Variability of atmospheric freezing level height derived from radiosonde data in China during 1958-2005 and its impact to cryosphere changes

YanJun Guo , YinSheng Zhang

1. National Climate Center, China Meteorological Administration, Beijing 100081, China

2. Institute of Tibetan Plateau Research, Chinese Academy of Sciences, Beijing 100085, China

Variability of atmospheric freezing level height derived from radiosonde data in China during 1958-2005 and its impact to cryosphere changes

YanJun Guo1*, YinSheng Zhang2

1. National Climate Center, China Meteorological Administration, Beijing 100081, China

2. Institute of Tibetan Plateau Research, Chinese Academy of Sciences, Beijing 100085, China

Atmospheric air temperature data from 92 stations in China’s radiosonde network were used to analyze changes in the freezing level height (FLH), glacier snow line, and ice edge from 1958-2005 (48 years) and to examine the impact of these changes on the cryosphere. In general, the FLH, glacier snow line, and ice edge exhibited latitudinal zonation, declining from south to north.Trends in the FLH, glacier snow line, and ice edge showed spatial heterogeneity during the study period, with prevailing upward trends. Temporally, the FLH, glacier snow line, and ice edge trends differed on various time scales.

freezing level height; glacier snow line; permafrost line; cryosphere; China

1. Introduction

Air temperature in the troposphere generally decreases with altitude, often reaching 0 °C over ground that is not frozen. The freezing level height (FLH, or 0 °C isotherm of free air) in the atmosphere is a critical parameter that influences the cryosphere in high mountain and high altitude areas by causing phase change in water in the cryosphere(Harriset al., 2000; Hoffmann, 2003; Francouet al., 2004;Coudrainet al., 2005; Vuilleet al., 2008). In particular, the mass balance of glaciers depends on the extent of ice melting and sublimation, and on the correlation of the permafrost distribution with temperature variation. Diaz and Graham(1996) noted a significant rise in FLHs in the tropics during 1958-1990, related to sea surface temperatures (SSTs) in the east-central equatorial Pacific. In the American sector of the tropics, the strongest relationship between FLH and SST was found for the SSTs preceding the FLH by about 3 months (Diazet al., 2003). The largest changes in FLH have been documented in recent decades, along with significant warming in high mountain regions (Diazet al., 2003).

China’s radiosonde network began observations in the 1950s and now has more than 100 stations. Recent works have examined radiosonde temperature time series from every station in China and have developed techniques to improve the data by quality controlling and homogenizing the time series (Guoet al., 2008; Guo and Ding, 2009). In this study, we examined the variation in FLH over China during 1958-2005, which was calculated from the homogenized radiosonde temperature time series. Furthermore, we studied a number of indicators of climatic variations in the cryosphere of China, including changes in the glacier snow line and ice edge. The data and methods are discussed in the next section, followed by our results. In the final section we discuss and summarize our major findings.

2. Material and methods

2.1. Radiosonde data

The FLH can theoretically be deduced from vertical profiles of temperature and geopotential height in free air. Ra-diosonde observations provided by the Chinese National Metrological Information Center (NMIC)/China Meteorological Administration (CMA) formed the basis of this analysis. Considering the amplitude of FLH variation, we used data for five mandatory pressure levels: the ground surface and 850, 700, 500, and 400 hPa. These levels were observed twice daily at 00 UTC and 12 UTC. The 00 UTC and 12 UTC series were combined into a merged radiosonde time series for the final homogenization procedure; sets of merged series were considered missing if either the 00 UTC or the 12 UTC series were missing. Seasonal anomalies were computed with reference to 1971-2000.

The 116 stations in the radiosonde network are distributed throughout China (Figure 1). We examined the data availability for each station and included as many stations as possible. Gaffenet al. (2000) demonstrated that the proportion of missing data is a key parameter in the reliability of a radiosonde time series. Guo and Ding (2009) found that 30% missing data is the critical value determining the usability of a time series from the Chinese radiosonde network.Thus, based on a maximum missing data fraction of 30%,we selected the optimal network (Figure 1, open circles).The analysis yielded a nominal radiosonde time series network of 92 stations for 1958-2005.

