With an area of 2.5 × 106
and mean elevation over 4000 m above the sea level, the Tibetan Plateau (TP) is the largest and highest plateau in the world [1
]. The TP is also the source region of several important rivers in Asia such as the Yellow, the Yangtze, the Mekong, the Salween, and the Brahmaputra River, and is therefore called the ‘Water tower of Asia’ [2
]. The TP provides fresh water to an area spanning about 5.6 million km2
and over 1.4 billion population [3
], and discharge in this area is important for domestic, agricultural, and industrial use of downstream regions.
Due to the uniquely complex terrain and high elevation, glacier and snow cover are extensively developed on the Tibetan Plateau [4
]. Investigations have shown that there are 36,763 glaciers with a total area of approximately 49,873 km2
on the TP [6
], which is the largest outside of the Polar regions [7
]. Meanwhile, snow cover is widely distributed over the TP. The snow-covered area over the TP is established in autumn and persists to the following spring, and even to summer in the western and southeastern parts at high elevation [8
]. Glacier and snow are indicators of the weather and climate in and around the TP [9
], and very sensitive to global warming. The TP has undergone significant climate warming since the mid-1950s, especially in winter [4
]. Annual warming rate over the TP during 1961–2013 was measured at 0.3 °C/decade, about twice the global rate during the same period [11
]. Global warming has accelerated the melting of the glacier and snow over the TP [12
]. According to the First and Second Glacier Inventory Datasets of China, the glacier area on the TP has retreated by 23% over the past 50 years on the whole [13
]. The Himalaya shows maximum glacial reduction in areal extent, while glaciers in the interior TP are moderately stable [1
]. In respect of snow, its long-term variations under global warming have also been widely focused [11
]. Due to the complex physical mechanisms behind snow cover and heterogeneous terrain over the TP, no consensus at present has yet been reached about the changing trend of snow cover during the past several decades in this region.
Snow and glacial melt are important hydrological processes in the TP [5
], therefore, glacier recession and changes in snow cover under recent climate change are expected to affect the hydrological cycle and subsequently influence the quantity and seasonal distribution of basin runoff [10
]. Meanwhile, rainfall is also a key factor contributing to streamflow in this region, especially in the eastern and southeastern TP. In Northwest China including TP, glacier retreat in 1990s has caused an increase of more than 5.5% in local river runoff [20
]. Glacier retreat in the Himalayas has also greatly contributed to increased runoff in recent decades [21
]. Based on hydrological data and Mann-Kendall analysis, the study of Chang et al. [22
] indicates that glacier and snow melt runoff caused by warming climate, as well as precipitation, might be major drivers for the increasing streamflow in the source region of Yangtze River. Furthermore, in this region, Yao et al. [23
] conducted a study by employing a statistical function between glacier volume and glacier area to calculate glacier runoff; the results suggest an increasing trend of 2.02 × 108
/10a for glacier runoff during 1986–2009, which accounts for 17.5% of the total runoff changes in this period. In addition, Zhang et al. [24
], by applying a modified degree-day model in the Tuotuo River basin located in the headwaters of the Yangtze River, found that in the 1990s, a two-thirds increase in river runoff was caused by the increase of glacier runoff. Meanwhile, on the basis of the statistical mechanics and maximum entropy principle model, studies on the upstream of the Shule River basin reveal that the impact of glacier runoff on river flow has increased during 1957–2010, especially after 2000 [6
]. In addition, in this region, Zhang et al. [17
], by adopting a macroscale hydrological model, indicated that the increase in glacier runoff accounts for approximately half of the increase in river runoff from 1971 to 2012. In the upper reach of the Heihe River basin, under precipitation rise and heavy cryosphere melting, the river runoff increases quickly after 1998 [25
]. Furthermore, in the upstream of Mekong and Salween River basins, it is found that rainfall runoff has the largest impact on changes in river runoff, while there is moderate influence from ice melt runoff [26
]. However, most of the above studies are based on statistical analysis of the measured data, to evaluate the influence of melt water on river runoff change in recent decades. There are still relatively few researches employing hydrological processes simulation-based method to comprehensively and systematically quantify the respective role of glacier runoff, snowmelt runoff, and rainfall runoff in affecting streamflow trend over the TP river basins.
The Lhasa and Niyang River Basins are the two large tributaries of the upper Brahmaputra River basin. Changes in water resources in the two basins have an impact on not only local social and economic development, but also affect water supply for agricultural and industry activities in downstream areas such as navigation or operation of hydropower plants. There are some researches analyzing the connections between streamflow, melt runoff, and climate [12
]. However, to what extent the melt and rainfall runoff control the change trend of river streamflow still remains unclear in the two regions, especially in the relatively more glaciated Niyang River Basin.
In this study, the degree-day model is coupled with the Variable Infiltration Capacity (VIC) model, i.e., the VIC-glacier model [30
] is employed in the Lhasa River Basin (LRB) and upstream of the Niyang River Basin (UNRB) (Figure 1
). An advantage of the hydrological modeling approach is the possibility of deriving certain components, such as soil moisture or snow runoff, which are not actually observed with long time-series in large scale basins. There are three main objectives in this study: (1) to quantify the contributions of rainfall, snowmelt, and glacier melt runoff to the total runoff of the two basins; (2) to study the response of the three runoff components to recent climate change during years 1963–2012; (3) to investigate the hydrological role of each runoff component in controlling the streamflow trend over the two focus basins.
To fully understand potential changes in the hydrological regime in the Tibetan Plateau basins under global continuous warming, the VIC-glacier model was employed in the LRB and UNRB, to analyze the responses of rainfall runoff and melt water to recent climate change, and also quantify their controlling roles in the trends of river streamflow during 1963–2012. The hydrological model was calibrated using observed streamflow and remotely sensed data to reduce simulation uncertainties in the focus basins.
The rainfall runoff is the dominant water resource, which accounts for over 70% of the total runoff in the two basins. Glacier runoff contributes 5.9% and 17.9% to the overall runoff in LRB and UNRB, respectively. In respect of snow melt runoff, its contribution to total runoff is about 12% in the two basins.
The simulations indicate that there is a significant increasing trend in glacier runoff for the two basins during 1963–2012, especially in the period of the 2000s, when the ice melt water exhibited a large increment, up to about 45% relative to the baseline period. Snowmelt runoff displays a general decreasing tendency in these regions. In respect of rainfall runoff, both basins suggest a rising tendency in the study period.
In the LRB, increasing river flow during the past decades is attributed predominantly to rainfall runoff rise, which contributes over 88% to the overall increases in basin runoff. For the more glaciated UNRB, the influence from melt water on streamflow is growing, and the rise in glacier runoff accounts for up to 41% in controlling total runoff trend during 1963–2012. With regard to snowmelt runoff, its role is relatively small in affecting total runoff tendency.
The runoff components in the glacierized LRB and UNRB have been quantified in this study, which will help improve our understanding of the variations in hydrological regime under change in climate warming. This study may shed light on hydrological modeling in cryospheric regions like the TP basins and other similar regions. The findings from our results may provide beneficial reference for local governmental decision-making in water resources management.