•  
  •  
 

Section

Physical Sciences

Abstract

Forest degradation is one of the leading sources of carbon emission globally. In Africa, particularly in Tanzania, forest degradation is driven by selective logging and forest fires. Monitoring degradation over large areas is very challenging due to the lack of accurate and reliable methods. This study assessed degradation in miombo woodlands and coastal forests using a synergy of remotely sensed data acquired from sentinel-1 and sentinel-2 achieved by integrating their complementary strength where sentinel-1 enhanced detection of structural changes and sentinel-2 provided spectral insight into vegetation. The 3I3D, unsupervised algorithm analysing trends for three vegetation indices (3I) in a three-dimensional feature space (3D), alongside Random Forest, Support Vector Machines to identify, classify, and map degradation indicators from July to December 2022. Results show that Random Forest had an overall accuracy of 91.6% and a Kappa statistic of 86.5%, while Support Vector Machine achieved only 58.7% accuracy. Burnt areas showed higher reflectance due to exposed soil and ash, while selective logging exhibited intermediate reflectance and radar backscatter. Non- degraded areas had lower reflectance and higher radar backscatter. Ruvu South experienced the highest disturbance with more burnt and logged areas compared to Liwale and Morogoro. Thus, demonstrating the effectiveness of the synergy of Sentinel-1 and 2 in forest degradation monitoring, providing essential insight for conservation strategies in Tanzania

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.