Abrupt Change in the First Tropopause Height over Xinjiang in 1960J1999

Transcription

1 Advances in Climate Change Research Articles Article ID: (2006) Suppl Abrupt Change in the First Tropopause Height over Xinjiang in 1960J1999 Zhang Guangxing 1, Li Juan 1, Cui Caixia 1, Xin Yu 2 1 Institute of Desert Meteorology, China Meteorological Administration, Urumqi Xinjiang , China; 2 Meteorological Bureau of Bortala Mongolian Autonomous Prefecture, Bole Xinjiang , China Abstract: Based on the1960j1999 daily radiosounding data from twelve stations in Xinjiang, which is divided into four sub-areas, the Altai Mountains, the northern and southern slopes of Tianshan Mountains and the Kunlun Mountains by climate features and latitude distribution, trends of the annual mean heights and spatial differences of the first tropopause in the past 40 years were analyzed, and tests of the abrupt change of mean heights conducted by Mann-Kendall and successive running t-test. The results showed that there was an overall increasing trend of mean height for the first tropopause over Xinjiang during 1960J1999, in particular during 1987J1999. From the view of the interdecadal change, the mean height obviously decreased from the 1960s to the early 1970s, and afterwards gradually increased with an accumulated height increment more than 100 m. The successive running t-test showed that a remarkable abrupt change of mean height occurred in the Altai Mountains in 1972, the southern slopes of the Tianshan Mountains in 1976, and the northern slopes in 1979, but no remarkable change happened in the Kunlun Mountains. The mean height obviously increased all over Xinjiang except the Kunlun Mountains. Although there were various differences in four sub-areas of Xinjiang, the increasing trend of mean height was rather consistent from the end of 1980s to 1999 over Xinjiang. Key words: the first tropopause height; change trend; analysis of abrupt change Introduction The tropopause, about several hundred to two thousand meters depth, is a transition layer between the troposphere and the stratosphere, and it includes the first tropopause (polar category) mainly over the middle and high latitudes and the second tropopause (tropical and subtropical category) over the low latitudes. The first tropopause never fails to fascinate human beings because it has the striking features of atmospheric thermodynamic and dynamic, and chemical structures [1], and of the abrupt change of temperature lapse rate, and the sudden decrease in air humidity. The first tropopause is a thick layer suppressing the upward development of cumulonimbus and blocking the vertical exchange of aerosols and vapor between the troposphere and the stratosphere. The principal studies have recently focused on the substance exchange between the Received: November 30, 2005; revised: April 3, 2006 Corresponding to Zhang Guangxing ( zhangguangxin@163.com) 2006 National Climate Commission of China. All right reserved. troposphere and the stratosphere [2] and the element change of the tropopause in individual regions [3J5]. Santer et al. [6] demonstrated the fact that the tropopause height rose several hundred meters all over the world from 1979 to 1999 with the radiosounding data, the reanalysis data and the climatic model incorporating natural as well as anthropogenic factors. Our concern is how the tropopause changed and its response to the global climate change in the past 40 years in the Xinjiang Uygur Autonomous Region, as a climatic sensitive area. The existing discussions on Xinjiang climatic changes were mostly based on surface meteorological elements [7], and few on the significant pressure levels in the middle troposphere [8], which is less influenced by local observation surroundings. The regional climate variation would affect the global change and the climate change in the low troposphere would impact the upper tropospheric change through dynamic and thermodynamic interaction, and vice verse. So it will be more perfect that the statues of the meteorological elements of ground surface as well as the lower and upper troposphere are considered at the same time in understanding the regional climate. Adv. Clim. Change Res., 2006, 2 (Suppl. 1): 41J45 41

