Climate change impacts
on NATURAL SYSTEMS AND PATTERNS


2.6 RIVERS



Type of
knowledge
Results and interpretation
Observation and analysis methods
References
Reconstructions

Large alpine valleys and their piedmont to the Mediterranean Sea:
Large-scale geomorphological phenomena have been recorded during the Little Ice Age and are related to the propagation downstream of sedimentary masses mobilised from the 14th century in high alpine watersheds. It leaded to a durable modification of the landscape and the hydromorphological behavior of streams. The behavior analysis of torrents near the catchment area heads reveals that they were subjected to perceptible fluctuations during the 19th century depending on climatic fluctuations.


An extension of the active minor bed during the Little Ice Age has been identified on several sections of the Rhone river and its tributaries; some experienced a clear progradation of a “sedimentary wave” (bedload transport), accompanied or not with a river metamorphosis (from a meandering river to a braided river). These events, classified according to their increasing scale, began in the 14th century for the Rhone river in Lyon, which is under influence of the sedimentary contributions of the Ain river. Note that the Isère river metamorphosis is more premature downstream Grenoble (Drac influence) than upstream the city, which is less influenced by sedimentary flux of the Isère and Arc high watersheds due to greater distance. The phenomenon also concerns the Camargue at the beginning of the 18th century. It seems that the phenomenon reached its peak in the 18th century, although the 19th century is known for its high floods and alpine torrential events. The relative respite of the 19th century might be linked to the effect of the embankment or simply a progressive return to calmer period after the very active 18th century.

This study demonstrates the strong vertical variability of the river beds when they are subjected to a repeated and quick alternation of weak and strong torrentiality phases. It also suggests that the Little Ice Age was heterogeneous concerning climate and geomorphological appearances recorded in the active torrential environments.

The studies of river paleo-dynamics performed for about fifteen years on the basis of old maps and archives, have been analysed.

Bravard 2000 - P

French southern Alps and Mediterranean region:
In second-order tributaries of the middle Durance catchments, phases of decrease in water discharge and sediment load correspond to soil formation, forest expansion in valleys floors, and floodplain incision. The period from 12 000 to 6500 yr BP corresponds to higher river discharge marked by a major sediment accumulation in valley floors, i.e. the so-called ‘Principal Postglacial Infilling’ (PPI; Jorda, 1985). The second half of the Holocene, after 6500 yr BP, is characterised by a relative decrease in sediment deposition and by a general trend towards floodplain incision. Superimposed on this general trend, detritic crises marked by coarser deposits appear to punctuate the whole Holocene period, in particular during the period from 12 000 to 6500 yr BP. Moreover, interbedded in the massive alluvial and colluvial infillings accumulated in the valleys during the early to mid-Holocene period, layers including many subfossil tree trunks (Pinus silvestris sp.) were found and dated by radiocarbon. The good preservation of the trunks and their often in-life position indicate that they were rather quickly buried. Their growth patterns, reconstructed from tree-ring observations, suggest successive rapid environmental changes: an increase in the flooding frequency resulting in an increase in sediment accumulation over the growth sites of the Pinus trees and their death after less than a century.

Palaeoenvironmental records document a change from wetter to drier conditions over the Holocene in the Mediterranean region. As pointed out by Kallel et al. (1997) and Ariztegui et al. (2000), the formation of sapropel 1 in the Mediterranean Sea between 9000 and 6800 cal yr BP coincided with a lake-level Holocene optimum in northern Africa and the western Mediterranean region. The beginning of the trend towards incision in southern Alps valleys after 7600–7000 cal yr BP could be synchronous with the cessation of the sapropel 1 deposition.

Results from lake-level, carbon-isotope and foraminifera-faunal studies are consistent with the reconstruction of cooler and moister summers in the Mediterranean at 6000 yr BP based on the modern pollen analogue technique constrained or unconstrained with lake-level data and AGCM simulations [see references in the study]. The position of the subtropical anticyclone (STA) over the North Atlantic ocean further north during the early to mid-Holocene would block less the westerly flow over southern Europe in summer in comparison with the present-day conditions. The winter and summer temperature could have been 2°C less than at present and the difference between precipitation and evapotranspiration could have been 50–200 mm greater than at present in the western Mediterranean region (Cheddadi et al., 1997). The Holocene pattern of variations in river discharge in the southern Alps and glacier tongue locations in the northern Alps fully agrees with the reconstruction of cooler and moister conditions in the western Mediterranean and warmer and drier conditions in the northern Alps during the first part of the Holocene (Huntley and Prentice, 1988; Cheddadi et al., 1997).

Century-scale periods of decrease in the fluvial activity in the Mediterranean region were in phase not with cooling periods marked by rises in lake level and glacier advance in central Europe, but with a glacier retreat in the northern Alps and lake-level lowering in the Jura. They occurred at ca. 11 500, 10 500, 9000, 7000, 4000, 3000, 2000 and 800 cal yr BP.

Generally speaking, a synchronicity between rises in lake levels in the Jura and increases in fluvial activity in the southern Alps valleys marked by buried Pinus silvestris trees or gravel and pebble accumulations can be observed over the Younger Dryas and Holocene periods except for the mid-Holocene. The chronology of this mid-Holocene phase of interruption of sediment accumulation has still to be detailed. The available radiocarbon dates place it after 6920 ± 190 yr BP, i.e. ca. 7750 cal yr BP, and before 5240 ± 190 yr BP, i.e. ca. 5600 cal yr BP (Rosique, 1996; Sivan, 1999). These correlations suggest that wetter conditions marked by high lake levels in central Europe corresponded to a higher frequency of stormy precipitation inducing floods, gravel deposits and buried trees in the northern Mediterranean area. Furthermore, it is noteworthy that the decreases in fluvial activity in the southern Alps recorded at 11 500, 9000 and 7000 cal yr BP do not have an equivalent in the aridification phases identified by Jalut et al. (2000) from the deciduous/sclerophyllous pollen ratio. The general synchronicity observed between century-scale climatic oscillations marked by increasing river discharge in the southern Alps, rises in lake levels in the Jura and glacier advances in the Alps could have resulted from an alternatingly southward–northward displacement of the Atlantic Westerly Jet.

Over the last 20 years, systematic investigations have been carried out in the French southern Alps valleys to reconstruct morphogenic processes associated with regional palaeohydrological changes (Jorda, 1985; Gautier, 1992; Rosique, 1996; Miramont, 1998; Miramont et al., 1999, 2000). The synthesis is based on (1) a compilation of data from 25 sites in the middle Durance valley and its tributaries and (2) more than 90 dates from radiocarbon dating and archaeological materials (Miramont et al., 2000; Jorda et al., in press). Most of data are from second-order tributaries of the middle Durance, located in the so-called ‘Terres Noires’, i.e. a Jurassic marl highly sensitive to erosion.

Correlations are presented over the Younger Dryas and Holocene periods between southern and central Europe using the fluvial activity record reconstructed by Miramont et al. (1999, 2000) in the southern Alps and the lake-level record established in the Jura (Magny, 1998, 1999).

Magny & al. 2002 - A

Northern French Alps - Rhone basin upstream Lake Le Bourget:
Clastic record (magnetic susceptibility) of Lake Le Bourget during the last 150 years was compared with the cumulative length of five large alpine glaciers located in the Mont-Blanc massif and in the Swiss Alps documented by Vincent et al. (2005). Clastic sedimentation was higher in Lake Le Bourget when these glaciers were either increasing or decreasing from three periods of culminating length (i.e. at the end of 19th century; between AD 1920 and AD 1931; between AD 1983 and AD 1995). The MS signal in Lake Le Bourget shows three significant peaks around AD 1880, AD 1930, AD 1950 and a small plateau between the 1980s and the 1990s. This suggests that since AD 1870, clastic supply in Lake Le Bourget has resulted from bedrock erosion associated with Mont-Blanc glacier fluctuations in the Arve River following the end of the Little Ice Age. If an anthropogenic imprint exist, it seems that it only impact the amplitude of the signal and not the timing. It seems that the flooding of the Rhone River into the lake results from climatic factors.
The comparison of Swiss glacier fluctuations and periods of higher lake levels in Western Europe during the last 3500 years (Holzhauser et al., 2005) with the continuous record of Rhone River flooding activity in Lake Le Bourget (the magnetic susceptibility signal) highlights periods where increased flooding activity matches with periods when glaciers where near their maximum extent and higher lake levels: between 2800 and 2400 cal BP; 1500–1300 cal BP; 1200–1050 cal BP; 900–800 cal BP and 700–220 cal BP. Periods of increased flood activity occurring between 2200 and 2000; 1950–1600 and 1050–950 cal BP, are, on the other hand, not contemporaneous to any clearly documented glacier or lake level fluctuations in the alpine range, but took place after wet periods associated with higher Rhone River torrential activity (cf. Arnaud et al., 2005): the so-called Iron Age Hydrological crisis (1200–1550 cal BP), the Roman Wet Period (between 2100 and 1900 cal BP) and the High Middle Age Wet Period (1500–1200 cal BP).
Between 3500 and 9400 cal BP, comparisons with the few continuous reconstructions of glacier or hydrological activities during the Holocene suggest that Rhone River flood activity in Lake Le Bourget during this period has been above all sensitive to wet (and often cooler) periods in the Alps (favouring higher lake levels in mid-Europe). These wet periods were contemporaneous to glacier fluctuations, especially after the Holocene Optimum (a period known as the Neoglacial). Only one period of higher lake level from Magny (2004) is not clearly documented in this study: it correspond to a period dated between 4800 and 4850 cal BP, matching (within dating uncertainties) a glacier advance of the Miage glacier documented by Deline and Orombelli (2005) between 4600 and 4800 cal BP in the Mont-Blanc Massif and the Rotmoss II phase of glacier advance in the Alps (Maisch et al., 2000).

