Improving Model Predictions to Support the 2029 Louisiana Coastal Master Plan
USGS and partners will identify the causes for the lack of predictive power in some models included in the 2023 Louisiana Coastal Master Plan, and suggest ways to improve the models in future Louisiana Coastal Master Plan modeling exercises.
The Science Issue and Relevance: Through the Louisiana Coastal Master Plan, the Louisiana Coastal Protection and Restoration Authority (CPRA) aims to preserve the state’s coastal ecosystems, natural resources, and economic development while considering future environmental change. Initiated in 2012, the Louisiana Coastal Master Plan uses a suite of modeling tools to explore the potential impacts of restoration and management projects along the Louisiana coast and is updated every six years to incorporate new knowledge into the planning process. During the processing of the 2023 Louisiana Coastal Master Plan, modeling exercises revealed that the predictive power of the statistical habitat suitability models was lacking for some fish and shellfish species. In collaboration with CPRA, USGS and partners aim to identify the causes for the lack of predictive power in some models and suggest ways to improve the models in future Louisiana Coastal Master Plan modeling exercises.
Methodology for Addressing the Issue: The current habitat suitability models for fish, shrimp, and blue crab used in the Louisiana Coastal Master Plan utilize both generalized linear models and generalized additive models to predict species responses to temperature, salinity, and vegetation on the landscape. In this project, USGS and partners will explore the input data to test for violations of assumptions or signs of missing important predictor variables. For example, temperature trends on the landscape may be trending warmer and this may be changing the species’ responses to temperatures across time as the species adapts to a new climate. The statistical models currently used assume a constant relationship between the species and temperature, so it is important to address this trend in the model if it exists within the monitoring data used to build the models. If issues are discovered, ways to improve the models will be identified and suggested to CPRA for incorporation into the 2029 Plan.
Future Steps: The suggested model improvements will allow for more accurate predictions of future ecosystem changes and provide a mechanism for Louisiana’s decision makers to prioritize and sequence restoration and projection projects using best available science. The modeling tools will be used to predict project outcomes and compare modeled results under different environmental scenarios to identify the best suites of projects that achieve the state’s goals.
Model Improvements for Louisiana’s 2023 Coastal Master Plan
Morphology Modeling in Support of the Louisiana Coastal Master Plan
USGS and partners will identify the causes for the lack of predictive power in some models included in the 2023 Louisiana Coastal Master Plan, and suggest ways to improve the models in future Louisiana Coastal Master Plan modeling exercises.
The Science Issue and Relevance: Through the Louisiana Coastal Master Plan, the Louisiana Coastal Protection and Restoration Authority (CPRA) aims to preserve the state’s coastal ecosystems, natural resources, and economic development while considering future environmental change. Initiated in 2012, the Louisiana Coastal Master Plan uses a suite of modeling tools to explore the potential impacts of restoration and management projects along the Louisiana coast and is updated every six years to incorporate new knowledge into the planning process. During the processing of the 2023 Louisiana Coastal Master Plan, modeling exercises revealed that the predictive power of the statistical habitat suitability models was lacking for some fish and shellfish species. In collaboration with CPRA, USGS and partners aim to identify the causes for the lack of predictive power in some models and suggest ways to improve the models in future Louisiana Coastal Master Plan modeling exercises.
Methodology for Addressing the Issue: The current habitat suitability models for fish, shrimp, and blue crab used in the Louisiana Coastal Master Plan utilize both generalized linear models and generalized additive models to predict species responses to temperature, salinity, and vegetation on the landscape. In this project, USGS and partners will explore the input data to test for violations of assumptions or signs of missing important predictor variables. For example, temperature trends on the landscape may be trending warmer and this may be changing the species’ responses to temperatures across time as the species adapts to a new climate. The statistical models currently used assume a constant relationship between the species and temperature, so it is important to address this trend in the model if it exists within the monitoring data used to build the models. If issues are discovered, ways to improve the models will be identified and suggested to CPRA for incorporation into the 2029 Plan.
Future Steps: The suggested model improvements will allow for more accurate predictions of future ecosystem changes and provide a mechanism for Louisiana’s decision makers to prioritize and sequence restoration and projection projects using best available science. The modeling tools will be used to predict project outcomes and compare modeled results under different environmental scenarios to identify the best suites of projects that achieve the state’s goals.