Water cycle changes during the past 50 years over the Tibetan Plateau: review and synthesis

130 Cold Region Hydrology in a Changing Climate (Proceedings of symposium H02 held during IUGG2011 in Melbourne, Australia, July 2011) (IAHS Publ. 346, 2011). Water cycle changes during the past 50 years over the Tibetan Plateau: review and synthesis YINSHENG ZHANG 1 & Y. GUO 2 1 Key Lab. of Tibetan Environment Changes and Land Surface Processes (TEL), Institute of Tibetan Plateau Research, CAS, China yszhang@itpcas.ac.cn 2 National Climate Centre, China Meteorological Administration, China Abstract The evidence for water cycle changes during the past 50 years on the Tibetan Plateau (TP) is synthesised by analyses of the meteorological observations and reanalysis data, and review of relevant studies. Robust warming has been evident, and decreasing wind speed has led to a weak atmospheric forcing. Snow depth decreased and the active layer depth increased in the permafrost region. In response to these changes, evapotranspiration slightly increased due to a wetter ground surface. Inhomogeneous changes in precipitation result in uncertainties regarding trends in river discharge over the regions and basins. Key words water cycle; Tibetan Plateau; climate change 1 INTRODUCTION In response to global warming, a critical question in hydrology is: if the climate warms in the future, will there be an intensification of the water cycle? and, if so, what evidence is currently available to address this problem? Empirical evidence for ongoing intensification of the water cycle would provide additional support for the theoretical framework that links intensification to warming (Huntington, 2005). To understand changes in the hydrological environment, it is necessary not only to precisely investigate the process at a local scale, but also to review long-term and large-scale data to obtain background knowledge. The Tibetan Plateau (TP), the largest geomorphologic unit (Fig. 1) on the Eurasian continent, and highest plateau in the world, contains an Asian water tower ; a large amount of water is stored in the world s highest and largest plateau. It is vital to understand whether supplies to these water resources have experienced any changes during recent global warming. Climate warming on the plateau during recent decades has been suggested by meteorological observations and ice core records. The climatic warming seems more evident on the plateau than the global mean trend during the past 50 years (Yao & Zhu, 2006). There are five hydrological basins on the TP: Yangtze River, Yellow River, Southwest, Northwest, and the Plateau closed basin. More than 70 metrological observatories and some hydrological stations are located on the TP. These data mostly represent natural processes because human water use is negligible and can thus be ignored in these gauged catchments. These data have been widely used in estimations of variations in runoff and evaporation. Various satellite and reanalysis data provide useful information of the water cycle, such as precipitation and soil moisture. This work briefly reviews the current state of science regarding historical trends in hydrological variables, including precipitation, runoff, water vapour, soil moisture, evapotranspiration and snow cover. Data are often incomplete in spatial and temporal domains, and regional analyses are variable and sometimes contradictory; however, most studies do indicate an ongoing change of the water cycle. 2 TEMPERATURE CHANGE Recent analysis of air temperatures (Liu & Chen, 2000) shows that the major portion of the plateau has experienced statistically significant warming since the mid-1950s. Temperature anomalies during 1960 2009 averaged for the different basins on the TP revealed a robust and synchronous warming over space and time (Fig. 2). The linear rates of annual mean temperature increases for Copyright 2011 IAHS Press

