Ecological Forecasting

Science Center Objects

The Ecological Forecasting research described below is conducted and managed under the USGS Applied Landscape Ecology and Remote Sensing project and partners.

Decision Support for Wetland and Wildlife Management

Sacramento National Wildlife Refuge, CA

Sacramento National Wildlife Refuge, CA (Credit: Kristin Byrd, USGS. Public domain.)

In partnership with Point Blue Conservation Science (Lead PI Matt Reiter), we are researching dynamic cropland and wetland habitats across the Central Valley of California so that water allocations may optimize multiple ecosystem benefits. Working with The Nature Conservancy and U.S. Fish and Wildlife Service, this project will develop annual forecasts and long-term projections of the spatial and temporal availability of wetland habitats, dependent species, and groundwater recharge to prioritize and strategically place new wetland habitat throughout the Central Valley. This work is conducted through 1) integration of a time series of Landsat-derived maps of habitat quality and extent and species distribution models, and 2) integration of Landsat-derived maps, the USGS Central Valley Water Evaluation and Planning model and the USGS LUCAS land change model, to produce spatially explicit projections of potential wetland and cropland habitats. The USGS Basin Characterization Model is used to determine where enhancement of wildlife habitat and groundwater recharge can co-occur. This project is funded by the NASA Applied Sciences Ecological Forecasting Program, the USGS Land Change Science Program, and the Bay-Delta Priority Ecosystem Sciences Program.

aerial image

AVIRIS image of Rush Ranch, Suisun Marsh, CA (Credit: Kristin Byrd, USGS. Public domain.)

Forecasting Tidal Marsh Habitat Change across Landscapes

We developed a feasibility study to integrate remote sensing estimates of peak biomass and suspended sediment concentration (SSC) to the Marsh Equilibrium Model (MEM), a marsh elevation forecasting model, to project how coastal marsh habitat and dependent species respond to sea level rise (SLR) at the landscape scale. With remote sensing inputs to MEM, scientists will be able to generate regional maps of coastal vulnerability and habitat suitability for a set of SLR scenarios. This feasibility study took place at Rush Ranch, a brackish marsh in Suisun Bay, CA. We compared Landsat 8, World View-2 and hyperspectral sensors to determine the best data source for mapping biomass and SSC. We explored the feasibility of a regional MEM-based forecasting tool for decision support through a substantial stakeholder engagement program with our end user partners, the National Estuarine Research Reserve System (NERRS). This project was funded by the NASA Applied Sciences Ecological Forecasting Program and the USGS Land Change Science Program.