Understanding Climate-Agriculture Interactions through AI-Enhanced Earth Observation

The intensifying impacts of climate change on agricultural systems represent one of the most pressing interdisciplinary challenges facing the Earth science community today. Rising temperatures, shifting precipitation regimes, more frequent compound extremes (droughts, heatwaves, floods), and the slow-onset processes of soil degradation and salinization are transforming the conditions under which the world’s food is produced. These changes do not occur in isolation: they emerge from complex interactions between the atmosphere, biosphere, hydrosphere, and human land-use systems, and they unfold across multiple spatial and temporal scales. Understanding these coupled dynamics, and translating that understanding into actionable knowledge, requires a genuinely cross-disciplinary approach that brings together climate science, Earth observation, data science, and agricultural research.Recent advances in satellite remote sensing (Sentinel-1/2, Landsat, MODIS, PlanetScope, and emerging hyperspectral and SAR platforms) combined with artificial intelligence and machine learning are reshaping our capacity to observe and interpret climate-agriculture interactions. These technologies are no longer ends in themselves; they are powerful tools that enable us to detect early signals of climate stress on crops, quantify land-atmosphere feedbacks in managed landscapes, and uncover patterns in soil moisture, biomass, and ecosystem productivity that conventional methods cannot reach. AI-driven fusion of multi-source observations is helping bridge long-standing gaps between fine-scale agronomic processes and coarse-resolution climate models.This Research Topic invites contributions that use these emerging tools to advance our scientific understanding of how climate change reshapes agricultural systems, and how those systems, in turn, influence regional climate and Earth surface processes. We welcome original research, methodological advances, and reviews covering, but not limited to:Remote sensing and AI-driven detection of climate stress on crop phenology, biomass, and yieldCoupled climate-land-agriculture modelling and the role of Earth observation in model evaluationSoil moisture, evapotranspiration, and water-cycle dynamics in agricultural landscapes under changing climateClimate extremes (droughts, floods, heatwaves) and their cascading impacts on agricultural ecosystemsLand-atmosphere feedbacks, carbon and energy fluxes in managed lands, and contributions to regional climateMachine learning approaches for downscaling, gap-filling, and integrating climate and agricultural Earth observation dataClimate adaptation pathways and the role of Earth observation in informing policy and decision supportOpen data, FAIR principles, and reproducible workflows in interdisciplinary climate-agriculture researchBy bringing together perspectives from climate science, Earth observation, geospatial data science, and agricultural systems research, this collection aims to advance both the fundamental understanding of climate-agriculture interactions and the methodological frameworks needed to study them at scale. We particularly encourage interdisciplinary contributions that build bridges across traditionally separate communities, fostering the kind of integrated science the climate-food nexus demands.The Research Topic aligns with Frontiers in Earth Science’s commitment to publishing high-impact, open-access research that supports a sustainable and climate-resilient future. Topic editors: Songyan Zhu, MOUSONG WU Source: https://www.frontiersin.org/research-topics/82416/understanding-climate-agriculture-interactions-through-ai-enhanced-earth-observation