Sampling of fire regimes with satellite active fire data: scan pattern, land cover data, and diurnal sampling effects

Edward J Hyer, Naval Research Laboratory, edward.hyer@nrlmry.navy.mil (Presenter)
Jeffrey S Reid, Naval Research Laboratory, reidj@nrlmry.navy.mil

Satellite observations of fire are widely used because they provide a combination of relatively frequent observations and domain coverage that are not available from other data sources. These products have greatly aided both scientific research and operational applications, because they provide consistently processed data for large areas and long term periods. However, consistent processing is not the same thing as consistent information. The view the products provide of spatial and temporal patterns of fire in different vegetation types is systematically biased by many different properties of the satellite constellation, satellite data processing, and interactions between the satellite observations and the fire regime on the ground. To infer the true properties of fire regimes in different ecosystems using these data, these biases must be quantitatively understood. This paper provides results from a series of experiments aimed at quantifying how specific properties of satellite fire detection products affect the observed patterns of fire activity at landscape scales. Numerous biases are described and quantified, including variations associated with ground instantaneous field of view, diurnal sampling of fire activity, and interactions with land cover data products. Progress is described toward improving accounting for these biases in smoke emissions estimates, and hard limitations of the current fire observation constellation are discussed.

Presentation: 2011_Poster_Hyer_170_32.pdf (3317k)

Presentation Type:  Poster

Session:  Other   (Wed 10:00 AM)

Associated Project(s): 

• Related Activity

Poster Location ID: 170