A 14-m long piston core (LDB04) was retrieved from Lake Le Bourget, NW Alps (France), in order to provide a continuous record of flooding events of the Rhone River during the Holocene. The selection of the coring site was based on high resolution seismic profiling, in an area with limited mass wasting deposits and accumulated proximal Rhone River inter- and underflow deposits. The age-depth model of this core is based on (i) 14 AMS radiocarbon dates, (ii) radionuclide dating (137Cs) and (iii) the identification of historical data (flood events, eutrophication of the lake). A multi-proxy approach was used to track the evolution and origin of clastic sedimentation during the Holocene, in order to identify periods of higher hydrological activity in the catchment area. Spectrophotometry was used to detect fluctuations in clastic supply and the study of clay minerals (especially the Illite crystallinity index) allowed locating the main source area of fine grained clastic particles settling at the lake after flood events. This dataset highlights up to 12 periods of more intense flooding events over the last 9400 years in Lake Le Bourget and shows that the main source area of clastic particles during this period is the upper part of the Arve River drainage basin. This part of the catchment area drains several large glaciers from the Mont-Blanc Massif, and fluctuations in Rhone River flood supply in Lake Le Bourget is interpreted as resulting essentially from Mont-Blanc Glacier activity during the Holocene. The comparison of clastic sedimentation in Lake Le Bourget with periods of increasing land use and periods of Alpine glacier and mid-European lake level fluctuations, suggest that the core LDB04 clastic record in Lake Le Bourget is a continuous proxy of the Holocene hydrological history of the NW Alps.

Debret & al 2010 - A
Observations

Europe:
In Europe, river flow has tended to increase where precipitation has increased and to decrease where precipitation has decreased in recent years.  Flow variation from one year to the next appears to be more closely linked to changes in regimes of precipitation than to temperature changes.  In a major part of Europe […],  seasonal flow variation from spring to winter has been tied to not only a change in total precipitation but more precisely to an increase in temperature: the precipitation fell more often as rain than snow and so reached the rivers more quickly then previously.

  IPCC 2001 - R

France:
In France, a statistical analysis of discharges at 140 gauging stations from 1975 to 1990 show a reduction of snow-melt regimes to the benefit of “transitional” regimes and to a marked irregularity in the seasonality of regimes.

  Krasovskaia et al., 2002 in Bravard 2006 - P

Switzerland:
Streamflow has been significantly increasing in the winter period, especially the winter annual maxima (about 60% of the stations). To a lesser degree this increase affects other quantiles as well, especially in the latter period of study. Looking at the whole period 1931-2000, we see that upward and downward trends are balanced in the low quantile range.

Streamflow has been significantly decreasing in the summer period, in particular in the summer low streamflow quantile range. In this case, the whole period 1931-2000 shows the most substantial change in the streamflow regime. In the 70-years analysis, downward trends are dominant, while for the 30 and 40-years periods, upward trends are dominant. The spatial distribution of trend significance for the recent period 1971-2000 for winter maximum streamflow are shown.

Spring and autumn act as transition periods between winter and summer. They show a change from the downward trends of 70-year period to predominating upward trends of the recent period. Spring trend results are most similar to the annual case. Autumn appears to be more influenced by summer in the 70-year period (almost only descending trends) and by winter (almost only increasing trends) in the other two periods.

On an annual basis, the seasonal trends compensate each other, which results in more upward trends in lower quantiles and more downward trends in maximum and upper quantiles. The percentages of significant annual trends are decreasing with shorter record lengths.

The trend results for streamflow in this study are in agreement with recent studies regarding precipitation trends in Switzerland conducted by Widmann & Schär (1997) and Frei & Schär (2001), where significant increases in winter precipitation were observed. However, it appears that changes in precipitation can only partly explain the observed trends in streamflow.

The data used in this study are mean daily streamflow of 54 currently operating gauging stations in Switzerland, with continuous records which are not affected by major anthropogenic influences and can be considered independent in space (from base network of Pfaundler, 2001). The analysis was conducted for three periods of study: 1971-2000 for 54 sites (70 years), 1961-2000 for 35 sites (40 years), and 1931-2000 for 17 sites (30 years) of the base network.

Mean daily streamflow data were analysed for trends with the Mann-Kendall nonparametric trend test. To find shifts in the distribution of daily streamflow, a range of quantiles on annual and seasonal bases were studied for three different observation periods. Two approaches in the trend analysis have been used: the trend analysis was conducted on the original data series; and series with positive lag-1 serial correlation were prewhitened by applying a first order autoregressive filter to the data prior to trend analysis.

Bîrsan & al 2003 - P

Ebro - Rhone - Pô basins:
Actually no significative trend has been detected in historical date (year, season, day) of rains and discharges hundred years old and more, in France, West Europe and other countries.

  Duband 2003 - P

Rhone catchment:
Concerning river discharge, statistical tests applied to 8 gauging stations of the Rhone River downstream Geneva demonstrated that hydrology is stationary. However two types of ruptures are apparent, one locally in 1891, due to artificial developments at the outlet of Lake Geneva, the second one at the end of the 1970’s, with the occurrence of wet decades throughout the basin, following a period (1940-1975) of lull. A new cycle rich in strong floods has occurred in recent years, similar to the late 21st period, but no effect of global change having been detected yet.

 

 

Sauquet et Haond 2003 in Bravard 2006 - P

Swiss Alps:
Statistically significant trends were identified for each station on an annual and seasonal basis and for different streamflow quantiles. A general increase in annual streamflow has been observed, mostly due to increases in winter, spring and autumn runoff. Most dominant changes have occurred in the winter season. Winter streamflow has increased over the whole distribution, but especially markedly for maximum flows: at least 63% of the basins show a statistically significant increase in winter maxima in all periods. On the other hand, increases in spring and autumn runoff are concentrated mostly in moderate and low flow quantile ranges. Trend behaviour in the summer season is different. Both decreasing and increasing trends are found concentrated in moderate and low flow quantiles rather than summer maxima. Summer season behaviour is important because summer runoff provides most water on an annual basis in basins with an Alpine influence. Most of the trends in winter high flows, and spring, summer and autumn moderate and low flows were field significant. However, significant trends were spread across the country and found in all of the different runoff regime regions.

Trends in daily precipitation in the same study periods were generally not as significant as those in streamflow. Streamflow changes could not be explained on the basis of changes in precipitation alone, especially for the period after 1961. Trends in air temperature point towards a general increase in minimum daily temperatures and a decrease in the maximum daily temperatures, leading to a significant decrease in the temperature range. Perhaps most importantly, strong increasing trends appear in the number of days with minimum daily temperature greater than 0 °C (up to 50% of all stations in some cases), which are concentrated in the winter and spring seasons. The authors speculate that the observed increases in winter runoff are due to a shift of snowfall into rainfall. Similarly, low and moderate flow increases in the spring could be explained in part by increased and earlier snow melt due to the observed air temperature rise.

Correlation analyses of streamflow trends with basin attributes show statistically significant relationships between streamflow trends and mean basin elevation, glacier and rock coverage (positive), and basin mean soil depth (negative). The correlations are generally strongest for the moderate flow ranges, and decrease for extreme flows. For extreme flows it appears that factors other than general basin properties play a key role in runoff production. The positive correlation between streamflow trends and basin attributes related to altitude points to the higher vulnerability of mountain basins to changes in precipitation and air temperature. This seems to be supported by the fact that most consistent statistically significant correlations are observed in winter and spring (especially higher streamflow quantiles in spring), when we expect the influence of precipitation and temperature on runoff production in mountain basins to be strongest. It has been shown that temperature changes are particularly significant for basins below 1000 m a.s.l. in Switzerland because the winter mean temperature at this altitude is around 0 °C and even small variations in temperature may determine whether precipitation falls as rain or snow (Scherrer et al., 2004).

The relationships between streamflow and watershed attributes suggest that the most vulnerable environments are, not surprisingly, mountain basins. For example, a strong correlation was found between streamflow trends and glacier coverage for low and moderate streamflow in the spring and summer. The authors expect that this is indicative of snow and ice melt in mountain basins which contributes to steady but low streamflow.

One of the suggestions from this work is that the observed increase in winter, spring and summer streamflow in glaciated basins is due to warmer temperatures, more precipitation, as well as more snow and ice melt. Altogether the results suggest that mountain basins are the most vulnerable environments from the point of view of climate change, because of their watershed properties which promote fast runoff and because of their fundamental sensitivity to temperature changes which affect rainfall, snowfall and snowmelt.

