Predictive Modeling of Climate Change Impacts on Farm Income and Food Security in Haryana, India: An Adaptation-Scenario and Optimization Analysis.
Year: 2026
Type: book
Abstract: Haryana is one of the big surplus production states of the country in terms of grain production but yet it is very much susceptible to the effects of increase in temperature, variability of monsoon and intensive cultivation of rice-wheat on groundwater. An integrated predictive-modeling chain is proposed that integrates downscaled climate projections of SSP2-4.5 and SSP5-8.5 scenarios, a hybrid crop-yield engine (DSSAT-CERES and machine-learning based ensemble), a Ricardian net-farm-income function, a composite Food Security Index (FSI), and linear-programming (LP) adaptation-optimization module. The ensemble yield model (XGBoost) was the best among all models in all five major crops and four agro-climatic zones in terms of predictive ability (R² = 0.91, RMSE = 0.22 t ha⁻¹). Under SSP5-8.5, wheat and rice yields are projected to decrease by 14-23% by the 2050s-2080s, which would be a 21% decrease in state average net farm income and a reduction in FSI from 0.61 to 0.56. The most fragile area is the south west of the country which is dry. With the net farm income (17.3% above BAS) from the changed cropping reallocation for LP optimisation and the transformational adaptation of heat-tolerant crops, the change in sowing dates, micro irrigation or crop diversification, the FSI returns to 0.70. How predictive modelling and optimization can be combined to provide a quantitative climate risk assessment and prioritization of adaptation investments down to the sub-state level is illustrated in the diagram..
Authors: Ashutosh Ashutosh, Manju Dahiya
Institution: Galgotias University
Language: English
Subfield: Agronomy and Crop Science
Domain: Life Sciences
Open Access status: gold
OpenAlex ID: W7204183172