Modelling glacier variation and its impact on water resource in the Urumqi Glacier No. 1 in Central Asia
Graphical abstract
Introduction
Meltwater from glaciers on high mountains is the lifeline for local and downstream residents, economy and ecosystems in the arid Central Asia (Ding et al., 2006; Immerzeel et al., 2010; Ren et al., 2017). Glaciers do not only provide valuable water, but they also act as important buffer against drought in dry seasons and years (Pritchard, 2017). Since glacier- and snowmelt are very sensitive to global warming (Huss et al., 2014), glaciers in most regions are experiencing accelerated melting, thinning, and retreating (Bhutiyani et al., 2008; Thayyen and Gergan, 2010; Immerzeel et al., 2012; Duethmann et al., 2014). This rises up huge risks to regional water resources, which are important to sustainable development of economy, society and environment (Immerzeel et al., 2010; Kraaijenbrink et al., 2017). Predicting glacier evolution and its impacts on hydrology is an urgent and critical question in practice as well as for scientific research.
Energy balance model is a physically based approach to calculate snow and ice melt by using energy fluxes to and from the snow/glacier surface (Wang et al., 2017). However, this approach requires many types of measurements, which in most cases is hardly available. Although various approaches has been proposed to derive energy components from conventional meteorological observation (Yang and Koike, 2005), the deriving processes would inevitably bring large uncertainty. Moreover, heterogeneous surface, e.g. debris cover, also substantially modify albedo, glacier melting and movement. Such high complexity hinders implementation of the energy balance approach into practice (Fujita and Sakai, 2014).
Temperature-index model is a relative simple approach by only using air temperature as a lumped index representing the energy budget (Zhang et al., 2007). Moreover, air temperature is one of the most conventional meteorological measurement and it has relatively constant relationship with elevation (lapse rate) allowing us to interpolate from in situ observation to spatial distribution (Wang et al., 2016). Therefore, the temperature-index model has been widely used in snow and glacier melting simulation in many regions around the globe (Matthews et al., 2015; Tarasova et al., 2016). A parameter, called a degree-day factor, which empirically links air temperature with snow and glacier melting, needs to be calibrated by observed data. Additionally, it is easy to extend the temperature-index model with other variables, e.g. albedo, shortwave radiation and topography (Gao et al., 2017a, Gao et al., 2017b).
Location of Glacier terminus and area are changing with climate. Therefore, coupling glacier dynamic into glacier hydrological model is necessary to assess glacier impacts of climate change on water resources. However, simulating glacier responses is complicated, due to complex characteristics of ice, subglacial topography and roughness, and lack of data in glacier mountainous regions (Zhang et al., 2015).
Generally, there are two type approaches to simulate glacier dynamics, i.e. physically based ice flow models and empirical models. Physically based ice flow models considers the movement of glacier as viscous flow (Li et al., 2012; Zhang et al., 2015), described by Stokes equation, where ice movement is controlled by glacier geometry (e.g. length, slope, width and bed undulation) and ice characteristics (ice temperature and debris). This type of models needs tremendous amount of data, which limit its wide implementation, especially in ungauged basins. Moreover, the models based on the Stokes equation are expensive in terms of time and computing resources, which are not realistic in most cases. However, empirical models require less data and computing resources and they work well. For example, the area-volume model (Zhang et al., 2012) estimates the change of area by the change of ice volume, which approximately equals to the calculated glacier mass balance (GMB) assuming temporally stationary glacier area. This model can only estimate the change of glacier area in a lumped way, without spatial distribution information.
The ∆h-parameterization approach (Huss et al., 2010; Li et al., 2015; Seibert et al., 2017) is another empirical approach to update the glacier surface elevation and area based on the simulated GMB from glacier hydrological model. The change of ice thickness in different elevations (∆h) is estimated based on its empirical parameterization with the normalized elevation range. The ∆h-parameterization approach is developed from observed data in 34 glaciers in Switzerland, and used in many other glaciers in the world (Etter et al., 2017), but more vigorous test is still needed in Central Asia with long-term and in-situ multi-disciplinary measurements.
Regarding the water resources from glacier, with temperature increasing, ice melting is accelerated resulting in the loss of glacier mass balance and the increase of melting water in the early stage. However, with the retreat of glacier, less glacier area contributes to melting water, leading to a decrease of water resources. Therefore, a “tipping point” exists which represents the peak of glacier melting water and the turning point from the increase to decrease of melting water. Accurately predicting the tipping point is essential to make decisions on water resources management in the regions highly depending on glacier melt, to take measures to adapt to this change and assure long-term water security. This paper integrated a spatially distributed hydrological model (FLEXG) and a glacier retreat model (∆h-parameterization), and tested the new model in the Urumqi Glacier No. 1 catchment.
Due to the harsh environment of the glacial and mountainous region in Central Asia, e.g. high altitude, lack of oxygen and ineffective transportation, continuous and long-term field measurement is extremely difficult. Lack of high quality in situ observation has been a bottleneck for glaciological and hydrological studies in this region. The Urumqi Glacier No. 1 has the longest glaciological observation in China, from 1959 to present (Ye et al., 2005), and with most comprehensive measurements including glaciology, meteorology, hydrology, topography, ecology and pedeology and geology variables and parameters. The Urumqi Glacier No. 1 provides us with a unique opportunity to understand the impact of climate change on glacier dynamic, to develop and validate glaciological and hydrological models, and to assess climate change impacts on water resources.
