Vertical Distribution Pattern And Geographic Interpretation Of Local Scale Alpine Timberline Ecotone From The Perspective Of Uncertainty Analysis

As an important geographical and ecological boundary,timberline plays an important role in indicating the response to climate change.However,the staggered distribution of different types of vegetation patches in the timberline ecotone(upper limit is treeline,lower limit is timberline)presents a certain randomness,and the distribution boundary between timberline and treeline also has a certain ambiguity.At present,most researches simplify the timberline and treeline as a continuous curve,which is difficult to express and analyze the fuzziness of the timberline and treeline and the randomness of the distribution of different vegetation patches in the transition zone.The Mt.Namjagbarwa is the only mountain with the most complete vertical zone of mountain vegetation in China.The number of vertical zones of Baima Snow Mountain and Bogda Mountain is the same,and the elevation of the mountain is similar,while the height of timberline distribution is significantly different.This study uses multi-source remote sensing data to obtain the sample point data of timberline and treeline of Mt.Namjagbarwa,Baima Snow Mountain and Bogda Mountain,build the cloud model of timberline and treeline distribution height,quantitatively analyze the uncertainty of timberline and treeline distribution,and use multiple linear regression analysis to analyze the dominant factors of timberline and treeline distribution height.According to the "3En principle" of the cloud model,with Ex-En and Ex+En as the thresholds,the data points of the timberline and treeline are divided into two parts: the basic element and the peripheral element(spatially corresponding to the core area and the peripheral area of the timberline ecotone respectively),and multiple regression is carried out respectively to analyze the differences of the influencing factors in different areas within the timberline and treeline.The main results of this paper are as follows:(1)The expectation of the distribution height of the timberline and treeline in Mt.Namjagbarwa is 4054.61 m and 4103.05 m,respectively.The entropy of the treeline(207.59m)and hyper entropy(70.36m)are higher than the entropy of the timberline(191.17m)and hyper entropy(50.13m).The expected distribution height of timberline and treeline in Baima Snow Mountain is 4121.09 m and 4186.38 m respectively,and the expectation distribution height of timberline and treeline in Bogda Mountain is 2675.22 m and 3009.61 m respectively;The entropy(timberline 410.71 m,treeline 597.32m)and hyper entropy(timberline 66.22 m,treeline 280.86m)of the cloud model of timberline and treeline distribution height in Bogda Mountain are greater than those in Baima Snow Mountain(entropy: timberline 182.33 m,treeline 193.96m;hyper entropy: timberline 56.26 m,treeline65.86m),that is,the uncertainty of timberline and treeline distribution in Bogda Mountain is higher than that in Baima Snow Mountain.(2)Dryness is the highest contributing factor to the distribution height of timberline(50.26%)and treeline(44.11%)in Baima Snow Mountain,while the contribution rate of average temperature in July to the distribution height of timberline and treeline in Mt.Namjagbarwa and Bogda Mountain is the highest,44.01%,46.74%,48.15% and 60.59%respectively.The dominant factors of distribution height of timberline and treeline in different regions are different.(3)The influence of climate and terrain on the core and peripheral areas of timberline/treeline are different.The contribution rate of average temperature in July to basic elements(52.54%-57.09%)was higher than that of peripheral elements(4.10%-8.68%)in Mt.Namjagbarwa,and the contribution rate of summit syndrome to peripheral elements(23.13%-23.74%)was higher than that of basic elements(6.23%-7.17%).The contribution rate of dryness to the peripheral elements of the treeline of Baima Snow Mountain(57.70%)was significantly higher than that of the basic elements of the treeline(25.66%).The contribution rate of mean temperature in July to peripheral elements of Bogda Mountain(timberline 42.15%,treeline 64.68%)was higher than that of basic elements(timberline19.70%,treeline 43.17%).(4)There were differences in the uncertainty of the influencing factors at the timberline/treeline of Baima Snow Mountain and Bogda Mountain.The barycenter and uncertainty(variation range and variation degree)of the summit syndrome of Bogda Mountain is higher than that of Baima Snow Mountain,while the barycenter and uncertainty of the snow effect is lower than that of Baima Snow Mountain.In addition,the barycenter of dryness of Bogda Mountain timberline is smaller than that of treeline,and the uncertainty of dryness is significantly higher than that of Baima Snow Mountain.