Conclusion

The role of fuel wood in meeting a region’s requirement in the form of energy has increased the interest in mapping the extent of fuel wood availability in a region. In addition to environmental benefits, fuel wood offers many economic and energy security benefits. The mapping of fuel wood has continually been refined over the past decade or so as the methodologies, knowledge and the technology available for such methods and for such mapping have improved. The possibility of quickly accessing and processing large spatial databases offers a tremendous improvement. In spite of rapidly advancing computer technology and the proliferation of software for decision support, GIS and Remote Sensing can be used for fuel wood mapping.

Fuel wood mapping shows that Kolar is a bioresource deficit district. Kolar depends mainly on non-commercial forms of energy. Non-commercial energy constitutes 84%, met mainly by sources like firewood, agricultural residues and cow dung, while commercial energy share is 16%, met mainly by electricity, oil, etc. NDVI computation considering NIR and R bands of MSS data shows that 52.5 % land cover in Kolar district is under vegetation and 47.5% is under non-vegetation (such as soil, water, etc.). MLC is the best for classifying the remote sensing data with an accuracy of 94.67% which indicates that 46.69% is agricultural land, 42.33% is wasteland (barren land), 4.62% is built up, 3.07% plantation, 2.77% natural forest and 0.53% water bodies. This is based on the accuracy assessment matrices (confusion matrices) The area under vegetation (Agriculture, Plantation and Forests) ranges from 27.91% (Gudibanda), 33.35% (Bagepalli), 36.94% (Gauribidanur), 48.41% (Chintamani), 52.66% (Kolar), 53.56% (Sidlaghatta), 54.51% (Mulbagal), 56.36% (Chikballapur), 65.62% (Srinivasapur), 66.76% (Bangarpet), and Malur (71.71%).

Mapping of fuel wood tree Prosopis juliflora was carried out considering village as a sampling unit. Sub-sampling units were selected such that it includes variability in Prosopis juliflora cover. Overlaying this field data with the remote sensing data corresponding to the same region helped in identifying the spectral response pattern of the species. With the identification of spectral response pattern for the species (considering density and age), mapping was done for the entire Kolar taluk as well as for the neighbouring Gauribidanur taluk with the help of merged remote sensing data (LISS III MSS and PAN) in Kolar district. The spectral pattern of Prosopis juliflora was in the range of 98-105.  The map of Prosopis juliflora was verified using GPS and the accuracy of mapping was 88% and error due to omission and commission was 12%. The technique used in this work would aid regional energy planners and foresters in management decisions which are largely a function of monitoring forest stand volume by species.