| Abstract | Tree resources outside forests (TROF) is becoming increasingly important for economic and ecological functions. Its assessment can be better done while we use both ground base tree measurements as well as remote sensing tools. In the present study, Line Intersect Sampling (LIS) method and Landsat TM Satellite image classification with maximum likelihood algorithm were used for field measurements and spatial distribution of TROF respectively at Yombo and Chasimba villages of Tanzania. Consistency of the results was checked by comparing the area covered by TROF and cashew nut cover in the study area calculated from both the methods. LIS method is found to be better suited for TROF assessment as there are only 25% of plots without trees, increasing the average number of trees per plot. Landsat TM image cannot give high accuracy in classification of TROF because of large pixel size (30 x 30 m). In case of unavailability of high- resolution satellite imagery, classification accuracy can be enhanced using ground truthing data as well as visual interpretation of images implying local knowledge. TROF has covered about 41% of the land which shows the richness. Although there are limited new plantations, TROF is rich in stem volume per hectare 50 m3/ha. Thus, in the condition of massive deforestation in nearby Ruvu North Forest Reserve, TROF of the study area can play vital role for the better use of tree products and consequently livelihood of the people. As deforestation continues in Nepal, the Nepalese TROF can play a valuable role for enhancing sustainable development and people's livelihoods. By extracting the Tanzanian experience, we can apply the line intersect sampling method for TROF assessment in Nepal. |