DDF （dry dipterocarp forest） is importantly deciduous forest type in Thailand since it consists of important tree species for timber products and non-timber products. So, people would like to come to use these produ...DDF （dry dipterocarp forest） is importantly deciduous forest type in Thailand since it consists of important tree species for timber products and non-timber products. So, people would like to come to use these products for daily uses in this forest type. The main aim of this study is to evaluate significant biophysical factors for DDF distribution using factor analysis and to model DDF distribution using ENFA （ecological niche factor analysis）. In this study, 13 watersheds of Ping Basin in northern Thailand were selected as the study site based on availability of forest inventory data in 2007 from DNP （Department of National Parks, Wildlife and Plant Conservation）. Basic biophysical data for data analysis included forest inventory data （179 DDF plots）, 10 climatic data, three topographic data, and one soil data. For identification and evaluation of biophysical factors for DDF distribution using factor analysis, the first three factors, namely DDF-1, DDF-2 and DDF-3, had been extracted with 95.35% of total variance. These three components were used to predict DDF distribution based on HS （habitat suitability） with ENFA. In practice, the results were validated with AVI （absolute validation index） and CVI （contrast validation index） with validated forest inventory dataset. This evaluation shows that DDF-2 model is the best HS data consisting of four physical factors （mean annually temperature, mean monthly maximum temperature, mean monthly minimum temperature, and elevation）, which is able to effectively used for habitat suitability for DDF distribution prediction. It was found that habitat suitability for DDF distribution can be classified into four classes including high suitable habitat, moderate suitable habitat, low suitable habitat, and unsuitable habitat. As a result, DDF distributions with high suitable habitat are highly related with DDF forest inventory plots of DNP. Thus, the obtained output can be further used for DDF rehabilitation according to climate and topographic factors.展开更多
文摘DDF （dry dipterocarp forest） is importantly deciduous forest type in Thailand since it consists of important tree species for timber products and non-timber products. So, people would like to come to use these products for daily uses in this forest type. The main aim of this study is to evaluate significant biophysical factors for DDF distribution using factor analysis and to model DDF distribution using ENFA （ecological niche factor analysis）. In this study, 13 watersheds of Ping Basin in northern Thailand were selected as the study site based on availability of forest inventory data in 2007 from DNP （Department of National Parks, Wildlife and Plant Conservation）. Basic biophysical data for data analysis included forest inventory data （179 DDF plots）, 10 climatic data, three topographic data, and one soil data. For identification and evaluation of biophysical factors for DDF distribution using factor analysis, the first three factors, namely DDF-1, DDF-2 and DDF-3, had been extracted with 95.35% of total variance. These three components were used to predict DDF distribution based on HS （habitat suitability） with ENFA. In practice, the results were validated with AVI （absolute validation index） and CVI （contrast validation index） with validated forest inventory dataset. This evaluation shows that DDF-2 model is the best HS data consisting of four physical factors （mean annually temperature, mean monthly maximum temperature, mean monthly minimum temperature, and elevation）, which is able to effectively used for habitat suitability for DDF distribution prediction. It was found that habitat suitability for DDF distribution can be classified into four classes including high suitable habitat, moderate suitable habitat, low suitable habitat, and unsuitable habitat. As a result, DDF distributions with high suitable habitat are highly related with DDF forest inventory plots of DNP. Thus, the obtained output can be further used for DDF rehabilitation according to climate and topographic factors.
