Estimating Tallgrass Prairie Fuel Loads for Prescribed Rangeland Burning Using Landsat Images.

Prescribed rangeland burning is a management tool necessary to sustain the native tallgrass prairie ecosystem that spans eastern Kansas, Oklahoma, and Nebraska.  Unfortunately, smoke released from rangeland burning can impact air quality, especially in urban areas (e.g., Kansas City) where increases in ozone formation have been observed during the spring burning season.  BlueSky, a model developed by the USDA Forest Service, uses a combination of mesoscale weather forecasts (MM5) and modeled fuel/fire characteristics to predict smoke dispersion from forest fires.  The adaptation of this model to prescribed rangeland fires would be very useful for predicting (and mitigating) air quality impacts; however, the unique characteristics of rangeland fires need to be quantified as inputs to the model in order to obtain accurate model results.  Fuel loading is one of the most important inputs to the model but is difficult to quantify.  Interannual variation in primary productivity, grazing regime, and high spatial variability associated with landform, make it difficult to quantifying fuel loads at large scales.  The Normalized Difference Vegetation Index (NDVI) obtained from Landsat images may provide a tool for estimating fuel loading over large regions and provide input to Bluesky.  The objective of this study was to quantify the relationship between remotely sensed NDVI and standing biomass, as a way to estimate fuel loads for prescribed burning of the tallgrass prairie.  Approximately 60 points were selected covering nearly 1200 ha of contiguous tallgrass prairie.  For each point, 4 m-diameter circles were clipped in spring 2008 to estimate actual standing biomass.  Clipped biomass was compared to NDVI obtained from previous fall and summer Landsat images.  A linear relationship was used to predict fuel loads for the entire area based on NDVI.  A subset of plots were clipped in the fall and then again in the spring in order to estimate over-winter losses.  Results of this study can be used in the BlueSky smoke modeling framework to create a management tool that can be used to predict and hopefully minimize the impacts of prescribed burning on air quality.