Abstract
Climate change poses a direct threat to global food security, and tropical regions such as Indonesia, already exposed to floods and droughts, face more of that risk than most. This study assesses agricultural land suitability under two Representative Concentration Pathway (RCP) scenarios, RCP 4.5 and RCP 8.5, across three future periods (2030s, 2050s, and 2070s) in Trenggalek Regency. The analysis combines field data collection, climate modeling through a 12-model ensemble from the Climate Change, Agriculture and Food Security (CCAFS) initiative, and spatial suitability modeling. Model validation showed strong performance on both fronts, with an Overall Accuracy of 0.90 and Kappa Coefficient of 0.83 for the land-cover classification, and an R² of 0.85 with an RMSE of 1.24 for the regression linking climate variables, temperature and rainfall, to the land-suitability index. Climate projections based on this validated model indicate rainfall and temperature rising gradually under RCP 4.5 and considerably more sharply under RCP 8.5. Land suitability follows a similar trajectory, declining over time and most under the high-emission scenario, where highly suitable and suitable areas shrink markedly by the 2070s as temperatures exceed rice tolerance thresholds and rainfall becomes more extreme. Sustaining agricultural productivity through this shift requires adaptation strategies tailored to specific locations rather than applied uniformly.