Ｃｌｉｍａｔｅ　Ｃｈａｎｇｅ　ｉｎ　Ｍｔ．Ｑｏｍｏｌａｎｇｍａ　Ｒｅｇｉｏｎ　ｓｉｎｃｅ　１９７１

Ｚｈａｎｇｓ　Ｇｒｏｕｐ

Global climate change has profound influence on natural ecosystem and socioeconomic system and is a focus which governments, scientific societies as well as common people of various countries have paid much attention to. Observations indicate that there is obvious ascending tendency for the global average surface temperature during the last 100 years (IPCC,2001). The Tibetan Plateau (TP)is located in central Asia with a mean elevation of over 4000 m a.s. l. and an area of about 2,500,000 square km. It is the highest and most extensive plateau in the world, which has long been known as the roof of the world and has unique and complicated plateau climate. The TP not only is a sensitive and initiating area to climate change in China(Feng et al.,1998)but also was considered as a driver and amplifier of the global climatic change (Pan and Li, 1996). Studies (Liu and Hou,1998; Yao et al., 2000; Zhang and Zheng, 2004) on contemporary climate change in the TP Climate change in Mt. Qomolangma region since 1971 show that the annual average air temperature change over the TP has a good consistency, and there is a warming tendency since the 1960s. There exists an elevational dependency of the climatic warming trends in the TP and an increasing tendency of warming with increasing height. Compared with the change of air temperature, the variationof precipitation over the TP is more complicated. There are disputations in precipitation change. From the 1950s to early 1990s, the precipitation over the TP decreased along the Yarlung Zangbo River, but it increased in high elevation area of southeastern, southern and northern Tibet and northern Qinghai (Lin and Zhao, 1996); the annual rainfall trend in most parts of Tibet increased but the changing trend of rainfall decreased in Ngari district from 1971 to 2000 (Du and Ma, 2004); the main trend of precipitation over the TP is increasing but generally has a low significance level in the last 30 years (Wu et al., 2005).

Mt. Qomolangma region islocated in southern TP, central Himalayas with a mean elevation of 4000-5000 m a.s.l. The unique and various geographical conditions, sensitive and vulnerable environment there make it to be a perfect site for investigating the interaction of land-atmosphere system and the change of pattern and process of ecosystem in the background of global climate change. In 2005, the 30th anniversary of the mountaineering scientific expedition into Mt. Qomolangma in 1975, the Chinese Academy of Sciences organized a comprehensive scientific expedition with the subject“the response of Mt. Qomolangma region to the global change”The expedition did research work in a wide range of fields, including atmospheric physics, atmospheric chemistry, glacier and hydrology, biology, environment in Mt. Qomolangma region, the changing process in the height of Mt. Qomolangma. The samples and data obtained in this expedition will provide reliable scientific evidence for studying the environmental change in Mt. Qomolangma region.

The climatic conditions of Mt. Qomolangma region are widely different from those of the same latitudinal zone in eastern China,and the atmospheric circulation acting on here and the natural barrier of the Himalayas generate obvious regional difference between the north and south sides (Team of Scientific Expedition to Tibet, 1975). Being the highest region on Earth, there is a lack of systemic reports about contemporary climate change in Mt. Qomolangma region in recent decades. Using monthly average, maximum, minimum air temperature and monthly precipitation data from 5 weather stations in Mt. Qomolangma region from 1971 to 2004, the spatial and temporal patterns of the climatic change in this region will be analyzed in this paper. We hope the results of this study will be helpful to provide insight into the response of Mt. Qomolangma region to the global change.

1. Scope of the study area

The Mt. Qomolangma natural reserve includes Tingri, Dinggye, Nyalam and Gyirong counties, and there are weather stations only in Tingri and Nyalam counties. For reflecting the climate change in Mt. Qomolangma region more accurately, the scope of the study area in this paper is from Gyangze to the east, Gyirong to the west, national boundaries to the south and the Yarlung Zangbo River to the north, which includes 5 weather stations of Tingri, Nyalam, Shigatse, Gyangze and Lhaze(Figure 1).

2. Data and method

(1) Data processing

Monthly average, maximum, minimum air temperature and monthly precipitation data of Tingri, Nyalam, Gyangze and Shigatse from 1971 to 2004 and Lhaze from 1978 to 2004 are used in this paper. Because of the high correlation of climate change between Lhaze and Shigatse(the correlation coefficient of temperature is 0.91, the correlation coefficient of precipitation is 0.89, passing the 99.9 percent confidence test), we extend the data of Lhaze to 1971 using linear regression through a least-square fitting between Lhaze and Shigatse.

(2) Analyzing the trend of climate change Climatic linear trend analysis (Wei, 1999) was used in this paper and whether the trend is significant is detected.

