2. The State Key Laboratory of Cryospheric Sciences, Cold and Arid Regions Environmental and Engineering Research Institute, Chinese Academy of Sciences, Lanzhou, Gansu 730000, China;
3. Water Resources Section, Delft University of Technology, Delft, The Netherlands
Many studies have reported that the global climate has warmed in recent decades and have projected how the climate may continue to change in future decades and centuries (e.g., IPCC, 2007; Pachauri et al., 2014). Recent observations certified that increased global temperatures have led to more negative glacier mass balances (Dyurgerov and Meier, 2005; Oerlemans, 2005; Prasad et al., 2009; Sorg et al., 2012; Miles et al., 2013). It is expected that in China the average surface air warming will continue in the 21st century (China Meteorological Administration, 2006). Since 1986/1987, warming has led to the retreat of glaciers in the Hengduan Mountains, causing reductions of glacier area that have had a considerable impact on the variability of runoff (He et al., 2010; Li et al., 2010; Liu et al., 2010; Zhang et al., 2010; Pan et al., 2012). Changes in glacier runoff impact local water resources and the ecological environment.
In the Hengduan Mountains, glaciers are among the main sources of water that crucially contribute to the sustainability of hydroelectric power generation, agriculture, tourism, and other activities (Coudrain et al., 2005; Piao et al., 2010). Monsoonal temperate glaciers are highly sensitive to climate changes, and a small fluctuation of temperature can cause such glaciers to shrink or grow noticeably. Thus, these monsoonal glaciers are very direct and clear indicators of climatic change. Observation data have shown that river flows have increased in the Hengduan Mountains, reserves of water resources have been sharply reduced, and flood and debris flow are tending to increase (He et al., 2010; Zhu et al., 2012).
During the last 20 years, the Hydrologiska Byråns Vattenbalansavdelning (HBV)model, which was developed by the Swedish Meteorological and Hydrological Institute (SMHI), has become widely used for runoff simulations in Sweden (Bergström, 1992; Bergström et al., 2001; Seibert, 2005)and in about 40 other countries, sometimes in modified versions. A version of the HBV model, HBV Light, provides an easy way to use Windows version for research and education. Numerous studies have investigated the applicability of the HBV model in surrounding regions of the Hengduan Mountains (e.g., Zao et al., 2007; Jin et al., 2008; Gao et al., 2011; Xiang et al., 2011), and some studies with the HBV model have investigated the impact of climate change on hydrology and water resources (Bergström et al., 2001; Menzel and Bürger, 2002; Pilling and Jones, 2002; Arnell et al., 2003; Christensen et al., 2004; Steele-Dunne et al., 2008; Dush et al., 2009; Zhu et al., 2012).
Due to the limited number of past observations in the Hailuo Creek Basin, it was difficult to determine the long-term glacier mass balance and glacier runoff of the basin. In this study, the data of six meteorological stations were used to produce appropriate precipitation and temperature data, and the performance of the hydrology model was validated by forcing it with meteorological data for the reference period (1952-2009) . Comparing the simulated stream flow with observations, the impacts of climate change were analyzed using results from the successfully calibrated and validated model. The interrelations between temperature, precipitation, potential evapotranspiration, glacier runoff, and river runoff were studied.2 Study area, data, and methods 2.1 Basin
The Hailuo Creek Basin is a small (80.5 km2)monsoonal temperate glacier basin (Figure 1)located in the eastern part of Mt. Gongga, which is in the Hengduan Mountains. The runoff derived from snow and ice plays a crucial role in regional water resources in the Hailuo Creek River Basin (Cao, 1995; Xie et al., 1995, 2001; Li et al., 2004; Yin et al., 2008; Pan et al., 2012). During the rainy season, the basin is under the control of the Indian monsoon, with rainy wet and warm weather. In the dry season (October to the following May), the basin is under the control of the Westerly Circulation, with sunny, windy, dry, and cold weather. The basin encompasses seven glaciers: Hailuo Creek Glacier, Hailuo Creek Glacier No. 1, Hailuo Creek Glacier No. 2, and four other smaller glaciers; collectively, they cover an area of 29.66 km2 (Pu, 1994). During the study period, the annual precipitation at the glacier tongues (3, 000 m a.s.l.)was about 1, 960 mm, with a maximum from June to September, and the annual mean temperature was 4 ℃. The annual mean temperature at the equilibrium line altitude (ELA; 4, 900 m a.s.l.)was about −4.4 ℃, and the annual precipitation was 3, 000 mm (Su and Liu, 2002). In the non-glacier zone, the vegetation coverage rate was about 95%.2.2 Data
We used the 30-m-resolution Global Digital Elevation Model (DEM) (http://datamirror.csdb.cn), and a vector map of glaciers was obtained from the Chinese Glacier Inventory (Pu, 1994). Each elevation zone was specified according to the 30-m-resolution DEM, and we extracted the digital drainage network and basin boundaries with ArcGIS9.2 software. The glacier distribution map in the Hailuo Creek Basin was produced from the Chinese Glacier Inventory. The ArcGIS software assigned the portions of each elevation-vegetation zone in the glacier areas and non-glacier areas of the entire catchment area.
In this study, the meteorological data provided by the Chinese Meteorological Administration (http://cdc.cma.gov.cn), specifically from the Alpine Ecosystem Observation and Experiment Station of Mt. Gongga, were used to force the HBV model to examine the impact of climate change on hydrology in Hailuo Creek Basin. There are six national weather stations in Hailuo Creek: Kangding (101.58°E, 30.03°N), Jiulong (101.30°E, 29.00°N), Muli (101.16°E, 27.56°N), Xiaojin (101.58°E, 30.03°N), Yuexi (102.31°E, 28.39°N), and Xinlong (100.19°E, 30.56°N). Two automatic weather stations (AWSs)at 1, 600 and 3, 000 m a.s.l. were established in 1988. The air temperature and precipitation data were used as the input data in this study. According to the two AWSs, the temperature lapse rate was calculated as 0.64 ℃/100 m and the precipitation as 6.45%/100 m. Potential evaporation was approximately substituted with water surface evaporation, which was calculated by the Penman-Monteith model modified by FAO (Thornthwaite, 1951; Jensen et al., 1990; Liu et al., 2005; Jia et al., 2009; Zhu et al., 2012). The air temperature and precipitation data in the Hailuo Creek Basin from 1952 to 2009 were reconstructed using data recorded at the six meteorological stations described above (Figure 1, Table 1). The data of meteorological elements were obtained from the China Meteorological Data Sharing Service System (http://cdc.cma.gov.cn/home.do). Data pertaining to changes in the Hailuo Creek Glacier over the last several decades came from some previous studies (Li et al., 2010; Liu et al., 2010; Pan et al., 2012).
