Evaluation of the potential of Hyperion data to estimate wildfire hazard in the Santa Ynez front range, Santa Barbara, California

Abstract

Wildfire hazard is a worldwide problem with enormous environmental, economic, and social costs. Better methods are needed to assess risk to more effectively manage natural resources. Current models need spatially explicit, timely, information about surface conditions. Meeting this need requires development of remotely sensed model inputs. Critical remotely sensed parameters needed for wildfire hazard assessment include surface reflectance, canopy water content, species composition and fuel state. Semiarid shrublands have been difficult to assess and monitor using multispectral satellite data because their spectral bands and wavelength locations are not sensitive to these input parameters, however, such inputs have been demonstrated using AVIRIS data. We compare near simultaneous acquisition of Hyperion (12/06/01) and AVIRIS (14/06/01) data for retrieval of surface reflectance. For both instruments, field spectra were acquired to support reflectance retrieval. Hyperion and AVIRIS spectra are similar in shape and albedo but SNR is 5 times higher in AVIRIS. Fuel condition was assessed using the endmember fractions from spectral mixture analysis, with Hyperion and AVIRIS providing similar fractions. Canopy water content was compared using the 1200 nm liquid water band and two water indices, the Water Index and the Normalized Difference Water Index (NDWI). Results show poor Hyperion performance relative to AVIRIS.