Unprecedented Megafire Remote Sensing
BackgroundBetween climate change and nearly a century of fire exclusion, wildfires have become more extreme with respect to size, severity, complexity of behavior, and resistance to suppression. These fires are commonly classified as megafires and are at the extreme end of the historical range of variability. Because of their rare nature, we have limited data by which to investigate what drives and extinguishes them thus making them unpredictable and difficult to manage for, during, or after.
Over the past several decades the use of satellite remote sensing has provided a means to investigate the complex mosaics of these and other fires; but the spatial and spectral resolution of these data has been limiting. Two of the commonly used sensors are Landsat and MODIS. Landsat is a broadband sensor that offers high spatial resolution at 30m, but is limited with only one band in the thermal infrared. MODIS is another broadband sensor that offers high temporal resolution extending into the thermal infrared, but with coarse spatial resolution at 900 m. Because of the nature of broadband sensors, both sensors are limited in spectral resolution, thus making it difficult to tease out differences in spectral signatures beyond general classifications such as non-photosynthetic and photosynthetic vegetation, soil, and char.
The next generation of sensors currently being tested on airborne campaigns include hyperspectral imaging spectroscopy and multiband thermal infrared imaging. Until recently, it has been difficult to use these sensors to investigate the relationship between fire behavior and the pre- and post- condition of the landscape. In preparation for the Hypserspectral Infrared Imager (HyspIRI) mission, the Airborne Visual Infrared Imaging Spectrometer (AVIRIS) and the multi-band thermal infrared imager MODIS/ASTER (MASTER) have captured before, during, and after data of megafires. AVIRIS offers high spatial (~15 m) and spectral resolution (10 nm) in the visual to shortwave infrared. MASTER offers fine spatial resolution (35 m) with intermediate spectral resolution covering 25 bands in the visual to shortwave infrared, 3 bands in the middle infrared, and five bands in the thermal infrared, which allow us to disentangle emissivity from land surface temperature. Specific fire behavior applications for MASTER are discussed on the HyspIRI Website.
In addition to these datasets, high resolution LiDAR has captured the spatial structure of forests in areas that have burned both before and after fire. LiDAR provides high resolution (1 m) structural data such as canopy surface models, vegetation strata information, and high-resolution digital elevation models.
ObjectiveThis website is designed to provide rasters of information from before, during, and after these megafire events captured using these technologies. Spectroscopic (AVIRIS and MASTER) Level 2 data includes atmospherically, topographically-corrected, georecitified surface reflectance, while LiDAR level 2 data products include rasters of structural metrics. AVIRIS and MASTER Level 3 data includes operational metrics such as the normalized burn ratio (NBR) and the normalized difference vegetation index (NDVI). Detailed descriptions of the instruments and the processed data can be found under the data tab on the left side. These data have a broad array of applications and are offered free to the public, thus facilitating interdisciplinary collaboration.
Stavros EN, Tane Z, Kane VR, Veraverbeke S, McGaughey RJ, Lutz JA, and Ramirez C. (In Prep) Unprecedented remote sensing data from before and after California King and Rim Megafires. Ecology.