Grass-Cast indicates for ranchers and rangeland managers what productivity is likely to be in the upcoming growing season relative to their own county’s 34-year history. Ranchers and rangeland managers will need to combine the forecast information with their knowledge of local soils, plant communities, topography, and other conditions as part of their decision-making process.
Grass-Cast also provides ranchers with a view of rangeland productivity in the broader region to assist in larger-scale decision making and to determine where grazing resources might be more plentiful if their own region is at risk from drought.
Grass-Cast cannot tell the difference between desirable forage species and undesirable forage species, therefore it is important for producers to know what proportion of a pasture is occupied by weeds and how well those weeds respond to rain (or lack of rain) compared to the desirable species. Producers should monitor these different vegetation types to see if one is responding to the weather better than the other and adjust Grass-Cast’s productivity estimates accordingly.
Grass-Cast also does not directly account for local management practices, such as grazing intensity in previous years. Producers should therefore adjust Grass-Cast’s estimates accordingly.
Grass-Cast is a collaboration among the USDA Agricultural Research Service (ARS), the USDA Natural Resources Conservation Service (NRCS), the National Drought Mitigation Center (NDMC), Colorado State University, the University of Arizona, and the USDA Northern Plains Climate Hub (NPCH).
Funding for this project came from the USDA ARS and NRCS, and the National Drought Mitigation Center.
1. Dr. Bill Parton (lead researcher at Colorado State University) knows from existing studies that precipitation data can be used to estimate actual evapotranspiration (iAET), which correlates well with iNDVI (a vegetative growth index). In turn, iNDVI is a good predictor of above-ground net primary productivity (ANPP, a measure of plant biomass).
2. Parton’s research team quantified the relationships among these three variables by estimating statistical correlations from a historical dataset (1982-2012) for 10 sites across the Great Plains region.
3. To actually forecast grassland productivity in a given spring, Parton’s team combines observed daily precipitation (maximum and minimum) with seasonal precipitation outlooks from NOAA’s National Weather Service–Climate Prediction Center to predict iAET for the growing season.
4. Lastly, Parton’s team transforms predicted iAET into iNDVI and then into ANPP for the growing season. The outcome is a grassland productivity forecast map, or “Grass-Cast,” for 382 individual counties in the Great Plains region where grassland is the primary land cover (as opposed to forests or croplands). The map actually reports the average of 10 different ANPP forecasts based on various combinations of recent weather data and seasonal precipitation outlooks.
Release 1.0 generates the Grassland Productivity Forecast (Grass-Cast) maps for counties in the northern Great Plains. The Grass-Cast is updated every two weeks throughout the growing season, beginning in early May. Three versions of the map are generated every two weeks, showing how much production might be expected if a county receives above, near, or below-normal precipitation during the rest of the growing season (May, June, July).