Innovative modeling on the effects of low-temperature stress on rice yields.

The article focuses on improving the accuracy of rice crop models in simulating the impact of low-temperature events on rice yield, particularly in temperate and cold production regions. Through six years of controlled chamber experiments, the study identified that low temperatures reduce spikelet fertility, grain number per spike, and grain weight at different growth stages from panicle initiation to grain filling. The researchers enhanced the spikelet fertility algorithm of the widely used ORYZA model and developed new functions for grain number per spike and grain weight, incorporating stage-specific temperature sensitivities and cultivar-specific cold tolerance thresholds. When applied to ten rice growth models, these improvements significantly increased simulation accuracy and reduced uncertainty in predicting rice yields under low-temperature stress, as validated by experimental and field data across multiple rice varieties. This work provides a refined tool for assessing and adapting rice production to climate change challenges involving low-temperature stress.