EGU General Assembly 2023
© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.

Detecting Forest Fires by Using Remotely Sensed Data in Riau, Indonesia

Agustiyara Agustiyara and Balázs Székely
Agustiyara Agustiyara and Balázs Székely
  • ELTE Eötvös Loránd University, Department of Geophysics and Space Science, Budapest, Hungary (

This research aims to shed light on remote sensing data, focusing on remote sensing for forest fires which still largely separates these expertise techniques. In this situation, the use of sentinel data makes it possible to make assessments related to land and forest fires by assessing the land cover function of land fires. The first specific location shows that land fires are clearly visible, especially on the Rupat island which is part of Bengkalis Regency, Riau Province, Indonesia. In a general sense, Rupat Island is a small island with a peatland ecosystem. This becomes complex when various land functions and activities, such as the development of the oil palm plantation industry, protected forest areas, industrial plantation forest (HTI) company areas, peat land, and other land uses activities are found on this island. Forest fires cause extreme long-term damage to the environment, wildlife, flora, and property including forestry and agricultural holdings every year. Along with improving the detection of and response times to such fires, there is also a need to improve post-event delineation, assessment, and monitoring of the affected areas. Such post-event analysis can then feed back into strategies and policies for wildfire prevention, prediction, mitigation, and response. However, the detection of such fires by these tools considers the accuracy in terms of the exact location and extent of land classification and burnt areas. The use of statistically significant remote sensing, the research process two products between 2019 and 2020. The research use data equation through the Sentinel-3 data, where the detection of land fires that are clearly visible in the "fire detection" image by performing a data algorithm to ensure that the fire point is no cloud cover. Sentinel-2 data was also used to explain the loss of vegetation on peatlands in the area of land fires, which clearly shows changes in burnt areas. With the same combination of analyses, sentinel-1 data was also used to clarify the land cover in the fire area, where the classification algorithms (forest) and other functions in sentinel-1 data were identified. Therefore, the use of remote sensing primarily aims to highlight the importance of data fusion and integrate it into the multiple factors and motives for forest and land fires.

How to cite: Agustiyara, A. and Székely, B.: Detecting Forest Fires by Using Remotely Sensed Data in Riau, Indonesia, EGU General Assembly 2023, Vienna, Austria, 24–28 Apr 2023, EGU23-7744,, 2023.

Supplementary materials

Supplementary material file