Figure 1 Location of radiosonde stations in China

Heterogeneity often exists in both instrumental climate records and radiosonde time series. Hence quality control(QC) and homogenization are necessary when using radiosonde data (IPCC, 2007). Many statistical methods have been developed to detect and correct inconsistencies in data sets, such as those caused by the use of different instruments and data correction methods. We employed a hydrostatic method (Collins, 2000) for the QC and a two-phase regression method (Easterling and Peterson, 1995) for the data homogenization. Previous studies showed that these methods are suitable for the Chinese radiosonde network (Guoet al., 2008; Guo and Ding, 2009).

2.2. The freezing level height (FLH)

The lowest five levels in the radiosonde time series(corresponding to the ground surface, 850, 700, 500, and 400 hPa) were examined for a transition to temperatures below 0 °C. The FLH was estimated for each snapshot by reverse interpolation of the temperature profile at each station to find the geopotential height of the 0 °C isotherm. The algorithm checked for zero crossings in the temperature profile between the ground surface and 400 hPa. If a single zero crossing existed, its altitude was taken as the freezing level. Two additional special cases were considered: no zero crossings (T＜0 °C throughout the entire profile) and multiple zero crossings due to temperature inversions. In the case whereT＜0 °C throughout the column, the freezing level was flagged as missing. In the case of multiple zero crossings,the locations were flagged and only the lowest FLH value was stored. The height of the freezing level was then obtained through linear interpolation between the geopotential heights of the transition levels. The mean monthly and annual FLH were also calculated.

2.3. Glacier snow line

The glacier snow line is defined as the altitude where the glacier mass balance is equal to zero, which means that solid precipitation is consumed by melting. It is also called the equilibrium line altitude (ELA) in glaciology. In this work,we calculated the ELA by a power function using the summer mean FLH as follows:

herehis the FLH in summer, andaandbare regression coefficients. Liuet al. (2000) found a correlation ofa=2,968.93 andb=0.09888 in the Qilian Mountains in China withR2=0.9877.

Equation(1)does not account for precipitation, which could cause problems with the resulting estimation. However, precipitation at the ELA has been found to correlate closely with air temperature (Liet al., 2008). In a monsoon climate region such as China, precipitation mainly occurs in summer and is not as important as air temperature to causing glacier fluctuation (Zhang, 1998).

2.4. Permafrost edge

Permafrost is defined as soil or rock that remains at or below 0 °C for at least one year. The permafrost edge is the limit of the permafrost distribution. Much effort has been made to deduce air temperature criteria for the existence of permafrost.Jianget al. (2003) demonstrated that annual mean air temperatures of -1.8 °C are required for permafrost to develop. Accordingly, we deduced the following formula for the permafrost edge altitude (PEA) calculation using the FLH:

here LAP is the lapse rate in free atmosphere, which can be computed from the radiosonde temperature and height.

3. Results

3.1. Annual mean FLH, glacier snow line, and permafrost line over China

Figure 2 shows distributions of the mean FLH, glacier snow line, and ice edge during 1958-2005. White areas in the three panels indicate an absence of stations (Figure 1).Generally, the FLH showed latitudinal zonation and declined from south to north. The FLH averaged about 1,000 m over most of northern China and rose steeply to 5,000 m at the margins of the tropics.

Figure 2 Distribution of mean freezing level height, glacier snow line, and ice edge in China during 1958-2005

The glacier snow line declined from south to north with a similar latitudinal dependence on the FLH. Several glaciers are closely monitored in China (WGMS, 2008). The average ELA of Glacier No. 1, located at 86.49°E, 43.06°N,was 4,049 meters above sea level (m a.s.l.); that of the July 1 Glacier, located at 99.45°E, 39.14°N, was 4,670 m a.s.l.; and that of the Donkemadi Glacier, located at 92.05°E, 33.04°N,was 5,600 m a.s.l. Compared with the climatology of the glacier snow line (shown in Figure 2), our results closely match the observations.

The averaged IEA during 1958-2005 ranged from 1,000 to 5,500 m a.s.l. The spatial pattern of the IEA also exhibited latitudinal zonation like that of the FLH and glacier snow line. Due to topographic effects, the spatial variation was rather homogenous in northeastern China, but sharp in central China. The IEA maximum reached 5,500 m a.s.l. in the central Tibetan Plateau.

