{"accessLevel": "public", "bureauCode": ["010:12"], "contactPoint": {"@type": "vcard:Contact", "fn": "Courtney D Killian", "hasEmail": "mailto:ckillian@usgs.gov"}, "description": "This dataset represents spatial predictions of hydrochemical facies (HCF) across the conterminous United States (CONUS) using a Random Forest Classification model. Predictions were generated for the water table, the bottom of the potable drinking supplies (DS), and depths of 200 and 400 meters below the DS. Two prediction modes are provided: (1) probability surfaces, consisting of multi-band rasters where each band represents the probability of a specific hydrochemical facies, and (2) categorical surfaces, representing the most probable facies per 1-kilometer grid cell. The files are generated by the workflow available in 1) Workflow: Predicting groundwater hydrochemical facies in three dimensions across the conterminous United States with random forest classification. The final file is written to the './results' directory. \nSix hydrochemical facies were modeled including Calcium-Magnesium Bicarbonate (CaMg-HCO3), Calcium-Magnesium Sulfate (CaMg-SO4), Chloride (Cl), Mixed, Sodium-Potassium Bicarbonate (NaK-HCO3), and Sodium-Potassium Sulfate (NaK-SO4). For probability outputs, rasters include the most probable facies class as an index (integer codes 1\u20136) and the maximum probability value per grid cell, providing measures of classification confidence. All rasters are georeferenced and share a consistent spatial resolution and extent, enabling integration with other hydrogeologic and geospatial datasets.\nThis dataset supports quantitative and spatial analysis of groundwater chemistry patterns, uncertainty assessment, and depth-dependent facies transitions, offering a robust foundation for hydrogeologic modeling and decision-making.", "distribution": [{"@type": "dcat:Distribution", "accessURL": "https://doi.org/10.5066/P13ZYL2G", "description": "Landing page for access to the data", "format": "XML", "mediaType": "application/http", "title": "Digital Data"}, {"@type": "dcat:Distribution", "description": "The metadata original format", "downloadURL": "https://data.usgs.gov/datacatalog/metadata/USGS.698a0f41b66b012c9961ad72.xml", "format": "XML", "mediaType": "text/xml", "title": "Original Metadata"}], "identifier": "http://datainventory.doi.gov/id/dataset/USGS_698a0f41b66b012c9961ad72", "keyword": ["CONUS", "U.S. Geological Survey", "USA", "USGS", "USGS:698a0f41b66b012c9961ad72", "United States of America", "environment", "geoscientificInformation", "groundwater", "groundwater quality", "hydrogeology", "machine learning", "mathematical modeling", "random forest classification"], "modified": "2026-03-25T00:00:00Z", "publisher": {"@type": "org:Organization", "name": "U.S. Geological Survey"}, "spatial": "-134.1087, 19.1726, -59.5906, 58.5534", "theme": ["geospatial"], "title": "2) Results: Groundwater hydrochemical facies predictions in three dimensions from Random Forest Classification, conterminous United States"}