Study On The Monitoring Models For Forest Resource Volume

Forest resources volume change is one of the important content of forest resource monitoring, andis also the main basis for forestry management. The second investigation of forest resource is to find outthe present forest management condition owned by state units and county-level, and the main purpose isto make up formulation of annual forest cutting quota, design forestry project planning, make upformulation long-term development planning, prepare forest management plan, evaluate the effect offorest management, manage forest resources assets，so that to provide the scientific basis.The traditional first and second investigation costs much time, and requires a lot of manpower andmaterial resources. Furthermore, such investigation surveys every5-10years, and the long time intervalmakes it difficult to meet the changing forest management needs. In view of this, this study designed todredge up the data from second investigation, and construct some forest volume estimation modelson the basis of statistical methods and growth equations so that to estimate the annual forest resourcevolume quickly and accurately, and to provide decision-making basis for the forestry managingdepartment.This study uses the data from second investigation of forest resource in Huangyan District, TaizhouCity, Zhejiang Province in2007. Combined with statistical forecast as well as growth model forecastingmethod, we construct models to estimate volumes of the main tree species in Zhejiang Province, so thatto monitor the dynamic changes of forest resources, in order to monitor forest resources timely.Based on the data of small class in this second investigation, we constructed volume regressionmodels of Masson pine, Sugiki, Kashiwagi Mioki, to research the forest resources in Huangyan Districtof Taizhou City, Zhejiang Province. These models use the growth equation and multivariate statistics,the stand volume as the dependent variable, and the tree growth factors and stand for the relevantvariables. After the model has been constructed, we evaluate the precision and feasibility of thesemodels. Since the result meets the requirements in the premise, we used the models to predict the nextyear stand volume of Huangyan district. The main work and conclusions are as follows:First of all, we determined the dominant tree species, and analyzed the unit forest volume of everydominant tree species according to the data of this second investigation, and rejected some abnormaldata which were not suitable for modeling. Second, we found out a linear relationship existing between unit volume and variables through theanalysis by dot method and correlation coefficient. And the further linear regression analysis showed thefollowing estimating equation between unit volume and variables.（1）The linear relationship of Masson Pine unit volume estimation model as:Y=40.361+11.452YBD+7.920LN-1.202HB+0.734PWII-0.649YP-0.564PWI;（2）The linear relationship of Chinese Fir unit volume estimation model as:Y=47.102+11.108YBD+7.286LN-1.831HB-2.539TR;（3）The linear relationship of Cypress unit volume estimation model as:Y=53.436+13.198YBD+5.200LN-2.511HB-2.050PWII.Third, we used accuracy test and F test to evaluate the three dominant tree models, and found outthat the average accuracy of the models were significantly higher than80%, and F test result was alsovery good, so these models can be used to estimate the forest resources volume in Huangyan District.On the basis of above study, we used these models to estimate the forest resources volume of thesubsequent years.