The present results show some connections between observed streamflow trends and changes in atmospheric variables, but the authors do not claim that these completely explain the variability in streamflow. What remains unexplained are trends in streamflow due to continuous changes in other watershed properties, such as land cover and land use, natural and anthropogenic changes to the fluvial system, and others. Although they selected watershed to be as much undisturbed as possible, and they expect that most of the (natural) watershed changes occur on much longer timescales than those studied here, they recognise that they do contribute to hydrological variability.

This study presents a statistical analysis of trends in mean daily streamflow records from 48 watersheds in Switzerland with an undisturbed runoff regime. Data were analysed for trends with the Mann–Kendall nonparametric trend test. To identify shifts in the distribution of daily streamflow, a range of quantiles on annual and seasonal bases were studied for three different observation periods (1931–2000, 1961–2000 and 1971–2000). The field significance of trends was analysed by a Monte Carlo bootstrap procedure. Identified trends in streamflow were related to observed changes in precipitation and air temperature, and correlated with watershed attributes.

Data included daily precipitation, and daily minimum and maximum air temperature measured at climate stations of the MeteoSwiss network. The authors looked not only at trends in different quantiles of the precipitation distribution, but also at changes in the number of wet days (defined here as days having at least 0.1 mm of measured precipitation) and the total amount of precipitation. To support direct comparisons with streamflow, the analyses were conducted for the same three study periods.

Birsan & al. 2005 - A
Swiss Alps:
In the Swiss Alps, in 2003, rivers fed by glaciers experienced higher annual average flow than normal and accumulation basins downstream of frozen sites filled rapidly. This can be explained by the acceleration in the melting of glaciers during the extreme hot weather period. 

The line at which the ice melt dominates rose indeed to altitudes never before recorded (over 3500m).  In glacier free zones, river flow was, inversely, lover than normal.  Lakes were influenced in the same was as rivers by the metrological conditions but their size had a modifying effect.  Low capacity lakes whose tributaries were not fed by glaciers were the most affected.
  ProClim 2005 - R
Switzerland:
For the rivers located North of the Alps, with unbuilt catchments area, the peak flow increased in winter.
  Birsan & al. 2005 in Frei & Widmer 2007 - E
Southern Germany:
The long-term behaviour of the highest runoffs can be characterised as follows:
• When examining the annual series from 70 to 150 years duration, the majority of annual highest runoff levels do not show significant changes.
• When examining the last approx. 30 years the highest runoffs show increasing trends at many gauges.
• The frequency of winter floods has increased since the 70s with the exception of Southern Bavaria.
• The monthly flood runoffs in the winter half year since the 70s are higher than in the time before the 70s.

In summary, it can be established that only in the last 30 to 40 years do the examined runoff time series demonstrate regional increases in flood runoffs.

Investigations of long series of hydrometeorological and hydrological measurements available were systematically carried out for Baden-Württemberg and Bavaria on the basis of a large data set within the context of KLIWA. The long-term behaviour of the flood runoffs, the mean runoffs, the regional and heavy precipitation, the air temperature, the evaporation and the snow cover period were analysed for time periods in the 20th century.

The investigation of the long-term behaviour [of flood runoffs] included determination of any linear trends present in the time series of the annual and monthly highest runoffs. The annual and monthly highest runoff values at 107 gauges, which have long observation series since at least 1931, formed the basis for the trend investigations. Furthermore 51 gauges with shorter time series, i.e. with observation start after 1932, were included in the analysis.
Hennegriff & al 2006 - A
Ecrins massif (Torrent du Glacier Noir):
Comparison of 16-day monitoring periods in July 2003 and July 2004 showed that mean daily AT measured 1 km from the snout of the Glacier Noir was 1.2 °C higher in 2003, mean Q of the Torrent du Glacier Noir was 2.3 times greater and the SSL (suspended sediment loads) was between 3.1 and 4.1 times greater in July 2003, than for the same period in the 2004 ablation season. There is an increase in SSC during the 2004 observation period which is less apparent in 2003, most likely because higher ATs and consequently higher Q earlier in the 2003 melt season had removed available sediment before the study took place in July.

It invites parallels to be drawn between the results from this study and likely impacts of higher alpine summer temperatures as a result of global warming. Such changes in SSLs have important implications for sediment transfer from mountain to piedmont zones, for stream ecology, for downstream channel stability and flood risk and for the hydro-electric industry.
In this study streamflow (Q) and suspended sediment concentrations (SSC) in a proglacial stream in the Ecrins National Park were monitored in the summers of 2003 and 2004.
SSC and Q were continuously monitored for 16-day periods during July 2003 and July 2004. SSC was monitored by automated pump sampling during diurnal events in each season and supplemented by a 10 min turbidity record. Q was monitored at a range of flows and a rating curve used to convert a 10 min water level record into Q. Air temperature (AT) was also logged at 10 min intervals throughout the study.
Stott & Mount 2006 - A
World:
On a world wide scale, the majority of maximum annual flow readings (70%) do not represent any significant statistical trend.  The trend of the other readings is divided almost equally between an increase and a decrease.  There is not therefore, at the present time, any convincing proof of any kind of long term increase in the force of river floods.
Study of trends in long time series of annual maximum river flows at 195 gauging stations worldwide.
Svenson & al. 2006 - P
France:
At the French scale, no coherence in the spatial distribution of the detected anomalies, or generalised evolution of discharges or floods, have been found. No coherent and generalised climatic trend has been detected in the evolution of the hydrometric extremes pattern.

Evolutions have been detected for only 5 of the 15 hydro-climatic regions:
• weaker low water levels in the Alps and the nival wave seems to be more premature in the Northern Alps;
• increase of the nival module of the glacial stations and earlier occurrence of the maximum basic discharge;
• Stronger low water in Pyrenées and Basque country. Pluvial floods also seem to decrease in Pyrenées;
• increase of annual maxima in the Northeast region;
• Stronger floods in oceanic region (Northern France), apparently related to the filling of the fluvial water tables.

In the Northern Alps, changes concern snow cover floods and low waters. The melting wave occurs earlier and the melting peak decreases. Low waters are less severe, with a significant increase of the annual daily minima and a significant decrease of the volume deficit. The duration of low waters also seems to decrease, although not significantly.

According to the climatic evolution maps of the IMFREX project, the observed warming is significant for a large number of stations and for several parameters (notably the number of days warmer than 25°C or the number of frosty days). On the other hand, the precipitation evolution is much less marked and is significant only for rainfall amounts on several days. The observed climatological evolutions are insufficient to explain the extreme hydrometric evolutions, notably because of the strong non-linearity of the rain-temperature-discharge relation.
Set of 192 long series of daily discharges, only with series considered as not influenced, covering at least 40 years of observation. Around thirty samples were extracted for every station from descriptive variables of high-waters, low-waters and water pattern. Every sample was subjected to a stationnarity test chosen according to the series autocorrelation, the expected type of distribution and the sample length.

A series of 15 hydro-climatic regions was established (Renard, 2006) by crossing a classification based on floods and low waters seasonality with a pluviometric zoning of Météo France. The regional coherence was studied by using regional variable, from the regional version of the Mann and Kendall test (Douglas et al., 2002; Yue and Wang, 2002), or with the regional test of deviation proposed by Renard.
Lang & Renard 2007 - P
Switzerland:
Precipitations show a slight increasing trend (about 113 mm over hundred years) for 303 precipitation record series covering the 1901-2005 period. However, this trend is not significant because of the large variability of this parameter. Given that the evaporation trend is almost identical to the precipitation one (increase of 99 mm per hundred years ), discharge remained almost constant over the last hundred years.

Since the 1960s, temperature of all Swiss large rivers shows a clear increasing trend. This evolution is more marked for the Plateau streams than for the Alpine ones (Jakob et al., 2007). The increase in water temperature occurs at the same time as the average atmospheric temperature increase.
Bibliographic review North & al. 2007 - R: OFEV
Swiss Alps (upper Aare and upper Rhône basins):
Annual total runoff from the (near-) ice-free basins of Allenbach and Grande Eau generally reflected the temporal variation of precipitation, rising to maxima in the late 1970s–early 1980s and late 1960s respectively. P 11–10 at Sion was reasonably correlated with Q 1–12 of both Grande Eau and Allenbach. Year-to-year fluctuations of runoff in the two basins were broadly parallel until the 1990s. Mean decadal flow in the Allenbach and Grande Eau declined by 11% and 16% respectively from 1977–86 to 1997–2006.

Runoff in the Massa, which drains the most highly icecovered of the glacierized basins (~66% glacierized), mimicked the cyclical pattern of variation in T 5–9 at Sion. The correlation between T 5–9 and Q 1–12, whilst generally strong, was greater the shorter the time period considered. High discharge in 1928 resulted from a warm spring with precipitation slightly above average. Rising temperatures, together with increasing precipitation, led to runoff generally increasing over subsequent years, with a maximum in 1947. Declining summer air temperatures next reduced runoff to a minimum in 1978. During the second warming cycle (1980s-2006), temperatures exceeded those experienced in the 1940s. However, mean discharge in the warmer 1990s–2000s failed to exceed that of the 1940s–50s.

