Near Real Time Wildfire Monitoring using Google Earth Engine

Abstract

Wildfire situation of Thailand is currently increasing of damage and count. Thus, the monitoring and damage assessment of natural disaster provide valuable and useful information for natural disaster management and mitigation. This study presented the guideline of the application of Google Earth Engine (GEE) for monitoring and analysis of wildfire by using satellite imagery of Terra/Aqua in the system of MODIS from database of the Fire Information for Resource Management System (FIRMS) by the National Aeronautics and Space Administration (NASA). The present study focused on the presentation of using GEE platform for the assessment and display the amount and cumulative count of hotspots in near real-time and automatic. In addition, it can determine the area of interesting (AOI) and time series. The results of the study presented the most of hotspot areas. It is useful for monitoring and planning of the protection of wildfire. In this study, the accuracy assessment was done by comparing with the locations of wildfire in year 2020 in the study area including 1) Amphoe Muang, Nakhonnayok province, 2) Amphoe Phow and Amphoe Mae-tang, Chiang Mai province. The study indicated that the values of total accuracy were 82.86%, 64.38% and 80.52%, respectively and the Kappa coefficient of correlation were 0.66, 0.61 and 0.62, respectively. It means the results are in good level. As a result, the application of GEE platform for monitoring wildfire in near real-time and automatic can be used as an efficient tool for monitoring the risk area of wildfire.