Evaluation of Forest Productivity and Governance on the Preservation of Tropical Rain Forests in Kalimantan Using The NGWR-TS Nonparametric Geospatial Method

Abstract

Tropical rainforests in Kalimantan are experiencing very rapid degradation, until 2019 the area of forest in Kalimantan has continued to shrink, so it is necessary to evaluate the handling of tropical rainforests by increasing forest conservation in Kalimantan. The purpose of this study is to carry out an experimental investigation about factors that support the preservation of tropical rainforests in 56 Districts/ Cities in Kalimantan. Data that used in this study were secondary data from Forestry Agency Indonesia. The data were 56 Districts/Cities from 5 provinces in Kalimantan. Forest Data in Districts/City were influenced by spatial dependence and spatial heterogeneity. Spatial dependence and spatial heterogeneity could be overcome by using one of the spatial statistical methods, namely the NGWR-TS Nonparametric Geospatial method. The method is the development of a nonparametric spline truncated regression that considers geographical or spatial factors. The results of the study obtained the best NGWR-TS modelling using the Gaussian kernel geographical weighting function with a minimum CV was 25.15. Using NGWR-TS with order m = 2 and point knot 2 with a minimum value of GCV was 1.57. The model had a value of R2 = 94.12, so variables that significantly affected the preservation of tropical rainforests amounting to 94.12% were the area of Protected Forests, nature reserves and conservation, Production Forests, Area of each Districts/City, Population Density, Rate Economic Growth and Regional Development Index.