Water cycle changes during the past 50 years over the Tibetan Plateau: review and synthesis 131 Fig. 1 Distribution of hydrological basins and meteorological stations on the Tibetan Plateau (TP). 2.0 1.5 Yangtze River South-west Plateau closed basin Yellow River North-west Anomaly (oc) 1.0 0.5-0.5-1.0-1.5 1960 Annual mean air temperature (oc) 1965 1970 1975 1980 1985 1990 1995 2000 2005 Fig. 2 Variation of air temperature anomalies during 1960 2009, averaged for different basins on the TP. the different basins during 1960 2009 were about 0.30, 0.24, 0.28, 0.45 and 0.38 oc/decade for the Yangtze River, Yellow River, Southwest, Northwest and Plateau closed basin, respectively. These changes exceed the warming rates for the Northern Hemisphere and the same latitudinal zone for the same period (IPCC, 2007). Such robust warming may lead to significant changes in the water cycle in this cryosphere dominant region (Yao & Zhu, 2006). 3 CHANGES IN WATER VAPOUR AND PRECIPITATION Recent satellite measurements indicate trends towards increasing water vapour in the tropical atmosphere during 1993 1999 that were consistent with model predictions (Minschwaner & Dessler, 2004). Table 1 summarizes the decadal changes of surface vapour pressure and precipitation during 1960 2009 on the TP. Surface vapour pressure, an indicator of atmospheric water content, synchronously increased among the five basins since 1990, but with different amplitudes. On a global basis, precipitation over land increased by about 2% over the period 1900 1998 (Dai et al., 1997; Hulme et al., 1998). Regional variation in precipitation changes was highly significant. For example, zonally averaged precipitation increased by 7 12% between 30 N and 85 N, and by 2% for the regions between 0 S 55 S. Precipitation has decreased substantially in some regions (IPCC, 2007). Precipitation over the TP has different changes relative to atmospheric

132 Yinsheng Zhang & Y. Guo Table 1 Decadal mean anomalies of vapour pressure and precipitation for the five basins on the TP. Mean Yangtze Yellow Southwest Northwest Plateau closed basin Vapour pressure (hpa) 1960 69 6 8 4 0.17 7 4 1970 79 9 0.10 4 0.13 8 0.10 1980 89 8 3 5 0.12 5 9 1990 99 0.10 9 6 0.12 0.10 0.13 2000 09 0.14 9 6 0.15 0.19 0.30 Precipitation (mm) 1960 69 11.2 0.4 5.9 18.5 12.6 23.6 1970 79 8.2 6.8 2.2 14.5 6.2 11.6 1980 89 6.9 28.9 10.5 6.3 13.2 5.8 1990 99 2.4 20.2 6.0 6.9 6.8 1.3 2000 09 16.6 0.4 22.5 17.5 14.3 42.6 Bold type notes positive anomalies water content (Table 2). Regional precipitation mainly experienced positive anomalies during the decades 1980 89 and 2000 09, with the exceptions of the Southwest basin and Yangtze River basin. In the period of 1990 99, precipitation decreased in most of the basins, but increase in the Southwest basin. The changes in precipitation, clearly, were rather inhomogeneous both in time and space. 4 CHANGING IN WIND, SUNSHINE AND CLOUDINESS Decreases in pan evaporation (P-EVA) have been observed on the TP (Fig. 3(a)), similar to the dominant trend over the world (Peterson et al., 1995). Such decreases seem inconsistent with the observed trends of increasing temperature and precipitation, resulting in an evaporation paradox (Brutsaert & Parlange, 1998). Several analyses have suggested, however, that decreasing P-EVA is consistent with increasing surface warming and an acceleration of the water cycle (Roderick & Farquhar, 2004). Various mechanisms have been suggested to explain the apparent paradox, e.g. it has been suggested that decreases in solar irradiance, resulting from increasing cloud coverage and aerosol concentrations, and decreases in diurnal temperature range, would cause decreases in pan evaporation (Peterson et al., 1995; Roderick & Farquhar, 2002). From our results, however, decreases in P-EVA may be primarily due to significant decreases in wind speed (WS) (Fig. 3(b)) and in sunshine duration (Fig. 3(c)), although the increase in lower cloudiness shows a complex trend (Fig. 3(d)). 5 CHANGES IN SNOW COVER AND ACTIVE LAYER DEPTH It is difficult to evaluate the hydrological impact of snow cover changes on the TP, due to lack of supports to process studies. Table 2 summarizes decadal changes of annual maximum snow depth, which is considered to be the most important index of the influence of snow cover on hydrological processes. In the period of 1980 1989, the entire TP showed positive anomalies of maximum snow depth, implying more melting water penetrating the soil or forming runoff directly in the next spring. Decrease in the maximum snow depth was evident since 1990, as demonstrated by Qin et al. (2006). Significant shrinkage of the permafrost has been reported on the TP (Yao & Zhu, 2006). Compared to other changes, active layer depth in the permafrost region is more important to hydrological processes, because it alters the hydraulic features of the surface layer. Table 2 summarizes decadal changes in the active layer depth. The positive anomalies reveal a significant thickening of the active layer, which may be due to climate warming.

