Forest Change in Cambodia

Cambodia

In Depth Coverage

Assessing Forest Change in Cambodian Protected Areas

The objective of this research is to assess the spatial and temporal scales of forest change in Cambodian Protected Areas.  The methods will combine Remote Sensing and GIS in the context of Landscape Ecological theory. Understanding the scales at which forest change is occurring will inform the scale of management required for protected areas, and provide regionally specific methods for monitoring forest change utilising remote sensing. This is especially significant in the context of the UN-REDD programme likely to be adopted at the Climate Change Conference in Copenhagen 2009.

  • Link With Climate Change

    Open

    Forests are a major contributor to the sustainable development of Cambodia, and continued deforestation and forest degradation will have significant economic and environmental implications. Climate change is one of these potential impacts. Deforestation and forest degradation can reduce removal of CO2 from the atmosphere, increase surface temperature, and reduce evapotranspiration. Altered land use of deforested areas may also contribute to climate change: large-scale plantations may reduce overall biomass and carbon stocks, and large-scale commercial agricultural operations may reduce soil fertility and subsequently the ability to support carbon stores.

  • Objectives

    Open

    The objectives of this research are to:

    • Quantify the temporal and spatial scales at which forest change occurs;
    • Quantify the rates of forest change (e.g. deforestation and conversion);
    • Identify the landscape spatial relationships that may contribute to forest change;
    • Develop multi-temporal and multi-platform remote sensing change detection methods that will be suitable for ongoing monitoring;
    • Provide a base line map of forest cover in the case study site for future monitoring.

     

  • Work plan

    Open

    This project will be undertaken in the following stages:

    1. Background review of literature in Remote Sensing/GIS for forest change detection in the context of Landscape Ecology.

    2. Initial field site visit to Phnom Kulen National Park and establishment of specific study areas . Ground reference survey will be undertaken to coincide with initial remote sensing acquisition.

    3. Initial field data assessed and preliminary report and publication.

    4. Internship for Kevin Davies with International Institute for Geo-Information Science and Earth Observation, Netherlands.

    5. Final field work. Extensive reference survey of forest change in Phnom Kulen National Park. Liaison with APSARA and Ministry of Environment. Initiate collaboration with Royal University of Phnom Penh. Base map of Phnom Kulen completed.

    6. Final analysis of results, final report, and PhD thesis preparation. Publications prepared and submitted. Remote sensing monitoring methodology developed and delivered to ASPARA and Ministry of Environment.

  • Schedule

    Open

    The work plan will be scheduled as follows:

    Stage 1. Mar 2009 - Ongoing: Background

    Stage 2. Dec 2009 - Jan 2010: Initial Field Work

    Stage 3. Feb 2010 - May 2010: Preliminary Report

    Stage 4. Jun 2010 - Aug 2010: Internship

    Stage 5. Oct 2010 - Mar 2011: Final Field Work

    Stage 6. Apr 2011 - Mar 2012: Final Report and Thesis
     

  • Technical and Scientific Approach & Methods Proposed

    Open

    The methodology will be undertaken in the following stages:

    Field Work
    Significant field reference data will be collected at all case study sites. This will involve undertaking botanical survey with local expertise in conjunction with APSARA and Ministry of Environment. Transects at random locations will be undertaken with differential GPS. Field reference data will be collected at or close to planned image acquisition dates.

    Historical Change Detection
    Archived Landsat MSS/TM/ETM+/SPOT/ASTER/ALOS will be acquired where available to assess long term patterns of forest change in the case study site. Multivariate Alteration Detection/Maximum Autocorrelation Factor (MAD/MAF) will be the primary change detection method to avoid the need to radiometrically normalise imagery. Appropriate methods for dealing with multi-temporal co-registration and varying pixel size will need to be addressed.

    Spatio-Temporal Analysis
    In order to quantify the spatial and temporal dynamics of forest change, landscape ecology metrics will be generated from the change detection results using Fragstats. Multiscale analysis (scale variance method) will also be undertaken to ascertain at which scales the processes driving forest change are operating. Bayesian analysis will be employed to assess the probabilistic relationship between deforested areas, forest patches, degraded forest classes, distance to protected area boundaries, other landscape classes (roads, rivers, human habitation etc.) and topography.

  • Results

    Open

    The following results will be produced:

    • Quantitative assessment of rates of forest change at the case study site;
    • Multi-temporal maps of forest change at the case study site to indicate past and current forest use practices;
    • Assessment of the spatial risk factors associated with forest change in a protected area;
    • Assessment of the spatial and temporal scales of forest change and recommendations for management scale;
    • Detailed methodology for future monitoring utilising remote sensing and GIS in the context of Landscape Ecological theory.

  • Deliverables

    Open

    The following will be delivered:

    • Preliminary Report
    • Detailed Final Report
    • Forest base maps for future monitoring of the case study site
    • Multi-temporal maps of forest change for the case study site
    • Monitoring Methodology Manual
    • Applicable journal articles published as part of this project
    • PhD Thesis on Forest Change in Cambodian Protected Areas 

  • Use of Satellite Imagery and GIS Solutions

    Open

    Archived and newly acquired SPOT imagery is preferred over other sensors due to its suitability for monitoring vegetation.

    ENVI+IDL is the preferred processing software. The required MAD/MAF libraries have been developed in IDL, and customisations to this library are expected for this project.

  • Local Actions

    Open

    This project is expected to deliver quantitative data on forest change, and rigorous methods for future monitoring of forest change in the Phnom Kulen protected area. All reports, publications and the thesis will be made available to APSARA and the Ministry of Environment to enable them to improve monitoring of forests in this protected area.

  • Miscellaneous information / interesting details on project

    Open

    The initial case study site is Phnom Kulen National Park, Siem Reap Province. This area is adjacent to the World Heritage Site of Angkor. Prior to the protected area being established, forest concessions resulted in large scale deforestation in the surrounding foothills. The area continues to face increased pressure as a result of tourism and the expansion of Siem Reap town. The area is significant because of its natural and cultural heritage, local communities depend on the protected area for forest products, and being part of the catchment for the Siem Reap river it provides significant environmental services.

     

  • Region Name

    Open
    Siem Reap Province

  • Partners involved in project

    Open
    University of Sydney, School of Geosciences

    APSARA (National Authority for the Protection and Management of Angkor and the Region of Siem Reap)

    Cambodian Ministry of Environment