Geoinformatické alternativy stanovení C-faktoru modelu RUSLE

Summary

English name:
GEOINFORMATIC ALTERNATIVES CALCULATION OF C-FACTOR IN RUSLE MODEL

The aim of thesis was searching of geoinformatics alternatives for assessment of vegetative cover, which enters the RUSLE model as a value of C-factor. The first step was the gathering of foreign articles which deal with the determination of C-factor, and obtain appropriate data. As source were free available Landsat 5 TM images from USGS, and images from ADC cameras and sensor Hyperion provided by the Department of Geoinformatics.

The main part of this thesis was about writing of theoretical research that deals with the description of the model USLE, RUSLE and RUSLE2. The USLE model were described by various factors, including the description of the approach according Janecek, whose method associated with the field evaluation is used in our area. There were also described differences between method USLE and both RUSLE and RUSLE2, and detail description of C-factor.

The thesis also presents theories for various geoinformatics alternatives for assessment of C-factor, which were found in foreign articles. The main approaches that have been shown to work include determination of C-factor using vegetation indices (NDVI, SAVI, mNDVI, ...), image classification and Linear Spectral Mixture Analysis (LSMA).

In the practical part, selected approaches were tested over various specific data. It was an approach with using NDVI, SAVI and SATV of images from Landsat 5 TM, as well as approach with the use NDVI above multispectral images of ADC camera and use the same approach of hyperspectral images from EO-1 Hyperion.

The result, in addition to the theoretical research, is also evaluation of approaches using tabular values, image histogram and visual comparison.

© Antonín BENC | 2013 | katedra GEOINFORMATIKY | Přírodovědecká fakulta | Univerzita Palackého v Olomouci