Machine Learning and Remote Sensing for Hydroclimatic Extremes and Environmental Risk

Hydroclimatic extremes are increasing in frequency and severity in many regions, driven by climatic shifts, land-use change, rapid urbanization, and rising environmental stress. Understanding how these extremes evolve, interact with landscape processes, and generate cascading environmental risks remains a major scientific challenge. Concurrently, the availability of high-resolution satellite data, open-access Earth observation archives, and advanced computational methods has expanded the potential to monitor environmental dynamics at scales previously impossible. Machine learning techniques—ranging from deep learning to complex network-based models—are increasingly used to analyze large geospatial datasets, detect emerging patterns, and predict extreme events with greater accuracy. Remote sensing products from multispectral, hyperspectral, radar, and thermal platforms offer invaluable information on hydrology, vegetation, soil moisture, land deformation, and surface processes.



Hydroclimatic extremes—ranging from floods and droughts to heatwaves and intense precipitation—are accelerating under global environmental change, placing unprecedented pressure on ecosystems, infrastructure, and societies. Traditional modelling approaches often fall short in capturing the complexity, nonlinearity, and multi-scale nature of these events. At the same time, rapid advances in machine learning, remote sensing, and Earth observation provide a transformative opportunity to improve detection, prediction, and risk assessment of hydroclimatic extremes across diverse landscapes. This issue aims to bring together innovative studies that harness the power of data-driven modelling, satellite observations, geospatial analytics, and hybrid physical–ML approaches to better understand hydroclimatic extremes and their impacts. We seek contributions that develop or apply machine learning and remote sensing tools to improve forecasting skill, characterize hazard interactions, map environmental vulnerabilities, and support decision-making. The issue will enable forward-looking platform that advances scientific insight, operational capabilities, and climate resilience.



To gather further insights into the application and integration of machine learning and remote sensing for hydroclimatic risks, we welcome studies focused on a variety of hydrological and environmental themes, but not limited to the following areas:



• Detection and prediction of extreme rainfall, droughts, heatwaves, and monsoon variability.

• Integration of diverse satellite platforms (optical, radar, multispectral, hyperspectral, LiDAR, thermal) with in-situ data.

• Development and validation of machine learning and hybrid physics–ML models for forecasting and characterization of hazards.

• Mapping of land–atmosphere interactions and hydrological impacts

• Assessment of water quality and ecosystem vulnerability to hydroclimatic extremes.

• Environmental risk mapping and support tools for decision-making in climate resilience.

• Case studies applying geospatial analytics to quantify land-use influences and cascading hazards.



We invite a wide array of manuscripts, including original research articles, methodological developments, data reports, reviews, and perspective pieces, especially those that bridge hydrology, climate science, environmental engineering, geospatial analytics, and risk assessment.