Glaciation (%)

Figure 18.5 Coefficients of variation of modelled discharge in the months July and August at the test sites Tuyuksu, Rofenache, Abramov and Vernagtbach, for current glacier extent and without glaciers. Shown is the polynomic trend of second order

Runoff scenarios for Vernagtbach


1. Oct 1. Nov 1. Dec 1. Jan 1. Feb 1. Mar 1. Apr 1. May 1. Jun 1. Jul 1. Aug 1. Sep

1977/78 simulated

2x CO2, glacier reduction by 50%

2x CO2, present-day glacierisation 2x CO2, without glaciers e g d ly ai d n a e

Runoff scenarios for Tuyuksu region

1. Oct 1. Nov 1. Dec 1. Jan 1. Feb 1. Mar 1. Apr 1. May 1. Jun 1. Jul 1. Aug 1. Sep

Runoff scenarios for Tuyuksu region e g

1982/83 simulated

2x CO2, glacier reduction by 50%

2x CO2, present-day glacierisation 2x CO2, without glaciers


- s

m b







h c


is di








e M



Runoff scenarios for Abramov region

1. Oct 1. Nov 1. Dec 1. Jan 1. Feb 1. Mar 1. Apr 1. May 1. Jun 1. Jul 1. Aug 1. Sep

1974/75 simulated

2x CO2, glacier reduction by 50%

2x CO2, present-day glacierisation 2x CO2, without glaciers

Figure 18.6 Calculated daily discharge of the reference year with substantial glacier melt and of a climate scenario after the doubling of CO2, for three different steps of deglaciation

The effect of the climate change on river streamflow was simulated for the current glacier extent and for two stages of deglacierisation: after an areal reduction by 50% and after complete melting.

These results in Central Asia are compared with former studies in the Alps (BayFORKLIM 1999), where the same runoff scenarios, based on comparable climate changes, have been generated for the Rofenache in Austria. To assess the influence of the degree of glaciation, additional scenarios for the heavily glaciated basin of Vernagtbach, a subbasin of Rofenache, have been created for this study.

Figure 18.6 shows examples for three test sites, including the hydrographs of the reference year and the 2 x CO2-scenarios for current glacier extent and for two steps of deglacierisation.

General effects of climate warming

The reaction of the river hydrographs follows the same principles in the Alps and in Central Asia, as the general mechanism of seasonal and long-term water storage and release are similar on every glacier.

Under current glaciation, discharge begins earlier in the year and rises towards summer, increasing the flood


Sep Summer

□ Average summer DCool, wet summer


■ . ill i







□ Summer with high glacier melt

□ Summer with low glacier melt




Glacier No. 1

Figure 18.7 Effect of climate change (prognosed by the GISS model for the doubling of CO2) and reduction of glaciated area by 50%, related to summer runoff of the two reference years risk. This case has to be regarded as hypothetical because a current glacier extent is not realistic after such a climate change. If the glacierised area is reduced by 50%, snowmelt still begins one month earlier and is more intense, but the summer peaks are mostly reduced to the same level that was already observed in the reference year. The complete disappearance of glaciers yields a water shortage in summer. Under these conditions, the river hydrograph is controlled by groundwater release and rainfall events only and it drops down during dry periods.

The extent of the glacier degradation effect on river discharge varies in the different research basins. These differences are shown in Figures 18.7 and 18.8 for the two modelled steps of deglacierisation and are discussed in the following section.

Reduction of glaciated area by 50%

Figure 18.7 shows the monthly hydrological responses after climate warming and areal reduction, related to the two reference years.

