A Review on Remote Sensing Applications on Wildfires Studies

Abstract

Wildfires are among the most prominent disturbances in forest and grassland ecosystems during different seasons and is considered as a natural hazard. They can also become a threat to property, human life, and the economy. Annually 350 to 450 million hectares burn globally releasing 1.5 to 3 petagrams of carbon and other components that impact the atmosphere Leblon et al. (2016).  This review examines the use of remote sensing technologies to study wildfires across different regions. Amazon rainforest generates over 20% of earth’s oxygen but has recently shifted from carbon sink to carbon emitter due to frequent wildfire patterns. Spatial analysis, forest density estimation, and hot spot analysis, are significant for observing regional variations in wildfire patterns. Monitoring large area become cost-and- time effective by using remote sensing imagery. They provide periodic scenarios and multiple approaches to help analyze factors that affect forest fires. A wide range of fire detection sensors like MODIS, VIIRS and Sentinel and various indices such as normalized burned ratio (NBR), normalized burned thermal ratio (NBTR) and burned area index (BAI) are widely employed (Payra et al., 2023). Significant growth has been noticed in wildfires occurrence via remote sensing, which is driven by advancements in detection methods and data acquisition. NOAA- AVHRR, NDVI and surface temperature images have been used for monitoring, experimental studies with microwave data and synergisms between spectral domain, such as combining microwave with optical or thermal infrared data show promising results in forest fire monitoring. However further investigation is required to develop operational multi-sensor system that are both effective and economically viable.