A Time Series Analysis of Forest Cover and Land Surface Temperature Change Over Dudpukuria-Dhopachari Wildlife Sanctuary Using Landsat Imagery

G N Tanjina Hasnat

doi:10.3389/ffgc.2021.687988

Abstract

Forest cover change is an important criterion as it affects the environmental balance whereas land surface temperature is a significant parameter within the earth climate system. Spatio-temporal change of forest cover can be detected and land surface temperature can be retrieved by applying remote sensing technology. The present study aimed to capture the impact of forest cover change on land surface temperature in Dudpukuria-Dhopachari Wildlife Sanctuary (DDWS), Bangladesh, using multi-spectral and multi-temporal satellite data. To avoid the biasness in the calculation, leaf flash time was targeted for collecting Landsat images from United States Geological Survey (USGS) Earth Explorer and, based on availability, images were collected purposively which ones had closer time period:1990 (March 5, 1990), 2000 (February 5, 2000), 2010 (February 24, 2010) and 2020 (March 23, 2020). Unsupervised classification was applied over the images Landsat 4–5 Thematic Mapper (TM), 7 Enhanced Thematic Mapper Plus (ETM+), and 8 Operational Land Imager (OLI) and Thermal Infrared Sensor (TIRS) data for detecting forest cover change. To retrieve the land surface temperature, Mono Window Algorithm (MWA) method was applied over similar images. Maximum forest degradation was observed in 2010 and the change found was 17% as compared to 1990. After 2010, the forest started to flourish. Land surface temperature dramatically changes over the time period. The highest land surface temperature in the forested area was observed in 2020 (32.2°C) and it was changed 7.7°C from that of the 1990 (24.5°C). In every 10 years, almost 2.3°C–3.0°C temperature change was detected. In the first three decades, a reverse relationship was observed between land surface temperature and forest cover; however, in the last decade, land surface temperature was found to increase with the increase of forest cover. Thus, the results of the study revealed that land surface temperature may not be relevant with the local forest cover change directly. It can be estimated from the results that local forest cover change may have limited impact on local temperature rather than global forest cover change, whereas global warming could play a vital role in changing land surface temperature locally as well as globally.

Highlights

Remote sensing is an important tool to analyse the spatial and temporal changes of land coverage (Petropoulos et al, 2015; Chew et al, 2016; Sekertekin and Bonafoni, 2020)
The area extends from N 22◦09 to N 22◦22 and E 92◦05 to E 92◦10 (Shahadat et al, 2014; Hossain et al, 2017). This evergreen and semi-evergreen tropical natural forest was declared as a protected area of Bangladesh in 2010 and covers 4,717 ha of land area (Shahadat et al, 2014; Bangladesh Forest Department (BFD), 2021)
Images were classified into two broad classes: natural forest coverage, and cultivated and bare land

Summary

Introduction

Remote sensing is an important tool to analyse the spatial and temporal changes of land coverage (Petropoulos et al, 2015; Chew et al, 2016; Sekertekin and Bonafoni, 2020). This is a potential technology for inventory and monitoring of world’s natural resources (Goetz and Dubayah, 2011; Sharma et al, 2021). It maintains the environmental and ecological balance (Liu et al, 2020; Yuan et al, 2020) by preventing unwanted climate change and keeping ecosystems balance between human beings, plants, animals, and other abiotic components. Forest cover changes and damages are eventually assessed by applying remote sensing technology (Zeng et al, 2020)

Methods

Results

Discussion

Conclusion