Modeling of Percentage of Canopy in Merawu Catchment Derived From Various Vegetation Indices of Remotely Sensed Data

Abstract

The research was aimed at studying Percentage of Canopy mapping derived from various vegetation indices of remotely-sensed data int Merawu Catchment. Methodology applied was by analyzing remote sensing data of Landsat 7 ETM+ image to obtain various vegetation indices for correlation analysis with Percentage of Canopy measured directly on the field (PTactual) at 48 locations. These research used 11 (eleven) vegetation indices of remotely-sensed data, namely ARVI, MSAVI, TVI, VIF, NDVI, TSAVI, SAVI, EVI, RVI, DVI and PVI. The analysis resulted models (PTmodel) for Percentage of Canopy mapping. The vegetation indices selected are those having high coefficient of correlation (>=0.80) to PTactual. Percentage of Canopy maps were validated using 39 locations on the field to know their accuracies. Percentage of Canopy map (PTmodel) is said to be accurate when its coefficient of correlation value to PTactual is high (>=0.80). The research result in Merawu Catchment showed that from 11 vegetation indices under studied, there were 6 vegetation indices resulted high accuracy of Percentage of Canopy maps (as shown in the value of coefficient of correlation as >0.80), i.e. TVI, VIF, NDVI, TSAVI, RVI dan SAVI, while the rest, namely ARVI, PVI, DVI, EVI and MSAVI, have r values of < 0.80.