“Hyperscale” Modeling to Understand and Predict Temperature Changes in Midwest Lakes
Many inland waters across the United States are experiencing warming water temperatures. The impacts of this warming on aquatic ecosystems are significant in many areas, causing problems for fisheries management, as many economically and ecologically important fish species are experiencing range shifts and population declines. Fisheries and natural resource managers need timely and usable data and tools in order to understand and predict changes to lakes and their biota.
A previous Northeast CSC-funded project modeled lake temperatures to help state agencies in the Midwest understand trends in walleye and largemouth bass populations and predict lake-specific fish populations under future climate scenarios. These results have been extremely valuable for decisions and management strategies at the state scale, and this new project will expand these efforts and will focus on lake temperature products at the three spatial scales of most interest to state agency stakeholders: lake, state, and region (for Minnesota, Wisconsin, and Michigan).
The project researchers will use a “hyperscale” modeling approach that will build upon the multi-state modeling framework developed in the earlier project to increase model accuracy for high priority managed lakes. This approach will use all observations of temperature that exist for every study lake in the region and use machine learning techniques to uncover biases in models used for lakes with many observations. The project will generate an improved assessment of aquatic habitat for lake fisheries, and will provide estimates of contemporary thermal habitats to be used by state partners to estimate the distribution and abundance of ecologically and economically important fish species. Deliverables for this project include: 1) hind-casted lake temperature profiles (1979-present), 2) summary outputs from the thermal models, and 3) individually tuned lake models for managers to use for testing and predicting future conditions under different climate change scenarios.
- Source: USGS Sciencebase (id: 598de688e4b09fa1cb146372)
Many inland waters across the United States are experiencing warming water temperatures. The impacts of this warming on aquatic ecosystems are significant in many areas, causing problems for fisheries management, as many economically and ecologically important fish species are experiencing range shifts and population declines. Fisheries and natural resource managers need timely and usable data and tools in order to understand and predict changes to lakes and their biota.
A previous Northeast CSC-funded project modeled lake temperatures to help state agencies in the Midwest understand trends in walleye and largemouth bass populations and predict lake-specific fish populations under future climate scenarios. These results have been extremely valuable for decisions and management strategies at the state scale, and this new project will expand these efforts and will focus on lake temperature products at the three spatial scales of most interest to state agency stakeholders: lake, state, and region (for Minnesota, Wisconsin, and Michigan).
The project researchers will use a “hyperscale” modeling approach that will build upon the multi-state modeling framework developed in the earlier project to increase model accuracy for high priority managed lakes. This approach will use all observations of temperature that exist for every study lake in the region and use machine learning techniques to uncover biases in models used for lakes with many observations. The project will generate an improved assessment of aquatic habitat for lake fisheries, and will provide estimates of contemporary thermal habitats to be used by state partners to estimate the distribution and abundance of ecologically and economically important fish species. Deliverables for this project include: 1) hind-casted lake temperature profiles (1979-present), 2) summary outputs from the thermal models, and 3) individually tuned lake models for managers to use for testing and predicting future conditions under different climate change scenarios.
- Source: USGS Sciencebase (id: 598de688e4b09fa1cb146372)