(3) Climatic jump analyzing Accumulated variance analysis methods (Wei, 1999) was employed to analyze the climatic jump in this paper.

3.Results

Mt. Qomolangma region is sensitive to climate change, and the complicated topography of the Himalayas and strong solar radiation form unique climate and environmental characteristics there. In the background of global climate change, there are a series of responses, such as temperature increase, glacier melting, etc. in Mt. Qomolangma region.

3.1 Temperature variation in Mt. Qomolangma region

3.1.1 Interannual variations of temperature in Mt. Qomolangma region, with a large body and high elevation, reach as much as 1/3-1/2 height of troposphere. Because of the high elevation, the temperature there is low. From 1971 to 2004, annual surface air temperature of 5 stations in Mt. Qomolangma region ranges from 3℃to 6.8The 34-year-averaged regional mean temperature is about 5℃ which is far lower than that in the same latitudinal zone in eastern China.

From 1971, the annual average surface air temperature clearly shows an increasing trend in Mt. Qomolangma region(Figure 2), which is consistent with previous studies that demonstrated a warming tendency since the 1960s in the TP (Zhang and Zheng, 2004; Lin and Zhao, 1996; Wu et al., 2005; Liu and Chen, 2000; Cai et al., 2003; Wei et al., 2003; Wang et al., 2004; Du, 2001; Li et al., 2003). But the warming is morenotable in Mt. Qomolangma region, with an increasing trend of about 0.234℃per decade (passing the 99.9 percent confidence test). Tingri, the highest mountain station among the 5 high-elevation stations, has the strongest warming trend of 0.302℃ per decade (passing the 99.9 percent confidence test).

The interdecadal variations of annual average temperature of Tingri, Nyalam, Gyangze, Shigatse and Lhaze indicate that the warming in Mt. Qomolangma region is weak from the 1970s to 1980s and the temperature of Tingri even decreased, but the temperature increased markedly after 1990. Compared to the 1970s, the temperature increased within a range from 0.44℃ to 0.84℃since 2000 (Table 1).

The annual average, minimum and maximum air temperature in Mt. Qomolangma region show obvious ascending tendency for the interannual change in recent 34 years (Figures 3a, 3c and 3e). The linear rates of temperature increase are 0.234, 0.306 and 0.201 per decade respectively, indicating a strong warming trend.

The linear rate of annual minimum temperature increase is higher than that of annual average and mean maximum; therefore we can conclude that the warming is influenced more markedly by increase of minimum temperature. In addition, there was a sudden change of the annual average, minimum and maximum air temperature accu0.201 per decade respectively,

indicating a strong warming trend. The linear rate of annual minimum temperature increase is higher than that of annual average and mean maximum; therefore we can conclude that the warming is influenced more markedly by increase of minimum temperature. In addition, there was a sudden change of the annual average, minimum and maximum air temperature accumulated variance in Mt. Qomolangma region in 1992 (Figures 3b, 3d and 3f). The warming was slowly before 1992 and the temperature subsequently increased significantly.

3.1.2 Seasonal variations of temperature Monsoon circulation remarkably influences the climate of Mt. Qomolangma region, which was under the control of westerly in the winter half year and warmnd humid circulation in thesummer half year, therefore it is very cold in the winter half year and warm in the summer half year. Figure 4 shows statistics of the monthly temperature averaged for the 5 stations in the most recent decades. In comparison with the 1970s, the monthly mean temperatures since 2000 increased by nearly 1 in the winter half year (October-March), but not obviously change in the summer half year (April- September). Thus, Mt.

Qomolangma region has undergone recent warming, especially in the winter half year. In terms of the scientific expedition, the growing season of Mt. Qomolangma is from June to September. The interdecadal variations of temperature in June and July are weak. Although there is an obvious increase in August and September, the increase is more obvious temperature in non-growing season generally.

3.1.3 Comparison analysis of the temperature changes Shrestha (1999) analyzed maximum temperature data from 49 stations in Nepal on the south of Mt. Qomolangma for the period 1971-1994 (Figure 5). Compared with the trend of maximum temperature change on the north of Mt. Qomolangma (Figure 3e), we can find that on both sides of Mt. Qomolangma, there are concurrent temperature changes, which shows an obvious warming trend. The significant warming started in 1978 on both sides of Mt. Qomolangma.

