Study On The Estimation Of The Forest Stock Volume Based On The Multi-source Data Of The Forest Second Type Inventory

The main task of the forest inventory for planning and designing(referred to as the forest second type inventory)is to find out the types of forest resources and the forest types identification and access to quantitative forest stand parameters.The survey methods and investigation techniques of the traditional forest second type are backward,and the survey period is long(once every 10 years).It takes time and effort to form annual dynamic data,and it is difficult to meet the dynamic monitoring needs of forest resources under the new situation.Therefore,there is an urgent need to rapidly monitor the forest second type inventory through the modern Internet of Things formation technology,big data remote sensing technology and 3S technology to improve the efficiency of investigation and monitoring.On the basis of previous studies,combined with the theory of multivariate statistics,this paper uses the subcompartment data of the forest second type inventory in 2014 and the data of the sample plots of the national forest inventory in 2018 in Keqi(keshi ketengqi)and focuses on the GF1 and the GF2 and Landsat8 satellite data and 3S technology and other modern means to study the extraction of the main stand factors of the forest second type inventory according to the actual needs,mainly subcompartment forest stock volume and subcompartment forest category estimates;based on this,using C#language,ArcGIS Engine and plug-in technology,a matching subcompartment factors extraction system is developed and verified experimentally.At the same time,the established KNN model is used to invert the Tuohe Forest Farm,and the inversion results are compared with the measured values in precision.Through the systematic verification of the 2194 subcompartment data in Keqi(keshi ketengqi),it is concluded that the weighted average relative accuracy of the subcompartment stock volume extraction results is 81.8%,and the relative average precision of the subcompartment forest type extraction results is 72.65%,and the subcompartment stock volume and the spatial distribution of forest categories are generated.When comparing the inversion results with the measured data,it is concluded that the average relative accuracy of the 1101 forest land subcompartment stock volume extraction results in the Tuohe Forest Farm is 88.04%,and the relative average precision of the subcompartment forest type extraction results is 70.91%.The extraction effect is good and meets the requirements of the forest second type inventory technical regulations,can be widely used in the national forest second type inventory production practice.