3.2. Trend during 1958-2005

Linear trends of the FLH, glacier snow line, and ice edge over the last 48 years (1958-2005) were computed and are shown in Figure 3 (in units of m/decade). Over the whole of China, 72 stations showed positive trends and 20 showed negative trends in FLH during 1958-2005. Significant positive trends were irregularly distributed over the area. Extreme positive trends (more than 50 m/decade) were found at several stations in Inner Mongolia and northeastern and southern China. An extreme negative trend was found in western China, with a rate of -30 m/decade.

Figure 3 Distribution of linear trend during 1958-2005 for freezing level height, glacier snow line, and ice edge in China.Blue or red color denotes negative or positive with significance level above 95%.

The positive and negative trends of the glacier snow line during 1958-2005 varied in distribution across China. Fifty-four stations (59%) in the network had positive trends.The most pronounced rise in the glacier snow line was found over the Tibetan Plateau and far northwestern China.

More significant positive trends were found for the change in the permafrost edge over China. During the study period, the ice edge showed a positive trend at 76 stations(83%), and at 60% of these stations the positive trend was significant at the 95% level. The most extreme positive trend was 12 m/decade in central eastern China.

3.3. Decadal changes

To investigate the interannual changes in the FLH, glacier snow line, and ice edge in China, we summarized the decadal mean changes relative to those in 1971-2000 during the decades of 1960-1969, 1970-1979, 1980-1989,1990-1999, and 2000-2005 (Figure 4). Compared to the glacier snow line, the FLH and ice edge showed isochronous variation on decadal scales. Since 1980, both the FLH and glacier snow line maintained positive increases up to 2005.Average anomalies of FLH were 37.0 m and 41.0 m for 1990-1999 and 2000-2005, respectively. Average anomalies of the ice edge were 59.8 m and 37.0 m for 1990-1999 and 2000-2005, respectively.

The glacier snow line fluctuations were rather gentle and irregular on decadal scales compared to those of the FLH and ice edge. Average anomalies of the glacier snow line were 2.6, -14.2, 2.3, 6.4, and -7.0 m for 1960-1969,1970-1979, 1980-1989, 1990-1999, and 2000-2005, respectively, much lower than those of the FLH and ice edge in the same decades.

3.4. Nationwide average time series of FLH, glacier snow line, and ice edge

We averaged the time series of annual FLH, glacier snow line, and ice edge anomalies during 1958-2005 over the selected 92 stations in China (Figure 1). Nationwide, the FLH, glacier snow line, and ice edge showed similar variations during 1958-2005. All of the time series had downward trends from 1958 to 1968 and upward trends up to 2005. We calculated the trends by least-squares linear fitting derived from averaged time series; the results indicated trends of 13.5, 1.2, and 28.8 m/decade for the FLH, glacier snow line, and ice edge, respectively.

Figure 4 Decadal mean anomalies of the freezing level height, glacier snow line, and ice edge in China during 1960-2005

Figure 5 Variation in the anomalies of the freezing level height, glacier snow line, and ice edge in China during 1958-2005

4. Discussion and summary

Several previous works in western China suggested that the FLH might serve as an indicator of climate change through its impact on the cryosphere. Zhanget al. (2009)found that the sudden increase in the FLH over the western Tianshan Mountains was correlated with rapid glacier melt and maximal negative balance in glacial amount. Maoet al.(2004) demonstrated that the FLH could be an important factor for forecasting flooding in the Aksu River, where glacier and snow covers exist in headwater regions. Furthermore, Wanget al. (2008) concluded that the average discharge in the Hotan River Basin in western China responded on interannual and interdecadal scales to changes in regional FLH.

We used air temperature at four levels from China’s radiosonde network to analyze changes in the FLH, glacier snow line, and ice edge during a recent 48-year period and investigated the impact of these changes on the cryosphere.We examined radiosonde time series from 92 stations selected from the entire national network. Generally, the FLH,glacier snow line, and ice edge exhibited latitudinal zones and declined from south to north. The trends in the FLH,glacier snow line, and ice edge during 1958-2005 showed spatial heterogeneity and prevailing upward trends. Temporally, the FLH, glacier snow line, and ice edge trends varied on different time scales.

This study was funded by the Major State Basic Research Development Program of China (973 Program) under Grant No. 2010CB951701 and No. 2010CB428606, and by the Natural Science Foundation of China (No. 41071042 and No. 40775045). It was also supported by the Innovation Project of the Chinese Academy of Sciences(KZCX2-YW-BR-22), and special finance support from the China Meteorological Administration (GYHY200906017).

