Soybean, an essential food crop, has witnessed a steady rise in demand in recent years. There is a lack of high-resolution annual maps depicting soybean-planting areas in China, despite China being the world's largest consumer and fourth-largest producer of soybean. To address this gap, we developed the novel Regional Adaptation Spectra-Phenology Integration method (RASP) based on Sentinel-2 remote sensing images from the Google Earth Engine (GEE) platform. We utilized various auxiliary data (e.g., cropland layer, detailed phenology observations) to select the specific spectra and indices that differentiate soybeans most effectively from other crops across various regions. These features were then input for an unsupervised classifier (K-means), and the most likely type was determined by a cluster assignment method based on dynamic time warping (DTW). For the first time, we generated a dataset of soybean-planting areas across China, with a high spatial resolution of 10 m, spanning from 2017 to 2021 (ChinaSoyArea10m). The R² values between the mapping results and the census data at both the county and prefecture levels were consistently around 0.85 in 2017–2020. Moreover, the overall accuracy of the mapping results at the field level in 2017, 2018, and 2019 was 77.08 %, 85.16 %, and 86.77 %, respectively. Consistency with census data was improved at the county level (R² increased from 0.53 to 0.84) compared to the existing 10 m crop-type maps in Northeast China (Crop Data Layer, CDL) based on field samples and supervised classification methods. ChinaSoyArea10m is very spatially consistent with the two existing datasets (CDL and GLAD (Global Land Analysis and Discovery) maize–soybean map). ChinaSoyArea10m provides important information for sustainable soybean production and management as well as agricultural system modeling and optimization. ChinaSoyArea10m can be downloaded from an open-data repository (DOI: https://doi.org/10.5281/zenodo.10071427, Mei et al., 2023).

Rice is the most important staple food in Asia. However, high-spatiotemporal-resolution rice yield datasets are limited over this large region. The lack of such products greatly hinders studies that are aimed at accurately assessing the impacts of climate change and simulating agricultural production. Based on annual rice maps in Asia, we incorporated multisource predictors into three machine learning (ML) models to generate a high-spatial-resolution (4 km) seasonal rice yield dataset (AsiaRiceYield4km) for the 1995–2015 period. Predictors were divided into four categories that considered the most comprehensive rice growth conditions, and the optimal ML model was determined based on an inverse probability weighting method. The results showed that AsiaRiceYield4km achieves good accuracy for seasonal rice yield estimation (single rice: R²=0.88, RMSE = 920 kg ha⁻¹; double rice: R²=0.91, RMSE = 554 kg ha⁻¹; and triple rice: R²=0.93, RMSE = 588 kg ha⁻¹). Compared with single rice from the Spatial Production Allocation Model (SPAM), the R² of AsiaRiceYield4km was improved by 0.20, and the RMSE was reduced by 618 kg ha⁻¹ on average. In particular, constant environmental conditions, including longitude, latitude, elevation and soil properties, contributed the most (∼ 45 %) to rice yield estimation. For different rice growth periods, we found that the predictors of the reproductive period had greater impacts on rice yield prediction than those of the vegetative period and the whole growing period. AsiaRiceYield4km is a novel long-term gridded rice yield dataset that can fill the unavailability of high-spatial-resolution seasonal yield products across major rice production areas and promote more relevant studies on agricultural sustainability worldwide. AsiaRiceYield4km can be downloaded from the following open-access data repository: https://doi.org/10.5281/zenodo.6901968 (Wu et al., 2022).