You are here:
Integration of Process Models and Remote Sensing for Estimating Productivity, Soil Moisture, and Energy Fluxes in a Tallgrass Prairie Ecosystem
Citation:
Goodin, D., L. Luo, Bob Mckane, B. Barnhart, J. Halama, P. Pettus, K. Djang, AND A. Brookes. Integration of Process Models and Remote Sensing for Estimating Productivity, Soil Moisture, and Energy Fluxes in a Tallgrass Prairie Ecosystem. Presented at American Association of Geographers Annual Meeting, San Francisco, CA, March 29 - April 02, 2016.
Impact/Purpose:
This presentation at the American Association of Geographers (AAG) annual meeting will describe collaborative research by Kansas State University remote sensing scientists and ORD-NHEERL-WED ecosystem modelers. We will describe our approach for using high-resolution satellite data to test the accuracy of an ecosystem model (VELMA) for predicting spatial and temporal variations in surface fuels in response to climate and rangeland management in the 12,000 square mile Flint Hills tallgrass prairie in Kansas. Characterization of uncertainty in modeled surface fuels through remote sensing methods is important for assisting the Kansas Department of Health and Environment (KDHE), which is collaborating with EPA Region 10 and ORD to assess potential air quality impacts associated with seasonal rangeland prescribed burning. Under prime conditions, such burning can impact communities across a multi-state region, including urban centers such as Kansas City, Tulsa, Lincoln and Omaha. The AAG annual meeting brings together remote sensing experts from around the world, and is an excellent opportunity to gather feedback and learn about emerging technologies useful to our project.
Description:
We describe a research program aimed at integrating remotely sensed data with an ecosystem model (VELMA) and a soil-vegetation-atmosphere transfer (SVAT) model (SEBS) for generating spatially explicit, regional scale estimates of productivity (biomass) and energy\mass exchanges in the Flint Hills tallgrass prairie ecosystem of Kansas and Oklahoma, USA. Our analysis will focus on use of remotely sensed indices derived by combing optical and thermal reflectance (i.e. "triangle" methods) and land cover classification to both evaluate and refine estimates of prairie canopy properties made using the ecosystem and SVAT models. Various methods for integrating these techniques will be compared and evaluated. Applications to management issues, especially air quality issues related to controlled burning of the Flint Hills grassland, will also be discussed.