Process-guided deep learning water temperature predictions: 5c All lakes historical prediction data

Multiple modeling frameworks were used to predict daily temperatures at 0.5m depth intervals for a set of diverse lakes in the U.S. states of Minnesota and Wisconsin. Process-Based (PB) models were configured and calibrated with training data to reduce root-mean squared error. Uncalibrated models used default configurations (PB0; see Winslow et al. 2016 for details) and no parameters were adjusted according to model fit with observations. Deep Learning (DL) models were Long Short-Term Memory artificial recurrent neural network models which used training data to adjust model structure and weights for temperature predictions (Jia et al. 2019). Process-Guided Deep Learning (PGDL) models were DL models with an added physical constraint for energy conservation as a loss term. These models were pre-trained with uncalibrated Process-Based model outputs (PB0) before training on actual temperature observations. Zip files for each lake contain four files, one for each of PB, PB0, DL, and PGDL.

Data and Resources

Field Value
accessLevel public
bureauCode {010:00}
catalog_@context https://project-open-data.cio.gov/v1.1/schema/catalog.jsonld
catalog_conformsTo https://project-open-data.cio.gov/v1.1/schema
catalog_describedBy https://project-open-data.cio.gov/v1.1/schema/catalog.json
identifier db922a51-1426-4d26-a49c-1d15f84e56df
metadata_type geospatial
modified 2020-08-20
old-spatial {"type": "Polygon", "coordinates": [[[-94.2609062308, 42.5692312673], [-87.9475441739, 42.5692312673], [-87.9475441739, 48.6427837912], [-94.2609062308, 48.6427837912], [-94.2609062308, 42.5692312673]]]}
publisher Climate Adaptation Science Centers
resource-type Dataset
source_datajson_identifier true
source_hash f15387a96ffb88cccca38848753dfdfc06a31a47
source_schema_version 1.1
spatial {"type": "Polygon", "coordinates": [[[-94.2609062308, 42.5692312673], [-87.9475441739, 42.5692312673], [-87.9475441739, 48.6427837912], [-94.2609062308, 48.6427837912], [-94.2609062308, 42.5692312673]]]}
theme {geospatial}
Groups
  • AmeriGEOSS
  • National Provider
  • North America
Tags
  • amerigeo
  • amerigeoss
  • ckan
  • climate-change
  • deep-learning
  • geo
  • geoss
  • hybrid-modeling
  • machine-learning
  • modeling
  • national
  • north-america
  • reservoirs
  • temperate-lakes
  • temperature
  • thermal-profiles
  • united-states
  • us
  • usgs-5d915c8ee4b0c4f70d0ce520
  • water
isopen False
license_id notspecified
license_title License not specified
maintainer U.S. Geological Survey (Point of Contact)
maintainer_email jread@usgs.gov
metadata_created 2025-11-21T23:19:45.299915
metadata_modified 2025-11-21T23:19:45.299919
notes Multiple modeling frameworks were used to predict daily temperatures at 0.5m depth intervals for a set of diverse lakes in the U.S. states of Minnesota and Wisconsin. Process-Based (PB) models were configured and calibrated with training data to reduce root-mean squared error. Uncalibrated models used default configurations (PB0; see Winslow et al. 2016 for details) and no parameters were adjusted according to model fit with observations. Deep Learning (DL) models were Long Short-Term Memory artificial recurrent neural network models which used training data to adjust model structure and weights for temperature predictions (Jia et al. 2019). Process-Guided Deep Learning (PGDL) models were DL models with an added physical constraint for energy conservation as a loss term. These models were pre-trained with uncalibrated Process-Based model outputs (PB0) before training on actual temperature observations. Zip files for each lake contain four files, one for each of PB, PB0, DL, and PGDL.
num_resources 8
num_tags 20
title Process-guided deep learning water temperature predictions: 5c All lakes historical prediction data