Spatio-temporal patterns and driving mechanisms of rice biomass during the growth period in China since 2000

China is a major rice-producing nation and understanding changes in rice planting areas and growth periods and the mechanisms driving rice biomass is important to ensure effective agricultural strategies and food security. In this study, rice planting areas from 2000 to 2021 in China were extracted using Savitzky-Golay filtering and the absolute distance index. Using MODIS remote sensing image data, we determined the rice growth period based on the inflection points of the enhanced vegetation index. In addition, the cumulative biomass of rice during the growth period since 2000 was determined based on the Carnegie-Ames-Stanford approach model. Finally, the key driving mechanisms and coupling relationships among the driving factors of rice biomass during the growth period were determined using geographically weighted regression and structural equation models. The results showed that the accumulation of net primary productivity (NPP) during the rice growth period in Northeast China, the middle and lower reaches of the Yangtze River, and Northwest China increased significantly (P < 0.01). The accumulation of NPP during the rice growth period in Huang-Huai-Hai, South China, and Southwest China decreased significantly (P < 0.01). In addition, Central Inner Mongolia, Qinghai-Tibet Plateau, and Loess Plateau have less rice planting area and no significant variation pattern in the accumulation of NPP during the rice growth period. In the nine major agricultural regions of China, precipitation, temperature, and evapo-transpiration were among the most influential meteorological factors affecting rice biomass. For the inland areas with less precipitation, drainage density was an increasingly important influencing factor. Areas with high topographic relief were more likely to be affected by the slope. In areas with relatively poor soil, the fertilizer application rate influenced rice biomass growth. Our results provide insights into the spatio-temporal patterns of rice biomass and its driving mechanisms, and demonstrate the efficacy of using remote sensing data at a macro scale.