%0 Journal Article
%T Keskmise ruumilise lahutusega talvistelt satelliidipiltidelt tehtavate lageraiealade pindalahinnangute t psust m jutavad faktorid
%A Julia Budenkova
%J Forestry Studies / Metsanduslikud Uurimused
%@ 1736-8723
%D 2008
%I 
%R 10.2478/v10132-011-0052-3
%X The objective of this study was to investigate the influence of attributes, describing clear-cut patch size, age, shape, nearest neighbours and habitat conditions on classification results on medium spatial resolution Landsat Thematic Mapper (TM) satellite images. Image differencing was the tool to reveal significantly changed areas in forests as forest clear-cuts. The Landsat TM satellite image pair used in the study was imaged in late winter in March in 1996 and in 2005 respectively. The study area was V ru State Forest District in South-East Estonia. Clear-cuts as significantly changed areas in forests were discerned from non-changed areas with thresholding of a two-date difference image. The results of the statistical analyses showed that the strongest influence on classification results had the attributes like "the arithmetical mean of clear-cut patch pixel values on the difference image" and "the relative boundary length with forest" (Table 1). Minor influence had "the relative boundary length with coniferous forest" and "clear-cut area to perimeter ratio". The age of clear-cut (i.e. years passed since clear-cut logging) showed no statistically significant influence on classification results provided that the age of clear-cut does not exceed 10 years. There was also no statistically significant influence of habitat conditions on the satellite images made in plain snow cover conditions. This finding supports the classification of clear-cuts on winter images without any prior stratification according to habitat conditions. The set of influential attributes remained the same when the classifications were performed on two more liberal and on two more conservative thresholding levels (Table 2). From this we may conclude that no clear-cut parameter is more sensitive than another to erroneously determined threshold values. Landsat TM winter images in bands 1-4 of the visible and near infrared spectral regions were found almost equally suitable for clear-cut area estimations (Table 3). The difference in the area of a single clear-cut patch represented in the forestry database to that classified from a Landsat TM image was about a sixth of the patch size. This implies the utility of medium resolution satellite images in clear-cut activity assessments in particular areas but not so much the applicability of these images for single patch area estimations.
%K remote sensing of forests
%K Landsat TM winter images
%K forest change detection
%K forest clear-cuts
%U http://versita.metapress.com/content/50203845335257h3/?p=40ebe27e0a9d4ee1b9e0c57f7b6a621a&pi=1