The Rhône drains the second most highly glacierized basin of the study. Correlation of T 5–9 at Sion with runoff in the Rhône was not as strong as for flow in the Massa, whereas correlation of P 11–10 with runoff was more positive for the Rhône, for respective periods. Runoff in the two rivers was strongly correlated. Discharge in the Rhône also increased from the cooler 1960s–70s to the warmer 1980s–2000s but by only 12%, whereas quinquennial average annual runoff in the Massa increased by almost 50% between 1974–78 and 2001–05, accompanying a rise in mean T 5–9 from 16.1°C to 18.1°C. In fact, runoff at Gletsch increased from 1977–86 to 1987–96, before decreasing in 1997–2006.

The pattern of temporal variation of discharge in the Lonza (~38% glacierized) was to an extent intermediate between that in the Allenbach (0%) and the Rhône (~54%). Runoff in the Lonza increased between 1977–86 and 1987–96, before decreasing in 1997–2006.
The study basins are located in the upper Aare and upper Rhône basins in Switzerland. Percentage glacierization of basin area was taken from the contemporary annual Hydrologisches Jahrbuch der Schweiz (e.g. BAFU, 2006). The Allenbach basin is ice-free. The Grande Eau basin is almost ice-free. Total glacier-covered areas of about 32 and 22 km2 in the Lonza and Rhône basins declined by about 10% and 7.5% respectively between 1977 and 2002. Percentage glacierization of the Massa basin (recalculated to take into account catchment area change) reflects a loss of total icecovered area of about 8 km2 (6%) from 136.6 km2 between 1934 and 2002, of which about three-quarters had disappeared by 1957.

The Rhône was gauged in a natural cross-section at Gletsch for 1894-1903 and 1920-1928 periods, before use of a flume structure from 1956. The Lonza is gauged from 1956. Following dam construction, from 1965 the Massa was gauged at Blatten-bei-Naters, upstream of the former station at Massaboden, reducing basin area by 3.47%. Runoff is unlikely to have been proportionally reduced.

Total annual runoff was taken as the total for the calendar year (Q 1–12). Mean May–September air temperature (T 5–9) was used to indicate heat-energy availability for melting. Total annual precipitation between November in one year and October in the next (P 11–10) reflects build-up of winter snowpack and summer rainfall which affect total annual discharge between January and December (Q 1–12) in the second year.
Collins 2008 - A
German Alps:
Caspary (2004) underlines that the discharge regimes of the rivers Donau, Enz, Kocher and Alp in South West Germany show statistical instationarities in their time series because of the relative accumulation of extreme events since the 1990s. E.g. a discharge event with a reoccurrence interval of 100 years in the reference period 1932-1976 of the river Enz at the gauge of Pforzheim equals a discharge event with a reoccurrence interval of 30 years in the reference period 1932-2002.
Bibliographic review Staffler & al. 2008 - A
Alps:
The water amount in alpine watersheds is determined by numerous factors:
• the climatic variability at a decadal scale (NAO, AO…);
• the temperature increase at a centennial scale;
• the extreme events;
• the water use for diverse socioeconomic activities.

Changes in the one of these factors, notably induced by a warmer climate, will modify the water quantity and quality, as well as the seasonal character of discharges, which is currently observed in the alpine range.
  Beniston 2007 - C

French Alps:
In the Northern Alps, the detected changes concern nival floods and low waters. The melting wave occurs earlier and the melting peak decreases. Low waters are less severe, with a significant increase of the annual daily minima and a significant decrease of the volume deficit. The duration of low waters also seems to decrease, although not significantly.

In the Southern Alps, the severity of low water events also decreases; on the other hand the nival melting wave does not show coherent changes for the five tested stations. Three glacial regime stations show a decrease of the basic discharge during the melting peak, and a rather clear increase of the nival season total module. This last result is coherent with the glacier retreat in the French Alps (Vincent, 2002).

Additional tests from the MORDOR simulation model allowed bringing into light a link between the temperature increase and a seasonality modification of stream flows related to melting.

A statistical analysis of 200 long hydrometric series has been carried out within the framework of a PNRH national project and the B. Renard's doctoral thesis (2006). 13 hydrometric stations were studied for the French Alps sector.

Lang 2007 - C1

Europe:
Variability of atmospheric circulation is thought to be the most important factor causing annual and decadal variability of fresh water fluxes from the continents. The analysis of long-term links between atmospheric forcing and winter precipitation and sensitivities of annual mean and maximum winter discharges observed at regional scale across Europe shows significant correlation between atmospheric circulation variability, described by the NAO, AO, FWC and SLPD indices, and peak and mean river discharges. West circulation over Europe is causing increased discharges in most basins. River basins have different sensitivities to atmospheric circulation variability across Europe, depending on position on the continent, local climate, vegetation and other basin conditions. The results show that annual maximum discharges are more sensitive to variability of atmospheric circulation than mean discharges. Mean discharges vary on average between 8 and 44%, while peak discharges vary between 10 and 54% per unit index change. Discharges in Iberia and Scandinavia are more sensitive than those in central and northwest Europe. Discharge closely follows variability of atmospheric circulation. Compared with FWC and SLPD, the NAO and AO indices have only limited use for analyzing climate impacts in river basins in northwest Europe.

This study presents an analysis of long-term (>30 years) links between atmospheric forcing and winter (December–February) precipitation, and sensitivities of annual mean and maximum winter discharges observed at up to 608 stations across Europe. Links to four atmospheric indices are examined: the North Atlantic Oscillation (NAO) index, the Arctic Oscillation (AO) index, the frequency of west circulation (FWC) as described by the subjective Großwetterlagen classification, and the north to south sea level pressure difference across the European continent (SLPD). The sensitivity of mean and peak river discharges to atmospheric forcing was estimated using a simple linear least squares regression function.

Bouwer & al. 2008 - P

Slovenia:
A direct comparison with the 1961-1990 reference period water balance shows that the quantities of precipitation were almost the same in the 1971-2000 period, but that the quantity of evaporation has increased and runoff has decreased. It shows a significant change in the evaporation and runoffs. Evaporation is higher by 11%, and the runoff is lower by 6%.

The authors analyzed and reconciled the water balance elements of precipitation, evaporation and runoff for the selected areas in Slovenia for the 1971–2000 period. The runoffs derived from the discharges at water gauging stations were compared with the runoffs calculated using the water balance equation. The correctness of the relationships between the elements of the water balance was reviewed with the help of balance error analysis.

Frantar & al. 2008 - P

World:
Atmospheric warming and enhanced melting of glaciers is already resulting in changes in the glacial contribution to run-off in mountain basins around the world. The enhanced melting of glaciers leads at first to increased run-off and discharge peaks and an increased melt season, while in the longer time frame glacier wasting can be so severe that it results in decreased run-off. Glacier basins with a decreasing run-off trend have been observed in south-central British Columbia, at low elevations in the Swiss Alps and in the central Andes of Chile, which is probably a combined effect of reduced melt from seasonal snow cover as the snow line rises, and relevant glacier area losses. In contrast, significant run-off increases are reported in Alberta, north-western British Columbia and Yukon in Canada, in highly glacierized basins in the Swiss and Austrian Alps, the Tianshan Mountains and Tibet in central Asia and in the tropical Andes of Peru. The run-off increase within these basins is closely related to observed temperature rise, indicating that there is an unequivocal signal of enhanced glacier melting under the present warming trends. In future warming scenarios, glacier run-off should start to decrease even in high-altitude basins, affecting water availability.

Alps (Switzerland and Austria):
The available run-off records for the Alps in Switzerland and Austria show a run-off increase in recent decades in highly glacierized basins due to warmer temperatures that result in snow line rise, glacier wastage and enhanced melt. Basins with intermediate glacierization already show a declining run-off trend, indicating that areal reduction of glaciers prevails over enhanced melt due to thinning. Finally, basins with low glacierization also show a declining run-off trend that is driven both by the area reduction of glaciers and by the precipitation decrease in recent decades.

A collection of fifteen published data series and one unpublished data series from glacierized basins are reviewed. Data include basins from Canada, the Alps (Switzerland and Austria), central Asia (Tianshan Mountains and central Tibet, China) and the Andes of Peru and Chile.

Casassa & al. 2009 - A

Germany:
The compilation of the trend analyses for German rivers shows that there is no unambiguous pattern of flood trends across Germany.
In this study, significant flood trends (at the 10% significance level) were detected for a considerable fraction of basins. In most cases, these trends are upward; decreasing flood trends are rarely found and are not field-significant. Marked differences emerge when looking at the spatial and seasonal patterns. Basins with significant trends are spatially clustered. Changes in flood behavior in northeast Germany are small. Most changes are detected for sites in the west, south and center of Germany. Further, the seasonal analysis reveals larger changes for winter compared to summer. Both, the spatial and seasonal coherence of the results and the missing relation between significant changes and basin area, suggest that the observed changes in flood behavior are climate-driven.

In this study, flood time series are derived and analyzed for trends for 145 discharge gauges in Germany. A common time period of 52 years (1951– 2002) is used. In order to obtain a country-wide picture, the gauges are rather homogeneously distributed across Germany. Eight flood indicators are studied, which are drawn from annual maximum series and peak over threshold series.

