Climate change has been shown to influence lake temperatures in different ways. To better understand the diversity of lake responses to climate change and give managers tools to manage individual lakes, we focused on improving prediction accuracy for daily water temperature profiles in 68 lakes in Minnesota and Wisconsin during 1980-2018.
The data are organized into these items:
- Spatial data - One shapefile of polygons for all 68 lakes in this study (.shp, .shx, .dbf, and .prj files)
- Model configurations - Model parameters and metadata used to configure models (1 JSON file, with metadata for each of 68 lakes, indexed by "site_id")
- Model inputs - Data formatted as model inputs for predicting temperature
- Training data - Data used to train or calibrate predictive models
- a. Lake Mendota training data - Tables with 1 row per date and depth, with the corresponding observed water temperature (3 comma-delimited files)
- b. Sparkling Lake training data - Tables with 1 row per date and depth, with the corresponding observed water temperature (3 comma-delimited files)
- c. Historical training data for 68 lakes - Tables with 1 row per date, depth, and site_id, with the corresponding observed water temperature (1 comma-delimited file)
- Prediction data - Predictions from PGDL, DL, and PB models
- a. Lake Mendota predictions - Tables with 1 row per date, a column for predicted temperature at each depth, and experiment metadata (10 comma-delimited files)
- b. Sparkling Lake predictions - Tables with 1 row per date, a column for predicted temperature at each depth, and experiment metadata (10 comma-delimited files)
- c. Historical predictions for 68 lakes - Tables with 1 row per date and depth, with the corresponding observed water temperature (4 comma-delimited files for each lake compressed into 68 zip files)
- Model evaluation - test data and overall assessment of model performance
- a. Lake Mendota evaluation - Tables with 1 row per date, a column for predicted temperature at each depth, and experiment metadata (3 comma-delimited files)
- b. Sparkling Lake evaluation - Tables with 1 row per date, a column for predicted temperature at each depth, and experiment metadata (2 comma-delimited files)
- c. Historical evaluation for 68 lakes - Tables with 1 row per date and depth, with the corresponding observed water temperature (4 comma-delimited files for each lake compressed into 68 zip files)
This research was funded by the Department of the Interior Northeast and North Central Climate Adaptation Science Centers, a Midwest Glacial Lakes Fish Habitat Partnership grant through F&WS, an NSF Expedition in Computing Grant 1029711 to the University of Minnesota, a postdoctoral fellowship awarded to J Zwart under NSF EAR-PF-1725386, as well as a seed grant from the Digital Technology Center at the University of Minnesota. Access to computing facilities was provided by the University of Minnesota Supercomputing Institute and USGS Advanced Research Computing, USGS Yeti Supercomputer (https://doi.org/10.5066/F7D798MJ). We thank North Temperate Lakes Long-Term Ecological Research (NSF DEB-1440297) and Global Lake Ecological Observatory Network (NSF #1702991).