|Elevation (m)||Non-glacier area (%)||Glacier area (%)|
|2, 600-3, 100||0.010||0.044||0.044||0.006||0.000||0.000|
|3, 100-3, 600||0.015||0.045||0.030||0.010||0.000||0.000|
|3, 600-4, 100||0.015||0.039||0.035||0.011||0.000||0.000|
|4, 100-4, 600||0.010||0.030||0.023||0.037||0.000||0.000|
|4, 600-5, 100||0.015||0.012||0.010||0.043||0.020||0.000|
|5, 100-5, 600||0.003||0.008||0.007||0.040||0.032||0.010|
|5, 600-6, 100||0.013||0.006||0.005||0.039||0.031||0.004|
|6, 100-6, 600||0.019||0.017||0.010||0.034||0.017||0.003|
|6, 600-7, 100||0.032||0.030||0.020||0.005||0.003||0.002|
|7, 100-7, 514||0.038||0.030||0.030||0.001||0.002||0.000|
Hailuo Creek Basin water stage was recorded by an automatic water gauge every 15 min. The flow velocity was measured manually by a current meter. Combined with the water stage data, the relative discharge could thus be calculated. Through several measurements at different water stages, the relationship between water stage and discharge was established. With this relationship, the daily discharge was calculated from 1994 to 2009 and was used for HBV model calibration.
The parameter values of threshold temperature, degree-day factor, and maximum value of soil moisture storage were observed. Threshold temperature was observed by a small-sized automatic weather station in 2011. Degree-day factor was calculated by glacier mass balance measurement data during 2008-2010. Soil samples were taken from different drilling depths. These samples were packed in an aluminum box, then baked in an oven for 12 hours continuously at a temperature of 105 ℃.2.3 Methods
Data from direct long-term monitoring were unavailable for the basin, and it was therefore necessary to reconstruct the historical runoff. The daily air temperature and precipitation in the Hailuo Creek Basin were reconstructed for the period 1952-1994 based on data of meteorological observations in the Hailuo Creek Basin during 1994-2009 and data from the six weather stations around the Hailuo Creek Basin during 1952-2009 (Figure 1).
The Mann-Kendall trend test (Mann, 1945; Kendall and Gibbons, 1990), one of the most widely used nonparametric tests for detecting trends in time series, was applied to analyze the temperature and precipitation data. This test can also be used to evaluate whether there is a significant discontinuity in data collected over a period of time, so it was also used to evaluate whether there was a sharp change in temperature and precipitation at different sites in Hailuo Creek Basin and other weather stations.
Kriging interpolation is the ideal geostatistics method to analyze the spatial variation of meteorological elements (Yamamoto, 2005; Xu et al., 2006; Zhu et al., 2013). The advantages are: (1) the degree of spatial similarity is assessed by the mean-variance based on spatial statistics, and the error can be theoretically estimated point by point, which does not have the boundary effects of regression analysis; and (2) using the structural characteristics of the data, missing data can be added by the method of linear unbiased optimal estimation. The influence of topography can be considered at the same time, while in other interpolation methods it cannot.
Comparing the reconstructed and observed monthly temperatures (R2=0.869, p <0.01) (Figure 2a), the average RE (Random error)was 7.8% from 1994 to 2002. Comparing the reconstructed and observed monthly precipitation (R2=0.88, p <0.01) (Figure 2b), the average RE was 9.4% from 1994 to 2002. The error of the reconstructed temperature values was small, but the error of the reconstructed precipitation values was generally large. Considering the great variation of precipitation in mountainous regions, this simulated result was fairly satisfactory (The average RE was 27.6%). Using the above method, the annual precipitation at Hailuo Creek Basin from 1952 to 2009 was reconstructed, and this reconstructed precipitation and temperature data provided the required forcing data for the HBV Light hydrology model that was used to simulate stream flow in the Hailuo Creek catchments.2.4 Model description
The HBV Light conceptual runoff model was coupled with a more detailed snow and glacier melt subroutine employing the degree-day approach (Braun and Aellen, 1990). In this study, the HBV Light model was applied to model the runoff depth and glacier runoff in Hailuo Creek Basin, which considered various aspect classes in each elevation belt (Hottelet et al., 1993). It was also coupled with Monte Carlo and generic algorithms to estimate the parameters automatically (Seibert, 1999).
The model simulates daily discharge using daily rainfall, temperature, and monthly potential evaporation as input data. Thirteen parameters need to be estimated (Table 2). This model includes four routines: the snow and glacier melt routine, the soil routine, the response routine, and the routing routine. First, the changes of precipitation and temperature with elevation were calculated using the two parameters PCALT (precipitation lapse rate)and TCALT (temperature lapse rate). Precipitation was divided into snow and rain, depending on whether the temperature was above or below a threshold temperature (TT, ℃). In the snow and glacier routine, the amount of melt water was calculated with the degree-day method (Braithwaite and Olesen, 1984), which used positive degree days (while daily mean temperature was above 0 ℃)multiplied by a factor. This model also considered the influence of different aspects and the different melt of glaciers and snow (Hottelet et al., 1993). Refreezing of melt water (Fountain and Tangborn, 1985; Fujita et al., 1996; Fujita et al., 2007)was also considered in this model. In the soil routine, rainfall and snowmelt (Pi, mm)were divided into water filling the soil box and groundwater recharge (Re, mm)depending on the relation between water content of the soil box (SM, mm)and its largest value (FC, mm). Actual evaporation from the soil box equaled the potential evaporation if SM/FC was above LP, while a linear reduction was used when SM/FC was below LP (Bergström and Forsman, 1973). The recharge divided by actual evaporation came into the response routine. In the response routine, two linear tanks were used to control the outflow. In the end, this runoff was finally transformed by a triangular weighting function defined by the parameter MAXBAS. The Nash-Sutcliffe Efficiency coefficient (Reff; Nash and Sutcliffe, 1972) and the coefficient of determination (R2)were used for assessment of the simulations. More details can be found in the model's user manual (Seibert, 2005).
