UC Davis

Agriculture faces intensifying challenges from climate change, threatening both perennial and annual crops that are crucial to global food and nutrient security. This dissertation advances the modeling of climate impacts on agriculture by developing and applying a novel statistical framework for perennial crops, and by scaling and calibrating a process-based model for annual crops, thereby improving yield projections under climate change and informing adaptation planning.The first study develops a novel modeling framework that integrates climate modeling, horticultural science, and statistical yield modeling, and applies it to California almonds. Results show that increasing minimum temperatures and humidity during the bloom and pollination stage, along with heat stress during the growing stage, are primary drivers of yield losses. Climate change is projected to reduce almond yields by up to 49% by 2100 under the high warming scenario (SSP585). However, sustained innovation gains could more than offset climate damages, highlighting the joint role of technological progress and climate adaptation in perennial systems.The second study expands AquaCrop, a process-based crop model developed by the Food and Agriculture Organization, for regional-scale applications. A flexible Python-based gridded implementation is developed and calibrated using two decades of county-level U.S. maize yield data, which substantially improves model performance compared to the default field-scale calibrated AquaCrop. This county-level calibration approach enables AquaCrop to capture both spatial and temporal dynamics of observation data more effectively than existing approaches, demonstrating the value of regional calibration for large-scale applications and establishing a scalable framework for future continental and global assessments.The third study applies the county-level calibrated AquaCrop to examine U.S. maize production under climate change. Driven by CMIP6 climate projections, simulations indicate that U.S. maize production could decline by 17% by the end of the century under the high warming scenario (SSP585), with particularly severe impacts in the Corn Belt region. Analyses of adaptation strategies show that irrigation, fertilization, and spatial relocation of maize systems can potentially offset climate damages, but feasibility is constrained by water availability, land competition, and resource efficiency. County-level projections provide actionable insights for agricultural policy, crop insurance, and farm-level planning.