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

The primary aim of this study is to prioritize investment required for scaling up climate-smart agriculture (CSA) technologies across different districts of Telangana state, which is in the semi-arid region of India. First, we analysed the trade-offs between expected agricultural income and its deviation across districts under drought and normal weather scenarios. The conventional MOTAD model was extended with various climate-smart technologies to assess their role in minimizing the trade-offs under various weather scenarios. A district-level panel dataset on cost of cultivation and crop production for 11 major crops under six different climate-smart technologies and farmers’ practices for five years (2010-11 to 2014-15) has been used. The dataset comprised a collation of official statistics on cost of cultivation, focus group interviews with farmers over the years, and data from experimental plots of Regional Agricultural Research Stations. The analysis reveals that the adoption of CSA technologies is likely to reduce production risk by 16% compared to farmers’ traditional practices (FTPs) while achieving optimum levels of crop income. Under a scenario of higher probability of drought, production risk is likely to increase by 12% in the state under FTPs; the adoption of CSA technologies could reduce the risk by 25%. The study suggests increasing investments in farm ponds and un-puddled machine transplanting in rice to minimize the risk-return trade-offs under a higher drought frequency scenario. Finally, the study generates evidence for policymakers to make informed investment decisions on CSA in order to enhance farming system resilience across districts in the semi-arid state of Telangana, India.