|Parameter||Definition||Units||Value range||Value||Calibration method|
|TT||Threshold temperature||℃||−1.78||Field observation|
|DDF||Degree-day factor||mm/ (℃·day)||5||Field observation|
|CFR||Refreezing coefficient||Dimensionless||0-0.6||0||Monte Carlo|
|Soil moisture routine|
|FC||Maximum value of soil moisture storage||mm||300||Field observation|
|LP||Fraction of FC above which actual ET equals potential ET||Dimensionless||0.92||Field observation and calculation|
|BETA||Shape coefficient (parameter that determines the relative contribution to runoff from rain or snowmelt)||Dimensionless||1-5||1.04||Literature (Booij, 2005; Steele-Dunne et al., 2008)|
|CET||Correction factor for runoff||C_1||0.3||Field observation and calculation|
|K1||Recession coefficient (upper box)||day−1||0.01-0.2||0.1||Literature (Booij, 2005; Steele-Dunne et al., 2008)|
|K2||Recession coefficient (lower box)||day−1||0.001-2||1.85||Literature (Booij, 2005; Steele-Dunne et al., 2008)|
|PERC||Maximum rate of recharge between the upper and lower groundwater boxes||mm/day||4||Field observation and calculation|
|MAXBAS||Length of triangular weighting function in routing routine||day||1-5||1.26||Literature (Booij, 2005; Steele-Dunne et al., 2008)|
To obtain a proper set of parameters, each parameter was given a reasonable range according to the literature and observation data. A set of parameters was then automatically obtained by employing the Monte Carlo method (Hornberger et al., 1986; Seibert, 1997; Uhlenbrook et al., 1999). Observed hydrological data for 1995, 1996 and 2002 were selected for calibration (Figure 3). In addition, data for 1994, 1997 and 1998 were used for validation.
In the calibration period, Reff and R2 were 0.88 and 0.86, respectively, during 1995; 0.89 and 0.86 during 1996; and 0.88 and 0.86 during 2002 (Figure 3). In the validation period, Reff and R2 were 0.76 and 0.71, respectively, during 1994; 0.85 and 0.79 during 1997; and 0.74 and 0.70 during 1998 (Figure 4). This was acceptable on a daily scale, especially considering the uncertainties in both observation and simulation data in such an alpine area.
For a well-defined parameter, the upper boundary should have a distinct peak, while in an ill-defined parameter the upper boundary would have a broad plateau. In the Hailuo Creek catchments, TT, DDF, FC, LP, CET, and PERC were the best-defined parameters by field experimentation. Figure 4 shows the simulated stream flow in the Hailuo Creek Basin within a year during the calibration period, using the observed meteorological data to force the HBV Light conceptual hydrology model. Figure 4demonstrates that the model was capable of reproducing the observed runoff depth quite well. The ensemble spread was relatively small compared to the dynamic range of values. The observations fell within the ensemble spread on almost all of the days. It can be seen that there were some discrepancies between the simulated and observed runoff values, particularly during the winter peaks when it was often underestimated.3.2 Simulation
Comparing the simulated and observed cycles of stream runoff calculated from the calibration period of 1994-1998 (Figure 5), the HBV Light conceptual hydrology model was clearly not perfect, even when forced with good calibration data. A simulated month of summer flow was compared with observed stream flow data, and the mean annual error was 9% and the summer error was 7% (Figures 3, 4). The mean error in winter, spring, and autumn was 17%, demonstrating that the model was capable of simulating the summer and annual observed flow. Considering the dominant role of glacier runoff, the accuracy of the degree-day model possibly decreased with increasing temporal resolution (Hock, 2003). In summer, the observed monthly stream flows generally fell within the ensemble, but in winter all ensemble elements were generally biased with respect to the observed data: winter flows were considerably overestimated in 1995 and 2002 and significantly underestimated in the 1994, 1996, 1997 and 1998. Summer flows were considerably underestimated in 1994, 1996, 1997, and 1998.3.3 Validation of past hydrology
This study revealed that the watershed parameters did not change when the HBV Light hydrology model parameters were used for a catchment. The stream flow was generated using past climate data (1952-2009) from observations at the six meteorological stations and modified by Kriging interpolation. Dynamically downscaled precipitation and temperature data were used to run the HBV Light hydrology model. The simulated flow was compared with the observed flows in the reference period. An ensemble of the seasonal cycle of mean monthly flow in the Hailuo Creek catchments was validated against observed stream flow data (Figure 5). The simulated results indicated that the annual runoff was in a significantly increasing trend (50.11 mm/a)during 1951-2009 (Figure 6).
The runoff in the Hailuo Creek Basin increased significantly after 1994 (Figure 6); the average annual runoff from 1994 to 2009 was 2, 107.48 mm, which was 1, 034.38 mm more than in the period of 1952 to 1994. The air temperature at Mt. Gongga area also increased significantly after 1994 (0.017 ℃/a), whereas the precipitation and potential evaporation decreased by 55 mm and 103 mm, respectively, between 1999 and 2004. However, the runoff in the corresponding period increased by 1, 073.1 mm, suggesting that the precipitation reduction did not lead to a runoff reduction. Past research showed that the runoff has changed considerably in the Mt. Gongga area since 1994, about two to eight years later than the abrupt climate change in 1986/1987 (Li et al., 2010). Before 1996, the runoff peak values in the Hailuo Creek Basin occurred in August, the precipitation peak values occurred in June, and runoff in September was higher than that in April, although precipitation rates in September and April were basically similar. The Hailuo Creek Basin is a small and rocky mountainous region, so the hysteresis between runoff with precipitation and temperature is caused by melted water because, in contrast to precipitation, ice and snow can supply runoff after a long ablation period. After 1996, the hysteresis between runoff with precipitation and temperature was not marked. Our results indicated that as the climate warmed, glacier melting rates accelerated and ablation area expanded, which resulted in meltwater increases. Meltwater is one of the principal runoff suppliers in the Hailuo Creek Basin, so it can be concluded that the variation of meltwater has a great influence on the seasonal variation of runoff. A monsoon glacier is weakened by the seasonal temperature changes, which leads to the remarkable reaction of glacier melting and temperature decreases. Weak temperature changes result in non-linear melting increases, which accelerate the loss of glaciers and cause changes in the characteristics of water circulation.4 Discussion 4.1 The meteorological factors influencing runoff
Runoff is influenced by various meteorological factors which affect each other; thus, the complex effect mechanism of meteorological factors on runoff has been a difficult aspect in hydrology research (Bergström et al., 2001; Menzel and Bürger, 2002; Pilling and Jones, 2002; Arnell et al., 2003; Christensen et al., 2004; Steele-Dunne et al., 2008; Dush et al., 2009; Zhu et al., 2012). In order to explore the reasons for the change of runoff in the Hailuo Creek Basin, three primary meteorological factors from the six meteorological stations were chosen to study the correlation between the runoff and meteorological factors (Table 3). As shown in Table 3, the correlations among the temperature, precipitation, potential evaporation, and runoff were positive in spring, summer, and autumn, while the correlations among potential evapotranspiration, temperature, and runoff were negative in winter. The controlling factors influencing runoff were different in different seasons. Runoff was greatly influenced by temperature in spring, summer, and autumn, while in winter the major influencing factor was the potential evapotranspiration. As for the whole year, temperature, precipitation, and potential evaporation were important influencing factors.
