Smart Satellite Tools Boost Forage Farming in Arid Lands 

A new study in Water Research presents an artificial intelligence (AI)-powered remote sensing framework to precisely map forage cultivation potential across northern China's hs, especially the middle reaches of the Yellow River. This study identified optimal forage belts at the kilometer scale, offering robust data and decision-ready tools to support ecological protection, sustainable agriculture, and national feed and food security.

Led by Dr. WANG Shudong at the Aerospace Information Research Institute, Chinese Academy of Sciences (AIRCAS), the study was jointly conducted by Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters and the Department of Earth and Environmental Science at the University of Pennsylvania.

Northern China's drylands face twin challenges: scarce water resources and the need to secure both feed and food supplies. To address these, this study developed a cross-scale multi-source fusion framework that integrates satellite observations, ecohydrological model outputs, and field measurements, minimizing the need for dense in-situ sampling.

Using multi-source satellite data and mechanistic models of water-balance and crop-growth, researchers generated  high-quality training samples, and applied ensemble and transfer learning to retrieve key production factors, including irrigation water use, vegetation net primary productivity, and soil organic carbon—with retrieval accuracies exceeding 90%. Distribution alignment and quantile mapping reduced regional bias by 43%, achieving over 85% positional accuracy for optimal forage belts.

The framework moves beyond single-metric assessments by treating forage-planting as a spatial optimization problem that jointly balances water consumption, soil-carbon sequestration benefits, and forage production capacity. By quantifying ecological gains, economic returns, and water costs on a unified scale, it identifies priority plots and optimal input-output ratios for efficient allocation of labor, resources, and funding.

Replicable and cost-effective, the approach supports ecosystem recovery and high-quality agricultural development under stringent water constraints. It is also adaptable to other dryland regions, such as the Inner Mongolia-Ningxia ecotone, and the oasis margins of the Hexi Corridor-Tarim Basin, and similar environments: the African Sahel, South Asia, and West Asia.