Investment planning to minimize climate risk in agricultural production: an optimization model for a semi-arid region in India

This study undertakes a prioritization of investments for upscaling context-specific climate-smart agriculture (CSA) technologies across several districts in the semi-arid Telangana state of India. We first analysed the trade-off between expected agricultural income in these districts and deviation from it under drought and normal weather scenarios. We extended the conventional MOTAD model with climate-smart technologies to assess their role in minimizing this trade-off under various weather scenarios. For our analysis, we relied on a five-year (2010–2011 to 2014–2015), district-level panel dataset on the cost of cultivation and crop production of 11 major crops under six different climate-smart technologies and farmers’ traditional practices (FTPs). This dataset included a collation of official statistics on the cost of cultivation, focus group interviews with farmers over the years and data from experimental plots operated by regional agricultural research stations. We found that the adoption of CSA technologies in Telangana, compared to FTPs, led to a 16% reduction in production risk while achieving optimum levels of crop income. Under FTPs, in a scenario of the high probability of drought, the production risk is increased by 12% while the adoption of CSA technologies reduced this risk by 25%. Our study suggests that increasing investments in farm ponds and unpuddled machine transplanting of rice can minimize the risk-return trade-off in a higher drought frequency scenario. The study fills a critical gap by generating evidence, and developing a robust planning tool, for policymakers to make informed decisions on investments in resilience-enhancing CSA