การประยุกต์ใช้เทคโนโลยีภูมิสารสนเทศเพื่อการวิเคราะห์แหล่งที่อยู่อาศัยของสัตว์ป่า

Abstract

This study investigates “the utilization of geospatial technology for wildlife habitats in Mae Ping national park.” It has three primary objectives: 1) to classify the utilization of land use in the park area; 2) to study the distribution and density of wildlife and; 3) to analyze the relationship between environmental factors and wildlife habitats in the park area. This study is observational research aiming to examine the causal correlation between the distribution of wildlife and their habitats. The main research tools in this study include the global navigation satellite system, remote sensing, geographic information system, and statistical methods. The land use classification from the Landsat 8 satellite, through the analysis supervised classification, found that approximately 95% of the Mae Ping national park is covered by the forests and water resources, which are suitable for wildlife habitats. Most of which are apparently the deciduous dipterocarp forest and followed by the dry evergreen forest; while the small area of mixed deciduous was found in the highlands. Connecting the investigation of wildlife distribution and density through the wildlife trace location, data was collected by SMART Patrol Monitoring Center, Forest Conservation Area 16 (Chiang Mai). The findings revealed that Mae Ping national park’s wildlife is clustered, especially in the central-eastern part of the area with density of 1.80 per square kilometer. When considering the type of wildlife, it was disclosed that wild boars are the most common wild animals, followed by barking deer and wild deer, respectively. The outcomes based on the analysis of the causal relationship between environmental factors and wildlife habitats were found to comprise spatial relationships through cluster analysis, which employed the grouping technique (or Partitioning Around Medoids–PAM). The grouping of wildlife habitat could be classified into three main areas; Chi-Square statistics was tested the statistical relationship. The results indicated that the distance from the water resource, the distance from saltlick, altitude, slope, the distance from road, and the distance from the tourist attraction has a significant relationship to the wildlife habitat at the statistical significance of p<0.05. The habitat simulation of each wildlife species can be modified from the multivariate conditional logistic regression; it was unveiled that wild deer habitat is seen in a deciduous dipterocarp forest and close to the saltlick. The barking deer’s habitat area appears in the highland area, low slope, close to the water source, but far from the road. The habitat of the wildfowl is far from the tourist attraction. The habitat area of hog deer is low on slopes and far from the roads. The habitat of the hedgehog is close to the water resource but far from the saltlick. The habitat of the antelope is close to the pond but far from the road. Wild boar habitats are close to the water source in areas with low slopes. The habitat of palm civet is not in a steep area.