Camp Fire processed Landsat 8 images, pre-fire, during-fire, post-fire

The Camp Fire rapidly spread across a landscape in Butte County, California, toward the city of Paradise in the early morning hours of 8 November 2018. GEE acquires L8 data products from USGS that include Real-Time Tier 1 DN values, representing scaled, calibrated at-sensor radiance, and Level-1 Precision Terrain (L1TP) processing. We carried out additional processing on the pre- and post-fire images, including an illumination correction to account for effects of steep topography, and radiometric normalization to ensure homogeneity between images. We also identified and processed high-quality images for both the pre-fire condition(25 Oct 2013), and the post-fire burn scar (26 Dec 2018). Using the tools we provided in the paper, we estimate that, over the first hour, the Camp Fire was consuming ~200 ha/minute of vegetation with a linear spread rate of 14 km over the fire’s first 25 minutes, or ~33km/hr. We briefly discuss broader use of remote sensing and geospatial analysis for fire research and risk management. A visualization app (sliderApp) that includes pre-fire, active fire, and post-fire images: https://caralyngorman.users.earthengine.app/view/camp-fire-sliding-map-3-9-19

2018年11月8日清晨，美国加利福尼亚州巴特县境内的营溪大火（Camp Fire）迅速向天堂市蔓延，覆盖大片区域。谷歌地球引擎（Google Earth Engine，GEE）从美国地质调查局（United States Geological Survey，USGS）获取Landsat 8（L8）数据产品，此类产品包含实时一级DN值（Real-Time Tier 1 DN values）——对应经缩放、定标后的传感器入瞳辐射亮度，以及一级地形精确处理（Level-1 Precision Terrain，L1TP）成果。研究团队对火灾前、灾后影像开展了额外预处理，包括针对陡峭地形影响的光照校正，以及保障影像间辐射均一性的辐射归一化处理。团队还筛选并处理了两组高质量影像：分别为火灾前影像（2013年10月25日）与灾后烧痕（burn scar）影像（2018年12月26日）。借助论文中提供的分析工具，研究团队估算得出：营溪大火在发生后的首小时内，以每分钟约200公顷的速率吞噬植被；火灾前25分钟的线性蔓延距离达14公里，对应蔓延速率约33公里/小时。本文还简要探讨了遥感与地理空间分析技术在火灾研究及风险管理领域的更广泛应用。配套一款可视化应用（sliderApp），涵盖火灾前、火灾发生中及灾后影像，访问链接：https://caralyngorman.users.earthengine.app/view/camp-fire-sliding-map-3-9-19