2 Zhang Guangxing et al.: Abrupt Change in the First Tropopause Height over Xinjiang In recent years, scientists have focused not only on regional climatic changes but also on their patterns, especially on their abrupt changes, such as the climatic features transformed from warm/dry to warm/wet over Northwest China [9J12]. Based on the radiosounding data over Xinjiang of China in 1960J1999, we explored the height variation of the first tropopause and its abrupt change in this paper, to better understand the regional climatic change. 1 Data and methods The radiosounding observations twice a day at 0800 and 2000 BST are selected at 12 radiosonde stations with relatively stable station location and the continuous observation from 1960 to 1999, whose sounding meteorographs have been strictly examined and corrected by China Meteorological Administration (CMA) according to the national measurement criteria, and their observing errors were within the tolerance limitations. In view of the climatic feature and the distribution of the 12 radiosonde stations, Xinjiang region is divided into four sub-areas, the Altai Mountains, the northern and southern slopes of Tianshan Mountains, and the Kunlun Mountains. The annual mean tropopause height is obtained from the daily twice observations. In order to avoid the one-sidedness of single station records, the regional mean of single station values within a sub-area is calculated to get the regional annual mean height of the sub-area, and to obtain the annual mean tropopause height of whole Xinjiang. The abrupt change of variables is tested by the detection of the Mann-Kendall rank (M-K) and the moving t-test technique. 2 Interannual variations of first tropopause heights Xinjiang has a large spatial span in the direction of north-south and east-west with two great basins (Tarim and Turpan Basins) inserted among three mountains (the Altai, Tianshan, and Kunlun Mountains). This terrain feature brings up different weather systems and regional climate in various sub-areas. Therefore, the first tropopause heights over the whole Xinjiang as well as over each of four subareas are analyzed in this paper. 2.1 Whole Xinjiang Figure1 shows the overall rising trend of the first tropopause height for whole Xinjiang, and the slope of the linear fitting equation is The five-year moving mean curve indicates that the first tropopause fluctuated at relative higher altitudes at the beginning of the 1960s, declined suddenly in the middle 1960s, wavily and steadily rose from about 1968 to the beginning of the 1980s, followed by descending obviously till the middle 1980s and apparently going up from 1987 to Altai Mountains The slope of the linear fitting equation over the Altai Mountains (Fig.2) is , twice larger as compared to that for whole Xinjiang (Fig.1), demonstrating a more apparent rising trend of the first tropopause height. The five-year moving average curve mainly shows an overall ascending trend accompanied with several small descending fluctuations for short periods. Figure 2 illustrates a descending trend from the 1960s to the beginning of the 1970s, a wavily ascending trend till the middle of the 1980s, next a descending trend over a short period again till the end of the 1980s and then a remarkably ascending trend from 1988 to the end of the 1990s y = x Fig.1 Variations of the annual mean height of the first tropopause in the whole Xinjiang region y = x Fig.2 Variations of the annual mean height of the first tropopause in the Altai Mountains 2.3 Northern and southern slopes of Tianshan Mountains Figure 3 shows rather evidently an overall ascending trend of the first tropopause height over the northern slope 42 Adv. Clim. Change Res., 2006, 2 (Suppl. 1): 41J45

3 Advances in Climate Change Research of Tianshan Mountains. The slope of the linear fitting equation is , which is larger than that of the whole Xinjiang and slightly smaller than that of the Altai Mountains, meaning that the slope is getting smaller and smaller from the north to the south, and the rising amplitude is also alternatively getting smaller and smaller. The fiveyear moving mean curve displays an overall ascending trend accompanied with smaller fluctuations. It is shown from Fig. 3 that the first tropopause height fell from 1960 to 1968, then rose till 1984, suddenly dropped in the next two years and rather obviously lifted from 1986 to Variations of the first tropopause height over the southern slope of Tianshan Mountains show the similar tendency (Figure omitted) to that for the northern slopes, but its slope is smaller so that it suggests further that the amplitude of the first tropopause height is getting smaller and smaller from the north to the south in the past 40 years y = x Fig.3 Variations of the annual mean height of the first tropopause in the northern slopes of the Tianshan Mountains 2.4 The Kunlun Mountains Influenced by the topography of Tibetan Plateau, the climatic feature in the Kunlun Mountains is entirely different from the other sub-areas of Xinjiang. The fitting slope of the first tropopause height equals to