Climate change is escalating the risks related to hydro-meteorological extremes. This preface introduces a special issue originating from a European Geosciences Union (EGU) session. It highlights the challenges posed by these extremes, ranging from hazard assessment to mitigation strategies, and covers both water excess events like floods, landslides, and coastal hazards and water deficit events such as droughts and fire weather. The collection aims to advance understanding, improve resilience, and inform policy-making.

This study integrated the high precision remote sensing data and ground observations to produce 30-meter spatial resolution maps of four key ecosystem services in China – net primary productivity, soil conservation, sandstorm prevention, and water yield – for the years 2000, 2010, and 2020. Through comparison and cross-validation with other data sources, this dataset has been proven to have significant advantages in accuracy and applicability.

The Unsolved Problems in Hydrology (UPH) initiative has emphasized the need to establish networks of multi-decadal hydrological observatories to tackle catchment-scale challenges on a global scale. This opinion paper provocatively discusses two endmembers of possible future hydrological observatory (HO) networks for a given hypothesized community budget: a comprehensive set of moderately instrumented observatories or, alternatively, a small number of highly instrumented supersites.

Machine learning models outperform simple mechanistic models in predicting pollinator visitation rates. We use deep learning to infer rules from land cover maps to estimate pollination services globally. Results suggest deep learning can improve predictions by identifying complex patterns in landscape composition, especially in data-rich but knowledge-poor areas. The challenge is to make deep learning algorithms more interpretable so that experts can validate prediction rules for pollination.

This work includes vegetation (VCI) and water content index (WCI) series from two semiarid rangeland areas in Spain. Based on then, a Z-score for both was calculated to use it as an anomaly index. In this way, we associated negative anomalies with drought episodes. Then, we study the relations of these negative anomalies to see if it is possible to use WCI as an alarm of agronomic drought (VCI negative anomaly). The description of the behaviour of both areas and their comparison are made.