The spatial distribution and size of forest regeneration areas are the basis for an analysis of where climatically unfavourable sites for forest regeneration are likely to occur if the area is clear felled. Knowledge about thermal variations is of interest in forestry, as avoidance of surface temperature extremes is important to regeneration success.

Forest clear-cutting is mapped by means of multitemporal analysis of Landsat MSS data. The method is based on comparing the reflectance in MSS band 2 (0.6-0.7 pm) from a forest area before and after clear-cutting. The data used are two Landsat MSS images from October 1978 (Landsat 2) and August 1983 (Landsat 4). In order to examine the relationship between owner and extent of clear-cutting, the spatial distribution and size of clear-cuts are linked to a digital map of land ownership. By overlay analysis the area of every clear cut is estimated for each owner. An attempt is also made to measure the amount of contiguous open areas resulting from clear-cutting adjacent to water, arable land, mires and older clear-cuts (young plantations). These open areas are computed by using filtering operations on raster images.

For mapping local-scale temperature variations in wooded land (forest, regeneration area and mire) the utility of both daytime and nighttime Landsat-5 TM thermal images, combined with digital elevation data, are analysed. The analysis includes three main parts: 1) studies of satellite-derived surface temperatures for different types of land cover, 2) studies of the influence of topography upon the temperature distribution of clear cut areas with emphasis on the nighttime data and 3) detection of the most important parameters causing low surface temperatures on clear cut areas, by integrated analysis of both physical data and data from the forest management plan. Finally the results from the various analyses are used for verification of the reliability of the results.

The results showed that 6% of the area covered by forest in 1978 had been clear-cut during the period 1978-1983. Within the region (3125 km^) there were 29 fresh clearcuts £ 100 ha, of which seven were > 200 ha. Clear-cuts of the size 1-10 ha however dominated. The amount of contiguous open areas expanded during this period, which meant an increase of areas vulnerable to local climate changes with accompanying potential regeneration problems. The results of the analyses of the thermal images showed that Landsat-5 TM thermal images recorded at night during the time of early sunset provide good possibilities for detecting low surface temperature areas on open ground at the local-scale, although neither the time of recording nor the spatial resolution was optimal. The thermal pattern of regeneration areas shown in the nighttime thermal image was highly correlated to the local topography. The regeneration data available supported the assumptions that clear cut areas with low temperatures, located in the nighttime thermal satellite image, are difficult to reforest and that thermal nighttime data, combined with digital elevation data, can be used to indicate unfavourable sites for reforestation of clear cut areas.

With the relations shown between surface temperatures derived from a Landsat TM nightime thermal image and the terrain parameters elevation, aspect and slope angle, the use of a digital elevation model as a complementary tool in site-related forestry for risk assessment of regeneration failures seems promising. This has been presented in the study in form of a risk assessment map.