|Runoff||Precipitation||Mean temperature||Potential evaporation|
|Autumn||0.921||0.999 (**)||0.999 (*)|
|Annual||0.891 (**)||0.910 (**)||0.886 (**)|
|** p<0.01; * p<0.05.|
In spring the correlations among precipitation, temperature, and runoff were positive. The increase rates of precipitation and temperature were 0.21 mm/a and 0.005 ℃/a, respectively (Figure 7), which caused the increase of runoff. In summer, the correlations among the precipitation, temperature, and potential evapotranspiration were not significant, failing the p <0.05 significance test. The correlations among the temperature, precipitation, potential evapotranspiration, and runoff were positive. In the past 50 years, the precipitation and temperature in the Hengduan Mountains increased slightly (Figure 7). Thus, it seems that the increase of runoff was also due to the increase of temperature and decrease of potential evapotranspiration.
In autumn the correlations among precipitation, temperature (p <0.01) , and potential evapotranspiration (p <0.05) and runoff were positive. The increase rates of precipitation, temperature, and potential evapotranspiration were 0.2394 mm/a, 0.1530 ℃/a, and 0.6970 mm/a, respectively (Figure 7). We can see that the precipitation and temperature increases were evident in autumn, which would result in the increase of glacier meltwater. In winter, the correlations among the temperature, potential evapotranspiration, and runoff were negative, while the correlation between precipitation and runoff was positive. The change rates of precipitation, temperature, and potential evapotranspiration were 0.0132 mm/a, 0.0119 ℃/a, and −0.1370 mm/a, respectively (Figure 7). In the past 50 years, precipitation, temperature, and potential evapotranspiration showed slightly increasing trends in autumn and winter (Figure 7). The influence of potential evapotranspiration in winter was more obvious than in other seasons, so the major reason for the increase of runoff in winter lies in the rising precipitation and the constant supply of groundwater. Overall, temperature and precipitation generally showed an increasing tendency in the past 50 years, while potential evapotranspiration showed a decreasing tendency. The major factor that influenced runoff was temperature, while the increase of precipitation and decr ease of potential evapotranspiration did not obviously influence runoff.4.2 The relationship between glacier change and runoff change 4.2.1 Glacier change
Loss of glacier mass has characterized Hailuo Creek glaciers during the past 79 years (Heim, 1936; Zhang and Su, 2002; He et al., 2003a, b; He et al., 2010) (Table 4 and Figure 8). The annual retreat rates of Hailuo Creek Glacier and Hailuo Creek No. 1 Glacier were 31.1 m and 21.6 m, respectively, during 1930-1966. The glaciers were stable in 1966-1983. Hailuo Creek No. 1 Glacier had little change and its average annual retreat rate was 11.8 m/a during 1984-2009 (although its annual retreat rate was 27.5 m/a in 1981-1990) . In contrast, the accumulated retreat of Hailuo Creek Glacier has been remarkable since the mid-1980s, being 20.5 m/a in 1983-2006. The glacier elevations have also changed obviously. The rate of rise of the altitude in Hailuo Creek Glacier was 2.5 m/a in 1936-2006, while the rate of Hailuo Creek Glacier No. 1 was 10.8 m/a in 1981-2006. Our results showed that there was a significant positive correlation between glacier change and runoff change in the Hailuo Creek Basin (r = 0.864, p <0.05) . Since the mid-1980s, glacier retreat had led to obvious increases in runoff, and the runoff has increased sharply since 1994.
|Glacier||Period||Advance/Retreat (m)||Altitude of front (m)||Mean annual change (m)||Source|
|Hailuo Creek Glacier||1930.12-1966.12||−1, 150||2, 900 (1966)||−31.1||(Heim, 1936; Zhang and Su, 2002)|
|1966.12-1983.01||−200||2, 920 (1982)||−11.8||(Zhang and Su, 2002; He et al., 2003a, b)|
|1983.01-1989.12||−147.8||2, 940 (1989)||−21.1||(He et al., 2003a, b)|
|1990.04-2004.03||−274||2, 980 (1994)||−18.2||(He et al., 2003a, b)|
|2004.03-2009.06||−55||3, 005 (2009)||−11.0||Local studies|
|Hailuo Creek Glacier No. 1||1930.12-1966.12||−800||−21.6||(Zhang and Su, 2002)|
|1966.12-1981.01||Stationary||Stationary||(Zhang and Su, 2002)|
|1981.01-1990.12||−250 to −300||3, 600 (1994)||−25 to −30||(Zhang and Su, 2002)|
|1991.01-2009.06||Rapid retreat||3, 880 (2006)||Rapid retreat||Local studies|
It was difficult to quantitatively evaluate the relationship between glacier mass and climate change in the Hailuo Creek Basin because of the lack of long-term climate data in the glacier areas. Aizen et al. (1994) considered that the glacier mean mass balance was −138 mm/a based on the meteorological data of the Kangding meteorological station in 1952-1990 (All mass balances are quoted as mm of water equivalent, that is, in the same units as runoff). Xie et al. (1995, 2001)considered that the glacier mean mass balance was −240 mm/a based on the climate data of the Luding meteorological station in 1960-1993. Li et al (2010) considered that the glacier water-mass balance was −240.6 mm/a according to the water balance method in 1959-2004.
