Temporal variability of the non-steady contribution from glaciers to water discharge in western Austria

Summary

A long series of negative glacier mass balance years during the last decades influenced the run-off characteristics of glacierized catchments in the Austrian Alps. For balanced conditions, run-off from catchments containing glaciers shows increased values for warm and dry periods, but the annual sum generally equals the total basin precipitation. In contrast, the recent series of negative glacier mass balances reduced the storage volume of the glaciers, providing additional water in the rivers for many years. The existing Austrian Glacier Inventories allowed the determination of very accurate glacier volume changes between 1969 and 1998. These data were used to calculate the excess discharge for two catchment basins in western Austria (one unaffected by hydro power management, one with hydropower management) for the time period between the inventories which then was compared to the accumulated river run-off from the individual basins. The inter-annual distribution of the total excess could be determined by using existing mass balance series from several glaciers in the different basins as scaling functions. In addition, a degree-day approach was used to provide information about the role of above-average glacial melt water on a monthly basis. Considering different gauging stations along the rivers, it was found that the amount of excess discharge during the entire period was in the range of 1.5–9% of the total discharge, depending on the relative degree of glacier coverage (4–40%). For summer months only, this fraction increases to 3–12%. In individual months, however, the relative importance of excess melt can reach more than 25% in a highly glacier covered catchment (40%), but it can also contribute up to 20% for catchments with a glacier coverage of 8–15%.

Introduction

The hydrograph of rivers with glacierized catchments generally differs from those of catchments without ice resources. This influence of glacier mass balance on streamflow and its importance for water management and hydropower planning was already described 50 years ago (Kasser, 1959). Recent models show that the projected change in climate will have a negative impact on hydropower performance in Switzerland, due to a change in the timing of water availability (Schaefli et al., 2007). Snow melt on glacier surfaces is delayed in contrast to snow melt on rocks and soil, and ice is transported from high elevations with less melt to lower levels with stronger melt. Consequently, hot and dry periods, when river levels are usually declining, intensify glacier melt and provide additional water for river run-off. This compensating effect was described in detail by Röthlisberger and Lang (1987). In general, the fraction of glacial melt water in alpine river discharge is rather small on an annual basis. For moderate mass balance years glacial melt water provided about 10% for the Rofenache in Vent (41% glacier coverage) and about 4% for the Inn in Innsbruck (4% glacier coverage, Wendler, 1967, Lanser, 1959). Even for stronger negative mass balance years, the contribution to the annual runoff for the Rofenache amounts only to about 21% (Braun et al., 2000). During the main melt period (June–September), however, some rivers are strongly influenced by the daily melt cycle of the upstream glaciers, which needs to be taken into account for long-term hydrological planning and natural risk assessment (e.g. Ala Archa which has 36% glacier coverage and 70% melt water during July and August with a moderate negative glacier mass balance, Hagg et al., 2006). In the case of a balanced mass exchange on the glaciers, the discharge due to ice melt is only shifted seasonally within the hydrological year. In late spring, river discharge is usually dominated by snow melt, whereas during the summer more water is released from glaciers (Escher-Vetter, 2000).

For periods of glacier retreat, however, melt water from glaciers provides excess discharge, not available in years with zero or positive glacier mass balance. Since the middle of the 19th century alpine glaciers have lost a considerable amount of their area and volume (Müller et al., 1976, Maisch et al., 1999, Paul et al., 2004, Bauder et al., 2007, Lambrecht and Kuhn, 2007, Groß, 1987, Rentsch et al., 2004, Finsterwalder, 1953) and a further rise in mean summer temperatures will continue this trend. There is a wealth of model studies, investigating the potential future streamflow in glacierized catchments (e.g. Singh and Kumar, 1997, Braun et al., 2000, Barnett et al., 2005, Schaefli et al., 2007, Huss et al., 2008). Usually, these investigations use a period of measured discharge for calibration of their runoff model, before simulating the future runoff scenarios. Future glacier melt and glacier extent is incorporated in different degrees of sophistication, from simple index methods and estimates (e.g. Braun et al., 2000) to distributed melt models (Hock, 1999) and more physically based temporal glacier evolution (Huss et al., 2008). Chen and Ohmura (1990) conclude that the area change of the ice coverage is the most important factor influencing the long term runoff trend in glacierized catchments. Studies by Collins, 2006, Collins, 2008 and a review of a number of investigations (Casassa et al., 2009) demonstrate that the effect of glacier melt on river run-off is of high importance for glacierized catchments. Since the end of the Little Ice Age in the middle of the 19th century the recession of glaciers led to an increase of summer discharge in most mountain regions. The peak of this contribution, however, seems to have been passed already in the 1940s to early 1950s (Collins, 2008).

However, the temporal variability of the excess discharge during periods of glacier retreat for different timescales (months to decades), on top of the normal ice melt for balanced conditions, is so far not well known. In this study the effect of negative balance years on the temporal evolution of run-off from glacierized catchments will be analysed. The period of positive mass balance years in the middle of the 1970s is not considered here, because no information is available about the total mass gain in the regions during this period. This temporal storage of mass reduced the run-off for some years, while it had a small positive effect on discharge during subsequent years. In this analysis we are only concerned with the net mass loss of the entire period. For this purpose we compare calculated excess discharge derived from the two Austrian glacier inventories (1969 and 1998) in relation to the meteorological conditions for two of the main glacierized catchment basins in western Austria.