This paper is organized into five sections. In Section 2, we briefly introduced the study site – Urumqi Glacier No. 1 catchment, and the datasets used in this study including topography, glacier and hydrology data. Section 3 describes the methodology to couple a hydrological model (FLEXG) with a glacier retreat model (∆h-parameterization). In Section 4, we presented the results to reproduce hydrograph, and glacier measurements, i.e. the historical variation of glacier area, GMB and ELA. Eventually we predicted the glacier response to future climate change and assessed its impact on water resources, especially the tipping point of glacier melt water. Finally, the conclusions drawn from this study were summarized in Section 5.
Section snippets
Study site and historical data
The Urumqi Glacier is located in the headwaters of the Urumqi River, in the Xinjiang Uyghur Autonomous Region in northwest China (43o50’N, 86o49’E; Fig. 1). The Urumqi Glacier became separated into two small dependent glaciers in 1994 and they are referred to east and west branch here. In 2002, the two branches covers respectively 1.12 and 0.72 km2 and they lay between 3740 and 4490 m a.s.l. Daily runoff data from 1985 to 2004 was measured at the No. 1 gauge station, which lies 200 m downstream
Methodology
The integrated modelling framework of FLEXG and ∆h is presented as Fig. 2. In this framework, glacier mass balance (GMB) is the bond linking the glacier melting module in FLEXG and the ∆h-parameterization method which redistributed the estimated GMB to different elevations. The calculated GMB by FLEXG is the input for ∆h-parameterization. ∆h-parameterization redistributes the estimated GMB to different elevation bands, based on the empirical curves depending on the size of glaciers (Huss et
Reproducing historical hydrograph and glacier variation
Fig. 4 shows the observed and the modelled glacier area extent in 1994 and 2002. We found that the FLEXG-∆h model has the ability to reproduce glacier area variation. The simulated glacier area change of 1994–1980, 2002–1994, and 2002–1980 were 0.031km2, 0.099 km2 and 0.130 km2, which were comparable with the observed area change, i.e. 0.038km2, 0.113 km2 and 0.151 km2. Interestingly, in 1994 the west and east branches of the Glacier No. 1 were separated, which was well captured by the FLEXG-∆h
Conclusions
In this study, we integrated a spatially distributed hydrological model (FLEXG) and a glacier retreat model (∆h-parameterization) to assess the impact of climate change on glacier thinning, retreat, and its influence on hydrology and water resource. The coupled model (FLEXG-∆h) was tested in the Urumqi Glacier No. 1 catchment with the longest and most comprehensive measurements in China. Besides, the model was validated by hydrological and glaciological data, including daily runoff, annual
Acknowledgement
Firstly, we would like to thank the great pioneers on glaciological studies in the Urumqi Glacier No. 1, who dedicated their lifelong energy to this arduous but glorious career. Without their tough work on the long-term field measurement, this paper can never happen. This study was supported by the National Key R&D Program of China (2017YFE0100700), the Key Program of National Natural Science Foundation of China (No. 41730646), and financially supported by the Key Laboratory for Mountain
References (41)
- et al.
Climate change impacts on future snow, ice and rain runoff in a Swiss mountain catchment using multi-dataset calibration
J. Hydrol.
(2017) - et al.
Decomposition of the mean squared error and NSE performance criteria: implications for improving hydrological modelling
J. Hydrol.
(2009) - et al.
High uncertainty in 21st century runoff projections from glacierized basins
J. Hydrol.
(2014) - et al.
Integrating a glacier retreat model into a hydrological model – case studies of three glacierised catchments in Norway and Himalayan region
J. Hydrol.
(2015) - et al.
An overview of black carbon deposition in high Asia glaciers and its impacts on radiation balance
Adv. Water Resour.
(2013) - et al.
The future of distributed models: model calibration and uncertainty prediction
Hydrol. Process.
(1992) - et al.
Changing streamflow patterns in the rivers of northwestern Himalaya: implications of global warming in the 20th century
Curr. Sci. India.
(2008) - et al.
The retreat of glaciers in response to recent climate warming in western China
Ann. Glaciol.
(2006) - et al.
Variations in the equilibrium line altitude of Urumqi glacier no. 1, Tianshan Mountains, over the past 50 years
Chin. Sci. Bull.
(2012) - et al.
The value of satellite-derived snow cover images for calibrating a hydrological model in snow-dominated catchments in Central Asia
Water Resour. Res.
(2014)
Modelling runoff from a Himalayan debris-covered glacier
Hydrol. Earth Syst. Sci.
Modeling the runoff and glacier mass balance in a small watershed on the Central Tibetan Plateau, China, from 1955 to 2008
Hydrol. Process.
The importance of aspect for modelling the hydrological response in a glacier catchment in Central Asia
Hydrol. Process.
Use of auxiliary data of topography, snow and ice to improve model performance in a glacier-dominated catchment in Central Asia
Hydrol. Res.
The WMO solid precipitation measurement intercomparison
Comparing large-scale hydrological model simulations to observed runoff percentiles in Europe
J. Hydrometeorol.
Future high-mountain hydrology: a new parameterization of glacier retreat
Hydrol. Earth Syst. Sci.
Climate change will affect the asian water towers
Science
Hydrological response to climate change in a glacierized catchment in the Himalayas
Clim. Chang.
Impact of a global temperature rise of 1.5 degrees celsius on Asia's glaciers
Nature
Cited by (51)
Diagnosing the impacts of landscape characteristics on hydrologic signatures in the Krycklan catchment in Sweden using a flexible hydrological model
2024, Physics and Chemistry of the EarthHydrological response to climate change in a glacierized catchment in eastern Tien Shan, Central Asia
2024, Journal of Hydrology: Regional StudiesModeling the impact of climate change on streamflow in glacier/snow-fed northern Tianshan basin
2023, Journal of Hydrology: Regional StudiesThe role of subsurface ice in sustaining bacteria in continental and maritime glaciers
2023, Science of the Total Environment