Development Of Statistical Analysis Software For Forest & Apply On Research Of Canopy Precipitation Interception

The forestry statistical analysis software of this research can directly deal with Excel's electronic table, Access's data bases and net table data, and it is very convenient that the record of data is entered and editted, and possesses the functions of the forestry statistics such as variance, correlation and regression analysis, and the analysis efficiency is identical with the most excellent statistical analysis software SPSS, and finishes the necessaries of satisfying forestry data processing.One of hot spot being forest hydrology research is canopy precipitation interception, and this paper is on the foundation of the existing research achievements of analysis home and abroad, and uses the forestry statistical analysis software of development by oneself to go on the analyses of data , and major research result indicates :(1) Schima superba canopy is more interception loss than Cunninghamia lanceolata canopy. It is all bigger than Cunninghamia lanceolata canopy that the coefficient of one-way linear model for the interception loss and precipitation of Schima superba canopy, and showing under the identical rainfall circumstance that Schima superba canopy is more interception loss than Cunninghamia lanceolata canopy, and this has conformed to the real circumstance, and has also proved the law that rich forest canopy's interception loss is more than the coniferous forest canopy's.(2) Double square root model suits the survey data most. Can make out from the analysis result of one-way & non- linear model, double square root model all suits to Schima superba and Cunninghamia lanceolata, and imitative effect is fairly ideal, and showing that double square root model is one of the robust model in the experiment.(3) Throughfall rate, stem flow rate, interception loss rate and rainfall basically fit the hyperbola model. The hyperbola model has certain applicability when the relation of penetrating flow rate, stem flow rate, interception loss rate and rainfall is analyzed, but the imitative effect is not too good. The research can be done further to this aspect from now on.(4) Multinomial model can raise to the imitative effect. The result of synthetical various regression analysis can be made out, and the multinomial model all has certain raise to the imitative effect. This is normal, and the arbitrary function all can approach with dividing section multinomials because can know from infinitesimal calculus.(5) There is not the general experience model. The data analysis result indicates which does not have kind of model can not suffer from the influence of the seed of a tree...