Petrow & Merz 2009 - A

Alpine countries:
Many studies pointed to the sensitivity of the alpine hydrological system to climate change (Weingartner et al. 2003; Jasper et al. 2004; Birsan et al. 2005; Horton et al. 2006). A comprehensive trend analysis of the runoff regimes of 48 undisturbed catchment areas in Switzerland showed that the annual runoff increased in the last century (Birsan et al. 2005). This was mainly due to a higher discharge in winter (mainly high flows), but also increased runoff for autumn and spring (moderate and low flows) since 1960. In another study, Santschi (2003) analyzed specifically the winter discharge in 41 river basins with respect to gravel bed movements and destruction of the upper layer of the riverbed. For alpine rivers and rivers south of the Alps, a critical increase in winter discharge was not evident with the available data (Santschi 2003). In the Swiss Plateau and the Jura, however, Santschi (2003) found an increase of high flows in winter for 35% of the rivers after 1960. These trends of increased winter discharge are suspected to lead to downstream sediment transport and scouring of riverbed and therefore are assumed to disturb fish reproduction.

Long-term trends in sediment yields have been mainly evaluated for larger Alpine rivers. The sediment yield in the Danube catchment has increased by 30–50% in the period 1950–1980 and to some degree in the river Lech after 1965 (Summer et al. 1994; Walling 1997). This trend is assumed to be caused by changed agricultural management and changed precipitation patterns, respectively. Decreasing trends were reported in areas of reforestation in the French Alps (Descroix and Gautier 2002; Piégay et al. 2004; Liébault et al. 2005) or beneath hydropower plants (Habersack 1996; Weiss 1996; Walling 1997). Increasing trends in suspended sediment concentrations for peak flows have also been documented in some rivers (e.g. in the Danube by a factor up to 2; Summer et al. 1994), whereas in many cases no clear trend could be found (Zobrist et al. 2004).

Seasonal changes in river discharge due to changed precipitation patterns and earlier snowmelt may have changed the timing of the sediment transport downstream.

Bibliographic review (this article focuses on potentially increased levels of fine sediments going to rivers and their effects on gravel-spawning brown trout).

 

Scheurer & al. 2009 - A
Modeling
Europe:
A major question is the extent to which future flooding may be exacerbated by climate change. Arnell (1999) has modeled potential changes in hydrological regimes for mainland Europe, using four different GCM-based climate scenarios. While differences exist between the four scenarios, each indicates a general reduction in annual runoff in Europe south of around 50°N and an increase polewards of that. The decreases could be as great as 50% and the increases up to 25%.

Bibliographic review of studies using GCM.

Arnell 1999 in Goudie 2006 - A
Europe:
Calculations on a European scale indicate that under most climate change scenarios, northern Europe would see an increase in annual average stream flow, but southern Europe would experience a reduction in stream flow. In much of mid-latitude Europe, annual runoff would decrease or increase by about 10% by the 2050s, but the change resulting from climate may be smaller than “natural” multidecadal variation in runoff.
  IPCC 2001 - R
Rhine River:
Climate-change simulations project a warming in the Rhine River basin of 1.0-2.4°C over present values by the middle of the century (IPCC 2001). Hydrological simulations suggest that this warming will shift the Rhine river basin from a combined rainfall and snowmelt regime to a more rainfall-dominated regime, resulting in an increase in winter discharge, a decrease in summer discharge, increases in the frequency and height of peak flows, and a longer and more frequent periods of low flow during the summer.
  Middelkoop 2001 in Barnett & al. 2005 - A
World / extratropical basins larger than 200,000 km²:
Under idealized CO2 quadrupling, relative change in annual mean discharge range from -12% to +76%, with a median value of +30%. Relative changes in the 100-yr monthly maximum discharge generally are smaller and less variable, with a median of +15%.

Danube basin
(station: Orsova, Romania), after quadrupling of atmospheric CO2 :
- Relative change in annual mean discharge (dq) : -3%.
- Relative change in 100-yr annual maximum monthly discharge (dQ) : +16%.
- Annual probability of 100-yr flood (P, defined with respect to the control experiment) : 4.6%
  Milly & al. 2002 - A
French Alps:
The ratio between future and actual mean annual discharges presents a strong northward gradient, identical to the precipitation anomaly. In the North, the discharge increase lies between 10 and 50%. In the central part of the basin and over the Alps, the ratio is close to 1. In the south, the stream flow diminishes by 20 to 40%.

In the northern part, the discharges are stronger in winter (because the liquid precipitation increases) and spring. They reach values comparable to actual ones at the end of spring and during summer and are moderately lower during autumn. In the southern part, the low altitude rivers are influenced by a general reduction of the discharge during the whole year, and especially during autumn and spring.

Precipitation, evaporation and runoff increase (respectively of 13, 10 and 16%), but this average evolution is much contrasted between north and south, and depending from seasons. Runoff increases in the north (+50 to +150 kg m-2 per year) and decreases in the south (-50 to -150 kg m-2 per year).

Evaporation anomaly is more homogeneous and less strong (-50 to +50 kg m-2 per year). The higher increase in evapotranspiration occurs in the Alpine catchments, where it reaches 20% (reduction of snow cover duration due to higher air temperature and reduction of snow fall ranging between -18 and -36%).

The snow pack is the most sensitive hydrological component to the atmospheric warming.

The ratio of solid precipitation over total precipitation decreases significantly over the Rhone basin (-21%), particularly for low and medium altitude watersheds, and the duration of the snow cover reduces. The snow pack is strongly reduced in the southern part of the Alps, the maximum snow water equivalent (SWE) and snow cover duration diminish by 50%. In the northern Alps, the SWE reduction is weaker (-20%). The snow pack anomalies strongly depend on the elevation and are significant below 1000 m. The mean snow depth diminishes by 25 cm under 2000 m (-31%) but only by 16 cm at 3000 m (-11%). The contrast is particularly strong between 1500 and 2400 m. The reduction of the snow cover lifetime is 50 days (-26%) under 2000 m versus 40 days (-3%) at 3000 m.

For the Durance at La Clapière (2150 m), the snow pack appears a month later and disappears a month earlier. The SWE reaches a maximum value reduced by 30% and begins to decrease in March instead of April. The winter discharges are not modified, but the melt flash floods differ completely, occurring a month earlier and their amplitude diminishes by 30%. The summer discharges are lower (stronger evaporation). The same evolution affects the Buech river (in the southern Alps) and the Isere, but for the last, the amplitude of flash floods is unchanged (increase of spring rainfall).
The Rhone basin has been modeled at a resolution of 8 m using CIM (CROCUS-ISBA-MODCOU), a coupled system including a Soil-Vegetation-Atmosphere transfer scheme (SVAT) and a macroscale hydrological model. This tool has been validated from 1981 to 1994 by comparing the daily river flows simulated by the models with measurements from 145 gauging stations (daily discharge and alpine snow depth observations). Comparisons have shown that the mean annual discharge is reproduced with an error lower than 5% and the efficiency is higher than 0.8 for the largest subcatchments. The climatic results are obtained from ARPEGE (period 2054-2064) with a 2*CO2 scenario.
Etchevers & al. 2002 - A
Isère:
The Isère river catchments area would experience a decrease in maximum snow accumulation of between 30 and 50% with a quasi-complete disappearance during the summer season. The winter flow would increase with the precipitation increase and the melting peak shift of a month. The spring flows magnitude would be reduced in most cases. At the year scale, the scenarios are showing a slight increase or decrease.

Durance:
The impacts of climate change on snow cover in the Haute Durance river catchments are similar, with a lesser reduction due to the lower altitude. The annual flow would decrease in a faster way than in the North (from 0 to 20%). The flows would increase significantly in autumn and winter and would decrease during summer. The spring flows would occur a month earlier, their magnitude should remain constant or would decrease (from 10 to 25%), depending on the scenarios.

In both cases, snow cover will tend to decrease and the nival characteristic of the river catchments tend to soften.
This method uses the SAFRAN tools for the meteorological parameter analysis, the hydrological model MODCOU integrating the surface scheme ISBA and a simplified snow module inspired by CROCUS. These tools were validated on the Rhone river watershed, before being used in a climatic scenario (GICC-Rhône project). Six different “2*CO2” scenarios were used to estimate the impact of climate change on nival hydrology of certain watershed with nival regime. Four models were used among which two supplied scenarios with low (LR) and high (HR) resolution.
Etchevers & Martin 2002 - P
Europe:
A major question is the extent to which future flooding may be exacerbated by climate change. The proposed decrease in annual runoff in southern Europe is confirmed by Menzell and Burger's (2002) work in Germany, for they also suggest that peak flows will be very substantially reduced. A model by Eckhardt and Ulbrich (2003) for the Rhenish Massif in Germany found that in summer mean monthly groundwater recharge and stream flow may be reduced by up to 50%.
Bibliographic review of studies using GCM... Menzell & Ulbrich 2003 in Goudie 2006 - A
Rhine:
Models suggest that Rhine's discharge will become markedly more seasonal by the end of the century, with mean discharge decreases of about 30% in summer [Shabalova et al., 2003]. The decrease in the summer discharge is related mainly to a predicted decrease in precipitation combined with increases in evapotranspiration. Increases in winter discharge will be caused by a combination of increased precipitation, reduced snow storage and increased early melt. Glacier melting in the Alps also contributes to the flow of the Rhine. Once these glaciers begin to disappear, this contribution will diminish sharply, and will eventually cease.
Bibliographic review of studies using GCM...