Our analysis indicated that the runoff in Hailuo Creek Basin has seen a remarkable increasing trend over the last 58 years, especially since 1994. The annual mean runoff in Hailuo Creek Basin increased by 185.37 mm/a during 1952-1993, and it increased by 1, 961.37 mm/a in 1994-2009. Meltwater accounted for 51.1% of the total runoff. The glacier melt runoff increased by 94.72 mm/a in 1952-1993, which accompanied a loss in glacier mass balance, the annual mean value of which was −94.72 mm/a. The glacier melt runoff increased to 1, 002.26 mm/a in 1994-2009, which indicated that there was a severe loss in mass balance (−1, 002.26 mm/a)in 1994-2009. The general decrease of precipitation was offset by the decrease of potential evapotranspiration. With temperature increasing, the elevation of the ELA has risen and the ablation area has expanded. Contrastive studies found that the mass balance of the No. 1 Glacier in the Urumqi River Headwaters (43°05'N, 86°48'E)was −2, 359.8 mm/a in 1960-2006 (Tianshan Glaciological Station, 2007), that of the Xiao Dongkemadi Glacier Basin (33°04'N, 92°04'E)was −1, 584 mm/a in 2008-2012 (Gao et al., 2011), that of the Qiyi Glacier (39°30'N, 97°30'E)was −856.2 mm/a in 2012 (Gao et al., 2011), that of the Zhadang Glacier Basin (30°29'N, 90°39'E)was −1, 547 mm/a in 2005-2006 (Gao et al., 2011), and that of Hailuo Creek Basin (101°23'E, 29°50'N)was −19, 012.41 mm/a, indicating that low-latitude monsoon glaciers are more sensitive to climate warming.4.2.2 Changes in glacier melt water runoff and its impact on river runoff
The annual mean glacier melt value in the Hailuo Creek Basin in 1952 was 1, 457.75 mm, which was lower than the value of 2, 053.68 mm in 1973 and significantly lower than the 4, 138.17 mm in 2009; the temperature rise has resulted in a clear increase of glacial runoff in the Hailuo Creek Basin. The glacier meltwater share was increased along with the melting of the glacier. In a hydrograph study of separation of runoff, Cao (1995) concluded that glacier runoff accounted for 55.9% of the Hailuo Creek stream flow in 1990, 51.5% in 1994 (Li et al., 2004), and 70.4% in 2006 (Yin et al., 2008). The proportion of glacier runoff in the Hailuo Creek stream flow is increasing as glaciers shrink. Li and Su (1996) considered that groundwater accounted for 15%-20% of the Hailuo Creek stream flow, and it was obvious that the most of the increase of river flow derived from glacier retreat. Glacier runoff was stable during 1952 to 1994, and then displayed a marked increasing trend in 1994-2000; it was generally stable in 2000-2009.5 Conclusions
We studied the Hailuo Creek Basin in southwest China to investigate the impact of climate change on its hydrology. Meteorological, glacier change, and runoff data were collected from field experiments, and the parameters for the HBV Light model were derived from observation data and automatic calibration algorithms. To assess the influence of glacier change on stream flow, the characteristics and trends of mass balance and glacier change were also analyzed. The analysis indicated that the runoff in the Hailuo Creek Basin has increased markedly over the last 58 years, and the glacier mass balance displayed a marked decreasing trend during the period of 1930-2009, especially after 1991.The annual mean glacier melt value in the Hailuo Creek Basin in 1952 was 1, 457.75 mm, which was lower than the 2, 053.68 mm in 1973 and the 4, 138.17 mm in 2009. The temperature rise has resulted in a clear increase of glacial runoff in the Hailuo Creek Basin. The annual mean glacier melt in Hailuo Creek Basin during 2000-2009 was 2, 186.6 mm, which was higher than the annual mean proportion of 49.8% during 1961-2009. The annual mean of shared glacier melt water in the Hailuo Creek Basin was 51.1% in 1994 and 72.4% in 2009. Most of the increase of flow stream derived from glacier shrinkage.
The glacier mass balance, ELA, glacier runoff, and regional patterns of glacier retreat were used to assess the performance of the HBV model, and our modeling results in summer, autumn, and annually were acceptable (Efficiency coefficient >0.6) . However, some daily results in rainy seasons and modeling results in dry seasons have serious errors. There were uncertainties and errors in the modeling results due to the following factors. (1) The error and uncertainties of forced data in the model were due to the remote image resolution and many other factors, such as judgment errors of glacial area and the boundary, catchment basin boundaries, elevation, position, and slope. The spatial variability of precipitation is high in glacierized regions of high mountainous catchments, and in this study good spatial interpolation of precipitation data was difficult to obtain; the precipitation gradient may not reflect the real spatial distribution of precipitation. (2) The model parameters of field moisture capacity, soil moisture content, and the degree-day factors did not consider temporal and spatial variations. (3) The HBV model does not consider glacier changes during past decades, and all of our glacier data came from the Chinese Glacier Inventory, which may have led to overestimation of glacier runoff in later stages. The HBV model can well simulate the inter-annual runoff change, so we could therefore analyze its effects on water resources in Hailuo Creek.Acknowledgments:
This research was funded by the Chinese Postdoctoral Science Foundation (No. 2015M570864) , the Open-Ended Fund of the State Key Laboratory of Cryospheric Sciences, Chinese Academy of Sciences (No. SKLCS-OP-2014-11) , a project of the Northwest Normal University Young Teachers Scientific Research Ability Promotion Plan (No. NWNU-LKQN-13-10) , a project of the National Natural Science Foundation of China (Nos. 41273010, 41271133) , and a project of the Major National Research Projects of China (No. 2013CBA01808) .