In the alpine basins, there is a big discrepancy between the two model runs, which can be explained by the fact that the two reference years differ extremely in air temperature and glacier melt (Table 18.5). The scenario shows a higher increase in runoff for the cool reference year than for the hot one, where glacier melt and discharge were already high before the climate warming. The differences between the reference years are more drastic at Vernagtbach than in the larger catchment of Rofenache. This displays the influence of a higher degree of glaciation, as the same meteorological input


May i=r

□ Average summer DCool, wet summer

- ■

1 1 "'I p-LT1

Sep Summer

May n

□ Average summer DCool, wet summer


Sep Summer

□ Summer with high glacier melt □ Summer with low glacier melt



Glacier No. 1


Figure 18.8 Effect of climate change (prognosedby the GISS model for the doubling of CO2) and a complete glacier disappearance on river runoff in the investigation areas, related to summer runoff of the two reference years was used in both cases. At Vernagtbach, even a decrease in runoff in August and September can be observed for the hot reference year, where glacier melt was so intense that its volume cannot be exceeded with half of the glacier area after the climate warming. Tuyuksu and Abramov glacier also show a stronger response for the cooler reference years, but altogether the changes are more moderate, especially in the Tuyuksu region, where the small water yield can only be explained with the low glaciation (12.5%) and the high groundwater infiltration. Also at Abramov glacier, there is only a relatively slight hydrological response for this scenario, in view of the still substantial glaciation of 25.5%. This may be attributed to the high elevation interval in which this glacier is located (highest point: 4960 m a.s.l.), which results in relatively cold temperatures at the higher parts of the glacier. At Glacier No. 1, the short dataset does not include years in which summer temperatures differ noticeably. Two years with differing mass balance behaviour, however, were chosen for illustration (Table 18.5). On this summer accumulation glacier type (50% of the annual precipitation falls in July and August), ice melt is strongly controlled by the air temperature during precipitation events, because the aggregational state of precipitation controls the albedo on the glacier. Therefore, it is highly relevant for glacier melt, whether precipitation occurs on cooler or on warmer days. Under these conditions, the mass balance can differ significantly in years with similar meteorological mean values. The basin shows a strong response for the reference year with little ice melt.

Complete melting ofglaciers

The qualitative changes in monthly runoff because of a complete disappearance of glaciers are quite similar in all investigation areas, whereas there are quantitative differences (Figure 18.8). Noticeable changes begin in May or June, where all sites show an increase in runoff as a consequence of a more intense snowmelt up to the higher elevations. In the main ablation season, the absence of glacier melt leads to a remarkable reduction of discharge. In contrast to the above-described scenarios, the strongest effect is achieved for the hotter reference years with a more intense glacier melt. In the Alps, differences between the reference years are highest again, because ofthe large meteorological discrepancy. Monthly runoff decreases from July to September in all cases, but in the cooler reference years, this effect is compensated by the enhanced snowmelt in spring. At Rofenache, summer runoff for the scenario is even higher than in the cool reference year. The same effect can be observed at Tuyuksu glacier and Glacier No. 1. Abramov glacier shows a strong shortage of runoff in July and August for both reference years.

The hydrological effect of this glacier degradation is mainly controlled by the contribution of glacier melt to the total discharge of the reference year, and this contribution again is strongly influenced by the degree of glaciation and by the typical seasonal weather patterns, especially summer air temperature and precipitation. Table 18.5 shows the extent of water shortage in the research basins, the percentage of glacial meltwater to total discharge and the factors that control this portion.

Table 18.5 Hydrometeorological conditions of the reference years and change in discharge in the main ablation season after doubling of CO2 and the complete melting of glaciers

Catchment Area


Year Weather conditions (Jul-Aug) Percentage Glacier mass Change of

Temperature at mean glacier elevation [°C] (long-term mean)


of glacier melt in total runoff [%] (long-term mean)

balance, related to catchment area [mm]

discharge in July-August [%] (related to summer runoff)

Runoff coefficient (long-term mean)

Temperature at mean glacier elevation [°C] (long-term mean)


of glacier melt in total runoff [%] (long-term mean)

balance, related to catchment area [mm]

discharge in July-August [%] (related to summer runoff)



-0.1 (3.2)

