Study On Data Fusion And Models Research In The Hardwood Resource With Hunan Province

Forest resource data that it mains through the national annual forest inventory as continuous forest inventory and forest management inventory referred to two types of surveys, due to the different sets of survey data interval, sampling methods and accuracy requirement is different, leading to there are two sets of forest resources data for the same area. In recent years, the state forestry administration of forest resources management should be made to actively promote the reform of forest resources monitoring system optimization, and gradually built service efficient integration of forest resource monitoring system, and ultimately the national and local monitoring of forest resources," a game of chess," Forest Resources " a number of " distribution of forests," a figure." Integrated monitoring of core forest resources is a class of inventory to national and local Inventory Data effective data fusion. The key of forest resource data fusion is to make fully utilize the advantages of two types of data, accurate to choose appropriate background and suitable model.Fully analyzed in this study based on previous data fusion model for forest resource modeling and stand growth model, by the Hunan mainly hardwood, including pure broadleaf forest, mixed forest and mixed needle as the research object, using the forest resource continuous investigation of Hunan province from1994to2009, and2004and2009hardwood sample data from the use of stand growth model, built mainly in Hunan hardwood diameter, high and the number of trees per hectare change models. The main contents and conclusions are as follows:(1) The Hunan province according to the different need for growth and hardwood climate and species natural distribution is divided into the North, Middle East with Lake Basin hilly area, hilly area and Western mountain three partitions.(2) Each district according to different hardwood species groups, respectively, with six classical growth model for the growth of hardwood tree height to fitting each equation by comparing the coefficient of determination R2of each equation fitting and residual error MSE, determine the most appropriate for the partition hardwood species group of tree height growth optimal model of tree height growth model seek time first derivative, has been the district of different hardwood species group annual increment of tree height model.(3) For each partition hardwood pure forest by calculating the average diameter of each variable with a simple Pearson correlation coefficient and multivariate multicollin earity test, select the diameter of the growth is mainly due to the impact variables:average age, average tree height, stand density index, with a multi-linear regression models drawn each partition hardwood forest diameter growth model Diameter growth model for△D=Dt+1-Dt operation to get the each partition hardwood forest different diameter of successive growth model.(4) The each district through using the method of hidden to the growth of the age model to modeling of average diameter, got the average diameter of the partition mixed needle forest change models.(5) Data from2004selected sample each partition to correspond to sample hardwood data of2004and2009, respectively, to establish number of trees into the hardwood sector, and arrive the hardwood forest per hectare model.(6) Hunan with Jianghua County forest resources subcompartment data in2007, for example, using the partition model system,which study on implements the county hardwood results with the results of the traditional data fusion in2013.