PROGRESS IN GEOGRAPHY ›› 2021, Vol. 40 ›› Issue (8): 1406-1418.doi: 10.18306/dlkxjz.2021.08.013

• Articles • Previous Articles     Next Articles

Spatial characteristics and national differences of active fires derived from Visible Infrared Imaging Radiometer Suite (VIIRS) in Mainland Southeast Asia in the dry season during 2012-2019

LIU Ying1(), LI Peng2,3,*(), XIAO Chiwei2, LIU Jia2,3, YE Jingqiao4, SHI Dong1   

  1. 1. School of Geosciences, Yangtze University, Wuhan 430100, China
    2. Institute of Geographic Sciences and Natural Resources Research, CAS, Beijing 100101, China
    3. University of Chinese Academy of Sciences, Beijing 100049, China
    4. School of Geographic Sciences, Southwest University, Chongqing 400715, China
  • Received:2020-11-17 Revised:2021-02-22 Online:2021-08-28 Published:2021-10-28
  • Contact: LI Peng;
  • Supported by:
    National Natural Science Foundation of China(41971242);National Natural Science Foundation of China(42001226);Program for Bingwei Excellent Young Talents of Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences(2018RC201);Youth Innovation Promotion Association of the Chinese Academy of Sciences(CAS2020055)


The tropics is the hotspot of global active fires. It is of great significance to reveal the spatial characteristics and dynamic changes of active fires for assessing regional biomass burning and carbon emissions. Based on the active fire vector data (2012-2019) of Visible Infrared Imaging Radiometer Suite (VIIRS) Suomi National Polar-Orbiting Partnership (S-NPP) provided by the NASA's Fire Information for Resource Management System (FIRMS), the occurrence density and agglomeration degree of active fires in Mainland Southeast Asia were quantified and analyzed using kernel density and spatial autocorrelation methods, in order to reveal the spatiotemporal variations in active fire occurrence frequency within the dry season, especially in February and April, during 2012-2019. The results show that: 1) The occurrence of active fires was dominated by low density (79%), and the high density areas had the smallest proportion (4%). Active fires are more likely to occur in economically backward countries (Cambodia, Myanmar, and Laos) and regions (for example, eastern Cambodia and northern Laos). High kernel density areas tend to distribute in high elevation, inland, and border areas in particular. Cambodia, especially its northeastern region, was always an area with high occurrence density of active fires in the study period. 2) The changes in medium density area of active fires were concentrated between January and April, which was mostly distributed around the low and high density areas, while the changes in high density areas were concentrated between February and April, spatially gradually shifted from northeastern Cambodia to western and eastern Myanmar, northwestern Thailand, and northern and southern Laos. 3) The density of active fire occurrence showed significant spatial autocorrelation in Mainland Southeast Asia and the five countries during the dry season. The spatial clustering types of the kernel density of active fires are dominated by "high-high" and "low-low" agglomerations in the dry season, especially between February and April. The local spatial autocorrelation of the kernel density of active fires in Vietnam and Cambodia is stronger than those in Thailand and Laos.

Key words: active fires, VIIRS, kernel density, spatial autocorrelation, spatial distribution, Mainland Southeast Asia