Impacts of Climate Warming on Rice Growth and Production in Hainan Province, China: Insights from a Multi-Crop Model Ensemble

Hainan Province is the only tropical province in China and is an important tropical rice-producing region, playing a critical role in regional food security. Rising temperatures associated with climate change may adversely affect rice growth and yield in tropical and subtropical regions, increasing production risks. This study integrated meteorological data from six national agro-meteorological stations in Hainan Province with long-term observations of rice phenology and yield to assess future climate impacts using a multi-climate model and multi-crop model ensemble framework. The CERES-Rice, ORYZA2000, and WOFOST models were separately calibrated and validated using historical observations. Subsequently, CMIP6 projections from five GCMs under SSP1-2.6, SSP3-7.0, and SSP5-8.5 were used to simulate rice phenology and yield during the 2050s, 2070s, and 2090s. A multi-model ensemble mean was further used to integrate inter-model differences across scenarios and future periods. All three crop models showed satisfactory performance in simulating rice phenology and yield. Future warming generally advanced flowering and maturity dates of early rice, whereas late rice exhibited greater spatial and inter-model variability. Yield losses became more pronounced under SSP3-7.0 and SSP5-8.5, with late rice showing greater reductions than early rice. Sowing date adjustment partially offset climate-induced yield losses, although the optimal strategy differed between rice seasons and climate scenarios. Overall, the multi-model framework provided a robust basis for climate risk assessment and adaptation planning for tropical double-rice production in Hainan Province.