Simulating the Effects of Low-Temperature Stress During Flowering Stage on Leaf-Level Photosynthesis with Current Rice Models
Min Kang,Siyuan Wang,Zhenkai Xu,Chenzhe Xu,Jingwei An,Yu Zhang,Yaowen Zeng,Iftikhar Ali,Liang Tang,Liujun Xiao,Bing Liu,Leilei Liu,Weixing Cao,Yan Zhu
DOI: https://doi.org/10.1016/j.agrformet.2024.110087
IF: 6.2
2024-01-01
Agricultural and Forest Meteorology
Abstract:Photosynthesis is crucial for crop growth and yield, yet remains a limitation in enhancing crop production under global climate change. Crop models have been widely used to project the impacts of climate change on crop yields, including extreme temperature events. However, most crop simulation models performed poorly on leaf-level photosynthesis simulation under low-temperature stress (LTS). Here, two rice cultivars, Huaidao 5 and Nangeng 46, were treated under seven LTS intensities (from 27/21 °C to 12/6 °C) and five LTS durations (from 3 to 9 days) during the flowering stage for three years under environment-controlled conditions. The results showed that the observed maximum photosynthetic rate (Pmax) and initial light-use efficiency (ε) decreased with increasing intensities and durations of LTS, and Pmax and ε were linearly correlated (R2 = 0.94). Experimental data under different LTS have been tested by four widely used rice models, including RiceGrow, ORYZA2000, RICEPSM, and GEMRICE, indicating that the effects of LTS durations on Pmax and ε have not been captured well by most existing models. By integrating accumulated cold degree days (ACDD, °C·d) as the LTS index with a cultivar-specific low-temperature threshold parameter (Th), algorithms of temperature factor (FT) and ε were improved, and the effects of LTS on Pmax and ε were quantified with a double exponential decay function in this research. Compared with the original models, the normalized root mean square error (NRMSE) between observed and simulated photosynthetic rates in the four improved models decreased by about 80 % under LTS treatments, indicating the better performance of the improved models. In addition, the recovery factor of LTS damage to photosynthesis (FLTrecover) algorithm was developed to simulate the recovery of the leaf-level photosynthetic rate (P) after LTS treatments. This study could provide an effective quantitative tool for accurately estimating rice photosynthetic production and grain yield under future climate conditions.