Font Size: a A A

Study On Forest Ecosystem Health Diagnosis And Ecological Effect Of Landscape Pattern

Posted on:2021-11-26Degree:MasterType:Thesis
Country:ChinaCandidate:W C ShiFull Text:PDF
GTID:2543306725999199Subject:Geography
Abstract/Summary:PDF Full Text Request
Resources and environment are the foundation of human survival and development.In recent years,China has not paid much attention to the development mode and the utilization of resources while developing the economy rapidly,which leads to the drastic consumption of resources and causes great pressure on the ecological environment.Forest is a key factor in regulating ecological balance,and plays a major role in maintaining the ecological environment and supporting human survival and development.Therefore,to diagnose the health status of forest ecosystem,explore the ecological effect between the landscape pattern index and the health of forest ecosystem,and make rational use of the controllability of landscape pattern index is of great significance for the scientific guidance of land space planning,landscaping and landscape layout.Based on Landsat 8 remote sensing image,DEM,and second-level survey data of Xiamen forest resources,this paper constructed a Xiamen forest ecosystem health diagnosis indicator system,which aimed to reveal the current status of the forest ecosystem health in Xiamen and further explore the mechanism between forest health and landscape structure.To provide guidance for the regulation of forest health in Xiamen.Research indicates:(1)Based on Landsat 8 OLI remote sensing image,DEM and Xiamen forest resources survey data,13 indicators were extracted to construct a forest ecological health diagnosis system.After the establishment of the index system,Lagrange multiplier method was used to combine and optimize the weights determined by principal component analysis and entropy weight method,and the forest health diagnosis model was finally established as follows:FEHDM=0.0318X1+0.0532X2+0.0804X3+0.0978X4+0.0623X5+0.1142X6+0.0535X7+0.0352X8+0.0313X9+0.2129X10+0.0559X11+0.1038X12+0.0677X13.(2)Using the natural discontinuity method,the health status of the forest ecosystem in Xiamen was divided into four levels:healthy,sub-healthy,medium-healthy,and unhealthy.The results show that the health-level forest accounted for 22.79%,the sub-health-level forest accounted for 19.34%,the medium-health level Of forests accounted for 41.66%,and unhealthy forests accounted for 16.11%.Visibly,more than 80%of the forests were above the medium health level,and the forest ecosystem in Xiamen was in good health.(3)Extracting the characteristics of texture,spectrum and index of image,the forest species in Xiamen were classified into fir,masson pine,broad-leaved forest,moso bamboo and other species by decision tree and random forest classification methods.The results showed that the classification effect of random forest was better than that of decision tree.According to statistics,the area of masson pine was 188.19km2,which was about 30%of the forest area of Xiamen city.The area of broad-leaved forest was 160.19 km2,accounting for 24.95%.The proportion of fir and moso bamboo was about 10%,other tree species accounted for about 26%.Therefore,the largest proportion of forest tree species in Xiamen was masson pine and broad-leaved forest.(4)In this paper,patch density(PD),edge density(ED)and mean patch shape index(MSI)were selected to analyse the landscape pattern characteristics of forest ecosystem in Xiamen City and explored the relationship between landscape pattern index and forest ecosystem health from urban scale,township scale,block scale and tree species type scale.There was a positive response relationship between ED and forest health at urban scale,and a negative response relationship between ED and forest health at township scale and block scale.MSI on the scale of tree species type presented a positive response relation with forest health,while MSI on the scale of urban area and town presented a negative response relation with forest health.The research showed that the forest ecosystem in Xiamen was in good health,with more than 80%of the forest above the level of medium health,but it was also faced with a large risk of deterioration.There was a significant correlation between landscape pattern index and forest ecosystem health on the scales of urban area,township,block and tree species.Combining with the controllability of landscape pattern index,it was expected to provide reference for ecological planning in the later period of Xiamen and provide theoretical basis for the improvement of forest ecosystem health in Xiamen.
Keywords/Search Tags:forest ecosystem health, landscape pattern, ecological effect, remote sensing, Xiamen
PDF Full Text Request
Related items