Research Of Forest Health Assessment Based On The Complex Network

The health evaluation research of forest ecosystem is emerging field in the study of forestry and ecological protection, which plays am important guiding role in improveing the level of sustainable forest management. In recent years, the research of the complexity and self-organizing of forest ecosystem by use of complex network technology has been started. The complex network technology applied to the evaluation of forest health is the sublimation of from real physical model to abstract the logical model. Divided the forest ecological system of the large scale into a multilevel network for evaluation and early warning, which can be more objective and comprehensive understanding of the law of development of forest ecosystem health. It provides scientific basis for sustainable forest management.This paper uses the theory of complex networks as a guide and comprehensively used with system ecology, vegetation ecology, graph theory, mathematical statistics, database technology, geographic information systems and other areas of the discipline knowledge, to asseses and forewarns the health of the forest ecosystem. The specific research work and the results are as follows:(1) Through the quantitative selection of factor analysis and qualitative screening of index operability, an index system of forest health evaluation based on sub-compartment scale was constructed, including the integrity, stability and sustainability of three. The index system contains 8 indexes. The weight of each indicator was determined by using principal component analysis. Index is standardized by the relevant standards and formula. The health of sub-compartment was judged by using the comprehensive index method and forest management measures are proposed according to the evaluation results.(2) The landscape patches health index calculation model based on sub-compartment network and the regional forest health index calculation model based on the landscape patches network are constructed by complex network topology analysis and the analysis of the relation between the topological parameters, combining with the landscape edge effect.(3) Adjacent sub-compartment of the same advantage tree species group_ age group (YSSZ_LZ) merge into forest landscape patches. Within the landscape patches, according to the adjacency relations of sub-compartment, to build sub-compartment network and by using sub-compartment network to assess the health of forest landscape patches.(4) The intensity of edge effects between landscape patches was calculated and landscape patch network based on edge effect was builded and by using the landscape patches network to assess the forest health of regional and forest ecosystems health warning model based on the landscape patches network was proposed.(5) Dalingshan forest in Guangzhou as an example of forest health is assessed and forewarned, the evaluation into sub-compartment, landscape patches and area three dimensions. Forest health evaluation results in the study area:Sub-compartment of high-quality, health, sub-health, unhealthy, sick respectively are 139,44,20,11,2 in sub-compartment scale. Landscape patches of high-quality, health, sub-health, unhealthy, sick respectively accounted for the total area 75.57%,10.68%,10.13%,2.86%,0.76% in Landscape patch scale. The area of forest health index is 78.62 and forest health grade for health in regional scale. Forest ecosystem health warning results in the study area:89.14% of the area is green alert and 10.10% of the area is the blue alert and 0.76% of the area is yellow alert and orange and red alert landscape patches is 0. That is to say, there aren’t heavy warning and huge warning area in the study area.With the intensity of edge effect as landscape patch network, construct landscape patch network topology. It laid the foundation for further study of evolution and dynamics complexity of landscape patches network. According to the network topology characteristics analysis, the relation between the topological parameters analysis and landscape edge effect, build the health index calculation model based on complex network, which provides a new way of thinking and new method for multiple scale quantitative evaluation forest health.