|Aizen VB, Nikitin SA, Song GP, 1994. Model of the dynamics of the Hailuogou Glacier A. In: Xie Z, Kotlyakov VM (eds.). Glaciers and Environment in the Qinghai-Xizang (Tibet) Plateau (Ⅰ)—The Gongga Mountain. Beijing: Science Press121-132.|
|Arnell NW, Hudson DA, Jones RG, 2003. Climate change scenarios from a regional climate model: Estimating change in runoff in southern Africa. Journal of Geophysical Research, 108(D16): 4519. doi: 10.1029/2002JD002782|
|Bergström S.1992. The HBV model-its structure and applications. SMHI Hydrology, RH No. 4, Norrkoping, pp. 35. Bergström S.1992. The HBV model-its structure and applications. SMHI Hydrology, RH No. 4, Norrkoping, pp. 35.|
|Bergström SB, Carlsson M, Gardelin G, et al, 2001. Climate change impacts on runoff in Sweden-assessments by global climate models, dynamical downscaling and hydrological modeling. Climate Research, 16: 101–112. doi: 10.3354/cr016101|
|Bergström S, Forsman A, 1973. Development of a conceptual deterministic rainfall-runoff model. Nordic Hydrology, 4(3): 147–170.|
|Beven K, Binley A, 1992. The future of distributed models: Model calibration and uncertainty prediction. Hydrological Processes, 6(3): 279–298. doi: 10.1002/hyp.3360060305|
|Beven K, 1993. Prophecy, reality and uncertainty in distributed hydrological modelling. Advances in Water Resources, 16(1): 41–51. doi: 10.1016/0309-1708(93)90028-E|
|Braun LN, Aellen M, 1990. Modelling discharge of glacierized basins assisted by direct measurements of glacier mass balance. In: Hydrology in Mountainous Regions. Wallingford, UK: IAHS Publication99-106.|
|Braithwaite RJ, Olesen OB, 1984. Ice ablation in West Greenland in relation to air temperature and global radiation. Zeitsehrift fuer Gletscherkunde und Glazialgeologie, 20: 155–168.|
|Booij MJ, 2005. Impact of climate change on river flooding assessed with different spatial model resolutions. Journal of Hydrology, 303(1-4): 176–198. doi: 10.1016/j.jhydrol.2004.07.013|
|Cao Z, 1995. The charaeteristies of glaeier hydrology in the area of the Gongga Mountains. Journal of Glaciology and Geocryology, 17(1): 73–83.|
|Cao ZT, Cheng GW, 1994. Prelimary analyses of hydrological characteristics of Hailuogou Glacier on the eastern slope of the Gongga Mountain. In: Xie Z, Kotlyakov VM (eds.). Glaciers and Environment in the Qinghai-Xizang (Tibet) Plateau (I)—The Gongga Mountain. Beijing: Science Press143-156.|
|China Meteorological Administration, 2006. Climate and Environment in China. Beijing: Meteorology Press.|
|Christensen NS, Wood AW, Voisin N, et al, 2004. The effects of climate change on the hydrology and water resources of the Colorado River Basin. Climatic Change, 62(1): 337–363. doi: 10.1023/B:CLIM.scar-8-3-22713684.13621.1f|
|Coudrain A, Francou B, Kundzewicz ZW, 2005. Glacier shrinkage in the Andes and consequences for water resources. Hydrological Sciences-Journal-des Sciences Hydrologiques, 50(6): 925–932.|
|Dush SK, Kulkarni MA, Mohanty UC, et al, 2009. Changes in the characteristics of rain events in India. Journal of Geophysical Research, 114: D10109. doi: 10.1029/2008JD010572|
|Dyurgerov MD, Meier MF, 2005. Glaciers and the Changing Earth System: A 2004 Snapshot. Institute of Arctic and Alpine Research. University of Colorado. Boulder, Colorado, 117.|
|Fountain A, Tangborn W, 1985. The effect of glaciers on streamflow variations. Water Resources Research, 21(4): 579–586. doi: 10.1029/WR021i004p00579|
|Freer J, Beven K, Ambroise B, 1996. Bayesian estimation of uncertainty in runoff prediction and the value of data: an application of the GLUE approach. Water Resources Research, 32(7): 2161–2173. doi: 10.1029/95WR03723|
|Fujita K, Seko K, Ageta Y, et al, 1996. Superimposed ice in glacier mass balance on the Tibetan Plateau. Journal of Glaciology, 42(154): 454–460.|
|Fujita K, Ohta T, Ageta Y, 2007. Characteristics and climatic sensitivities of runoff from a cold-type glacier on the Tibetan Plateau. Hydrological Processes, 21: 2882–2891. doi: 10.1002/hyp.6505|
|Gaume E, Villeneuve JP, Desbordes M, 1998. Uncertainty assessment and analysis of the calibrated parameter values of an urban storm water quality model. Journal of Hydrology, 210(1-4): 38–50. doi: 10.1016/S0022-1694(98)00171-1|
|Gao C, Zhai JQ, Tao H, et al, 2009. Hydrological response to land use/landcover change in Chaohu Basin. Journal of Natural Resources, 24: 1794–1802. doi: 10.11849/zrzyxb.2009.10.013|
|Gao HK, He XB, Ye BS, et al, 2012. The simulation of HBV Hydrology Model in the Dongkemadi River Basin, headwater of the Yangtze River. Hydrological Process, 26: 1593–1603.|
|Gao X, Ye BS, Zhang SQ, et al, 2011. Glacier runoff variation and its influence on river runoff during 1961-2006 in the Tarim River Basin. Science China (Series D: Earth Sciences), 53(6): 880–891.|
|He YQ, Pu T, Li ZX, et al, 2010. Climate change and its effect on annual runoff in Lijiang Basin-Mt. Yulong Region, China. Journal of Earth Science, 21(2): 137–147. doi: 10.1007/s12583-010-0012-5|
|He YQ, Zhang ZL, Theakstone WH, 2003a. Changing features of the climate and glaciers in China's monsoonal temperate glacier region. Journal of Geophysical Research, 108(D17): 4530–4536. doi: 10.1029/2002JD003365|
|He YQ, Gu J, Zhang D, 2003b. What is the major reason for glacier retreat on Yulong Mountain, China?. Journal of Glaciology, 49(165): 325–326.|
|He XB, Ye BS, Ding YJ, 2009. Bias correction for precipitation measurement in Tanggula Mountain Tibetan Plateau. Advances in Water Science, 20(3): 403–408.|
|Heim A, 1936. The glaciation and solifluction of Minya Gongka. Geographical Journal, 87(5): 444–454.|
|Hock R, 2003. Temperature index melt modelling in mountain areas. Journal of Hydrology, 282(1-4): 104–115. doi: 10.1016/S0022-1694(03)00257-9|
|Hottelet C, Braun LN, Leibundgut C, et al, 1993. Simulation of snowpack and discharge in an Alpine Karst Basin. In: Young GJ (ed.). Snow and Glacier Hydrology, Proceedings of the Kathmandu Symposium. Wallingford, UK: IAHS Publication249-260.|
