The U.S. Geological Survey (USGS) has developed and implemented an algorithm that identifies burned areas in temporally-dense time series of Landsat Analysis Ready Data (ARD) scenes to produce the Landsat Burned Area Products. The algorithm makes use of predictors derived from individual ARD Landsat scenes, lagged reference conditions, and change metrics between the scene and reference conditions. Scene-level products include pixel-level burn probability (BP) and burn classification (BC) images, corresponding to each Landsat image in the ARD time series. The scene-level products are available through https://earthexplorer.usgs.gov. Annual composite products were derived from the scene level products. Prior to generating annual composites, individual scenes that had > 0.010 burned proportion were visually assessed as part of a quality assurance check. Scenes with obvious commission errors were removed. The annual products include the maximum burn probability (BP), burn classification count (BC) or the number of scenes a pixel was classified as burned, filtered burn classification (BF) with burned areas persistent from the previous year removed, and the burn date (BD) or the Julian date of the first Landsat scene a burned areas was observed in. Vectorized versions of the BF raster are also provided as shapefiles (BF_labeled) with attributes including summary statistics of BC, BD, BP, as well as the majority level 3 ecoregion (Omernik and Griffith, 2014) and count of pixels by each National Land Cover Database Category (Vogelmann et al., 2001; Yang et al., 2018) for each burned area polygon. These products were generated for the conterminous United States for 1984 through 2019 individually for Landsat TM (5), Landsat ETM+ (7), OLI/TIRS (8), and for all sensors combined. The products for each sensor combination and year are contained in a compressed tar file are available through here (USGS Science Base Catalog, https://doi.org/10.5066/P9QKHKTQ) and also at https://gec.cr.usgs.gov/outgoing/baecv/LBA/LBA_CU_C01_V01/
Additional details about the algorithm used to generate these products are described in Hawbaker, T.J., Vanderhoof, M.K., Schmidt, G.L., Beal, Y, Takacs, J.D., Falgout, J.T., Picotte, J.J., and Dwyer, J.L. 2020. The Landsat Burned Area algorithm and products for the conterminous United States. Remote Sensing of Environment, Vol. 244, https://doi.org/10.1016/j.rse.2020.111801
Eidenshink, J., Schwind, B., Brewer, K., Zhu, Z.L., Quayle, B., & Howard, S. 2007. A project for monitoring trends in burn severity. Fire Ecology, 3, 3-21
Hastie, T., Tibshirani, R., & Friedman, J. 2009. The Elements of Statistical Learning; Data Mining, Inference, and Prediction. (2nd ed.). New York, NY, USA: Springer
Omernik, J.M., Griffith, G.E., 2014. Ecoregions of the conterminous United States: evolution of a hierarchical spatial framework. Environmental Management 54, 1249–1266. https://doi.org/10.1007/s00267-014-0364-1
Vogelmann, J.E., Howard, S.M., Yang, L.Y., Larson, C.R., Wylie, B.K., Van Driel, N., 2001. Completion of the 1990s National Land Cover Data Set for the Conterminous United States from Landsat Thematic Mapper Data and Ancillary Data Sources. Photogrammetric Engineering and Remote Sensing 65, 650–662.
Yang, L., Jin, S., Danielson, P., Homer, C., Gass, L., Bender, S., Case, A., Costello, C., Dewitz, J., Fry, J., Funk, M., Granneman, B., Liknes, G., Rigge, M., Xian, G. 2018. A new generation of the United States National Land Cover Database: Requirements, research priorities, design, and implementation strategies. ISPRS Journal of Photogrammetry and Remote Sensing, 146, 108–123. https://www.sciencedirect.com/science/article/pii/S092427161830251X
Zhu, Z., & Woodcock, C.E. 2014. Automated cloud, cloud shadow, and snow detection in multitemporal Landsat data: An algorithm designed specifically for monitoring land cover change. Remote Sensing of Environment, 152, 217-234