The climate change of the TP is earlier than the east of China (Feng et al., 1998) and is regarded as the origin of climatic changes in China. Liu and Chen (2000) also suggested that the warming of the TP occurred earlier than global change and had a higher increase trend. By contrastive analysis of climate change in recent three decadesbetween Mt. Qomolangma region and various special scales, we draw similar conclusions. Using global gridded monthly temperature data from the CDC (NOAA Climate Diagnostics Center), we btain the time series of interannual variations of temperature in China and the globe (Figures 6c and 6d), which then were compared with the trend in Mt. Qomolangma region and Tibet (Du, 2001) (Figures 6a and 6b). The trend of temperature change of Mt. Qomolangma region is consistent with that of Tibet and the recent warming began in 1978. The warming in China began in 1984. Theglobal temperature increased twice obviously, one began in 1976 and the last one began in 1993. Thus, the warming over Mt. Qomolangma is earlier than elsewhere in China, even earlier than global average. The rate of increasing air temperature is 0.234 per decade in Mt. Qomolangma region, which is larger than 0.226 per decade in China and 0.148 per decade of global average in the same period. Some studies (Liu and Hou, 1998; Yao et al., 2000; Beniston et al., 1997; Diaz et al., 1997; Aizen et al., 1997; Giorgi et al., 1997) indicated that high-elevation area is more sensitive to global climate change and the global warming showed an altitudinal dependency. Therefore in the background of global warming, it is of essential scientific significance to the understanding of the

characteristics of climatic and environmental change at sky-high Mt. Qomolangma.

3.1.4 Response of glaciers to climate change since the 20th century, along with the warming, most glaciers on the Earth began to melt. The glaciers in Mt. Qomolangma region also showed significant changes under the global warming. Taking the biggest Rongbuk Glacier on the northern slope of Mt. Qomolangma as an example, from 1966 to 1997, the middle, east, and far east Rongbuk Glacier retreated 270m, 170m and 230m at rates of 8.7m/a, 5.5m/a and 7.4m/a, respectively (Ren et al., 1998). Compared with the former 30 years, the melting rates of the middle and east Rongbuk Glacier accelerated a little from 1997 to 2001 (Zhang and Zheng, 2004). Observation of the expedition into Mt. Qomolangma in 2005 showed that the upper limit of melting in east Rongbuk Glacier is 6400 m a. s.l., but it is 6350 m a.s.l. in 2002. The strong upward range of 50m indicated the quickly melting of glacier (Kang, 2005). Kang Shichang (2005) also suggested other evidence of the melting in east Rongbuk Glacier: expeditioners found a glacier cliff at 5600 m a.s.l. in 2002, but it totally disappeared at present, and also an ice lake nearby. The expedition set up 20 poles within a range of 6300-6500 m a.s.l. to measure the balance of glaciers account and only 4 remained, which fully indicated that the thickness of glacier was decreasing significantly. This is powerful evidence of the acceleration of glacier melting. Due to the lag of glaciers responding to climatic change and the more obvious warming in recent decade, the glaciers in Mt. Qomolangma region will keep melting.

3.2 Precipitation variation

3.2.1 Interannual variations of precipitation

The Himalayas are barriers to the warm and humid summer monsoon and make a great difference of precipitation between the north and the south of Mt. Qomolangma. Tingri, Lhaze, Shigatse and Gyangze in northern Mt. Qomolangma are of semiarid climate because of the barrier of Himalayas to humid monsoon.

The 34-year-averaged annual precipitation values in Lhaze and Shigatse, which are near the Yarlung Zangbo River, are 327. 4mm and 433.3mm and more than those in Tingri and Gyangze, which are 296.4mm and 292. 1mm. Nyalam in the southern part of Mt. Qomolangma is of subhumid climate and the 34-year-averaged annual precipitation value is 657.3mm. Figure 7 shows the precipitation variations of 5 stations during the recent 34 years and precipitations in Lhaze and Shigatse increased at a rate of about 3 mm/a (passing the 95

percent and 90 percent confidence test). Precipitations in Tingri and Gyangze are generally of little variations in fluctuations and that in Nyalam shows an obvious decreasing trend, but the significance level is low.

From the annual precipitation anomalies averaged for 4 stations in the north of Mt. Qomolangma (Figure 8a), it can be seen that the precipitation was plentiful in the 1970s, then it decreased in the early 1980s and the precipitation of 1982 was the least in the recent 34 years. The precipitation increased from the end of the 1980s to the early 1990s, and then it decreased in the mid-1990s and increased again since 1998. The precipitation from 1998 to 2000 was the most in the last 34 years, and itdecreased since 2001. The precipitation anomaly of Nyalam in the south of Mt. Qomolangma generally shows a decreasing trend (Figure 8b). The precipitation from the 1970s to 1980s shows an increasing trend and it was especially humid since the mid-1980s. A sharp drop of precipitation occurred in 1990 and persisted in the early 5 years of the 1990s. It was the driest in 1992, which was the same as in Nepal in the south of Mt.Qomolangma (Shrestha et al.,2000), and the precipitationincreased slowly subsequently.

