Time series of chaparral live fuel moisture maps derived from MODIS satellite data

Abstract

Wildfires in chaparral shrublands of southern California are a major hazard and important ecological disturbance agent. Fire managers typically monitor fuel moisture of chaparral shrublands to assess the risk of wildfires, using field-based sampling methods for a few small study areas located sparsely throughout southern California. Remote sensing provides the potential for deriving spatially explicit and temporally frequent data on live fuel moisture (LFM) conditions. The objective of this present study was to explore the potential for monitoring LFM with maps derived from Moderate Resolution Imaging Spectroradiometer (MODIS) data on the National Aeronautics and Space Administration (NASA) Terra Earth-observing satellite. A time series of MODIS surface reflectance data (MOD-09_A1) for San Diego County, California from Fall 2000 through 2003 was used to derive normalized difference indices, which were regressed against LFM data. A high degree of temporal co-variability was found, with three MODIS indices providing similar predictability. Regression relationships were inverted and applied to MODIS images to map LFM interval classes for chaparral areas of San Diego County. The spatial–temporal patterns of LFM maps suggest that, at a minimum, the MODIS can provide spatially explicit information that extends the utility of ground-based measurements of LFM data at a few sites.