Assessment of satellite orbit-drift artifacts in the long-term AVHRR FireCCILT11 global burned area data set

The recent 1982–2018 FireCCILT11 burned area (BA) product derived from NOAA AVHRR data is examined with regard to its suitability for long-term BA analyses. We focus on identifying NOAA AVHRR satellite orbit-drift artifacts within the FireCCILT11 BA time series since the occurrence of such artifacts...

Full description

Bibliographic Details
Published in:Science of Remote Sensing
Main Authors: Louis Giglio, David P. Roy
Format: Article in Journal/Newspaper
Language:English
Published: Elsevier 2022
Subjects:
Q
Online Access:https://doi.org/10.1016/j.srs.2022.100044
https://doaj.org/article/d29e7e885079463fbb606e27f851e8b5
Description
Summary:The recent 1982–2018 FireCCILT11 burned area (BA) product derived from NOAA AVHRR data is examined with regard to its suitability for long-term BA analyses. We focus on identifying NOAA AVHRR satellite orbit-drift artifacts within the FireCCILT11 BA time series since the occurrence of such artifacts can render any BA data set inappropriate for long-term analyses. We show that significant orbit-drift artifacts are present in the FireCCILT11 product over numerous large spatial patches located on every continent except Antarctica. In addition, the BA mapped by the FireCCILT11 in the United States is compared to independent ground-based records compiled by the National Interagency Fire Center. Prior to 2001, the FireCCILT11 product drastically overestimated BA within this region, to the point of falsely suggesting an abrupt radical change in fire regime at the turn of the century. Our findings indicate that caution is required when using the FireCCILT11 product for long-term BA studies, particularly within the tropics or the United States for the periods 1982–2000 and 2018. Studies requiring gridded BA data for these years, particularly at local and regional scales, should explicitly consider AVHRR orbit-drift artifacts that may result in FireCCILT11-reported BA that is substantially different (for example, several times greater) than the actual BA.