Ma et al. 2023 random forest water table depth and uncertainty

Water table depth estimates provided by a random forest model trained on historical USGS observations and Fan et al. 2013 water table depth dataset. This dataset includes long-term mean water table estimates and uncertainty at 1km resolution for the ParFlow-CONUS1 domain.

Data and Resources

Additional Info

Field Value
Last Updated July 9, 2025, 08:00 (UTC)
Created July 9, 2025, 08:00 (UTC)
accessRights Users must create a HydroFrame API account and register their PIN before using the hf_hydrodata package. If it's your first time using this package you will need to sign up for a HydroFrame account on the HydroFrame Signup Page (https://hydrogen.princeton.edu/signup; note: this only needs to be done once). Visit our HydroFrame PIN Page (https://hydrogen.princeton.edu/pin) to create a 4-digit PIN. After creating your PIN, you must register that PIN on the machine that you intend to use.
columnDataDict Access column data dictionary at: https://hf-hydrodata.readthedocs.io/en/latest/gen_ma_2023.html
creationMethod Long-term mean water table depth estimates were obtained using the median of tree outputs from the trained random forest model. The uncertainty was assessed based on the coefficient of variation of the tree outputs from the random forest model, which was calculated as the standard deviation of the tree outputs divided by their mean.
creatorEmail hydrogen-support@princeton.edu
creatorName HydroFrame Team
creatorWebsite https://hydroframe.org/
dataAuthType public
dataBbox [-121.47939483437318, 31.651836025255015, -76.09875469594509, 50.49802132270979]
dataProvenance [{"date":"2023-10-05","name":"Creation and publication of dataset"}]
dataType NumPy array
datasetPageUrl https://hf-hydrodata.readthedocs.io/en/latest/gen_ma_2023.html
datasetVersion v1.0
docsURL https://hf-hydrodata.readthedocs.io/en/latest/gen_ma_2023.html
doi https://doi.org/10.1111/gwat.13362
issueDate 2023-10-05
lastUpdateDate 2023-10-05
license other
otherLicense Copyright © 2024 The Trustees of Princeton University and The Arizona Board of Regents on behalf of The University of Arizona, College of Science Hydrology & Atmospheric Sciences. All rights reserved. hf_hydrodata was created by William M. Hasling, Laura Condon, Reed Maxwell, George Artavanis, Will Lytle, Amy M. Johnson, Amy C. Defnet. It is licensed under the terms of the MIT license.
pocEmail hydrogen-support@princeton.edu
pocName HydroFrame Team
pocWebsite https://hydroframe.org/
publisherEmail hydrogen-support@princeton.edu
publisherName HydroFrame Team
publisherWebsite https://hydroframe.org/
purpose Machine-learning-based water table depth estimates for ParFlow-CONUS1 domain.
spatialCov Grid: conus1 Spatial Resolution: 1000 meters XY Grid Spatial Extent: 3342 x 1888 LatLon Spatial Exent: -121.47939483437318, 31.651836025255015, -76.09875469594509, 50.49802132270979 Origin (meters): -1885055.4995, -604957.0654 Projection: +proj=lcc +lat_1=33 +lat_2=45 +lon_0=-96.0 +lat_0=39 +a=6378137.0 +b=6356752.31
spatialCovFormat text
spatialRes 1000 meter
status submitted
theme ["groundwater"]
uploadType dataset