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.

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.

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.

This dataset includes "test data" compiled water temperature data from an instrumented buoy on Lake Mendota, WI and discrete (manually sampled) water temperature records from North Temperate Lakes Long-TERM Ecological Research Program (NTL-LTER; https://lter.limnology.wisc.edu/). The buoy is supported by both the Global Lake Ecological Observatory Network (gleon.org) and the NTL-LTER. The dataset also includes Lake Mendota model erformance as measured as root-mean squared errors relative to temperature observations during the test period. This dataset is part of a larger data release of lake temperature model inputs and outputs for 68 lakes in the U.S. states of Minnesota and Wisconsin (http://dx.doi.org/10.5066/P9AQPIVD).

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.

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.

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.

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.

This dataset includes "test data" compiled water temperature data from an instrumented buoy on Lake Mendota, WI and discrete (manually sampled) water temperature records from North Temperate Lakes Long-TERM Ecological Research Program (NTL-LTER; https://lter.limnology.wisc.edu/). The buoy is supported by both the Global Lake Ecological Observatory Network (gleon.org) and the NTL-LTER. The dataset also includes Lake Mendota model erformance as measured as root-mean squared errors relative to temperature observations during the test period. This dataset is part of a larger data release of lake temperature model inputs and outputs for 68 lakes in the U.S. states of Minnesota and Wisconsin (http://dx.doi.org/10.5066/P9AQPIVD).

This dataset includes "test data" compiled water temperature data from an instrumented buoy on Lake Mendota, WI and discrete (manually sampled) water temperature records from North Temperate Lakes Long-TERM Ecological Research Program (NTL-LTER; https://lter.limnology.wisc.edu/). The buoy is supported by both the Global Lake Ecological Observatory Network (gleon.org) and the NTL-LTER. The dataset also includes Lake Mendota model erformance as measured as root-mean squared errors relative to temperature observations during the test period. This dataset is part of a larger data release of lake temperature model inputs and outputs for 68 lakes in the U.S. states of Minnesota and Wisconsin (http://dx.doi.org/10.5066/P9AQPIVD).