The Research Of Forest Ecosystem Health Evaluation On Wolong Natural Reserve

Health assessment of forest ecosystem plays an important role and significance for the protection and management on nature reserves. This paper takes the Wolong National Nature Reserve as the research object. It mainly uses Forest Resource Inventory Data in the research area and BP neural network theory, combining field investigation, established the forest ecosystem health evaluation index system and evaluation model of forest ecosystem health. The research takes the subcompartment as the evaluation scale, evaluated forest health in Wolong Nature Reserve, based on the GIS technology platform, assessment of forest ecosystem in this area and display the health status, it is not only beneficial to the management and development of the protected areas, but also provides a reference for the selection of the local wild panda habitat. The main conclusions are as follows:(1)Put forward and using the T-R-E-E analysis method, we constructed an index system, T-R-E-E represents respectively typical, resistance, energy, environment this 4 essentials elements, based on it, the author initial established the forest ecosystem health evaluation index system of Wolong Nature Reserve, (including giant panda habitat suitability, species diversity, forest pests and diseases etc.) these 9 classic indicators in total. Then qualitatively screened 19 indicators including slope, composition, dominant species etc. that can describe these nine specific indicators according to Forest Resource Inventory Data, after that used principal component analysis method quantitative screen indicators, in the end, determined the forest ecosystem health evaluation index system based on subcompartment of Wolong, the structure of the index system is 3 layers 4 elements and 6 specific indicators.(2)Combined with the entropy method and AHP method, weighted average evaluation index weight. In the element layer, the impact on forest ecosystem health of Wolong Nature Reserve from large to small is:energy> resistance> typical> environment, the weights were 0.3707,0.2733,0.2674,0.0887. In the index layer, the impact on forest ecosystem health of Wolong district from large to small is:dominant species> average DHB> composition> average tree height> soil types> slope, the weights were 0.2733,0.2479,0.1980,0.1228,0.0694,0.0887. Studies show that the dominant species is the most important indicator of the impact on Wolong district forest ecosystem health. It reflect the important degree of forest fire, this is consistent with the actual local situation.(3)Using MATLAB software to construct the structure of BP neural network model with 6×10×5, for the forest ecosystem evaluation of the Wolong Nature Reserve. Among them, there are 1583 high quality subcompartments,3842 healthy,826 sub-health,208 unhealthy and 21 disease, the area is 44760hm2,112780hm2,33620hm2,8242hm2,598hm2 respectively, accounted for the total area of the study area 22.38%,56.39%,16.81%,4.121%,0.299%. The model results show that the overall health status of forest ecological system in the research area is generally good.