Acta Geographica Sinica ›› 2020, Vol. 75 ›› Issue (9): 1907-1920.doi: 10.11821/dlxb202009007

• Climate and Ecological Environment • Previous Articles     Next Articles

Occurrence characteristics and response to El Niño of MODIS-based active fires in Indonesia during 2001-2018

LIU Jia1,2(), LIANG Yihang1,3, LI Peng1,2(), XIAO Chiwei1,2   

  1. 1. Institute of Geographic Sciences and Natural Resources Research, CAS, Beijing 100101, China
    2. University of Chinese Academy of Sciences, Beijing 100049, China
    3. School of Earth Science and Resource, Chang'an University, Xi'an 710000, China
  • Received:2019-10-23 Revised:2020-06-27 Online:2020-09-25 Published:2020-11-25
  • Contact: LI Peng;
  • Supported by:
    National Natural Science Foundation of China(41971242);Program for BINGWEI Excellent Young Talents of Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences(2018RC201);China Postdoctoral Science Foundation(2019M660777);Youth Innovation Promotion Association of the Chinese Academy of Sciences(CAS2020055)


The occurrence and development of global active fires under the background of global warming is an important research content. Based on the Moderate Resolution Imaging Spectroradiometer (MODIS) C6 active fire point vector data (2001-2018) of Indonesia provided by the Fire Information Resource Management System (FIRMS) of the USA, the occurrence probability, intensity and response to the El Niño of Indonesian active fires were quantified and analyzed using the GIS-based fishnet method at a 1 km×1 km grid system, in order to reveal the spatio-temporal variations of active fire occurrence frequency in Indonesia during 2001-2018. The results showed that: (1) A total area of 22.79% was affected by active fire at least once in the past 18 years, which is called as "fire-affected area", while the rest was not impacted by fire according to the MODIS observations. The proportions of "fire-affected area" in the five major islands in a descending order were 32.93% for Kalimantan, 31.44% for Sumatra, 17.16% for Sulawesi, 9.50% for Java and 7.58% for Papua, respectively. The occurrence of active fire was dominated by low probability (1/18-6/18), typically between August and October, and widely distributed in various islands, which was primarily seen in Kalimantan and Sumatra. The number of grids in a medium probability (7/18-12/18) was small but intensively distributed, especially in the eastern lowland plains of Sumatra and the southern and southwestern coasts of Kalimantan. The girds of high probability were sporadically distributed in Indonesian islands, mostly concentrated in Java and Sulawesi. (2) The intensity of active fire was dominated by level-one (once per year) and level-two (2-3 times/year), followed by level-three (4-7 times/year). The active fires of level-one to level-five (1-22 times/year) were widely distributed in various islands, and primarily seen in Kalimantan and Sumatra, while the active fires (23-165 times/year) of level-six were concentrated in Java, Sumatra and Sulawesi. It is worth noting that the distribution of grids with different intensities diminished from the center to the periphery. In addition, the intensities of active fires in different probability groups were different. The low-probability of active fire was mainly 1 time/year, followed by 1-3 and 2-7 times/year for the medium and high probability, respectively. In particular, the intensity of level-six active fire only was observed in high probability grids. (3) Indonesia's active fires between August and October had different sensitivities to strong and weak El Niño, showing increased frequency and area of active fires. The stronger an El Niño event is, the higher the response of active fire will be. The islands in a descending order of sensitivity to active fire were Sumatra, Sulawesi, Kalimantan, Java and Papua.

Key words: active fires, occurrence frequency, probability-intensity, El Niño, fishnet analysis, Indonesia