Forest change mapping in central Vietnam from 1988 to 2022 using Landsat time-series data

Kurzfassung

This study provides an in-depth analysis of forest cover changes in Thua Thien Hue, a province in central Vietnam, spanning the period from 1988 to 2022. Leveraging time series Landsat observations and employing the Continuous Change Detection and Classification - Spectral Mixture Analysis (CCDC-SMA) detection method, all geospatial analyses were conducted using the Google Earth Engine (GEE) platform. The findings reveal a significant decline in the area occupied by Stable forests, which decreased from 57.3% of the total area in 1993 to 48.1% in 2022. In contrast, degraded forests experienced an increase in coverage, rising from 12.4% to 18.9% over the same time period. Notably, forest loss reached a peak of 1.5% by the end of 2013. The overall accuracy of the forest change maps across different time periods was varied. The highest accuracy was achieved for the forest change map spanning 2003 to 2008, while the lowest accuracy was observed for the period from 1998 to 2003. The results suggest that several factors may contribute to the reduction in stable natural forest cover in the province, including urban expansion, population density growth, expansion of plantations, and the rise in water surface due to the establishment of hydropower plants. To mitigate the negative impacts of these drivers on forest cover, it is imperative to implement sustainable land and water resource management strategies that strike a balance between urban development and responsible forestry practices. While this study acknowledges certain limitations, such as the inherent complexity of forest dynamics, the findings underscore the need for further research to validate and reinforce the presented results. By gaining a deeper understanding of the factors influencing forest canopy cover changes in Thua Thien Hue, researchers can pave the way for informed decision-making and effective conservation efforts in similar tropical regions worldwide.