Monitoring Forest Disturbance Using Sentinel Data: A Case Study of Non-seasonal Time Series Approach

Abstract

Monitoring deforestation and reforestation dynamics is critical for forest conservation, particularly in biodiversity-rich regions, such as Khyber Pakhtunkhwa (KPK), Pakistan. The current study aimed to introduce a novel application of the PVts-β method for non-seasonal time series analysis using Sentinel-1 radar imagery. Moreover, the study also addressed limitations of traditional approaches, such as reliance on optical data and sensitivity to cloud cover. By quantifying deforestation (7,985 hectares) and reforestation (4,098 hectares) between 2018 and 2022, the study highlighted significant land cover changes. As compared to the existing methods, such as Breaks for Additive Season and Trend (BFAST) and Continuous Change Detection and Classification (CCDC), PVts-β offers advantages in computational efficiency and disturbance detection accuracy, as validated through performance metrics and sensitivity analysis. The results provide actionable insights for conservation strategies and policy-making, emphasizing the adaptability of the methodology to various regions and forest types. The current study advances remote sensing applications in forest monitoring, offering a robust framework to address global environmental challenges.