J1.9147, indicating drop tendency (Figure omitted). The five-year moving mean curve exhibits that the height of first tropopause dropped significantly before 1967, then lifted strikingly till 1972, subsequently fell till 1978, went up from 1978 to 1983, descended sharply till 1987, and then has lifted steadily since Although the overall trend in the Kunlun Mountains appeared fall, the steadily going up trend has been in common with the other sub-areas of Xinjiang since the end of the 1980s. The climate over Xinjiang has gotten continuously warmer during the same period. From the above, it is concluded that the rising amplitude of the first tropopause height was getting smaller gradually from the north to the south of Xinjiang and its height showed an overall descending trend in the Kunlun Mountains. 3 Interdecadal variations of the first tropopause height In order to highlight the significant change of the tropopause height over whole Xinjiang and its four subareas, the interdecadal variation features of mean height are shown in Fig.4. The first tropopause height for whole Xinjiang was the highest in the 1990s, the second in the 1960s and the lowest in the 1970s, and it was nearly 100 m higher in the 1990s than in the 1970s, with a mean ascending rate of about 34 m per decade. H /m the Altai Mountains the northern slopes of Tianshan Mountains the southern slopes of Tianshan Mountains the Kunlun Mountains the whole Xinjiang region 1960s 1970s 1980s 1990s Decade Fig.4 Interdecadal changes of the annual mean height of the first tropopause in Xinjiang and its four subareas (==H: height difference in comparison with the lowest value) The first tropopause height over Altai Mountains went up successively about 150 m from the 1960s (the lowest) to the 1990s (the highest) at a rising rate of about 36 m per decade. In the northern slopes of Tianshan Mountains, the tropopause height was the lowest in the 1960s, and it went up successively to the highest in the 1990s with a total increment about 120 m, at a mean ascending rate of 30 m or so per decade. In the southern slopes of Tianshan Mountains, the tropopause height experienced a process of high value (the 1960s)Jlowest value (the 1970s)Jhighest value (the 1990s), and the increment from the lowest to the highest was about 150 m, with the mean rising rate about 48 m per decade. The decadal variability of the first tropopause height over the Kunlun Mountains was not only more significant, Adv. Clim. Change Res., 2006, 2 (Suppl. 1): 41J45 43

4 Zhang Guangxing et al.: Abrupt Change in the First Tropopause Height over Xinjiang but also had its own unique feature as compared to the other sub-areas. The height dropped steadily from the 1960s to the 1980s, reached the lowest in the 1980s, and rose steeply about 110 m from the 1980s to the 1990s at a rising rate of 110 m per decade. Concerning the interdecadal variation of the tropopause height over Xinjiang, it still showed a common swelling-up tendency in the 1990s no matter how it changed in its own way in each sub-area. 4 Abrupt change It is seen from the above that the first tropopause height in Xinjiang changed over the past 40 years, and especially the change has been significant since the1980s. The abrupt change of the height is tested by the detections of the M-K rank (Fig.5) and the moving t-test technique (Figs.1J3) at the confidence level of Contour C1 and C2 in Fig.5 are statistics for the detection of M-K rank, if contour C1 intersects C2 at a point between the upper and lower critical lines, meanwhile the contour C1 surpasses the upper or lower critical line, then an abrupt change happened at the point ( time of the abrupt change) [13]. The detections of M-K rank of regional mean first tropopause height in different sub-areas show that in the Altai Mountains (Fig. 5) and Tianshan Mountains (Figure omitted) there was an abrupt change in 1967 or so. The abrupt change was the most significant in the Altai Mountains and the second in the northern slope of the Tianshan Mountains, then in the southern slopes, and steady variations instead of abrupt change in the Kunlun Mountains. The moving t- test technique is also used to confirm the analysis of abrupt changes by M-K rank detection. Statistics for M-K J1.96 C 1 C 2 95% the upper or lower critical value J3.92 Fig.5 Mann-Kendall abrupt change analysis of the annual mean height of the first tropopause in the Altai Mountains The confidence level of 0.05 was used in the significance test in the moving t-test technique, and statistics were obtained by successive moving five-year step in this paper. The results from Figs.1J3 show that except the Kunlun Mountains, an abrupt change of the first tropopause height happened over whole Xinjiang in 1979 (Fig.1), the Altai Mountains in 1972 (Fig.2), the northern slope of the Tianshan Mountains in 1979 (Fig.3), the southern slope of the Tianshan Mountains in 1976 (Figure omitted), but the abrupt change over whole Xinjiang was less significant than those in sub-areas. 