TY  - Data
T1  - AsiaRiceYield4km: Seasonal Rice Yield in Asia from 1995 to 2015
A1  - Zhang Zhao
DO  - 10.5281/zenodo.6901968
PY  - 2023
DA  - 2023-10-23
PB  - National Cryosphere Desert Data Center
AB  - Based on the annual rice map of Asia, this study integrated multi-source predictive factors into three machine learning (ML) models and generated a seasonal rice yield dataset with high spatial resolution (4 km) from 1995 to 2015 (AsiaRiceYield 4km). Divide the predictive factors into 4 categories, consider the most comprehensive rice growth conditions, and determine the optimal ML model based on inverse probability weighting method. The results showed that AsiaRiceYield4km had good accuracy in estimating seasonal rice yield (single grain rice: R2=0.88, RMSE=920 kg·ha-1; double cropping rice: R2=0.91, RMSE=554 kg·ha-1; three grain rice: R2=0.93, RMSE=588 kg·ha-1). Compared with the SPAM model, the R2 of Asian rice yield by 4km increased by 0.20 on average, and the RMSE decreased by 618 kg·ha-1 on average. Especially, constant environmental conditions, including longitude, latitude, altitude, and soil properties, contribute the most to the estimation of rice yield (~45%). At different growth stages of rice, the predictive factors of reproductive stage have a greater impact on rice yield prediction than those of nutritional stage and full growth stage. This dataset is a new type of long-term grid based rice yield dataset, which can fill the gap of high spatial resolution seasonal yield products in major rice producing areas and promote relevant research on global agricultural sustainable development.
DB  - NCDC
UR  - http://www.ncdc.ac.cn/portal/metadata/9c650428-8f4b-470a-8025-f40299efb763
ER  -