The New VIIRS 375 m active fire detection data product: Algorithm description and initial assessment
详细信息 查看全文
- 作者：Wilfrid Schroeder ; Patricia Oliva ; Louis Giglio ; Ivan A. Csiszar
- 关键词：Active fire detection ; S-NPP/VIIRS ; MODIS
- 刊名：Remote Sensing of Environment
- 出版年：5 March, 2014
- 年：2014
- 卷：143
- 期：Complete
- 页码：85-96
- 全文大小：3234 K
文摘
The first Visible Infrared Imaging Radiometer Suite (VIIRS) was launched in October 2011 aboard the Suomi-National Polar-orbiting Partnership (S-NPP) satellite. The VIIRS instrument carries two separate sets of multi-spectral channels providing full global coverage at both 375 m and 750 m nominal resolutions every 12 h or less depending on the latitude. In this study, we introduce a new VIIRS active fire detection algorithm, which is driven primarily by the 375 m middle and thermal infrared imagery data. The algorithm builds on the well-established MODIS Fire and Thermal Anomalies product using a contextual approach to detect both day and nighttime biomass burning and other thermal anomalies. Here we present the fire algorithm's design and implementation, including important information describing the input data characteristics and potential artifacts associated with pixel saturation and the South Atlantic Magnetic Anomaly, both found to affect the middle infrared channel data. Initial assessment using results derived from the global processing of the algorithm indicated small, although variable, commission errors (< 1.2%) for nominal confidence fire pixels. We achieved improved performance using the 375 m active fire data compared to the VIIRS 750 m baseline fire product, resulting in a 3 脳 and 25 脳 factor increase in the absolute number of fire pixels detected using day and nighttime data, respectively. Similarly, VIIRS 375 m fire data showed significantly superior mapping capabilities compared to current MODIS fire detection data with improved consistency of fire perimeter delineation for biomass burning lasting multiple days.