Romote Sensing Estimation Of Main Forest Survey Factors Of Cypress Forest In Hilly Area Of Eastern Sichuan Based On SPOT6 Image

It is important for scientific management and management of forest resources to know well forest environment and tree growth dynamics in time and predict the changing trend of forest resources.Cypress is one of the three evergreen conifer species widely distributed in Sichuan Basin,and plays an important role in maintaining regional ecosystem stability In this paper,SPOT 6 high-resolution remote sensing image is used to estimate the main forest survey factors of the average height,plant number density and hectare accumulation of cypress forest,and accurately know well the dynamics of cypress resourcesin the hilly area of eastern Sichuan,and provide basic data for its management.The paper takes the northern part of Yuechi County in Guang'an as the research area,and uses SPOT 6high-resolution remote sensing image as the data source.Combined with the ground survey data,the remote sensing image is preprocessed and supervised and classified to obtain the distribution information of the cypress forest in the study area.The single-band,band ratio,vegetation index,texture feature factor and topographic factor were selected as independent variables.Optimization of independent variable factors by variable projection importance criterion(VIP).The partial least squares regression method is used to construct the remote sensing estimation model of the average height,plant number density and hectare accumulation of the cedar forest.The main findings are as follows:(1)Elevation,the topographic factor,has the greatest impact on the average height,plant density and hectare accumulation estimation model,and participates in the prediction of all models,and is significantly positively correlated with them Texture information has a significant impact on the estimation of plant number density.Spectral information has a significant impact on the estimation of stand average height and hectare storage.(2)A multivariate linear regression model for the average height of Berlin stands in northern Yuechi County was established:H=2.66374-0.00431485×B1-0.00239331×B2-0.00154697×B3+5.0322×NDVI+0.0483273×RVI+0.0366861×Slope+0.00414446×Elevatio n,model R~2 is 0.7225,RMSE is 0.8161m,RE is 8.60%,overall accuracy is 91.39%.(3)A multivariate linear regression model for the number density of cypress stands in northern Yuechi County was established:N=181.437-0.309496×B2-0.39956×B3-472.537×CONT+205.704×COR-472.537×DIS-205.025×ENT+945.074×HOM+339.314×SM-1121.86×VAR+0.0371742×Elevation,model R~2 is 0.8288,RMSE is 134.1619 plants/hm~2,RE is9.71%,overall accuracy is 90.29%.(4)A multivariate linear regression model for the hectare accumulation of Berlin in northern Yuechi County was established:V=-36.0163-0.0646079×B1-0.0413764×B2-0.0483753×B3+54.6415×NDVI+0.344897×RVI+1.34407×Slope+0.109866×Elevation,the model R~2 is 0.8893,RMSE is 9.3372m~3/hm~2,RE is 15.74%,and the overall accuracy is84.26%.