New satellite observations are used to improve our understanding of agricultural land on a global scale. Innovative artificial intelligence methods are employed to combine the available information. This improvement is evaluated using independent observations and varies depending on the characteristics of instruments and models.

We investigate the potential of drone-based thermal cameras in the Swiss Alps and found they can misread surface temperatures when the camera warms or cools. Lab and field checks showed some correction methods exaggerate extremes. Calibrating on site with ground sensors gave the best results as well as the implementation of a site-specific validation. This improves temperature maps needed to track melting snow and ice, and unstable ground in rapidly changing mountains.

Subsidence, or the sinking of land, threatens the U.S. Gulf Coast, where millions face growing flood risks from sea-level rise. Measuring the subsidence is essential for coastal planning and climate adaptation. Satellite radar (InSAR) is widely used to measure subsidence, but studies often disagree. Comparing two recent datasets, we found widespread disagreements except in dense urban areas. Decision makers should use coastal InSAR products with caution until these discrepancies are resolved.

Land and sea surface temperatures, collectively called skin temperature, are available from satellite measurements since few decades now and provide a holistic view of the Earth's climate system. Here we review 5 different datasets of Tskin to show that climate change signature is clear from the different datasets with notable exception, such as a significant cooling in India, parts of Africa and the Pacific.

Earth Observation archives now encompass multispectral imagery, yet producing analysis-ready Sentinel-2 time series remains a critical bottleneck. We present an open-source Python package that builds data cubes from Spatio-Temporal Asset Catalog catalogues and integrates cloud masking, co-registration, and super-resolution in one workflow. Validated on high-performance computing, it enables rapid, reproducible cube construction and updates, improving temporal consistency in a braided river.

Water hyacinth (WHY) are one of the most costly aquatic invasive species. Regular monitoring of WHY would the management of infested waterboddies. We present an algorithm to derive daily maps of WHY from satellites, featuring an interpolation scheme, which led to 55% more daily WHY maps of Hartbeespoortdam Reservoir. The algorithm is light-weight and can be run on a cloud-computing platform. This allows applications in (sub-)tropical regions with limited access to computing power and/or funding.

The approach was developed to reduce false flood signals in urban areas while keeping data and processing requirements low. By using only commonly available satellite bands, the method provides clearer identification of water and wet surfaces.

The shrinking of glacier frees large areas worldwide in alpine environment which are progressively colonized by the vegetation. Monitoring this process often relies on in situ botanical observations which are challenging to obtain. Here, we quantify accurately the uncertainty of the vegetation cover measured by optical images from Unmanned Aerial Vehicles. We find that, despite biases, this increasingly used technology extend and complete the data collected during such field campaign.

This study reveals how orbital variations drive ENSO and climate. Introducing the index of geo-orbital oscillation (IRGON) in an integrated method of evidence, results show that annual changes in Earth–Sun distance create massive energy shifts in the hemispheres. Satellite radiative data confirms these oscillations trigger extremes, including the 2026 wildfires and the previous triple-dip La Niña. A relative geoenergetic equilibrium (RGE) hypothesis is proposed to address the complexities.

Monitoring sea ice is important for understanding climate change and supporting safe travel in the Arctic, but creating detailed maps usually requires time-consuming manual work. In this study, we trained a computer model to learn from large amounts of unlabelled satellite images before using only a small number of labelled examples. This approach produced accurate ice maps while greatly reducing the need for manual labelling, making large-scale monitoring more practical.