Warming Amplification In High Elevation Regions Of The World

Mountains and plateaus cover over one-fourth of the earth’s continental areas. Because of their close link with the livelihood of over 14% of the world’s population, their specificities in terms of complex topography, especially great altitudinal range, varying climate regimes, and consequently high biodiversity, and their important role in hydrological cycle and in influencing regional and global climates, the study of climate change in the high-elevation regions is of high scientific, ecological and societal interest. In the past decades, huge efforts have been made to explore the features and impacts of climate change in high-elevation regions. However, despite numerous studies, our understanding of (1) whether high-elevation regions are warming faster than their low-elevation counterparts, and (2) whether there is elevation-dependent warming (EDW) within a high-elevation region, remains uncertain. In another word, as defined in this study, on the issue of warming amplification in high-elevation regions of the world, the questions of (a) whether there is a regional amplification for a high-elevation region compared with its low-elevation counterpart, and (b) whether there exists an altitudinal amplification trend within a high-elevation region, remain to be answered. This study focuses on the examination of both regional amplification and altitudinal amplification in the high elevation regions across the globe with new methodology through collecting land surface air temperature data over the period 1961-2010 from all available meterological stations around the globe. The innovation and introduction of methods, and main conclusions are as follows:1 INNOVATION AND INTRODUCTION OF METHODS(1) Development of altitudinal warming component extraction (AWCE) method. First, the function equations for the extraction of warming components of altitude, latitude and longitude from the total warming rates at individual stations within a high-elevation region is developed; and then the AWCE method is used to test whether there is an altitudinal amplification trend in individual high-elevation regions. This is primarily because temperature trends in a high-elevation region are controlled not only by altitude but also latitude or latitude and longitude. If the total warming rates are directly used for the analysis of linear relationship between temperature trends and altitude, EDW trend can hardly be detected in any high-elevation region. This is especially true for the Tibetan Plateau. In brief, the application of AWCE method involves following four steps:(a) Unit transformation of altitude (in meter), latitude (in degree) and longitude (in degree) into km for each station; (b) Estimation of the effect coefficients of altitude, latitude and longitude (ECALT, EClat and ECLONG, respectively) on a regional scale using stepwise regression; (c) Extraction of the warming component of altitude (QALT) from the station warming rate (QTOTAL) for each station within a high-elevation region using the AWCE equation; and (d) Test of the altitudinal amplification trend for the region based on the QALT values extracted from the individual stations by regressing station’s QALTs against station’s altitudes.(2) Establishment of paired region comparison method and paired groups of station comparison method. On the question of whether a high-elevation region is warming faster than its low-elevation counterpart(s), there is no a standard reference region in the previous studies. Most studies have analyzed temperature trends in a specific high-elevation region and compared them to global or hemispheric trends. To improve the comparability, the paired region comparison method is established. First, relative to a specific high-elevation region, its low-elevation counterpart is identified as the neighboring region at the same latitudes and with the same land area. The regional temperature trend for the high-elevation region is then computed in comparison with that of its low-elevation counterpart. Similarly, for the study when all the high-elevation stations (>500m above sea level) are taken as a whole compared with their low-elevation counterparts, the paired groups of station comparison method is established. Here the low-elevation stations are defined as those located at the same latitudes. The regional temperature trend for the high-elevation stations is then computed in comparison with that of its low-elevation counterpart.(3) Introduction of stepwise regression. It has been suggested that the pattern of temperature change in the Alps is mainly controlled by both altitude and latitude. However, no statistical test has been done on either regional or global scale. Although at the scale of individual high-elevation regions this relationship may not be apparent, a significant relationship of temperature trends with altitude and latitude may be detected on the global scale. Hence stepwise regression analysis is introduced into this study for the test of correlations of temperature trends with altitude and latitude.2 MAIN CONCLUSIONS(1) Based on annual mean temperature (TMIN) series (1961-2010), the AWCE method is adopted to examine whether an altitudinal amplification trend exists for the 8 high-elevation regions (The Tibetan Plateau, Loess Plateau, Yunnan-Guizhou Plateau, Alps, Rocky Mountains, Appalachian Mountains, Andes and Mongolia Plateau). Results show a significant altitudinal amplification trend for all the regions tested, with an averaged altitudinal amplification rate of 0.19 (±0.09) ℃ km-1 50-yr-1 during 1961-2010. The paired region comparison method is used to analyze the temperature trends for 4 pairs of high- and low-elevation regions. Results show that the warming for the 4 high-elevation regions (The North Tibetan Plateau, East Loess Plateau, Southeast Rockies and Alps) is stronger than their low-lying counterparts. These results demonstrate that warming amplification in high-elevation regions is an intrinsic feature of recent global warming.