| As the kinds of sunlight, rainfall, air temperature, soil, water, terrain and so on are theinfluential factors for the tree’s growth, factor’s varieties(changes) will make correspondingexpression on tree’s annual rings. Based on this, the tree’s annual rings will record theinformation data of tree’s growth geographical environment. The thesis thinks that every tree isthe ’min-weather station’ of its growth environment, and conversely, by studying each“min-weather station”, we can get the geographical weather information of partialmin-environment.With collecting samples data of masson pine from Daming Mountain (15samples), DayaoMountain (17samples), Qing Mountain (5samples) and the area of Tianbao Reservoir (8samples),the author of the thesis studies and analyses these masson pines’ growth process, andtries to discover the idiographic rule of masson pines’ geographical environment change in orderto offer scientific suggestions for afforestation, water and soil conservation, economic forest,fruit tree cultivation and botanical garden in the area above through the studying methods offactor analysis, clustering analysis, artificial neural nets, fractal theory, complex fractal, waveletanalysis etc..Firstly, with the use of cluster and factor analysis methods in the statistics software SPSS,the writer researches the tree rings around Tianbao Reservoir, finds the tree rings can record suchgeographic environmental information as their slope directions, soil, light factor, water factor,and tempreture ect. and tries to access its varying patterns though analysis. In addition, thetree-ring data in Daming Mountain, Dayao Mountain and Qing Mountain are analyzed throughSPSS neutral network multilayer perception. At the same time, its geographic climate is judgedand geographic environment is classified according to the growth patterns of the trees in qingMountain and the models established by induction sensors.Secondly, with the data of average broadband of tree rings between2003and2013inDaming Mountain and Dayao Mountain, the researcher forecasts their average broadband in2014in the above two mountains though professional forcasting procedures compiled by deltafunctions of MATLAB artificial neutral network kit. The reason to use average broadband of treerings is that it can neutralize the differences in soil, physical geography and ecologic competitionof each sample tree and represent its region to some extent. Forecasted by the artificial neutral network model, the predicted value is5.9631cm in Daming Mountain,5.1295in DayaoMountain, both of which are bigger than those of the year2013. Therefore, indirect conclusionscan be made that climate in Guangxi Region in2014is better than that of2013.Furthermore, based on the theory of Lipschitz-exponent and fractal theory and the theory ofinformation complexity, the writer calculates characteristics of sudden-change the andcomplexity of tree-ring sequences in Daming Mountain and Dayao Mountain with the specialprocedure compiled with corresponding delta functions of MATLAB wavelet analyzing tool kit.By comparing the calculating results, the research indicates that characteristics of sudden-changethe and the complexity of Dayao Moutain is bigger that Daming Mountain,thus proving thegeographic factors such as water and heat in Daming Mountain stay relatively stable and regular.Finally, this thesis compiles special MATLAB procedures and makes sequential waveletanalysis on the average tree-ring sequences during the past36years in Daming Mountain inorder to research into periodicity and regularity of tree rings responding to the years changing.Since the varying tree rings function as the barometer of local hydrothermal environment andclimate situation, such study can also indirectly indicate the periodicity and regularity of thevarying climate and environment in Daming Mountain in the past36years. |
PDF Full Text Request