Research On The Compatible Biomass Models Of Pinus Massoniana In Different Regions

Based on the sample data of 151 stains of Pinus massoniana biomass in three provinces of Sichuan, Hubei and Chongqing as an example, the study used the dummy variable method to represent the different origin factors with qualitative code. The key of the research was to study the problem of compatibility of different origin factors and regional factors when the biomass and biomass of each sub model in independent fitting on Masson. The compatibility of the aboveground biomass and biomass of each sub model, aboveground biomass and tree volume model were researched by using the nonlinear simultaneous equations. Then focusing on the model incompatible problem, a dummy variable which represents the regional characteristics in different regions was introduced to establish the whole large area inland biomass and biomass, the sub Compatibility Model of biomass and volume, in order to improve the research methods and techniques of regional forest biomass in large scale model and provide the technical support for the realization of spatio-temporal distribution pattern of China's forest biological quantity real-time monitoring. The results show that:1) When the aboveground biomass and biomass of each sub independent fitting regression model, the aboveground biomass and the sub biomass compatibility model and aboveground biomass and tree volume biomass compatibility model were built, to introduce a dummy variable on behalf of the regional characteristics of the incompatible problems could not only solve the different area model, but also effectively improve the precision of prediction model, enhance the stability of the model.2) When the aboveground biomass and the sub biomass regression model were fitted, to introduce a dummy variable that represents the origin of factors was effective on the crown biomass, branch biomass and leaf biomass model, while it had little effect on aboveground biomass, stem biomass, stem biomass and bark biomass. Therefore, in the crown biomass, branch biomass and leaf biomass independent fitting regression model, to use a dummy variable method to introduce a origin factor in qualitative code is an effective force to improve the accuracy of the model and improve the prediction effect.3) To derive a new nonlinear simultaneous equation directly from the regression model fitted with the aboveground biomass and biomass of each item, is an effective force to solve the problem of compatibility between aboveground biomass and biomass of each sub model, and the ratio adjustment method of the overall prediction effect was better.4) Overall, whether it was an independent fitting model or a compatible model, the trivariate biomass models had the highest prediction accuracy, following by the bivariate biomass models, and univariate biomass models were the worst Among them, the above ground biomass, stem biomass, stem biomass and bark biomass were improved from univariate biomass models to bivariate biomass models, and branch biomass and leaf biomass, crown biomass were improved from bivariate biomass models to the trivariate biomass models.