Location: Rangeland Resources & Systems Research
Project Number: 3012-12610-001-007-S
Project Type: Non-Assistance Cooperative Agreement
Start Date: Sep 1, 2024
End Date: Aug 31, 2028
Objective:
Objective 1) Provide ranchers and rangeland managers in the Great Plains and Southwest biweekly forecasts (April 1 to August 31) of rangeland plant production using the Grass-Cast model for the 2025 to 2028 growing seasons.
Objective 2) Attempt to expand the Grass-Cast model to a new geographic area (e.g., a portion of the Great Plains tall-grass prairie region) to the extent available data allow.
Objective 3) Add any new ANPP data sets to the existing Grass-Cast database within Ag Data Commons (subject to permissions from original sources of data).
Objective 4) Publish papers which describe the correlations of rangeland plant production to climatic variables and rangeland production simulated with the Grass-Cast model for the Southwest US and Great Plains.
Objective 5) Explore the potential to improve Grass-Cast performance by using MODIS NDVI data (2000 to 2024), which will generate revised correlations of annual variation in NDVI to climatic variables (e.g., growing season precipitation and actual evapotranspiration) for the Southwest and Great Plains, as well as any new geographic area (e.g., a portion of the tall grass prairie region).
Approach:
For objective 1), the Cooperator and ARS will continue to collaboratively refine and improve Grass-Cast to estimate aboveground net primary productivity (ANPP) and support climate informed decision-making that can improve environmental sustainability in rangelands. The Cooperator will produce biweekly forecasts of ANPP for the Southwest and Great Plains regions for the 2025, 2026, 2027 and 2028 growing seasons. These biweekly forecasts (April 1 to August 31) are submitted to ARS for publication on the Grass-Cast website. Emphasis will remain on the Great Plains and Southwest.
But for objective 2), the Cooperator and ARS will explore the potential to expand to another region in the United States (e.g., the tall grass prairie ecosystem) to the extent that available data allow. This will strengthen and expand the utility of this decision tool for ranchers and land managers who use flexible stocking strategies and make climate-informed decisions to maximize beef production and economic resilience in response to climatic variability in rangelands within the footprint of Grass-Cast. These efforts will integrate current weather to date, seasonal projections of weather/climate, modeling, remote sensing, and biophysical relationships to produce high resolution (6 x 6 mile grids) maps of the predicted aboveground biomass every two weeks during the grazing season. The Cooperator will run simulations, resulting in updated maps every two weeks from April through August, and will incorporate suggested feedback from end users to improve user-friendly features of the decision tool.
For objective 3), the Cooperator will draft three peer-reviewed Grass-Cast manuscripts which describe: i) the Grass-Cast model’s past expansion to the Southwest U.S., ii) pilot efforts to expand Grass-Cast into a portion of the Great Plains tall grass prairie region, and iii)
correlations of climatic variables to the ANPP data sets underlying Grass-Cast (> 600 sites).
For objective 4), the Cooperators will add any eligible new information from the Grass-Cast data sets, e.g., annual variation in ANPP for > 80 sites; annual values of PPT and AET; cumulative growing season NDVI (MODIS and AVHRR data), into the existing database on Ag Data
Commons (to the extent permissions from original data sources allow).
For objective 5) Use the observed rangeland county-level growing season MODIS NDVI (2000 –2024) to calculate correlations of growing season NDVI to climatic variables (PPT and AET). These new correlations may eventually replace the current Grass-cast regressions that used the 1982-2015 AVHRR NDVI, if they perform satisfactorily. These new correlations of NDVI to climatic variables will include the current Grass-Cast regions as well as data from any expanded geography (e.g., portions of the tall grass prairie region) and may be used to develop a revised version of the Grass-Cast model (if they perform satisfactorily) during the last two years of the project.
