The Study Of Vegetation Dynamic And Responses To Climate Change Along The Appalachian Trail Corridor

The gradients of the Appalachian Trail (A.T.) in elevations and latitudes providea mega-transect to study environmental variations in the eastern United States. Basedon the land surface phenology (LSP) parameters extracted from long time seriesremote sensing data, we analyzed the distribution characteristic and variation ofNDVI and LSP in the whole study area, different ecoregions and vegetation typesfrom1982to2006, we analyzed the relationship between LSP and longitude, latitude,elevation, local climatic variables and large scale climate oscillations, and wedetailedly discusses the relatiationship between vegetation dynamic change andhydrothermal conditions (temperature and precipitation) and geography factor(longitude, latitude and altitude) in the A.T. corridor. Main conclusions are as follows:(1) The spatial distribution of hydrothermal is difference, the northeast part ofA.T. show high temperature and the southern is low. The highest rainfall mainlyconcentrated in the southern and central ecoregions. In the past25years, thetemperature increased in the A.T. corridor, but the precipitation variation trend is notobvious. The temperature increase with0.31℃/10a in the whole area. And the largestoccurred in M221ecoregion with0.37℃/10a. The trends of temperature is notobvious less than39°N, but temperature decreasing1.1℃with increasing latitude1°when higher than39°N. Precipitation have a decrease trend with latitude increasesbelow latitude37°N, but there is no trend when latitude higher than37°N. Thetemperature increasing at the eastern part of the United States, and predictstemperature show2~6℃up to2100.(2) Over the past25years, NDVI peak change presents the decreasing trends,and the variation trends are different under different landcover types, NDVI increasedwith increasing altitude and latitude. The growth condition of vegetation graduallydecreased. NDVI in different vegetation types also showed a decreasing trend. Thevariation trend of evergreen forest and deciduous forest are the largest, to illustrate thetwo vegetation types are affected by other factors influence. NDVI response togeographical factor also has a lot of features. The NDVI increase0.0062units with100m elevation increase when the altitude more than300m. NDVI decrease by0.01units with the latitude increase1°, and the variation trend is more intense in less than40°N.(3) Phenology for different ecoregions and landcover types showed different distribution characteristics and variation trends. For the whole study area, the start ofseason (SOS) delayed2.6days in25years, end of season (EOS) delayed10days, theextended length of season (LOS) is mainly determined by the EOS. For three provinceecoregions, only M211shows phenomenon advanced trend, other ecoregions showeddelayed trend, but variation trend is no significant. In the M211,221and M221, theEOS delayed for16days,4days and17days, respectively. High latitudes phenologyvariation trend more obvious higher than that of low latitudes. Bushes, evergreenforest and mixed forest of EOS have larger trends were0.84d/a,0.51d/a, and0.57d/a. The deciduous forest, grassland and cultivated land variation trend were small.SOS, EOS and LOS vary with latitude and longitude and altitude showed obvioustrend of increasing or decreasing. With altitude increasing100meters SOS delayed0.61days, EOS advanced0.97days, and LOS shorten1.58days. Variation ofphenology along the latitude in39°N have obvious difference, between34°N~39°N,SOS decreasing rapidly, EOS and LOS has a strong increasing trend. with latitudeincreasing1°SOS delayed1.47days, EOS delayed3.02days and LOS extended by4.48days. The SOS delayed3.56days, EOS advanced1.7days and LOS extended by5.27days with latitude increase1°when latitudes more than39°N. In high latitudesand altitudes area, the phenology variation characteristics close to Johns Hopkins’slaw.(4) The relationship between average of NDVI and temperature in the sameperiod were positively correlated, negatively related with the precipitation, but thecorrelation is not obvious. NDVI affected by the temperature in the winter and spring.A warmer winters and springs has a positive role for the average of NDVI. In winter,spring and summer, precipitation has inhibition effect on NDVI, autumn precipitationhas a promoting effect on NDVI. In the spring and fall, the NDVI have lag responseto temperature, there was no significant response in summer. The cumulative rainfallof up to five months in March has strong lag effect on NDVI.（5）Temperature is the main condition of phenology change, the precipitationalso has some influence on the low latitude area. The significance of annual averagetemperature for SOS was the highest, followed by the temperatures in winter andspring. Precipitation on vegetation SOS inhibited at low latitudes while mainlypromoted at high latitudes. Precipitation in spring has a significant correlationbetween vegetation growth while precipitation in winter has a highest negativecorrelated with SOS. SOS has some lag effect by temperature and precipitation.Autumn temperature has the greatest effect on EOS. Average annual, spring, autumn and winter precipitation has some impact on EOS, and there is a certain role inpromoting. All vegetation types in the SOS have a strong negative correlation totemperature, that is SOS advance when temperature increase. In summer, autumn andwinter, the EOS of shrubs, evergreen forest, mixed forest, deciduous forest has asignificantly higher positive correlation with temperature. While the EOS of shrub,evergreen forest, mixed forest and cultivated land has a great positive correlation withthe fall and winter precipitation. Phenology sensibility of the high latitude area(northeast part of the A.T.) to temperature is higher than the low-latitude region(southeast part the A.T.). The fluctuations of phenology showed a response to thecharacteristics of climate change in hot and cold, wet and dry year and responsedstrongly to the global extreme events at the same time. The1989SOS delayed andEOS advanced corresponded to the La Ni a in1988~1999while the1998SOSadvanced and EOS delayed is consistent with the El Ni o in1997~1998.(6) It is a distinct spatial difference of regional response of NDVI, phenologyand climate conditions to latitude, longitude and elevation. The temperature correlatedsignificantly with latitude and longitude, which are the mainly factors for temperature.There is no major factor affected on precipitation. It is similar to precipitation, thecorrelation coefficients for NDVI and each factor is very close. The latitude andlongitude are key control factors for SOS. The elevation is factor for EOS, and LOS iscontrolled by all of factors. The multiple linear regression analysis is applied torelationship of NDVI, phenology, precipitation and temperature conditions to latitude,longitude and elevation. The result of multiple correlation coefficients is:temperature>SOS>LOS>EOS>NDVI>precipitation, this can explain the correlationbetween NDVI, phenology, climate conditions and latitude, longitude and altitude.