Classifying tree species in satellite images is an important task for environmental monitoring and forest management. Here we present a dataset containing Sentinel-2 satellite pixel time series of individual trees intended for training machine learning models. The dataset was created by merging information from the German National Forest Inventory in 2012 with satellite data. It sparsely covers the whole of Germany for the years 2015 to 2022 and comprises 48 species and 3 species groups.

Agricultural, fishery, forestry and agro-processing activities produce large quantities of residues, by-products and waste materials every year. Here, we present a global organic matter database (OMD), the first of its kind, consolidating estimates of residues and by-products potentially available for use in a circular bio-economy. It also provides definitions, typologies and methods to aid consistent classification, estimation and reporting of the various residues and by-products.

Nitrogen is crucial in ocean food webs, but only some microbes can fix N₂ gas into a bioavailable form. Most are known only by their nifH gene sequence. We created a software workflow for nifH data and ran it on 944 ocean samples, producing a database (DB) that captures the global diversity of N₂-fixing marine microbes and the environmental factors that influence them. The workflow and DB can standardize analyses of past and future nifH datasets to enable insights into marine communities.

Sea ice forms a thin boundary between the ocean and the atmosphere, with complex, crust-like dynamics and ever-changing networks of sea ice leads and ridges. Statistics of these dynamical features are often used to evaluate sea ice models. Here, we present a new pan-Arctic dataset of sea ice deformations derived from satellite imagery, from 1 September 2017 to 31 August 2023. We discuss the dataset coverage and some limitations associated with uncertainties in the computed values.

This research presents a comprehensive synthetic dataset of bio-optical properties and radiometric quantities in the optical domain, resolved for all sun-view angular combinations, from ultraviolet to visible light, that provide aid in the development of satellite algorithms, including directional problems. The dataset will significantly enhance research on light behavior in water and support future hyperspectral missions. It has been made publicly available on Zenodo.

We introduce CAMELS-IND (Catchment Attributes and MEteorology for Large-sample Studies – India), which provides daily hydrometeorological time series and static catchment attributes representing the location, topography, climate, hydrological signatures, land use, land cover, soil, geology, and anthropogenic influences for 472 catchments in Peninsular India to foster large-sample hydrological studies in India and promote the inclusion of Indian catchments in global hydrological research.

We present a compilation of water optical properties and phytoplankton pigments from the surface of the Atlantic Ocean collected during nine cruises between 2009 and 2019. We derive continuous Chlorophyll a concentrations (a biomass proxy) from water absorption. We then illustrate geographical variations and relationships for water optical properties, Chlorophyll a, and other pigments. The dataset will be useful to researchers in ocean optics, remote sensing, ecology, and biogeochemistry.

In this study, we developed the second version of China's high-resolution soil information grid using legacy soil samples and advanced machine learning. This version predicts over 20 soil properties at six depths, providing accurate soil variation maps across China. It outperforms previous versions and global products, offering valuable data for hydrological and ecological analyses and Earth system modelling, enhancing our understanding of soil roles in environmental processes.

In the Holocene, deltas and coastal plains developed due to relative sea level rise (RSLR). Past coastal and inland water levels are preserved in geological indicators, like basal peats. We present a dataset of 712 Holocene water level indicators from the Dutch coastal plain, relevant for studying RSLR and regional subsidence, compiled in HOLSEA workbook format. Our new, internally consistent, expanded documentation encourages multiple data uses and to report RSLR uncertainties transparently.

We present the first comprehensive map of offshore lighting derived from low-light imaging satellite data. The empty sea provides a dark and uniform canvas upon which light detections can be aggregated for extended periods to reveal human lighting structures. The form of the structures only becomes apparent when data from 1 or more years are accumulated. Identifiable structures include fishing grounds, platforms, gas flares, anchorages, and transportation routes.

We propose three strategies for field seismic data curation, knowledge-guided synthesization, and generative adversarial network (GAN)-based generation to construct a massive-scale, feature-rich, and high-realism benchmark dataset of seismic facies and evaluate its effectiveness in training a deep-learning model for automatic seismic facies classification.

GRACE and GRACE-FO are unique observing systems to quantify mass changes at the Earth’s surface from space. Time series of these mass changes are of high value for various applications, e.g., in hydrology, glaciology, and oceanography. GravIS (Gravity Information Service) provides easy access to user-friendly, regularly updated mass anomaly products. The portal visualizes and describes these data, aiming to highlight their significance for understanding changes in the climate system.

We present atmospheric and oceanic data from the Atlantic References and Convection ship campaign with the Maria S. Merian from Mindelo to Punta Arenas observed with the integrated ship sensors; humidity and temperature profiler; ceilometer; aerosol instruments (Calitoo, Microtops, and DustTrak); radiosondes; uncrewed aircraft vehicles; and conductivity, temperature, and depth scans. The data include three complete profiles of the Intertropical Convergence Zone and a storm in the South Atlantic.

We created a 10 m resolution rice distribution map for East Asia in 2023 (EARice10), achieving an overall accuracy (OA) of 90.48 % on validation samples. EARice10 shows strong consistency with statistical data (coefficient of determination, R²: 0.94–0.98) and existing datasets (R²: 0.79–0.98). It is the most up-to-date map, covering the four major rice-producing countries in East Asia at 10 m resolution.

Long-term observation sites have been established in six Austrian locations, covering major ecosystem types such as forests, grasslands, and wetlands. The purpose of these observations is to measure baselines for assessing the impacts of extreme climate events on the carbon cycle. The collected datasets include meteorological variables, soil temperature and moisture, carbon dioxide fluxes, and tree stem growth in forests at a resolution of 15–60 min between 2019 and 2021.

In this study, we present a dataset for the Dischma catchment in eastern Switzerland, which represents a typical high-alpine watershed in the European Alps. Accurate monitoring and reliable forecasting of snow and water resources in such basins are crucial for a wide range of applications. Our dataset is valuable for improving physics-based snow, land surface, and hydrological models, with potential applications in similar high-alpine catchments.

The continuous uptake of atmospheric CO₂ by the ocean leads to decreasing seawater pH, which is an ongoing threat to the marine ecosystem. This pH change has been globally documented in the surface ocean, but information is limited below the surface. Here, we present a monthly 1° gridded product of global seawater pH based on a machine learning method and real pH observations. The pH product covers the years from 1992 to 2020 and depths from 0 to 2000 m.

Our study introduces a Landsat-based data cube simplifying access to detailed environmental data across Europe from 2000 to 2022, covering vegetation, water, soil, and crops. Our experiments demonstrate its effectiveness in developing environmental models and maps. Tailored feature selection is crucial for its effective use in environmental modeling. It aims to support comprehensive environmental monitoring and analysis, helping researchers and policy-makers in managing environmental resources.

This study developed a new approach to long-time continuous annual vegetation mapping from remote sensing imagery, and mapped the vegetation of the Qinghai–Tibet Plateau (QTP) from 2000 to 2022 using the MOD09A1 product. The overall accuracy of continuous annual QTP vegetation mapping reached 83.3%, with the reference annual 2020 data reaching an accuracy of 83.3% and a kappa coefficient of 0.82. The study supports the use of remote sensing data to mapping long-time continuous annual vegetation.

Bivalves such as mussels and oysters reflect the quality of the environment by filtering ambient water. We measured carbon and nitrogen chemical composition in bivalve tissues from 33 sites along French coastlines sampled since the 1980s. Thanks to such time series, this dataset allows us to track how marine species record changing climate, physical–chemical environment, and organic matter cycles and provide precious information on the coastal ecosystem response to global change.