181 (185)

7 (29)


-57 (-


0.64 (1.09)

11.7 km2 (80%)


4.4 (3.2)

215 (185)

41 (29)


-90 (-


1.65 (1.09)



-0.1 (3.2)

181 (185)

18 (25)


-25 (-


0.96 (1.16)

98.2 km2 (41%)


4.4 (3.2)

215 (185)

45 (25)


-74 (-


1.61 (1.16)



2.8 (3.6)

251 (244)

23 (17)


-43 (-


1.00 (1.02)

28.0 km2 (25%)


1.2 (3.6)

315 (244)



-10 (-


0.75 (1.02)



4.2 (4.2)

79 (70)

70 (66)


-86 (


1.27 (1.18)

55.5 km2 (51%)


2.4 (4.2)

129 (70)

53 (66)


-75 (-


0.81 (1.18)

Glacier No. 1


2.0 (1.8)

250 (185)

53 (39)


-70 (-


0.93 (0.81)

3.3 km2 (55%)


1.5 (1.8)

189 (185)

24 (39)


-59 (-


0.58 (0.81)

The strongest shortages of discharge occur at Vernagtbach and Abramov, whereby at Abramov the same values are reached for an average reference year as at Vernagtbach for a hot year. Therefore, the decrease is most dramatic at Abramov glacier. The summers in this region are typically dry, and therefore the contribution of glacial meltwater to total runoff is very large. The hydrological effect is also very high at Glacier No. 1, even in the year with a balanced glacier mass budget, although this test site may not be comparable to the others because of its small size. The model gives best results if no groundwater storage is assumed. Without glaciers, runoff only occurs after precipitation, and with this set of model parameters no baseflow can be generated. In the Tuyuksu region, the rather slight hydrological response to glacier disappearance can be explained by the lowest degree of glaciation and by the fact that two cool reference years had to be chosen, because of problems with the quality of data.


The investigations have shown that the conceptual HBV-ETH precipitation-runoff model is capable of simulating runoff in the continental climate of Central Asia. This is indicated by a good agreement between modelled and measured values of runoff as well as glacier mass balance. As a consequence, the degree-day method for calculating snow and ice melt can be applied successfully under these climate conditions. In high mountain regions of Central Asia, energy for melt is predominantly supplied by radiation and therefore highly correlated with air temperature (Ohmura 2001). Moreover, the dominance of low water vapour pressure favours the occurrence of evaporation and the formation of bright ice surfaces. The constantly high albedo values of ice lead to rather stable degree-day factors.

Model runs under contemporary conditions, but without glaciers, show an increase of the year-to-year variation of runoff, if only the main ablation season is considered. This supports the theory of runoff compensation, which explains the balancing effect of a moderate glacier cover on year-to-year runoff variability.

Runoff scenarios for a warmer climate and different steps of deglaciation display a similar general behaviour in Central Asia and the Alps. The most important features are the increase of discharge during snowmelt with rising temperatures and the shortage of summer runoff with the disappearing of glaciers. Quantitative differences are mainly due to the degree of glaciation, local weather patterns and glacier mass balance behaviour in the reference period. These factors have to be taken into account if the hydrological reaction of a glaciated catchment to climate changes is to be estimated properly. Local circulation patterns and their changes with climate variations are especially important determining factors. They control water balances of smallhead watersheds and their reaction to climate changes to a higher degree than any large-scale climate parameter such as continentality. Therefore, general statements about differences between mountains in different climate zones are very difficult to make.

In contrast, there are clear differences in the consequences of the hydrological changes for the lowland areas. The greater the hydrological difference between mountains and lowlands, the more the lowlands depend on mountain runoff and the more significant are changes in mountain hydrology. This means that the shortages in summer runoff lose part of their relevance at the margin of the Alps, where high summer precipitation guarantees a minimal runoff during the vegetation period. In contrast, the dry lowlands of the Central Asian mountains depend on glacier melt to a much higher degree, and they will face a serious water deficiency if the glaciers continue to shrink at the rate observed today.


