| Summary: | Vegetation spectral reflectance that recorded by Landsat TM satellite sensor has been known widely as a tool to identify object on the earth surface. Factors which influences to the spectral reflectance values are not only forest vegetation species but also vegetation age, topographic, slope aspect, percentage crown closure and slope steepness. The relationship between factors influence spectral reflectance and external factors have not known clearly yet, especially forest vegetation on Java area.
The Objective of the research was to assess factors that influence spectral reflectance especially teac forest in KPH Telawa District, Perhutani, Central Java Province. A multipectral supervised maximum likelihood classification algorithm and multiple linear regression analyses were used to extract digital Landsat TM data of KPH Telawa District area. Dependent variable is spectral reflectance values of teac forest vegetation and independent variable is factors influences spectral reflectance, such as, vegetation age, topographic, slope aspect. percentage crown closure and slope steepness. Spectral reflectance values are read from digital TM data from the screen that classified using maximum likelihood algorithm with ILW1S software on the IBM PC/AT compatible. Secondary and complementary data were obtained from Perhutani Telawa District and field survey.
The results from the multispectral digital classification was 19 classes of teac forest vegetation of Perhutani Telawa District successfully classified, i.e.,: Group of KU I, KU IV. KU VI, KU VII, KU VIII, Combined KU II, Ill, V, Sonokeling class, Mahoni class I, Mahoni class 2, Mahoni class 3, Open area I, Open area 2, Open area 3, Open area 4, Akasia class dan Kayu putih class, Burned area, Shrub and bush, Grasses and Low increarnent ten forest class with total classification accuracy of 96%. 0.3% of the total pixel were failed to classify. Based on statistical analyses using multiple linear regression the results was topograhic and slope aspect factors influenced significantly to the spectral reflectance of teac forest vegetation on band 2, 3, 4, 5 and 7 of Landsat TM digital data The Crown closure and vegetation age factors influenced significantly to the spectral reflectance of teac forest vegetation on band 2, 1 4. and 5 of Landsat TM digital data. While slope aspect influenced to the all bands used of this research with there is an interaction between topographic and slope aspect. The research conclusion was spectral reflectance of teac forest vegetation influenced strongly by all variable used,i.e.,
|
|---|