Collins WG, 2000. The operational complex quality control of radiosonde heights and, temperatures at the national centers for environmental prediction, Part I: Description of the method. Journal of Applied Meteorol ogy and Climatology, 40: 137-151.

Coudrain A, Francou B, Kundewicz ZW, 2005. Glacier shrinkage in the Andes and consequences for water resources. Hydrol. Sci. J., 50:925-932.

Diaz HF, Graham NE, 1996. Recent changes in tropical freezing heights and the role of sea surface temperature. Nature, 383: 152-155.

Diaz HF, Eischeid JK, Duncan C, Bradley RS, 2003. Variability of freezing levels, melting season indicators, and snow cover for selected high-elevation and continental regions in the last 50 years. Clim. Change,59: 33-52.

Easterling DR, Peterson TC, 1995. A new method for detecting undocumented discontinuities in climatological time series. Int. J. Climatol., 15:369-377.

Francou B, Vuille M, Favier V, Cáceres B, 2004. New evidence for an ENSO impact on low latitude glaciers: Antizana 15, Andes of Ecuador,280S. J. Geophys. Res., 109: D18106. DOI:10.1029/2003JD004484.

Gaffen DJ, Santer BD, Boyle JS, Christy JR, Graham NE, Ross RJ, 2000.Multidecadal changes in the vertical temperature structure of the tropical troposphere. Science, 287: 1242-1245.

Guo Y, Thorne P, McCarthy P, 2008. Radiosonde temperature trends and their uncertainties over eastern China. Int. J. Climatol., 28: 1269-1281.

Guo Y, Ding Y, 2009. Long-term free-atmosphere temperature trends in China derived from homogenized in situ radiosonde temperature series. J.Clim., 22(4): 1037-1051.

Harris NG, Gettys N, Bowman KP, Shin DB, 2000. Comparison of freezing-level ALHitudes from NCEP Reanalysis with TRMM precipitation radar brightband data. J. Clim., 13: 4137-4148.

Hoffmann G, 2003. Taking the pulse of the tropical water cycle. Science, 301:776-778.

Intergovernmental Panel on Climate Change (IPCC), 2007. Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge Univ. Press, Cambridge, U.K.

Jiang FC, Wu XH, Wang SB, Zhao ZZ, Fu JL, 2003. Basic features of spatial distribution of the limits of permafrost in China. Journal of Geomechanics, 19(14): 12-22.

Li X, 2008. Cryospheric change in China. Global and Planetary Change, 62:210-218.

Liu C, 2000. Glaciers and their distribution in China. In: Shi YF (Ed.). Glaciers and Their Environments in China: The Present, Past and Future.Science Press, Beijing. 9-53.

Mao W, 2004. Relationship of 0 °C level height and summer flood of Aksu River, Xinjiang. Journal of Glaciology and Geocryology, 26(6): 117-123.

Vuille M, Francou B, Wagnon P, Juen I, Kaser G, Mark BG, Bradley RS,2008. Climate change and tropical Andean glaciers—Past, present and future. Earth Sci. Rev., 89: 79-96.

Wang Y, 2008. Response of summer average discharge in the Hotan River to changes in regional 0 °C Level Height. Advances in Climate Change Research, 4(3): 151-155.

WGMS, 2008. Global Glacier Changes: Facts and Figures. In: Zemp M,Roer I, K??b A, Hoelzle M, Paul F, Haeberli W (Eds.). UNEP. World Glacier Monitoring Service, Zurich, Switzerland.

Zhang G, 2009. The response of the Glacier No. 1 to the height change of the 0 °C Level in summer at the riverhead of the Urümqi River, Tianshan Mountains. Journal of Glaciology and Geocryology, 31(6): 117-123.

Zhang Y, 1998. The response of glacier ELA response to climatic fluctuation on high Asia. Bulletin of Glacier Research, 16: 1-11.

10.3724/SP.J.1226.2011.00485

*Correspondence to: Dr. YanJun Guo, National Climate Center, China Meteorological Administration. No. 46, Zhongguancun Nandajie, Haidian District, Beijing 100081, China. Email: gyj@cma.gov.cn

11 June 2011 Accepted: 12 August 2011