This project was funded by the German Research Foundation (DFG, project BR1622/5-3) and supported by the Bavarian Academy of Sciences. The authors thank Igor Severskiy, Felix Pertziger, Ersi Kang and their colleagues for providing essential data. The helpful comments of the reviewers are gratefully acknowledged.


AgetaY, Higuchi K (1984) Estimation of mass balance components of a summer-accumulation type glacier in the Nepal Himalaya. Geografiska Annaler 66A(3): 249-255. Aizen VB, Aizen EM, Melack JM (1995) Climate, snow cover, glaciers, and runoff in the Tien Shan, Central Asia. Water Resources Bulletin 31(6): 1113-1129. BayFORKLIM (1999) Klimaanderungen in Bayern und ihre Auswirkungen. Abschlussbericht des Bayerischen Klimaverbundes, Miinchen, p. 90. Bergstrom S (1976) Development and Application of a Conceptual Model for Scandinavian Catchments. University of Lund Bulletin, Series A, p. 52. Braun LN, Renner CB (1992) Applications of a conceptual runoff model in different physiographic regions of Switzerland. Hydrological Sciences Journal 73(3): 217-231. Braun LN, Weber M, Schulz M (2000) Consequences of climate change for runoff from Alpine regions. Annals of Glaciology 31: 19-25.

Escher-Vetter H, Weber M, Braun LN (1998) Auswirkungen von Klimaanderungen auf den Wasserhaushalt hochalpiner, teilweise vergletscherter Gebiete. Abschlussbericht BayFORKLIM, on: CD-ROM ''Klimaerwärmung Gletscher - wie Verändern Sich die GebirgsabflUsse". Kommission für Glaziologie, Bayerische Akademie der Wissenschaften, München, ISBN 3769635000.

Hagg W (2003) Auswirkungen von Gletscherschwund auf die Wasserspende hochalpiner Gebiete, Vergleich Alpen - Zentralasien. MUnchner Geographische Abhandlungen 15: p. 91.

KasserP (1959) Der Einfluss von Gletscherrückgang und Gletschervorstoß auf den Wasserhaushalt. Wasser- und Energiewirtschaft 6: 155-168.

KazNIIMOSK (1999) Climate Change and a New Defence Strategy Against Mudflows and Snow Avalanches. National report on the impact and adaptation assessment for the mountain region of South and Southeast Kazakhstan and the Kazakh part of the Caspian Sea coastal sector. Netherlands climate change studies assistance programme, Kazakhstan climate change study, Vol. 1, Kazakh Institute for Environment Monitoring and Climate, Almaty, p. 86

Kommission für Glaziologie (2001) Gletschergebiet Tujuksu, Sailiski Alatau. Map 1:10'000, published in: Fluctuations of Glaciers 1995-2000, Vol. VIII, World Glacier Monitoring Service, Zürich, expected in 2003.

Nash JE, Sütcliffe JV (1970) River flow forecasting through conceptual models. Part I - a discussion ofprinciples. Journal of Hydrology 10(3): 282-290.

Ohmura A (2001) Physical basis for the temperature-based melt index method. Journal of Applied Meteorology 40: 753-761.

Rango A (1992) Worldwide testing of the snowmelt runoff model with applications for predicting the effects of climate change. Nordic Hydrology 23: 155-172.

Rothlisberger H, Lang H (1987) Glacial hydrology. In: Gurnell AM, Clark MJ [eds.], Glacio-Fluvial Sediment Transfer. Wiley, New York, pp. 207-284.

WGMS (1999) Glacier Mass Balance Bulletin No. 4. IAHS (ICSI)-UNEP-UNESCO-WMO, compiled by the World Glacier Monitoring Service, Zürich.

Was this article helpful?

0 0

Post a comment