 

Shabalova & al 2003 in Goudie 2006 - A
Thur and upper Ticino river basins (Swiss Alps):
Winter (DJF) runoff increased on average by 14% (Thur basin) and 31% (Ticino basin), whereas summer (JJA) runoff reduced by 16% (Thur basin) and 33% (Ticino basin). The annual mean runoff was reduced by about 6% (Thur basin) and 10% (Ticino basin) relatively to the control run. Note that, depending on the selected scenario, the projection results differed strongly; for the annual mean runoff, the changes were between -17 and +7% for the Thur basin and between -22 and +4% for the Ticino basin. In general, significant changes in runoff were projected in spring (early summer), mainly due to clearly changed snow conditions. As a consequence, the projected runoff peaks occurred earlier and substantially reduced relative to the peaks in the control runs. Note, furthermore, that changes in runoff were largest for areas where runoff is most strongly controlled by snow dynamics.

When using the SDT_P15 climate scenarios, the uniform all season increase in P (+15%) almost compensates the potential T-induced decrease in ET/PET and SWC. The SDT_P15-driven projections showed positive runoff changes not only for winter but also for summer, independently of the choice of scenario.
Possible future changes in the natural water budget relative to the 1981-2000 (Thur) and 1991-2000 (Ticino) baselines were investigated by driving the distributed catchment model WaSiM-ETH with a set of 23 regional climate scenarios for monthly mean temperature (T) and precipitation (P). The scenarios referred to 2081-2100 and were constructed by applying a statistical-downscaling technique to outputs from 7 global climate models.
Jasper & al. 2004 - A

World:
Near-surface air-temperature predictions from existing global climate models that are forced with anthropogenic increases in atmospheric greenhouse gas concentrations imply a high degree of confidence that future changes to the seasonality in water supply will occur in snowmelt-dominated regions.

The projected changes in temperature strongly imply future changes of seasonal runoff patterns in snowmelt-dominated regions. In a warmer climate, snow will melt earlier in the year than it did before and in some places [North America] this has already happened. Taken together, these impacts mean less snow accumulation in the winter and an earlier peak runoff in the spring.

On a global scale, the largest changes in the hydrological cycle due to warming are predicted for the snow-dominated basins of mid- to higher latitudes, because adding or removing snow cover fundamentally changes the snow pack's ability to act as a reservoir for water storage. Studies in various regions of the globe indicate that the stream-flow regime in snowmelt-dominated river basins is most sensitive to wintertime increases in temperature.

  Barnett & al. 2005 - A
Rhone:
According to simulations of CO2 concentration doubling, the whole of the Rhone, Saône and  l’Ardèche regions would encounter a decrease in average and low water levels,  but high flow  rate would increase.  For the river Durance, with a nival regime, a decrease in the melting peak and a less thin winter are observed.  However these conclusions vary greatly between the different scenarios.

The flows are likely to decreases during the period May to November, but the evolution in winter alters between the various scenarios.  The decrease in spring flow, resulting from the decrease in the nival components, is the strongest conclusion to be drawn from the study.  As snow caps begin to melt earlier and snow fall reduces, the strong spring follows are generally reduced and they occur 1 month earlier.  Winter flows increase significantly (higher rainfall) whereas summer flows reduce to 50% (drying out of the ground).  These general tendencies are observed throughout the different scenarios to varying degrees.
Evaluation of global climate change with a “2*CO2” scenario and 4 models of atmospheric circulation (low and high resolution).

Hydrological modeling and determination of 6 scenarios combining observed climatic variables (data provided by 131 stations distributed in the watershed for the 1981-1997 period) with simulated anomalies for temperature and precipitation.

Application of climate forcing scenarios to 5 different hydrological models (Modcou, ISBA-Modcou, Marthe, CEQUEAU, Ecomag) covering the French part of the Rhone river watershed.
GICC Rhône 2005 - R
Europe:
The answer of the European continental hydrological cycle is caracterised by an earlier nival melting peak and an intensification of the seasonal cycle of surface humidity, because of the intensification of the seasonal precipitation cycle.

The evolution of the surface hydrological cycle under the influence of the climate change has also been analysed by Douville et al. (2002). They conclude that the earlier snow cover melting should lead to an earlier runoff spring peak in Europe.

Within the framework of the international Coupled-Model Intercomparison Project (CMIP), Covey et al. analysed the results of 18 ocean-atmosphere coupled models simulating the climatic impact of a 1% per year increase of the atmospheric concentration of CO2.
Planton & al 2005 - A
Rhone catchments (France):
A detailed study has been performed on the potential impacts of climate change on the runoff regimes of 11 small catchments having glacier surfaces comprised between 0 and 50%, at altitudes ranging between 1340 and 2940 m, under different hydrological regimes (Horton et al., 2005 ; Schaeffli, 2005). Predictions were developed for a scenario of +1°C (expected for 2020-2049) and two scenarios considering two increased green house gas emissions (period 2070-2099: + 2.4 to 2.8 °C and +3.0 to 3.6°C, with rates higher in summer than for the average). The conclusion are the following for the +1°C scenario:

- A decrease of annual precipitations
- An increase of winter precipitations, with the risk of higher flood peaks
- A decrease of summer precipitations
- A strong decrease of ice-covered areas, due to the strong increase of summer temperatures. The regimes will be mainly driven by snow-melt during the Late 21 st century.
- A decrease in the amplitudes of discharge 
- A significant decrease of annual discharge (5-15% for the +1°C scenario) due to the reduction of precipitation, the increase of evapo-transpiration, the long term decrease of glacier surface and discharge.


Horton et al. (2005) predicted “a significant decrease of the total annual discharge and a shift in the monthly maximum discharge to earlier periods of the year due to the temperature increase and the resulting impacts on the snow melt processes”. At lower altitudes, “the influence of precipitations is more pronounced and the variability of the predicted climate change impact is mainly due to the large range of predicted regional precipitation change”.
  Horton et al. 2005 ; Schaeffli 2005 in Bravard 2006 - P
Rhone catchments (France):
- In the Doubs basin, the snow-rain regime shifts to rain regime with an increase of discharge in December and January, and a decrease in spring, without a significant change of the total yearly discharge.

- In the Saône basin (Mâcon), the rain regime remains the same, but discharges decrease in summer.

- In the Isère basin (Northern Alps), the maximum shifts from April to March, the winter maximum increases, and the summer minimum decreases by 50%.

- In the Southern Alps, during contemporary dry years, the Durance basin experiences a “precocious and excessively rapid snow melt… resulting in an early peak and correspondingly very weak summertime flows”. The simulated change forecasts “an annual reduction of river discharge and of the soil moisture, decreasing by as much as  30% below the present values”.

- However, if “the reduction of snowfall and earlier snow melting (increased air temperature) induced a decrease of the average snow depth by 50% and of the snow duration by more than one month”, snow pack at high altitude is less affected because even with the air warming, the average air temperature would remain below 0°C.

- The Ardèche river basin experienced a “significant reduction in summer flow” and a strong reduction of the soil water content, … “reflecting the heavy reduction of precipitation in that area” The GICC-Rhone programme extended these conclusions drawn from sub-watersheds to the larger area of the French part of the Rhone basin (Leblois et al., 2005):

- Average yearly discharge and low flows decrease (from May to November), but high discharges increase. Low flows may be reduced by 40-50% close to the outlet of the Rhone.

- Spring flow related to snow-melt decreases since the warming of the climate reduces snow depth and the duration of snow cover, and snow melt occurs one month earlier. 

- The behaviour of rivers in the winter depends on the different scenarios, but generally the increase of winter rainfall induces an increase of winter discharges.
The coupled ISBA-MODCOU model was used in three sub-watersheds and on the entire Rhone basin for a selected warm year, then tested for the prediction of change (Noilhan et al., 2000; Etchevers et al., 2001; Etchevers & Martin, 2002; Leblois, 2002; Leblois & Grésillon, 2005)

 

Bravard 2006 - P
Swiss Alps:
Runoff
changes
The simulated annual discharges show a significant decrease compared with the control period, for all catchments and for all climate experiments. The median decrease ranges between -26% (Drance de Bagnes) and -12% (Vorderrhein) under the B2 scenario, and between -30% (Drance de Bagnes) and -14% (Vorderrhein) under scenario A2. It should be noted that the simulated mean discharge decrease may be overestimated as a result of the highly simplified simulation of the glacier retreat.

The considerable decrease of the mean annual discharges has several reasons, the most evident being the decrease in mean annual precipitation. Another significant part of the discharge reduction is due to the predicted increase of evapotranspiration. This increase is expected to be less important in summers than in winters: even if there is a pronounced temperature increase in summer, summer evapotranspiration will be limited due to reduced precipitation.

The important variability of annual precipitation changes simulated by the different climate models and the significant variability of simulated changes in evapotranspiration explain the different magnitudes of mean annual runoff decrease. Note that if the maximum predicted discharge decrease is significantly higher for the A2 scenario than for the B2 scenario, then the median and minimum decreases obtained for both scenarios are quite similar. The variability highlighted is mainly due to inter-model differences.

