Study On Forest Resources Investigation Of Pinus Massoniana Rongan County Based On Kriging

This paper summarized the application status of forest resources survey sampling techniques and forest stock volume estimation,and discussed several sampling techniques which were based on classical statistical theory.Then using Kriging interpolation method to compare and analyze the traditional sampling method and the spatial stratified sampling method,it provided a reference value for sampling method and forest volume estimation.Forest resources survey which based on Kriging method,comprehensively used remote sensing information technology,combining a small amount of ground survey sample data to estimate the forest volume and forest coverage with a view to minimizing the workload of ground investigations.Based on high resolution airborne remote sensing image as the base,on which the layout of 100 m*100 m sampling box,first sampled,sampling results in the process of the sampling as prior knowledge,at the levels of reliability and precision of sampling 95%and 85%on the traditional simple random sampling,systematic sampling and spatial random sampling,spatial stratified sampling,sampling results.Through map of rongan county pine data to compare with the result of traditional space sampling and sampling,spatial stratified sampling was saving time and cost for the most optimal sampling method,analyzed and draw the following conclusion:(1)The spatial autocorrelation degree of masson pine forest area in Rongan County was very significant,which possessed the prerequisites for using spatial sampling survey.The result of Kriging interpolation was used as the stratification cofactor of spatial stratification,and the effect was remarkable.The spatial stratification was divided into three layers.Layer 1 had the largest forest cover,layer 2 had the second largest forest cover,and layer 3 had the least forest cover.(2)The relative error of simple random sampling was 13.5%,the relative sampling error was 12.85%,the relative random sampling error was 16.17%,the relative error of spatial sampling was 10.69%.The relative error of spatial stratified sampling was the smallest,the relative error of traditional sampling was second,and the error of random sampling in space was slightly larger than the relative errors of other samplings.Based on the results of various sampled Pinus massoniana forest coverings,it was consistent with the relative errors of these samples.(3)Assuming that the sampling cost of traditional sampling techniques was 100,the cost of random sampling in space was 82.2 and the cost of spatial sampling in stratification was 29.7.Obviously,the spatial stratified sampling was more cost-effective,followed by the random sampling of space,while the cost of traditional sampling techniques was more expensive than that of spatial sampling.(4)Based on the stratified sampling,it was estimated that the mass of Pinus massoniana forest in Rongan County was 2836400 cubic meters,and the forest volume per unit area was 65.83m3/hm2.According to the data of Guangxi Zhuang Autonomous Region in 2010,the estimation result was in line with the actual situation.