Earth, Atmosphere & Environment Sciences
Title : | Estimation of Above Ground Biomass Using Multi-frequency Sar Polarimetric Techniques in a Tropical Forest Ecosystem in the Western Ghats |
Area of research : | Earth, Atmosphere & Environment Sciences |
Principal Investigator : | Dr. Smitha Asok V, University Of Kerala |
Timeline Start Year : | 2023 |
Timeline End Year : | 2026 |
Contact info : | smithaasok@gmail.com |
Equipments : | Protimeter BLD5365 Soil Moisture Meter
Leica Disto D810 Touch Laser Distance Meter (Red - 200 metres)
Workstation Computer |
Details
Executive Summary : | Forest aboveground biomass (AGB) plays an important role in the study of the carbon cycle and climate change in the global terrestrial ecosystem and it’s estimation using remote sensing is an effective method for regional scale assessments. Tropical forest ecosystem is opulent in biomass and this study is intended to take the advantages of SAR remote sensing in biomass estimation in the Peppara and Neyyar Wildlife Sanctuaries in the Southern Western Ghats region, along with the selected vegetation classes in the surrounding forest. The work proposes to assess the AGB of the tropical forest ecosystem through the utilization of multi-frequency SAR polarimetric techniques coupled with extensive field measurements. SAR data from different sensors, C band of Sentinel-1, L band of ALOS PALSAR2, and L&S band of NISAR (if available) shall be utilized for the development of AGB prediction model primarily of Random Forest Model owing to its robust nature. SAR backscatter measurements especially for longer wavelengths of the EM spectrum have strong sensitivity to forest biomass and in monitoring carbon stocks. This method also has an edge over conventional optical remote sensing techniques owing to its all weather capability. Field data collection for the biophysical parameters of trees, including Tree Height, Diameter at breast height (DBH) and identification of tree species will be carried out for the selected vegetation classes in the study area. Forest being a complex ecosystem with complex incoherent ground scattering objects, this study will employ an incoherent polarimetric decomposition method, ie, Yamaguchi decomposition model to the selected SAR data for the extraction of polarimetric parameters.This is a four component scattering model based on the Coherency matrix of polarimetric SAR images wherein four sub matrices corresponding to the surface scattering, double bounce scattering, volume scattering and helix scattering mechanisms are generated from the Coherency matrix. Biophysical indices such as BMI (Biomass Index), CSI (Canopy Structure Index), VSI (Volume Scattering Index) and RVI (Radar Vegetation Index) will be computed from the coherency matrix to predict AGB. Finally geostatistical analysis for correlating AGB with various parameters extracted from the SAR dataset will be done to assess the relationship between biomass and each backscattering coefficients, radar indices and other polarimetric parameters. For validation, statistical measures including R2, root mean square error (RMSE), RMSE%, mean absolute error (MAE), bias and bias% will be calculated between the predicted AGB from model and the field calculated AGB. After validation process, the AGB map of the entire study area will be created from the model. The model is expected to have scalability in the tropical forest realm and shall be applicable to other terrains with similar vegetation characteristics as well. |
Total Budget (INR): | 25,19,240 |
Organizations involved