Science Service System

Summary of Proposal LAN3484

TitleAboveground biomass retrieval in sub-tropical forests - The potential of combined X-, C- and L-band SAR
Investigator Waqar, Mirza Muhammad - Chiba University, Center of Environmental Remote Sensing
Team Member
Mr. Waqar, Mirza Muhammad - Chiba University, Center of Environmental Remote Sensing
Prof. Tetuko Sri Sumantyo, Josaphat - Chiba University, Center of Environmental Remote Sensing
SummaryExecutive Summery: Carbon stored in the aboveground biomass (AGB) of trees is directly impacted by deforestation and degradation. Accurate biomass estimation and mapping is vital for better understanding of the carbon cycles over terrestrial ecosystems. Remote sensing methods are especially suitable for forest AGB estimation, where both active and passive optical remote sensing datasets and methodologies have been used. However, regional scale AGB estimation using coarse and medium resolution optical datasets becomes problematic in sites with complex forest structure, whereas small coverage area, non-frequent data availability, and topographical effects are problems with high-resolution data. Both low- and high-resolution datasets are also restricted by cloud cover. This research intends to study the use of X-band, C-band and L-band SAR datasets in synergy with other remote sensing methods and field observations, to develop region-specific algorithms and techniques for forest AGB estimation by integrating PolSAR and InSAR. Using the aforementioned techniques of forest AGB assessment can give a better picture of the biomass distribution, along with overcoming some of the limitations of other techniques. Proposed pilot study areas for this project is Forested area of Gilgit Baltistan. By implementing Pol-SAR and InSAR, forest growth during past one decade will be monitor and accurate AGB estimation will be carried out. Research Objectives: The prime goal of this research is to estimation accurate forest biomass by integrating multi-sensor data. 1. Developing region specific SAR pre-processing method. 2. Forest species classification by fusion of SAR and high resolution optical datasets. 3. Forest growth monitoring by combining time series SAR with high resolution optical data during past 5 years. 4. Using Pol-InSAR to estimate accurate above ground biomass. 5. Identify factor influencing biomass estimation accuracy. Datasets Requirement: Following datasets are required for this project: 1) Allometric Equations of different tree species in study area. 2) 2 pairs of L-band SAR data from archived ALOS-2 datasets and C-band RadarSat-2 satellite. 3) Time series TerraSAR-X single pol images 4) High-resolution optical imagery (e.g. WorldView, IKONOS) to understand and interpret the forest structure and spatial distribution in regions of interest. 5) High resolution DEM to correct topographical errors in SAR data. 6) Survey data and allometric equations, where available, for remote sensing data calibration. Expected Outcomes: 1) Region specific SAR preprocessing method. 2) Forest canopy height map using SAR InSAR. 3) Estimation of forest growth during past 5 years. 4) Pol-InSAR based accurate AGB of study area. 5) Forest species based classified map of study area. 6) Factor influencing accuracy of AGB estimation over study area based on uncertainty analysis.

Back to list of proposals

DLR 2004-2016