RiceExtent v1.0: a global rice-specific land cover product from multi-source data integration
Abstract
Accurate land cover information is fundamental for monitoring land-use dynamics, supporting sustainable agriculture, and informing environmental and climate-related decision-making; however, detailed and globally consistent crop-specific datasets, particularly for rice, remain limited. The study presents RiceExtent v1.0, a harmonized global rice-specific land cover dataset athigh (10 m) and moderate (250 m) resolution, developed by integrating verified local, national, and regional rice maps with ESA WorldCover v200 using a hierarchical multi-criteria dataset ranking and spatial merging approach that prioritizes source accuracy and spatial detail. Using an extent-based merging strategy that aggregates multi-source rice data, RiceExtent derives a maximum observed rice cultivation extent rather than annual mean estimates, thereby capturing double and triple cropping patterns, fragmented smallholder fields, and inter-annual variability often underrepresented in other products. The final product was evaluated using area-based statistical comparison metrics against Spatial Production Allocation Model (SPAM), RiceAtlas, and GloRice, along with USDA rice production statistics. Validation using independent ground-truth samples was performed using True Positive Rate (TPR). The developed dataset captures an estimated global rice area of ~ 1.67 million km² and shows broad agreement with official statistics and improving representation in Southeast Asia, South Asia, Africa and the Americas. RiceExtent v1.0, provided through an open repository ( https://dataverse.tdl.org/dataset.xhtml?persistentId=doi:10.18738/T8/JYAEHV ) with accompanying Python script, offers a robust resource for agricultural monitoring, crop and hydrological modeling, and climate impact assessments, while enabling collaborative refinement of global rice mapping.
Related
- Targeted CPSF6 recoding reveals global fitness advantages of SPAD integration targeting and cell type-specific roles for capsid binding in HIV-1 DNA synthesis and nuclear import
- Edible Insects for Future Protein and Product Development: Mapping Global Research Trends, Innovation, and Development Through Bibliometric Analysis
- Dry olive cake, an olive by-product, as a functional fiber source in diets for laying pullets: effects on growth and gut health.
- Global context integration for pharmacologically robust preclinical sleep stage classification
- Synapse-specific and plasticity-regulated AMPA receptor mobility tunes synaptic integration
BibTeX
@article{patel2026riceextent,
title = {{RiceExtent v1.0: a global rice-specific land cover product from multi-source data integration}},
author = {Akansha Patel and Arun Bawa and Seonggyu Park and Celray James Chawanda and Sang-Min Jun and Soon-Kun Choi and Sushil Kumar Himanshu and Raghavan Srinivasan and Jaehak Jeong},
journal = {Scientific Reports},
year = {2026},
month = oct,
doi = {10.1038/s41598-026-70276-0},
url = {https://www.nature.com/articles/s41598-026-70276-0},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - JOUR
TI - RiceExtent v1.0: a global rice-specific land cover product from multi-source data integration
AU - Akansha Patel
AU - Arun Bawa
AU - Seonggyu Park
AU - Celray James Chawanda
AU - Sang-Min Jun
AU - Soon-Kun Choi
AU - Sushil Kumar Himanshu
AU - Raghavan Srinivasan
AU - Jaehak Jeong
PY - 2026
DA - 2026/10/21
JO - Scientific Reports
DO - 10.1038/s41598-026-70276-0
UR - https://www.nature.com/articles/s41598-026-70276-0
AB - Accurate land cover information is fundamental for monitoring land-use dynamics, supporting sustainable agriculture, and informing environmental and climate-related decision-making; however, detailed and globally consistent crop-specific datasets, particularly for rice, remain limited. The study presents RiceExtent v1.0, a harmonized global rice-specific land cover dataset athigh (10 m) and moderate (250 m) resolution, developed by integrating verified local, national, and regional rice maps with ESA WorldCover v200 using a hierarchical multi-criteria dataset ranking and spatial merging approach that prioritizes source accuracy and spatial detail. Using an extent-based merging strategy that aggregates multi-source rice data, RiceExtent derives a maximum observed rice cultivation extent rather than annual mean estimates, thereby capturing double and triple cropping patterns, fragmented smallholder fields, and inter-annual variability often underrepresented in other products. The final product was evaluated using area-based statistical comparison metrics against Spatial Production Allocation Model (SPAM), RiceAtlas, and GloRice, along with USDA rice production statistics. Validation using independent ground-truth samples was performed using True Positive Rate (TPR). The developed dataset captures an estimated global rice area of ~ 1.67 million km² and shows broad agreement with official statistics and improving representation in Southeast Asia, South Asia, Africa and the Americas. RiceExtent v1.0, provided through an open repository ( https://dataverse.tdl.org/dataset.xhtml?persistentId=doi:10.18738/T8/JYAEHV ) with accompanying Python script, offers a robust resource for agricultural monitoring, crop and hydrological modeling, and climate impact assessments, while enabling collaborative refinement of global rice mapping.
SN - 2045-2322
ER -