5 Conclusions The first tropopause height over the whole Xinjiang rose overall in the past 40 years, and in particular has risen outstandingly since The difference between the maximum decadal mean altitude in the 1990s and the minimum in the 1970s was more than 100 m. For different climatic sub-areas, the principal tendency of the first tropopause height showed fall in the Kunlun Mountains area and rise in the other sub-areas; and significant rises have appeared in the whole Xinjiang except a slight rise in the Kunlun Mountains since the end of the 1980s. The rising amplitude reduced from the north to the south. The rising trend in the Altai Mountains was the most outstanding; however, an overall declining trend appeared in the Kunlun Mountains. Two different results were obtained when detecting the abrupt changes of the first tropopause height over Xinjiang during the 40 years. The detection of the M-K rank shows that the abrupt change happened in the Altai and Tianshan Mountains in 1967 or so and that the further in the north the location is, the more remarkable the abrupt change is. On the other hand, the t-test technique shows that the abrupt change of mean values happened in whole Xinjiang in 1979, the Altai Mountains in 1972, the northern slopes of the Tianshan Mountains in 1979 and the southern slopes in The two detection techniques both show that no abrupt change but steadily climatic variations took placed in the Kunlun Mountains area. References [1] [2] [3] Qiang Fu, Celeste M, Stephen G W, et al. Contribution of stratospheric cooling to satellite-inferred tropospheric temperature trends [J]. Nature, 2004, 429: 55J58. Yang Jian, Lu Daren. Progresses in the study of stratospheretroposphere exchange [J]. Advance in Earth Sciences, 2003,18 (3): 380J385 (in Chinese). Wang Liqiong, Zhang Lifeng. The annual variation feature 44 Adv. Clim. Change Res., 2006, 2 (Suppl. 1): 41J45

5 Advances in Climate Change Research [4] [5] [6] [7] [8] [9] analysis of meteorology elements of Taipei upper air [J]. Scientia Meteorological Sinica, 2001, 21(2): 193J199 (in Chinese). Li Guohui, Lu Daren, Tie X X. The impact of tropopause variation on ozone distribution in upper troposphere / lower stratosphere [J]. Chinese Journal of Space Science, 2003, 23(4): 269J277 (in Chinese). Zou Jinshang, Zhang Jiangqiu, Wang Bingzhong. The characteristics of temporal and spatial variation of tropopause over china and its controlling factors [J]. Scientia Meteorological Sinica, 1989, 9(4): 417J426 (in Chinese). Santer B D, Wehner M F, Wigley T M L, et al. Contributions of anthropogenic and natural forcing to recent tropopause height changes [J]. Science, 2003, 301(7): 479J483. Yang Qing. The climate change and the analysis on its trend in Xinjiang since recent 40 years [C]// Desert Technology 6th International Conference. Urumqi, China, Zhang Guangxing, Yang Lianmei, Yang Qing. Changing trend and abrupt change of the 0 level height in summer in Xinjiang from 1960 to 2002 [J]. Journal of Glaciology and Geocryology, 2005, 27(3): 376J380 (in Chinese). Tang Maocang, Gao Xiaoqing, Zhu Deqin. The evidences and the causes for the shift of climate pattern in northwest China during this century [J]. Journal of Glaciology and Geocryology, 2003, 25(2): 170J173 (in Chinese). [10] Shi Yafeng, Shen Yongping, Hu Ruji. Preliminary study on signal, impact and foreground of climatic shift from warmdry to warm-humid in Northwest China [J]. Journal of Glaciology and Geocryology, 2002, 24 (3): 219J226 (in Chinese). [11] Zhang Guowei, Wu Sufen, Wang Zhijie. The signal of climatic shift in Northwest China deduced from river runoff change in Xinjiang region [J]. Journal of Glaciology and Geocryology, 2003, 25(2): 183J187 (in Chinese). [12] Guo Ni, Zhang Jie, Liang Yun. Climate change indicated by the recent change of inland lakes in Northwest China [J]. Journal of Glaciology and Geocryology, 2003, 25 (2): 211J 214 (in Chinese). [13] Fu Congbin, Wang Qiang. The definition and detection of the abrupt climatic change [J]. Scientia Atmospheric Sinica, 1992, 16 (1): 111J119 (in Chinese). [14] Li Jianping, Chou Jifan, Shi Jiuen. Detection methods on the mean value jump of the climate [J]. Journal of Beijing Meteorological College, 1996, (2): 16J21 (in Chinese). Adv. Clim. Change Res., 2006, 2 (Suppl. 1): 41J45 45