| With the development of social economy,the influence of ecological environment resources on human being is more and more strong.The protection and development of forest is an important task in the protection and construction of ecological environment.At present,the domestic scholars mainly use the traditional sample survey method to study the forest health status,because the sample survey method requires a lot of manpower,weakness,and the time spent,so it can't respond to timely and effective changes in forest health.With the development of hyperspectral remote sensing technology,in the marine,geological,ecological,military,and many other areas to play its advantages.Hyperspectral data is characterized by a large number of bands,accurate and detailed extraction of terrain information.China launched its first environmental and ecological monitoring satellite in 2008,and the HSI hyperspectral data has the advantages of good spectrum,high spectral resolution and abundant information,and HSI hyperspectral data provide a new data source for forest health monitoring.Firstly,pretreatment of hyperspectral data,then extract the vegetation index which can reflect the growth of forest trees and plants and select the vegetation index by sensitivity analysis method.Secondly,to construct Forest Health Assessment Model of Dulaer Forest Farm in Aershan.Finally,the forest health index and the research area of the forest stand,terrain,climate,soil and other factors were divided into zonal statistics,and analyze the relationship between the two.This paper mainly draws the following conclusions:(1)From the aspects of vegetation coverage,photosynthesis intensity,vegetation spectral characteristics and so on,the vegetation index information was extracted from hyperspectral data,and the vegetation index was selected by sensitivity analysis.The results show that the vegetation index: Enhanced Vegetation Index,Normalized Difference Vegetation Index,Ratio Vegetation Index,Red Edge Normalized Index,Red Edge Index,Sum Green Index and Red Edge Position Index.Based on these,Forest Health Assessment System of Dulaer Forest Farm in Aershan City was constructed.(2)Invite experts to rate the various vegetation selection index,expert opinion after reunification by AHP to the vegetation index weight,and the assessment of forest health Dulaer Forest Farm in Aershan city model is established by using comprehensive index method,according to the model results obtained by using the Arc GIS software breakpoint classification tool natural forest health index is divided into four grades:health,sub-health,health,good health,and produce the spatial distribution map of forest health Dulaer Forest Farm in Aershan city.(3)The results obtained from the Forest Health Assessment Model established by hyperspectral remote sensing are compared with those obtained by the traditional sample survey method.It shows that the results obtained by this model are significantly related to the traditional sample survey method,The coefficient of determination R2 was 0.7849,P <0.01.The forest health index derived from this model is less than the traditional survey,but the trend is consistent.(4)Based on the data of forest stand,terrain,climate and soil,the relati onship between the factors and forest health index was analyzed.The corr elation analysis showed that the altitude,forest age,Precipitation betwee n the forest health index has significant positively correlated(P <0.01).T he Soil thickness was positively correlated with forest health inde x(P < 0.05).There was a negative correlation between a temperature an d forest health index(P <0.01).At the same time,Overlay analysis of the survey factorsand forest health index,and obtain the growth of forest tree s and plants under different conditions. |
PDF Full Text Request