount of external feed that HHs are forced to purchase to guarantee production.Further, as reported in studies by Sishodia et al.and the Central Ground Water Board , the region is currently experiencing high water scarcity, and the situation is likely to worsen.In addition to being risky, therefore, the CD system may be economically unviable in dryland regions , contrary to general perceptions.For this reason, the promotion of dairy farming among poor HHs should be a point of concern for development programs, especially in dryland regions.In the farming systems examined, higher revenues were associated with higher costs due to increased use of purchased inputs, credit, and animal healthcare services.If these costs cannot be limited, they offset revenues, hinder profits, and perpetuate the ‘poverty trap’.In this study, this situation is illustrated by the fact that HHs in the CWL and CD systems had high levels of credit and debt, due to insufficient income and low profits.Increasing credit and debt thus pose a risk, as they are likely to become intertwined with farming strategies aimed at simply adopting a system and continuing to farm.Over time, this situation often leads to a range of social-ecological consequences , all of which perpetuate vulnerability to climate change.Although the CSR system has adapted into a modernized version of traditional small ruminant production, it is likely to be subject to further constraint due to dwindling common property resources and the decreasing availability of private lands for grazing in the future.Moreover, the scarcity of grazing resources has deprived low-income HHs of alternative and profitable livelihood opportunities from goat rearing.

The decrease in native poultry rearing is having a similar impact on these HHs, despite the presence of a niche market.In dryland regions, the current reduction in livestock rearing is leading to income losses, while also translating into decreased dietary diversity during lean periods and the loss of a critical buffer in times of drought or dry spells, as crop production is highly vulnerable to such threats.Characterization and economic performance studies like this one thus provide insight into the socio-economic and ecological dimensions of farming systems and support a more customized approach to agricultural development in dryland regions.Following, we discuss some outcomes from this study under the perspective of the WDP policy, given that it is India’s leading strategy for the development of dryland regions.Firstly, this study shows that 86% of the HHs are now practicing intensive market-oriented farming.Intensive systems are often associated with increased specialization and low integration between crop and livestock production, resulting in high-water usage and the doctoral thesis by Kuchimanchi.We thus infer that water resources generated through the soil and water conservation measures due to WDPs in the region are apparently being over-utilized by some HHs thereby decreasing the availability of water throughout the year.Likely, this also explains why only 38% of the HHs in the region have been able to adopt the CD system, or why the CWL and CD systems limit crop production to the monsoon season.A second notable outcome of the study is that, while intensification and specialization in farming have increased agricultural production, it has not led to economic prosperity.For example, the average daily per capita income for HHs in the CWL, CD, and CSR systems were USD 0.2, USD 1.2, and USD 2.4, respectively.These income values are lower than the per diem wage rate of USD 2.5, as prescribed by the Indian Ministry of Labor and Employment , and the World Bank extreme-poverty threshold of USD 1.9 day/person.Lastly, while WDPs tend to promote specific farming systems , that has induced changes in land use, cropping patterns, and livestock rearing in terms of herd size, animal type, and purpose.