Water cycle changes during the past 50 years over the Tibetan Plateau: review and synthesis 133 200 150 ( a) P- EVA mm 100 50 0-50 -100-150 Tr end= - 24. 5 mm/ 10a -200 200 150 ( C) Sun shine hr 100 50 0-50 -100-150 Tr end= - 7. 2 hr / 10a -200 Tr end= - 0. 1 m/ s/10a Tr end= 0. 02 / 10a ( b) WS m/s ( d) Lower cl oudi ness / 10 Fig. 3 Meteorological anomalies during 1960 2009 on the TP, (a) pan evaporation (P-EVA), (b) wind speed (WS), (c) sunshine duration, and (d) lower cloudiness. 0. 6 0. 4 0. 2 0. 0-0.2-0.4-0.6 4. 0 3. 0 2. 0 1. 0 0. 0-1.0-3.0-4.0 Table 2 Decadal mean anomalies of maximum snow depth and active layer depth for the five basins on the Tibetan Plateau. Mean Yangtze Yellow Southwest Northwest Plateau closed basin Max depth of snow (cm) 1960 69 0.7 0.7 0.9 1.3 0.2 1.3 1970 79 0.1 0.1 0.1 0.3 0.1 0.1 1980 89 0.5 0.8 0.4 0.7 0.2 0.3 1990 99 0.1 0.7 0.3 0.1 0.3 2000 09 1.0 1.3 0.8 1.1 0.5 1.1 Active layer depth (cm) 1960 69 6.7 0.6 6.7 14.8 2.3 16.6 1970 79 2.1 3.6 0.5 1.8 4 9.8 1980 89 3.6 2.6 5.6 3.7 1.2 6.1 1990 99 3.5 3.7 4.9 3.0 2.9 7.8 2000 09 12.5 14.2 7.5 12.4 12.6 6.8 6 CHANGES IN SOIL MOISTURE, EVAPOTRANSPIRATION AND RUNOFF Due to lack of observations, we use ERA-40 reanalysis data to elucidate changes in surface water storage and evapotranspiration (ET). Figure 4 plots both annual ET and summer soil moisture in the 0 10 cm layer during 1960 2000. The averaged curves imply slight increases both in ET and soil moisture in the 0 10 cm layer, with linear increases of 1.0 mm and 0.6% per decade, respectively. Some previous reports (e.g. Brutsaert & Parlange, 1998), concluded that increasing water vapour concentration results in a warming-induced increase in ET. For the cryosphere, an increasing ET would be caused by increases in surface soil moisture, such as shown in Fig 4. Table 3 shows discharge trends for the three TP basins, cited from Cao et al. (2006) that includes the Mann-Kendall tests. A positive (negative) value of Zc indicates an ascending (descending) trend. Results suggest a decreasing discharge trend in the Yangtze and Yellow River basins during 1956 2000. However, an increasing trend in discharge was observed for the Southwest basin.

