Evaluating remote sensing datasets and machine learning algorithms for mapping plantations and successional forests in Phnom Kulen National Park of Cambodia.

This study develops a modelling framework by utilizing multi-sensor imagery for classifying different forest and land use types in the Phnom Kulen National Park (PKNP) in Cambodia. Three remote sensing datasets (Landsat optical data, ALOS L-band data and LiDAR derived Canopy Height Model (CHM)) were used in conjunction with three different machine learning (ML) regression techniques (Support Vector Machines (SVM), Random Forests (RF) and Artificial Neural Networks (ANN)). These ML methods were implemented on (a) Landsat spectral data, (b) Landsat spectral band & ALOS backscatter data, and (c) Landsat spectral band, ALOS backscatter data, & LiDAR CHM data. The Landsat-ALOS combination produced more accurate classification results (95% overall accuracy with SVM) compared to Landsat-only bands for all ML models. Inclusion of LiDAR CHM (which is a proxy for vertical canopy heights) improved the overall accuracy to 98%. The research establishes that majority of PKNP is dominated by cashew plantations and the nearly intact forests are concentrated in the more inaccessible parts of the park. The findings demonstrate how different RS datasets can be used in conjunction with different ML models to map forests that had undergone varying levels of degradation and plantations.

Evaluating the ability of community-protected forests in Cambodia to prevent deforestation and degradation using temporal remote sensing data.

Community forests are known to play an important role in preserving forests in Cambodia, a country that has seen rapid deforestation in recent decades. The detailed evaluation of the ability of community-protected forests to retain forest cover and prevent degradation in Cambodia will help to guide future conservation management. In this study, a combination of remotely sensing data was used to compare the temporal variation in forest structure for six different community forests located in the Phnom Kulen National Park (PKNP) in Cambodia and to assess how these dynamics vary between community-protected forests and a wider study area. Medium-resolution Landsat, ALOS PALSAR data, and high-resolution LiDAR data were used to study the spatial distribution of forest degradation patterns and their impacts on above-ground biomass (AGB) changes. Analysis of the remotely sensing data acquired at different spatial resolutions revealed that between 2012 and 2015, the community forests had higher forest cover persistence and lower rates of forest cover loss compared to the entire study area. Furthermore, they faced lower encroachment from cashew plantations compared to the wider landscape. Four of the six community forests showed a recovery in canopy gap fractions and subsequently, an increase in the AGB stock. The levels of degradation decreased in forests that had an increase in AGB values. However, all community forests experienced an increase in understory damage as a result of selective tree removal, and the community forests with the sharpest increase in understory damage experienced AGB losses. This is the first time multitemporal high-resolution LiDAR data have been used to analyze the impact of human-induced forest degradation on forest structure and AGB. The findings of this work indicate that while community-protected forests can improve conservation outcomes to some extent, more interventions are needed to curb the illegal selective logging of valuable timber trees.