Hydrological regime changes
All simulations under the future regional climate scenarios show the same significant changes of the hydrological regimes and the same trends of these changes. Summer discharge is found to reduce significantly, winter discharge increases, and the snowmelt-induced peak is shifted to earlier in the year and decreases in most cases. Owing to the substantial reduction of the glacierized areas, pure glacial discharge regimes tend to disappear. The changes are very similar for catchments having the same present regime, independently of their geographical situation. The changes predicted for the B2 scenario tend to be enhanced for A2. The median simulated shift of the maximum monthly or semi-monthly discharge is, for example, around half a month for B2 and an entire month for A2. The results are, however, highly variable for the different RCM experiments. For many catchments the main difference compared with the present regime types was the appearance of a small secondary peak in autumn that does not exist in present intra-alpine regimes, but only in south-alpine regimes.

It should be noted that the shift of a glacier- or snowmelt-driven hydrological regime to a future regime more driven by snowmelt and/or rainfall will be associated with a modification of the year-to-year variability of the discharges. For high-elevation catchments (catchments with current proportions of glacier cover higher than 15%), the variation coefficient of annual discharges is, for example, predicted to increase by around 85% for scenario A2. The interannual variability of seasonal discharges also changes significantly due to an important modification of the temporary water storage properties of these mountainous catchments.

For catchments with high elevations, the regime modifications are highly conditioned by the changes of the seasonal temperatures. The simulated maximum monthly discharge is, for example, highly correlated to the change in mean spring temperature. For catchments at lower altitudes, the influence of precipitation modification is more pronounced and the variability of the climate-change impacts is mainly due to the large range of predicted regional precipitation changes.
The 11 catchments analysed (Drance de Bagnes, Saaser Vispa, Lonza, Rhône at Gletsch, Weisse Lütschine, Minster, Tamina, Vorderrhein, Dischmabach, Rosegbach and Verzasca) were selected to represent the seven different hydro-climatic zones of the Swiss Alps and the different mountainous hydrological regimes occurring in these zones.

The simulation of catchment behaviour for the different time periods was carried out using a hydrological model, for the precipitation-runoff transformation, and a land-cover evolution model. The only land-cover change that is explicitly accounted for is the modification of the glacier-covered surface.

The glacio-hydrological model needs three input time-series, namely daily precipitation, daily potential evapotranspiration PET and mean daily temperature. For all catchments, the precipitation and temperature data are obtained from nearby meteorological stations. The hydrological discharge simulation is carried out at a daily time-step through a conceptual reservoir-based model called GSM-SOCONT.

The model has seven parameters to calibrate (four for catchments without a glacier): two degree-day factors for the ice and snowmelt computation, four time constants of the reservoirs, and the maximum storage capacity of the slow reservoir. Calibration and validation were undertaken for two consecutive 10-year periods (1971-1980 and 1981-1990 respectively, for the majority of the catchments). The calibrated model performs well for all catchments for both the calibration and the validation period.

Each RCM experiment consists of a simulation for the period 1961-1990 (control run) and a simulation for the period 2070-2099 (future run). For each RCM experiment, the boundary conditions were obtained from one of the three AOGCMs used in PRUDENCE. The AOGCM experiments were completed for the two SRES emission scenarios A2 and B2. In this study, nine RCMs are included. Outputs from 19 RCM experiments are available in total: 12 for the A2 scenario and seven for B2.
Horton & al. 2006 - A
France:
The hydrological models show that the observed discharge evolution (increase of annual maxima and intensification of floods) seems coherent with the precipitation behavior for the Northeast region, the rain-discharge transformation being apparently not responsible for this change. For the Alps, a link between the temperature increase and the seasonality modification of stream flows related to melting was highlighted.
Comparison of the observed evolution of discharges with that obtained from hydrological model: for the continental region, the pluvial floods evolution was obtained from the GR4J model (Perrin, 2000) and for the Alps, the nival floods evolution was obtained from the MORDOR model (Paquet, 2004).
Lang & Renard 2007 - P

Alps:
Remarkably increases in runoff and discharge volumes were computed for the Lavanttal region (Austria) when considering possible effects of climate changes (Regional Office of Carinthia, Department of Water Economy 2008).

An indirect effect of the increase of mean temperature is the rising altitude level for the limit between rainfall and snowfall. In areas of the Northern Alps below 1500 m a.s.l., an increase of flood peaks is expected in winter due to higher soil water contents, the rising of the rainfall/snowfall limit level and due to an increased liquid precipitation (KOHS, 2007).

Autonomous Province of Bolzano – South Tyrol (Italian Alps):
A high spatial variability of the sensitivity of the torrent catchments to specific impacts of climatic changes has been highlighted:
- The runoff of nearly all torrent catchments is expected to increase in summer;
- The runoff in winter is expected to increase only in torrent catchments having a high percentage of their total surface area below 2000 m;
- The bed load transport in summer is expected to increase in high mountain areas and is expected to decrease in catchments at submontane levels;
- The bed load transport in winter increases in a few mountain torrent catchments and does not change in the most catchments;
- In some catchments eroding younger deposits (weathered material), a decrease in extreme events is highlighted;
- In some catchments eroding older deposits an increase in extreme events is pointed out;
- The frequency of small scale debris flow and sediment transport processes is expected to increase in most of the torrent catchments.

Bibliographic review for Alps.

B
asic assumptions for potential future climate conditions:
– The daily mean temperatures in summer and winter are increasing;
– The mean sum of precipitation in summer is decreasing or remains constant;
– The mean sum of precipitation in winter is increasing;
– The intensity and frequency of short extreme rainfall events in summer and autumn is increasing.

It was assumed that the following factors are varying spatially and are relevant for the sensitivity of the torrent catchments to climate changes:
– Percentage of areas located between 1000 and 2000 m a.s.l. A threshold value of 50% of these areas respective to the total catchment area was chosen.
– Characteristics of bed load source areas: If the percentage of areas with older deposits to the total bed load source area exceeds 30% of the total catchment area, the torrent catchments were classified as torrents eroding older deposits.
– Available bed load source areas: The sensitivity of torrent activity against climate changes increases with a higher proportion of bed load source areas respective to the total catchment area.
– Permafrost degradation: Catchments were classified as sensitive, if more than 30% of the total catchment area is subjected to permafrost degradation.
– Areas with elevated surface runoff: The sensitivity of torrential rivers and rivers to climate changes increases with a higher proportion of areas with reduced water storage capacities respective to the total catchments area.

Staffler & al. 2008 - A

Dischma and Inn catchments (eastern Swiss Alps):
The major change that is visible when comparing the reference simulation discharge and the climate change scenarios discharges is a strong peak in early May in the climate change scenarios not visible in the reference. This is the beginning of snow melt in the catchment, which occurs a month earlier in both scenarios than in the reference case. There are two main observations to be made: (i) the snow melt discharge peak is not only earlier but also more pronounced for the Dischma catchment; (ii) the higher snow melt peak is more concentrated in late spring/early summer, while it is spread over a longer period under the current climate. This means that snow contributes to the discharge over a much shorter time, but at the same time, the probability of spring flooding might increase. The two scenarios are qualitatively similar, yet showing differences in the magnitude of snow melt discharge. The two scenarios merge again in June, indicating the time at which the influence of snow melt on discharge becomes negligible in the catchment for both scenarios. Over the summer, discharge is then determined by rainfall. In contrast, the reference run shows snow influence on discharge until early August—discharge from snow, therefore, being significant for a large part of the year.

For the Inn catchment, the spring discharge peak is not as pronounced as for the Dischma. The Inn catchment being much dryer than the Dischma catchment, particularly in winter, the spring discharge depends on snow built up during late fall. Since in the A2 scenario, these precipitation events occur as liquid precipitation for a much larger fraction of the altitudes, there is less snow available to melt in spring. Therefore, despite an earlier and shorter melt period for the Inn catchment, there is no increase in the amplitude of the spring discharge. Moreover, the lowest elevations are not able to build up a snow pack anymore, and thus, cannot contribute to the discharge in the spring. This is in contrast to the Dischma simulation which contains a much narrower range of elevations, hence the fraction of snow-free altitude bands is insignificant. Finally, the more consistent distribution of exposure in the Inn simulation means that shading effects do not have a strong impact on the discharge (a given altitude band melts more uniformly) compared to the Dischma catchment.

Another feature of the data is local discharge peaks in late autumn: in the reference simulation for both catchments, the precipitation events at that time of the year produce snow down to 1600 m a.s.l., not leading to a significant discharge increase. In the climate change scenario, no snow is seen below 2200 m a.s.l., and significant discharge peaks are created. Again, since the Inn catchment contains a significant proportion of low elevations, this effect is amplified and might increase the risk of flooding.

In this paper, the authors presented model simulations of climate change impact (scenarios A2 and B2) on snow and runoff in two Alpine catchments, Dischma and Inn (43 km² and 1945 km², respectively).

They used the model Alpine3D, which provides a detailed process representation especially for snow and appears, therefore, particularly suited for change studies. The model is based on a fully distributed application of SNOWPACK, which is an advanced model of snow cover development. The single point snow column description provided by SNOWPACK includes a parameterized module for vegetation. Snow and soil dynamics are numerically represented by a large arbitrary number of layers. The high resolution of the surface snow or soil layers allows a more accurate surface energy balance to be provided as well as allowing the vertical water transport in snow and soil to be described with a simple bucket scheme.

