Burned Area Evaluation Method for Wildfires in Wildlife Sanctuaries Based on Data from Sentinel-2 Satellite

被引:3
|
作者
Uttaruk, Yannawut [1 ]
Rotjanakusol, Tanutdech [2 ]
Laosuwan, Teerawong [2 ]
机构
[1] Mahasarakham Univ, Fac Sci, Dept Biol, Maha Sarakham 44150, Thailand
[2] Mahasarakham Univ, Fac Sci, Dept Phys, Maha Sarakham 44150, Thailand
来源
关键词
remote sensing; digital image processing; burned areas; spectral indices; random forest; LAND-SURFACE TEMPERATURE; SEVERITY; INDEX;
D O I
10.15244/pjoes/152835
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Wildfire is a kind of disaster that can damage living creatures and environment as well as cause small dust affecting to health of people. Wildfire can be naturally occurred and human's actions. This research aims to evaluate burned areas caused by wildfires in Omkoi Wildlife Sanctuary. The research was conducted by collecting data from Sentinel-2 Satellite for 4 years started from 2017 to 2020. Three formats of spectral indices, i.e., NBR, NDWI, and RBR were used for evaluating burned areas caused by wildfires. Validation was tested by using statistical methods. The results revealed that burned areas from 2017 to 2020 were 0.107 km2, 1.160 km2, 0.387 km2, and 1.031 km2, respectively. The larges burned area was found in 2018 followed by 2020, 2019, and 2017. For validation, it was found that total accuracy of 2017 was 93.33 % with Kappa statistics of 0.87 whereas total accuracy of 2018 was 78.33 % with Kappa statistics of 0.57. In 2019, total accuracy was 81.67 % with Kappa statistics of 0.63. In 2020, total accuracy was 76.67 % with Kappa statistics of 0.53.
引用
收藏
页码:5875 / 5885
页数:11
相关论文
共 50 条
  • [1] Development of a Burned Area Processor Based on Sentinel-2 Data Using Deep Learning
    Knopp, Lisa
    PFG-JOURNAL OF PHOTOGRAMMETRY REMOTE SENSING AND GEOINFORMATION SCIENCE, 2021, 89 (04): : 357 - 358
  • [2] Development of a Burned Area Processor Based on Sentinel-2 Data Using Deep Learning
    Lisa Knopp
    PFG – Journal of Photogrammetry, Remote Sensing and Geoinformation Science, 2021, 89 : 357 - 358
  • [3] A Deep Learning Approach for Burned Area Segmentation with Sentinel-2 Data
    Knopp, Lisa
    Wieland, Marc
    Raettich, Michaela
    Martinis, Sandro
    REMOTE SENSING, 2020, 12 (15)
  • [4] A Burned Area Mapping Algorithm for Sentinel-2 Data Based on Approximate Reasoning and Region Growing
    Sali, Matteo
    Piaser, Erika
    Boschetti, Mirco
    Brivio, Pietro Alessandro
    Sona, Giovanna
    Bordogna, Gloria
    Stroppiana, Daniela
    REMOTE SENSING, 2021, 13 (11)
  • [5] Burned Area Estimation Using a New Accuracy Verification Method Based on Sentinel-2 Images
    Chen, Yunping
    Lu, Chuangjiang
    Huang, Xuan
    Xie, Siyuan
    Sun, Yuan
    IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2023, 20
  • [6] ASSESSMENT OF BURNED AREA MAPPING METHODS FOR SMOKE COVERED SENTINEL-2 DATA
    Grivei, Alexandra-Cosmin
    Vaduva, Corina
    Datcu, Mihai
    2020 13TH INTERNATIONAL CONFERENCE ON COMMUNICATIONS (COMM), 2020, : 189 - 192
  • [7] Burned Area Classification Based on Extreme Learning Machine and Sentinel-2 Images
    Gajardo, John
    Mora, Marco
    Valdes-Nicolao, Guillermo
    Carrasco-Benavides, Marcos
    APPLIED SCIENCES-BASEL, 2022, 12 (01):
  • [8] Landsat and Sentinel-2 Based Burned Area Mapping Tools in Google Earth Engine
    Roteta, Ekhi
    Bastarrika, Aitor
    Franquesa, Magi
    Chuvieco, Emilio
    REMOTE SENSING, 2021, 13 (04) : 1 - 30
  • [9] OPTIMIZATION OF A RANDOM FOREST CLASSIFIER FOR BURNED AREA DETECTION IN CHILE USING SENTINEL-2 DATA
    Roteta, E.
    Oliva, P.
    2020 IEEE LATIN AMERICAN GRSS & ISPRS REMOTE SENSING CONFERENCE (LAGIRS), 2020, : 568 - 573
  • [10] A Practical Method for High-Resolution Burned Area Monitoring Using Sentinel-2 and VIIRS
    Pinto, Miguel M.
    Trigo, Ricardo M.
    Trigo, Isabel F.
    DaCamara, Carlos C.
    REMOTE SENSING, 2021, 13 (09)