3.2.2 Seasonal variations of precipitation

From the seasonal variations of precipitation shown in Figure 8c,it can be seen that on the northernslope of Mt. Qomolangma, the arid season, from November to March, is dominated by upper westerly winds, low precipitation and long sunshine hours. It is the rainy season from June to September and when it is overcast and rainy because of the significant influence of monsoon (Team of Scientific Expedition to Tibet,1975). The interdecadal variations of monthly mean precipitation is complicated in the north of Mt. Qomolangma. In the 1970s, the precipitation is the greatest in June. In the 1990s it was in August and since 2000, in July. The precipitation of Nyalam (Figure 8d) in the south of Mt.Qomolangma shows a doublepeak distribution of seasonal variation, which means that there is a peak value in August and September and a hypopeak value in February and March. The interdecadal variations of monthly mean precipitation of Nyalam are rather coincident and the precipitation of each month in the 1980s is universally more.

4. Discussion and conclusions

The main characteristics of climate change in Mt.Qomolangma region from 1971 to 2004 are:

The annual averaged surface air temperature clearly shows an increasing trend in Mt.

Qomolangma region, with a rate of about 0.234 per decade. Tingri, the highest mountain

station among the 5 high-elevation stations, has the strongest warming trend of 0.302 per

decade. The linear rate of annual minimum temperature increase is higher than annual average and mean maximum temperature and the warming is more obvious in the winter half year.

The rate of increasing air temperature in Mt. Qomolangma region is larger than global average (0.148=per decade) during the same period. The trend of temperature change in Mt. Qomolangma region is consistent with that of Tibet and the recent warming both began in 1978. But the warming in China began in 1984 and the global warming began in 1993.

Along with the warming, glaciers in Mt. Qomolangma region have melted strongly in the most recent 30 years. At present, the climate is still warming and the glaciers in Mt. Qomolangma region will keep melting.

Figure 6 Annual mean temperature anomalies of Mt. Qomolangma region(a), Tibet (b), China(c) and global average (d) (The thick solid curves represent the corresponding smoothed time series using a low-pass filter)

There are obvious differences in the precipitation trends between the north and south of Mt. omolangma. Precipitations of northern stations show an increasing trend but with low significance level. Precipitations of Nyalam in the southern part of Mt. Qomolangma shows a clear decreasing trend and a sharp drop of precipitation occurred in the early 1990s.

High-elevation areas are sensitiveto global climate change andthere is a tendency for the warmingtrend to increase with theelevation (Beniston et al., 1997;Diaz et al., 1997; Aizen et al.,1997; Giorgi et al., 1997).

From 1954 to 2002, the warmingin the central-southern Tibetis most obvious in China (Tao,2005). Our study showed thatTingri, with an elevation of4300m and a distance of morethan 100km to Mt. Qomolangma,has the strongest warming trend.Therefore, we can deduce that thewarming of Mt. Qomolangmahigh-elevation region is mostsignificant in China in the sameperiod, and the highest automaticmeteorological comprehensiveobservation station in the worldset up at the base camp of Mt.Qomolangma with a height of5032 m a.s.l. will play an importantrole in monitoring the globalclimate change. There are fewweather observation stations anddata in Mt. Qomolangma region.Combining the observation datafrom the weather stations inNepal and analyzing the differencesof climate change betweenthe north and south, we maydraw comprehensive and reliableconclusion on the trend andcharacteristic of climate changein Mt. Qomolangma region. Snowalbedofeedback in the TP maymagnify the signal of climatechange (Liu and Hou, 1998), andthe observation and analysis ofclimatic and environmentalchanges in sky-high Mt.Qomolangma region may provideeven earlier precursor signals forglobal climatic changes.(Excerpted from Vol.16,2006.3 ofJournal of Geographical Sciences)

Acknowledgements

The authors would like to thank Prof.Guo Yaxi and Dr. Luo Yong of China MeteorologicalAdministration for their generaldirection to this study. We thank TingriCounty government and Du Jun of MeteorologicalBureau of Lhasa for their kindlyproviding meteorological data used in thispaper. We acknowledge Prof. Shao Xuemeiand Dr. Zhang Xueqin of the Institute ofGeographic Sciences and Natural ResourcesResearch, and Prof. Liu Xiaodongof the Institute of Earth Environment, fortheir valuable suggestions.

* The author, Zhangs Group is one of research teamsin Institute of Geographic Sciences and NaturalResources Research (IGSNRR), CAS. It focuses onthe land use and land cover change, climate change,their impacts and driving forces on the TibetanPlateau. Correspondence should be addressed toZhang Yili (email: zhangyl@igsnrr.ac.cn).