CAMELSH/ERA5-Land (1): A Large-Sample Hourly Hydrometeorological Dataset and Attributes at Watershed-Scale for Contiguous United States

CAMELSH: ERA5-Land Hourly Forcing Dataset is an extension of the CAMELSH (Catchment Attributes and Hourly HydroMeteorology for Large-Sample Studies) framework, representing the first hourly large-sample hydrometeorological dataset developed for the contiguous United States (CONUS). This dataset comprises hourly meteorological variables derived from the ERA5-Land reanalysis product, aggregated across 9,008 hydrologic basins. The current version spans 46 years (1980–2025) and provides hourly data suitable for a wide range of hydrological and climatological applications, including long-term trend analyses, process-based modeling, model evaluation, and climate sensitivity assessments. The dataset includes 11 core atmospheric and land-surface variables, such as dewpoint temperature, 2-meter air temperature, soil moisture (four layers), net solar and thermal radiation, 10-meter wind components, surface pressure, total precipitation, snow water equivalent, and potential evaporation. These variables enable the derivation of key hydroclimatic indices related to precipitation intensity and seasonality, evaporative regimes, and snow dynamics. Meteorological variables were sourced from the ERA5-Land dataset via Google Earth Engine, with hourly means computed for all grid cells either fully contained within or intersecting each watershed boundary. The dataset is fully accompanied by spatial and temporal metadata and adheres to standardized units, facilitating seamless integration with CAMELS-based catchment attributes. In addition to the main NetCDF dataset for each year, we also provide the CAMELSH_shapefile.7z, a zipped shapefile that defines the boundaries of the 9,008 hydrologic basins included in CAMELSH. This shapefile enables spatial referencing and integration with other geospatial datasets and supports mapping, visualization, and basin-level analyses. Due to the 50 GB storage limitation per link on Zenodo, the dataset is divided into three separate repositories. Each repository contains data for a specific time period: (1) 1980–1995 (https://doi.org/10.5281/zenodo.15264376), (2) 1996–2010 (https://doi.org/10.5281/zenodo.15264814), and (3) 2011–2025 (https://doi.org/10.5281/zenodo.15264840). This dataset is developed and maintained by the HYDROWIT Lab at the University of Michigan. For inquiries or technical support, please contact Taeho Kim ([e-mail]). The dataset is distributed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. For further methodological details and background, users are encouraged to consult the following references: [1] Tran, V. N., Kim, T., Xu, D., Nguyen, T., & Ivanov, V. (2025). CAMELSH: A Large-Sample Hourly Hydrometeorological Dataset and Attributes at Watershed-Scale for CONUS. Scientific Data [In review]. [2] Tran, V. N. (2025). CAMELSH: A Large-Sample Hourly Hydrometeorological Dataset and Attributes at Watershed-Scale for Contiguous United States [Data set]. Zenodo. https://doi.org/10.5281/zenodo.15066778