To generate model input data representative of future climate, observed data were altered according to expected changes as predicted by a set of RCMs (Regional Climate Models) which participated in the PRUDENCE project (Prediction of Regional scenarios and Uncertainties for Defining European Climate Change Risks and Effects). A sub-set of 12 runs was selected to provide simulations for the changes in temperature, precipitation and longwave radiation between a reference period, 1961–1990, and a future period, 2071–2100, for the two IPCC greenhouse gas emission standard scenarios, SRES A2 and B2.

In order to explore the reliability of the method used in this study, a reference Alpine3D run was defined using unmodified meteorological input. The outputs of this reference run were compared to measurements in order to check the accuracy of the simulation (the results are considered satisfactory).

Bavay & al. 2009 - A

Swiss Alps:
A 20-glaciers sample including different glacier geometries, sizes and exposures allowed to investigate glacier response to climate change in South-eastern Swiss Alps (Engadine, Val Poschiavo, Val Bregaglia). Using a combined model for 3D glacier evolution and stream-flow runoff driven by regional climate scenarios, perspectives for the 21st century were generated. Glaciers in the study area are expected to retreat significantly over the next decades. Most of them will disappear in the second half of the 21st century. The hydrological cycle will be strongly affected by glacier retreat. A decrease in glacier area of 63% until 2050 and an increase in annual discharge over the next three decades were determined for catchments with high glacierization. By 2100, the model results indicated a shift in the hydrological regime and a 23% decrease in annual runoff attributed to increased evapotranspiration and strongly reduced glacier melt contribution.

Using a distributed accumulation and temperature-index melt model, mass balance time series were derived in seasonal resolution from 1900 to study the response of a 20-glacier sample to current climate warming. The model was calibrated using ice volume changes obtained from differentiating digital elevation models. In-situ point measurements of annual mass balance and winter accumulation were available for some glaciers, and long-term discharge records were used for model validation. A glacio-hydrological model is run into the future based on regional climate scenarios, providing estimates of the impact of climate change on glacier extent and the hydrological cycle in the south-eastern Swiss Alps.

Huss & al. 2010b - A
Hypothesis

Swiss Alps and pre-Alps:
The effects of global warming on the hydrological balance of the catchment area of a torrent may be summerized as fallow:
with precipitation tending to fall increasingly as rain rather than snow, increased discharge during the winter half of the year at altitudes between 1000 and 1500 m are likely. Thus pre-Alpine regions in particular are affected;
significant effects on the water cycle are observed in spring (March to April) owing to the reduced volume of melt water;
summer and autumn are still critical times, characterized by storms and concentrated heavy rainfall. This storm activity may be expected to bring an increase in peak flow rates, transportation of solids and mudflow events;
evapotranspiration will increase greatly (up to 30%) with higher air temperatures. A lack of snow in spring will also encourage premature evapotranspiration.

  Bader & Kunz 2000f - R: PNR31

Europe:
Glacier retreat has implications for downstream river flows. In rivers fed by glaciers, summer flows are supported by glacier melt (with the glacier contribution depending on the size of the glacier relative to basin area, as well as the rate of annual melt). If the glacier is in equilibrium, the amount of precipitation stored in winter is matched by melt during summer.

However, as the glacier melts as a result of global warming, flows would be expected to increase during summer—as water is released from long-term storage—which may compensate for a reduction in precipitation. As the glacier gets smaller and the volume of melt reduces, summer flows will no longer be supported and will decline to below present levels.

The duration of the period of increased flows will depend on glacier size and the rate at which the glacier melts; the smaller the glacier, the shorter lived the increase in flows and the sooner the onset of the reduction in summer flows.

  IPCC 2001 - R

World (cold regions):
Following a global analysis of future hydrological trends, Nijsssen et al. (2001) suggested that the largest changes in the hydrological cycle are predicted for the snow-dominated basins of mid to higher latitudes, and in particular marked changes are likely in the amplitude and phase of the annual water cycle (Arora and Boer, 2001).

Bibliographic review of studies using GCM. Nijsssen & al 2001 ; Arora & Boer 2001 in Goudie 2006 - A

Alps:
With the warming of climate, “minimal and maximal discharges will be observed more frequently than in present times during other periods of the year than it is presently expected”. In others words, prediction will be more difficult and the authors recommend the adoption of a probabilistic approach (Krasovskaia et al., 2002). However specialists consider that discharge regimes have not changed enough to justify any change in the policy of dam management (D. Duband, oral comm.).

  Krasovskaia et al., 2002 in Bravard 2006 - P

World:
The consequences for river runoff (of precipitation changes) are likely to affect not only the watersheds within the mountains themselves, but also in the lowland regions that are heavily dependent on these mountain resources.


For every °C increase in temperature, the snowline will rise by about 150 m. Shifts in snow-pack duration and amount as a consequence of sustained changes in climate will be crucial factors in water availability for hydrological basins.

  Beniston 2003 - R

Europe:
Even as summers become drier, the incidence of severe precipitation [on summertime flooding] could increase.

  Christensen & Christensen 2003 - A

Rhine River :
Climate-change simulations project a warming in the Rhine River basin of 1.0-2.4°C over present values by the middle of the century (IPCC, 2001). Hydrological simulations suggest that this warming will shift the Rhine river basin from a combined rainfall and snowmelt regime to a more rainfall-dominated regime, resulting in an increase in winter discharge, a decrease in summer discharge, increases in the frequency and height of peak flows, and a longer and more frequent periods of low flow during the summer.

  Barnett & al. 2005 - A

Alps:
The warm winter spells may have impacts, especially if the temperatures exceed the freezing point, inducing a melting of snow enhanced early runoff into river basins that originates in the mountains, avalanche risk for the exposed slopes.

  Beniston 2005a - A

World:
For rivers fed largely by ice melt, reduction in glacier volumes will have a particularly strong impact on dry-season river flows, and on the provision of downstream water.

  Kääb & al. 2005 - A

Alps:
The specific annual discharge of mountain rivers is higher than the specific discharge of extended watersheds including lowland areas. This results from higher precipitations, low evaporation rates, and by conditions favouring runoff. “The hydrological regime is strongly influenced by water accumulation in the form of snow and ice and the corresponding melting processes resulting in a pronounced annual cycle of the discharge. A modification of the prevalent climate and especially of the temperature can therefore considerably affect the hydrological regime and induce important impacts on the water management” (Horton et al., 2005). The recent increase in temperatures has probably already had consequences on river regimes.

Swiss Alps:
In Switzerland, “shifts in snow-pack duration and amount will be crucial factors in water availability » for runoff according to Beniston et al. (2003). The increase in winter temperatures will have clear consequences on the beginning of snowmelt and on the reduction of flow during the spring at low altitudes and on summer flow at the highest altitudes. The rarefaction of snow cover below 1000 m will reduce runoff. These shifts will affect river regimes with higher winter discharges. However, increased evaporation in winter may partly reduce runoff and river discharge.

Climate warming will increase the average discharge of rivers flowing from glaciers at first during the period of retreat, but then will decrease summer discharge, as rivers will progressively lose their glacial-type hydrological regime.

  Horton 2005; Beniston et al. 2003 in Bravard 2006 - P

 

Swiss Alps:
Paradoxically, the impacts associated with future extreme rainfall may be reduced because of the buffering effects of snowfall on rapid runoff. This is because future springs and falls will be colder than today’s summers with lower freezing levels.

  Stoffel & Beniston 2006 - A

Alps:
A temperature increase might have strong consequences on the water supply in alpine watersheds at a large-scale. The snow cover decrease, the generalised glacier retreat and the summer precipitation decrease can limit the buffer effect in terms of hydrological balance and thus the water availability of these European watersheds. Extreme rainfalls could increase in a warmer climate, but with a seasonal shift, so the snow would have a buffer effect and modulate the future flood risk.

  Beniston 2007 - C

World:
The critical stage associated with decrease of glacier run-off has already been attained in southern and central British Columbia, Alberta, and in low elevation/low glacierization basins in the Swiss Alps. In future warming scenarios glacier run-off decrease should become more predominant as glaciers experience severe wastage, affecting water availability of vast populations who rely on water resources that originate largely from glacier melt, particularly at the end of the summer and in dry years. More data and analyses are clearly needed to assess current run-off trends and relate them to glacier changes and to climate trends, which is essential for improving the forecast of future water availability from mountain regions.

A collection of fifteen published data series and one unpublished data series from glacierized basins are reviewed. Data include basins from Canada, the Alps (Switzerland and Austria), central Asia (Tianshan Mountains and central Tibet, China) and the Andes of Peru and Chile.

Casassa & al. 2009 - A

Alpine countries:
Seasonal changes in river discharge due to changed precipitation patterns and earlier snowmelt may have changed the timing of the sediment transport downstream. Sediment transport downstream alternating with sediment deposition on gravel beds can now be expected to occur also in winter and earlier in spring.

Bibliographic review (this article focuses on potentially increased levels of fine sediments going to rivers and their effects on gravel-spawning brown trout).

Scheurer & al. 2009 - A

[Top]

Legend of bibliographical references:
- A : Article (Peer reviewed publication)
- C : Comment
- E : Scientific study (unpublished)
- P : Proceedings
- R : Report
- Re : Experience Feedback
- T : Thesis
- W : Website