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

Description

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.

Resources

Name Format Description Link
0 https://doi.org/10.5066/F7D798MJ
0 https://doi.org/10.5194/gmd-12-473-2019
0 http://dx.doi.org/10.5066/F7DV1H10
0 http://dx.doi.org/10.1029/2019WR024922
33 1810.02880.pdf https://arxiv.org/pdf/1810.02880.pdf
55 The metadata original source https://data.doi.gov/harvest/object/64cbde31-a675-4e95-ad37-1ff3a25d89e3
0 http://dx.doi.org/10.5066/P9AQPIVD

Tags

  • reservoirs
  • hybrid-modeling
  • modeling
  • temperate-lakes
  • deep-learning
  • united-states
  • usgs-5d915c8ee4b0c4f70d0ce520
  • thermal-profiles
  • us
  • climate-change
  • water
  • temperature
  • machine-learning

Topics

Categories