(2) Based on annual mean minimum temperature (TMIN) and annual mean maximum temperature (TMAX) series (1961-2010) around the globe, the AWCE method is adopted to test whether an altitudinal amplification trend exists for the 6 high-elevation regions (The Tibetan Plateau, Loess Plateau, Yunnan-Guizhou Plateau, Alps, Rocky Mountains and Appalachian Mountains). Results show that for TMIN (Tmax), a significant altitudinal amplification trend is detected for 6 (4) of the high-elevation regions. The average magnitude of altitudinal amplification trends in TMIN (TMAX) for the 6 regions is 0.306±0.086℃ km-1 50-yr-1 (0.154±0.213℃ km-1 50-yr-1), substantially larger (smaller) than that (0.230±0.073℃ km-1 50-yr-1) for TMEAN·Paired region comparison method is used to analyze the temperature trends for 5 pairs of high-and low-elevation regions. Regional amplification is detected for 4 of the 5 high-elevation regions. Qualitatively, highly (largely) consistent results are observed for TMIN (TMAX) compared with TMEAN.In addition, when the two-factor-AWCE method is used for TMEAN, TMIN and TMAX for the 6 high-elevation regions (The Tibetan Plateau, Loess Plateau, Yunnan-Guizhou Plateau, Alps, Rocky Mountains and Appalachian Mountains), it is found that the magnitude of altitudinal amplification trend in each temperature index is highly consistent with that from the normal three-factor-AWCE method for every region tested. This indicates that the two-factor AWCE method can also be used for the estimation of altitudinal amplification trend in a high-elevation region. It also suggests that temperature changes in high-elevation regions are mainly associated with altitude and latitude.(3) Through the comparison of linear trends for high- and low-elevation stations, it is found that the warming is about 1.24 times greater for the 910 high-elevation stations (>500m) than their low-elevation counterparts in the past 50 years. With the elevation band method, a significant altitudinal amplification trend is detected for both 1961-2010 and 1976-2010 for the 910 stations, and the magnitude of altitudinal amplification trend is 1.39 times larger for the last 35 years than in the last 50 years. When stepwise regression method is adopted, it is found that there is a significant relationship of warming trends with altitude and latitude for the high-elevation stations. Further analysis show that the warming magnitudes are proportional to the temperature lapse rate in altitudinal and latitudinal directions for the 910 stations. According to Stefan-Boltzmann’s law, this means that the warming amplification in high altitude regions is mainly related to the energy balance in the altitudinal and latitudinal dimensions.Comparatively, the warming trend for the low-elevation stations at the Northern Hemisphere high-latitudes (Arctic Region) is 1.24 times that of the low-elevation stations at the low latitudes during 1961-2010. Although there is no latitudinal amplification trend in the Arctic Region, the latitudinal amplification trend is highly significant for the Northern Hemisphere low-elevation stations. Further analysis suggests that the warming trends for these low-elevation stations are positively related to the temperature lapse rate in the latitudinal direction. According to Stefan-Boltzmann’s law, it implies that the Arctic amplification is closely related to the energy balance in the latitudinal dimension.(4) The temperature trends are highly significant in all four seasons for both the high- and low-elevation stations during 1961-2010. While the warming in spring is slightly weaker for the 937 high-elevation stations than their low-elevation counterparts, it is at higher rate for the former than the latter in other three seasons. Using the elevation band method, a significant altitudinal amplification trend is detected for all four seasons. The amplification amplitudes in autumn and winter seasons (0.2790℃ km-1 50-yr-1,0.2756℃ km-1 50-yr-1) are larger than those in spring and summer seasons (0.1023℃ km-1 50-yr-1,0.1350℃ km-1 50-yr-1). The analysis of stepwise regression shows both a significant altitudinal amplification trend and a significant latitudinal amplification trend on the seasonal scale for the high-elevation stations.As a whole, with the innovation and introduction of the test methods, the study of EDW in high-elevation regions has been greatly advanced on both regional and global scale. It is found that warming amplification (both regional amplification and altitudinal amplification) occurs in high-elevation regions of the world generally. There is not only an altitudinal amplification trend but also a latitudinal amplification trend for the high-elevation stations as a whole. All these results have important implications for further research of climate change in high-elevation regions in terms of attribution of EDW, projection, impact assessment and adaptation options.