| Abstract | Changes in the biophysical, social and economic environment are critical issues for watershed development and planning. The impact of these changes will affect to people in various ways, depending on their available resources and opportunities. In order to develop sustainable watershed management, agriculture system and food security, effective land use monitoring is essential. It is necessary first to assess the land use pattern and its suitability for farming, so that protective measures can be taken and better decision-making can be achieved for policy planning and implementation. In the Mae Tun Watershed of Northwestern Thailand, the environment has been degrading rapidly over the last 30 years due to the commercial logging of hard woods and forest clearing for agriculture. The demand for food and fuel wood is increasing day by day because of high population growth. The net result is that the household food security is threatened due to diminishing availability and use of forest food resources less income earning opportunities for the rural poor and increased burdens on households in meeting their basic needs. In working out solutions to these problems, this piece of the study has attempted to use methodologies such as Satellite Imagery Analyses (SIA) and Geographical Information Systems (GIS) for investigating land use issues related to small-scale farming. The research results demonstrate that SIA and GIS are suitable techniques to monitor and manage natural resources and secure food for households. The analysis of land use changes provides information on land degradation due to increasing deforestation and expansion of shifting cultivation for rice and cash crop productions. The forest areas have decreased due to population rise and increase demand for food supply. The forest areas have been converted into degraded/open forest. It was found through questionnaire survey that crop yields have declined due to soil loss and nutrient depletion under shifting cultivation practices. The main problem in SIA for generating land use maps is the confusion of spectral responses from different features. The area is always covered by cloud, hence prohibited obtaining cloud free data during wet and dry seasons to distinguish deciduous and evergreen forests. Various band combination techniques and unsupervised classification have been tested to improve accuracy of the land use maps. The DEM, prepared from the topographic map was combined with image data for enhancing the accuracy. The result of the land use maps classification has improved greatly. Land degradation, soil erosion in particular, is a serious problem in the area, which is responsible for low crops yield due to reduced soil quality. The average soil loss of the area was higher than the permissible soil loss of 12 tons per ha per year, however only about 9 percent of the total cultivated area are affected by severe soil loss in terms of area coverage. The combined methodology of SIA and GIS has been developed for mapping land suitability classes for small-scale farming based on the available data for different factors taken into account in this study. Land use sustainability in watershed area implicitly depends on biophysical and a socioeconomic factors acting on with different intensity. It is not possible to take into account of all the factors involved in sustainability modeling of a given land use, due to huge amount of data involved and resources required. Land use sustainability can be modeled based on those factors, which indicate significant relation with land use. In case of Mae Tun watershed, there were twelve significant variables of both biophysical and socioeconomic nature. Of the biophysical nature are the land quality or land characteristics, and erosion severity, where as the sources of multiple cropping, labor availability, farm holding size, land tenure, production sold, credit, family income, off-farm income, accessibility, and education are the socioeconomic variables which were found significant in expressing the land use sustainability. Sustainability should be defined for specific purpose and time. Evaluation of prevailing situations of the study area indicated that about one fourth of the total agricultural area or twenty-five percent meet the sustainability requirements, of which none of the mapping units were found to have long-term sustainability. Planning for sustainable watershed land use, it is required that the developed land use alternatives and management practices that aim to protect forest area and to improved household food security for the study area under the scenario likely to occur in the future. The goal of the scenario named as "food production maximizing and optimizing" is to maximize and optimize the production for increasing the household food product while ensuring the conservation of forest resource and other adjacent ecosystems. Maximization and optimization, however looks simpler in terms of a process or technique, under minimal or partial intervention of policy measures towards soil and water conservation. It is meant here in the sense that emphasis be placed to protect the ecological quality of the land or soil, and thus improve the sustainability gradually. The suggested land use alternatives for the study area focuses on extending the sustainability levels of the land units by improving the biophysical status or land quality through cultivation according to land suitability and adopting appropriate agronomic measures, such as crop rotation, legumes inter-cropping, contour plantation, cover crops and mulching. It is also taken into consideration while developing land use alternatives that the level of household food product and farm income be maintained or improved. |