Essential climate variables (ECVs) are used to track critical attributes of atmospheric, oceanic, and terrestrial systems over time scales appropriate for analyzing their relationships with climate change. As part of a larger Climate Data Record (CDR) and ECV Development project, scientists at the USGS Geosciences and Environmental Change Science Center in Denver, Colorado, have led the development and validation of the Burned Area ECV algorithm. This algorithm automatically extracts burned areas from Landsat imagery, which has a spatial resolution of 30 m and a temporal resolution of 16 days or more, depending on cloud cover. Fires are frequently unreported; consequently, existing fire databases are often incomplete. Furthermore, these databases often have location errors, and records may be duplicated. The Burned Area ECV products will provide new and unique information about spatial and temporal patterns of fire occurrence that existing fire databases may lack, especially in areas such as the shrub and grassland ecosystems in the Great Plains and in the western and southeastern United States.
Validation protocols have been established incorporating and adapting the Committee on Earth Observation Satellites (CEOS) Land Product Validation (LPV) Subgroup's best practice guidelines in collaboration with the European Space Agency (ESA) Climate Change Initiative ECV Fire Disturbance (Fire_cci) validation team. Validation data collection is underway for a number of Landsat path/row scenes in the United States; the data will be used to quantitatively assess the accuracy of the Burned Area ECV and to identify areas for improvement in future algorithm versions.
Landsat Burned Area Essential Climate Variable products for the conterminous U.S. for (A) 1985–1994, (B) 1995–2004, and (C) 2005–2014.