134 Yinsheng Zhang & Y. Guo 6.0 4.0 Eva 0-10 cm soilmoisutre (%) 8.0 6.0 1 Anomaly of eva (mm) 2.0-4.0 4.0 2.0-4.0 Anomaly of soil moisture (%) 5.0-5.0 Anomaly of eva (mm) -6.0-6.0-8.0 1960 1965 1970 1975 1980 1985 1990 1995 2000-8.0-1 -5.0 5.0 1 anomaly of 0-10 soil moisture (%) -1 Fig. 4 Anomalies of ERA-40 annual ET (eva) and summer soil moisture in the 0 10 cm layer during 1960-2000 (left) and their correlation (right). Table 3 Trends in mean annual discharge and Mann-Kendall tests (after Cao et al., 2006). Basin First year Last year Series length Test Zc Yangtze River 1956 2000 45 1 Yellow River 1956 2000 45 0.30 Southwest 1956 2000 45 0.47 7 SUMMARY Figure 5 summarizes the results of this analysis as follows: 1 Robust and synchronous warming over space and time was a primary characteristic of climatic change over the TP. 2 Significant decreases in wind speed and sun shine, and also an increase in lower cloudiness caused weakening of ground surface potential evaporation, as indicated by a decrease in pan evaporation. 3 Precipitation has slightly increased since 1990, but with spatial inhomogeneity. 4 Due to global and regional warming, maximum snow depth decreased and active layer depth increased. 5 Although pan evaporation tended to decrease, ET still slightly increased in response to increasing surface soil moisture. 6 There remains substantial uncertainty with regard to trends in river discharge. Robust & synchro Warming! Atmospheric Water since 1990 Precipitation in 1990-99 & 2000-09 Inhomogeneous in tem & spa Hydro-forcing Sunshine Wind speed Ep Ground surface Snow cover Permafrost Vegetation Surface Water moisture ET Runoff? Fig. 5 Schematic changes of hydrological variables on the TP during 1960 2009, & for increasing and decreasing trends, respectively. Ep-pan evaporation, ET-evapotranspiration.

Water cycle changes during the past 50 years over the Tibetan Plateau: review and synthesis 135 It is necessary to mention that we did not account for changes in glacier mass. The glaciers in this region are shrinking and that may significantly affect river discharge and water resources over the TP (Yao & Zhu, 2006). Acknowledgements This work was supported by the Major State Basic Research Development Program of China (973 Program) under Grant no. 2010CB951701, the Natural Science Foundation of China (no. 595 40775045 and no. 41071042), the Innovation Project of Chinese Academy of Sciences (KZCX2-YW-BR-22). REFERENCES Brutsaert, W. & Parlange, M. B. (1998) Hydrologic cycle explains the evaporation paradox. Nature 396, 30. Cao, J., et al. (2006) River discharge changes in the Qinghai-Tibet Plateau. Chinese Science Bull. 51(5), 594 600. Dai, A., Fung, I. Y. & DelGenio, A. D. (1997) Surface observed global land precipitation variations during 1900 1988. J. Climate. 10, 2943 2962. Hulme, M., Osborn, T. J. & Johns, T. C. (1998) Precipitation sensitivity to global warming: comparisons of observations with HadCM2 simulations. Geophys. Res. Lett. 25, 3379 3382. Huntington. T. G. (2006) Sugar and syrup evidence for intensification of the global water cycle: Review and synthesis. J. Hydrol. 319, 83 95. IPCC (2007) Climate Change 2007: The Scientific Basis. Cambridge University Press, UK. Liu, X. & Chen, B. (2000) Climatic warming in the Tibetan Plateau during recent decades. Int. J. Climatol. 20(14), 1729 1742. Minschwaner, K., & Dessler, A. E. (2004) Water vapour feedback in the tropical upper troposphere: model results and observations. J. Climate 17, 1272 1282. Qin, D., et al. (2006) Snow cover distribution, variability, and response to climate change in western China. J. Climate 19, 1820 1833. Peterson, A. T., et al. (1995) Evaporation losing its strength. Nature 377, 687 688. Roderick, M. L. & Farquhar, G. D. (2004) Changes in Australian pan evaporation from 1970 to 2002. Int. J. Climatol. 24, 1077 1090. Yao, T. D. & Zhu, L. P. (2006) The response of environmental changes on Tibetan Plateau to global changes and adaptation strategy. Adv. Earth Sci. 21, 459 464.