|Hornberger GM, Cosby BJ, Galloway JN, 1986. Modelling the effect of acid deposition: Uncertainty and spatial variability in estimation of long-term sulphate dynamics in a region. Water Resources Research, 22(8): 1293–1302. doi: 10.1029/WR022i008p01293|
|IPCC (Intergovernmental Panel on Climate Change).2007. In: Pachauri RK, Reisinger A (eds.). Climate Change 2007: Synthesis Report. Contribution of Working Groups Ⅰ, Ⅱ, and Ⅲ to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Core Writing Team. IPCC, Geneva, 102. IPCC (Intergovernmental Panel on Climate Change).2007. In: Pachauri RK, Reisinger A (eds.). Climate Change 2007: Synthesis Report. Contribution of Working Groups Ⅰ, Ⅱ, and Ⅲ to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Core Writing Team. IPCC, Geneva, 102.|
|Jakeman AJ, Hornberger GM, 1993. How much complexity is warranted in a rainfall-runoff model?. Water Resources Research, 29(8): 2637–2649.|
|Jensen ME, Burman RD, Allen RG.1990. Evaporation and Irrigation Water Requirements. In: American Society of Civil Engineering Manual No. 70. Jensen ME, Burman RD, Allen RG.1990. Evaporation and Irrigation Water Requirements. In: American Society of Civil Engineering Manual No. 70.|
|Jia WX, He YQ, Wang XF, et al, 2009. Temporal and spatial change of the potential evaporation over Qilian Mountains and Hexi Corridor from 1960 to 2006. Advance in Water Science, 20(2): 159–167. doi: 10.3321/j.issn:1001-6791.2009.02.002|
|Jin XL, Zhang Q, Xu CY, 2008. Regionalization study of a conceptual hydrological model in the Dongjiang Basin. Journal of Lake Sciences, 20(6): 723–732.|
|Kavetski D, Kuczera G, Franks SW, 2006. Calibration of conceptual hydrological models revisited: 1. Overcoming numerical artifacts. Journal of Hydrology, 320(1-2): 173–186. doi: 10.1016/j.jhydrol.2005.07.012|
|Kendall MG, Gibbons JD.1990. Rank Correlation Methods. 5th Ed.. Griffin, London. Kendall MG, Gibbons JD.1990. Rank Correlation Methods. 5th Ed.. Griffin, London.|
|Kite GW, Kouwen N, 1992. Watershed modelling using land classifications. Water Resources Research, 28(12): 3193–3200. doi: 10.1029/92WR01819|
|Li W, Cheng GW, Luo J, et al, 2004. Features of the Natural Runoff of Hailuo Ravine in Mt. Gongga. Journal of Mountain Science, 22(6): 698–701. doi: 10.3969/j.issn.1008-2786.2004.06.010|
|Li J, Su Z, 1996. Glaciers of Hengduan Mountains. Beijing: Science Press.|
|Li Z, He Y, Yang X, et al, 2010. Changes of the Hailuo Creek glacier, Mt. Gongga, China, against the background of climate changes during the Holocene. Quaternary International, 218(1): 166–175.|
|Li J, Feng Z, 1984. The Quaternary glaciation in Hengduan Mountains. Journal of Lanzhou University,: 61–73.|
|Liu SQ, Graham WD, Jacobs JM, 2005. Daily potential evapotranspiration and diurnal climate forcings: influence on the numerical modelling of soil water dynamics and evapotranspiration. Journal of Hydrology, 309(1-4): 39–52. doi: 10.1016/j.jhydrol.2004.11.009|
|Liu Q, Liu SY, Zhang Y, et al, 2010. Recent shrinkage and hydrological response of Hailuogou glacier, a monsoon temperate glacier on the east slope of Mount Gongga, China. Journal of Glaciology, 56(196): 215–224.|
|Lu RR, Zhong XH, 1996. Block and burst of the water channels inside Hailuogou Glacier. Journal of Glaciology and Geocryology, 18(Suppl.): 257–263.|
|Mann HB, 1945. Nonparametric tests against trend. Econometrica, 13(3): 245–259. doi: 10.2307/1907187|
|Menzel L, Bürger G, 2002. Climate change scenarios and runoff response in the Mulde catchment (Southern Elbe, Germany). Journal of Hydrology, 267(1-2): 53–64. doi: 10.1016/S0022-1694(02)00139-7|
|Miles BWJ, Stokes CR, Vieli A, et al, 2013. Rapid, climate-driven changes in outlet glaciers on the Pacific coast of East Antarctica. Nature, 500: 563–566. doi: 10.1038/nature12382|
|Oerlemans J, 2005. Extracting a climate signals from 169 glacier records. Science, 308(5722): 675–677. doi: 10.1126/science.1107046|
|Pan BT, Zhang GL, Wang J, et al, 2012. Glacier changes from 1966-2009 in the Gongga Mountains, on the south-eastern margin of the Qinghai-Tibetan Plateau and their climatic forcing. The Cryosphere, 6(5): 1087–1101. doi: 10.5194/tc-6-1087-2012|
|Pang H, He Y, Theakstone WH, et al, 2007a. Soluble ionic and oxygen isotopic composition of a shallow firn profile, Baishui glacier No. 1, southeastern Tibetan Plateau. Annals of Glaciology, 46: 325–330. doi: 10.3189/172756407782871648|
|Pang HX, He YQ, Zhang NN, 2007b. Accelerating glacier retreat on Mountain Yulong, Tibetan Plateau, since the late 1990s. Journal of Glaciology, 53(181): 317–319.|
|Pilling CG, Jones JAA, 2002. The impact of future climate change on seasonal discharge, hydrological processes and extreme flows in the Upper Wye experimental catchment, Mid-Wales. Hydrological Processes, 16(6): 1201–1213. doi: 10.1002/hyp.1057|
|Pachauri RK, Allen MR, Barros VR, et al..2014. Climate Change 2014: Synthesis Report. Contribution of Working Groups Ⅰ, Ⅱ and Ⅲ to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Pachauri RK, Allen MR, Barros VR, et al..2014. Climate Change 2014: Synthesis Report. Contribution of Working Groups Ⅰ, Ⅱ and Ⅲ to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change.|
|Prasad AK, Yang KHS, El-Askary HM, et al, 2009. Melting of major glaciers in the western Himalayas: evidence of climatic changes from long term MSU derived tropospheric temperature trend (1979-2008) . Annales Geophysicae, 27(12): 4505–4519. doi: 10.5194/angeo-27-4505-2009|
|Piao S, Ciais P, Huang Y, et al, 2010. The impacts of climate change on water resources and agriculture in China. Nature, 467: 43–51. doi: 10.1038/nature09364|
|Pu J, 1994. Glacier Inventory of China (the Changjiang River Drainage Basin). Lanzhou: Gansu Culture Press117-129.|
|Seibert J, 1999. Regionalisation of parameters for a conceptual rainfall-runoff model. Agricultural and Forest Meteorology, 98: 279–293. doi: 10.1016/S0168-1923(99)00105-7|
