THE FILTERING OF SATELLITE IMAGERY APPLICATION USING METEOROLOGICAL DATA AIMING TO THE MEASURING, REPORTING AND VERIFICATION (MRV) FOR REDD

摘要

Since Reducing emissions from deforestation and degradation in developing countries (REDD) scheme has been manifested at the thirteenth Conference of the Parties (COP) of the United Nations Framework Convention on Climate Change (UNFCCC) in 2007, varied approaches of measuring, reporting and verification (MRV) carbon emissions from forest were generated. The suitability of different methods depend on many factors, like cost, accuracy, technologies, scale (sub-national to international level) and availability of data. Among those approaches the application of remote sensing technologies are likely to play leading role. Satellite images require involvement of data analysis. This study emphasizes the use of filtering for satellite images to overcome serious obstructions which degrade their qualifications, for instance, cloud and haze. Improved Local Maximum Fitting (LMF) algorithm was applied. Firstly, 'Min Max Filter' using meteorological data was introduced. In order to obtain the reliable NDVI value, acquired raw data (in this study, 8-days MODIS images of Suphanburi province in Thailand) should be selected meticulously before going on the further step of fitting model. The filtering methods are divided to 2 concepts, changing window size depends on the average monthly rainfall historical statistical data and the number of continuous cloud-days using cloud mask method. The advantages are to avoid the case that correct NDVI value may be eliminated by using too large window filter, or the noise cannot be eliminated by using too small window filter. Then, the filtered images were used to the next step of the time series fitting model with Fourier formula. As the result, the Improved Local Maximum Fitting can eliminate the influence of cloud and haze and increase qualities of original satellite images which can contribute to identify type of forests and to monitor forest condition, furthermore utilization for leakage monitoring.