|Seibert J.2005. HBV light version 2, User´s manual. Stockholm University. http://people.su.se/_jseib/HBV/HBV_manual_2005.pdf. Seibert J.2005. HBV light version 2, User´s manual. Stockholm University. http://people.su.se/_jseib/HBV/HBV_manual_2005.pdf.|
|Seibert J, 1997. Estimation of parameter uncertainty in the HBV Model. Nordic Hydrology, 28(4-5): 247–262.|
|Shi YF, Liu SY, 2000. Estimation of the response of glaciers in China to the global warming in the 21st century. Chinese Science Bulletin, 45(7): 668–672. doi: 10.1007/BF02886048|
|Shi YF, 2001. Estimation of the water resources affected by climatic warming and glacier shrinkage before 2050 in west China. Journal of Glaciology and Geocryology, 23(4): 333–341.|
|Sorg AF, Bolch T, Stoffel M, et al, 2012. Climate change impacts on glaciers and runoff in Tien Shan (Central Asia). Nature Climate Change, 2(10): 725–731. doi: 10.1038/nclimate1592|
|Su Z, Liu S, 2002. Glacier changes on the Hengduan Mountain and their response to climate. In: The Environment and Ecosystem in the Eastern Edge of the Qinghai-Xizang Plateau. Chengdu: Sichuan University Press102-109.|
|Su Z, Shi YF, Zheng BX, 2002. Quaternary glacial remains on the Gongga Mountain and the division of glaciation period. Advances in Earth Science, 17(5): 639–647.|
|Steele-Dunne S, Lynch P, McGrath R, et al, 2008. The impacts of climate change on hydrology in Ireland. Journal of Hydrology, 356(1-2): 28–45. doi: 10.1016/j.jhydrol.2008.03.025|
|Tianshan Glaciological Station.2007. Annals of Tianshan Glaciological Station in 2006. Lanzhou, Cold and Arid Regions Environmental and Engineering Research Institute. Tianshan Glaciological Station.2007. Annals of Tianshan Glaciological Station in 2006. Lanzhou, Cold and Arid Regions Environmental and Engineering Research Institute.|
|Thornthwaite CW, 1951. The water balance in tropical climates. Bulletin of the American Meteorological Society, 32: 166–173.|
|Uhlenbrook S, Seibert J, Leibundgut C, et al, 1999. Prediction uncertainty of conceptual rainfall-runoff models caused by problems in identifying model parameters and structure. Hydrological Sciences Journal - Journal Des Sciences Hydrologiques, 44(5): 779–797. doi: 10.1080/02626669909492273|
|Van der Perk, Bierkens MFP, 1997. The identifiability of parameters in a water quality model of the Biebrza River, Poland. Journal of Hydrology, 200(1-4): 307–322. doi: 10.1016/S0022-1694(97)00021-8|
|Wang S, Pu JC, Wang NL, 2011. Study of mass balance and sensibility to climate change of Qiyi Glacier in Qilian Mountains. Journal of Glaciology and Geocryology, 33(6): 1214–1221.|
|Wu X, Lin Z, 1983. Climatic change indicated by tree rings in Xiaozhongdian, Yunnan Province. In: The Exploring Collection in Hengduan Mountain. Kunming: Yunlan People Press.|
|Xie ZC, Su Z, Shen YP, et al, 2001. Mass balance and water exchange of Hailuogou Glacier in Mount Gongga and their influence on glacial melt runoff. Journal of Glaciology and Geocryology, 23(1): 7–15.|
|Xie Z, Su Z, Cao Z, 1995. Water and mass balance in the Gongga Mountain. In: Researches on Evolution, Environment Change and Ecosystems of Tibet Plateau. Beijing: Science Press340-346.|
|Xiang L, Liu XF, Hao LS, et al, 2011. Analysis of runoff characteristics of Luanhe River under various emission scenarios in the next century. Progress in Geography, 30(7): 861–867.|
|Xu Y, Chen YX, Wang JS, 2006. Using indicator Kriging to analyze and evaluate spatial distributions of soil water and salt in field. Advances in Water Science, 17(6): 477–482.|
|Yamamoto JK, 2005. Correcting the smooth effect of ordinary kriging estimates. Mathematical Geology, 37(1): 69–94. doi: 10.1007/s11004-005-8748-7|
|Yao TD, Li ZG, Yang W, et al, 2010. Glacial distribution and mass balance in the Yarlung Zangbo River and its influence on lakes. Chinese Science Bulletin, 55(20): 2072–2078. doi: 10.1007/s11434-010-3213-5|
|Yin G, Wang XD, Gao ZY, et al, 2008. Study of the hydrology of glacial runoff in Hailuogou Valley, Gongga Mountain by means of isotopic tracing. Journal of Glaciology and Geocryology, 30(3): 365–372.|
|Zhang J, He XB, Ye BS, et al, 2013. Recent variation of mass balance of the Xiao Dongkemadi Glacier in the Tanggula range and its influencing factors. Journal of Glaciology and Geocryology, 35(2): 263–271. doi: 10.7522/j.issn.1000-0240.2013.0032|
|Zao YJ, Zhang JX, Zhang SA, 2007. Application research of HBV model in Guanzhai Basin of Huai River. Journal of China Hydrology, 27(2): 57–59.|
|Zhang Y, Fujita K, Liu SY, et al, 2010. Multi-decadal ice-velocity and elevation changes of a monsoonal maritime glacier: Hailuogou glacier, China. Journal of Glaciology, 56(195): 65–74. doi: 10.3189/002214310791190884|
|Zheng BX, 1990. The glacier, environment and its changes since the last glaciation in West China. Quaternary Sciences, 2: 101–110.|
|Zheng BX, Ma Q, 1994. The glaciers variation, climate change and the river valley development in the Holocene on the Gongga Mountain. Acta Geographica Sinica, 49(6): 500–508.|
|Zheng BX, 2001. Study on the Quaternary glaciation and the formation of Moxi platform in the east slopes of Mount Gongga. Journal of Glaciology and Geocryology, 23(3): 287–291.|
|Zhang W, Su Z, 2002. Changes features of Hailuo Creek glacier. In: Environment and Ecosystem in the Eastern Edge of Qinghai-Xizang Plateau. Chengdu: Sichuan University Press113-120.|
|Zhou GP, Yao TD, Kang SC, et al, 2007. Mass balance of Zhadang Glacier in the central Tibetan Plateau. Journal of Glaciology and Geocryology, 29(3): 360–365.|
|Zhu GF, Shi PJ, Pu T, et al, 2013. Changes of surface soil relative moisture content in Hengduan Mountains, China, during 1992-2010. Quaternary International, 298(7): 161–170. doi: 10.1016/j.quaint.2013.01.012|
|Zhu GF, He YQ, Pu T, et al, 2012. Spatial distribution and temporal trends in potential evapotranspiration over Hengduan Mountains region from 1960 to 2009. Journal of Geographical Sciences, 22(1): 71–85. doi: 10.1007/s11442-012-0912-7|