Embed Size (px)
Citation preview
Environ. Res. Lett. 11 (2016) 054022 doi:10.1088/1748-9326/11/5/054022
LETTER
From dwindling ice to headwater lakes: could dams replace glaciersin the European Alps?
Daniel Farinotti1,5, Alberto Pistocchi2,5 andMatthiasHuss3,4
1 Swiss Federal Institute for Forest, Snow and Landscape Research (WSL), Birmensdorf, Switzerland2 EuropeanCommission, DG Joint ResearchCentre (JRC), Institute for Environment and Sustainability, Ispra, Italy3 Department ofGeosciences, University of Fribourg, Fribourg, Switzerland4 Laboratory ofHydraulics, Hydrology andGlaciology (VAW), ETHZurich, Zurich, Switzerland5 Author towhomany correspondence should be addressed.
E-mail: [email protected] and [email protected]
Keywords: glaciermelt, dam storage, climate change,mitigation
Supplementarymaterial for this article is available online
AbstractThe potential exploitation of areas becoming ice-free in response to ongoing climate change has rarelybeen addressed, although it could be of interest from thewatermanagement perspective. Herewepresent an estimate for the potential ofmitigating projected changes in seasonal water availabilityfrommelting glaciers bymanaging runoff through reservoirs. For the EuropeanAlpswe estimate thatby the end of the century, such a strategy could offset up to 65%of the expected summer-runoffchanges frompresently glacierized surfaces. Afirst-order approach suggests that the retention volumepotentially available in the areas becoming deglacierized is in excess of the volume required forachieving themaximal possiblemitigation bymore than one order ofmagnitude. Obviously, however,such a strategy cannot compensate for the reduction in annual runoff caused by glacier ice depletion.Our estimates indicate that by 2070–2099, 0.73±0.67 km3 a−1 of this non-renewable component ofthewater cycle could bemissing inAlpinewater supplies.
1. Introduction
Environments significantly influenced by the presenceof seasonal snow or glaciers are hotspots regarding theimpacts on water availability in response to expectedclimate change (e.g. Beniston 2003, Viviroli et al 2011,Mankin et al 2015). With warmer temperatures, boththe duration and the spatial extent of the seasonalsnow cover are projected to decrease (e.g. Barnettet al 2005, Steger et al 2013, Brutel-Vuilmet et al 2013)and glaciers are expected to retreat significantly (e.g.Zemp et al 2006, Radić and Hock 2011, Cogleyet al 2012). This is anticipated to have influences onwater resources at different time scales, and to affectrunoff seasonality in particular (for reviews, see deJong et al 2005, Schaner et al 2012, Radić andHock 2014): while increased discharge is expectedduring the winter and spring time, a significantdecrease is projected for summer and fall (e.g. Bavayet al 2009, Huss 2011, Berghuijs et al 2014). Glacier
recession, moreover, is known to have an effect onlonger-term water availability, with an initial phase ofincreased water yields in response to increased meltrates, and a later stage of decreased runoff due to thereduction in actual glacier size (e.g. Hock et al 2005,Farinotti et al 2012, Bliss et al 2014).
Glacier recession has generally been connotednegatively, with concerns emerging from the impactsof reduced water yields in summer (Piao et al 2010,Immerzeel et al 2010), the impacts of a changed land-scape in touristic regions (Scott 2006, Scott et al 2007),or the hazards induced by the destabilization of for-merly frozen grounds (Carey 2005, Kääb et al 2005,Fischer et al 2006) and the associated potential formore frequent glacier lake outburst floods (Richard-son and Reynolds 2000, Dussaillant et al 2010).Water-availability related concerns are particularly promi-nent in arid regions, such as Central Asia or parts ofSouthern America, in which the dependence on snowand glacier melt for regional supply is most important
OPEN ACCESS
RECEIVED
22December 2015
REVISED
30March 2016
ACCEPTED FOR PUBLICATION
28April 2016
PUBLISHED
20May 2016
Original content from thisworkmay be used underthe terms of the CreativeCommonsAttribution 3.0licence.
Any further distribution ofthis workmustmaintainattribution to theauthor(s) and the title ofthework, journal citationandDOI.
© 2016 IOPPublishing Ltd
(Hagg and Braun 2005, Kaser et al 2010). Such regions,moreover, often coincide with economically lessdeveloped areas, thus lacking the potential capabilityof mitigating the impacts through engineering solu-tions (Orlove 2009).
The potential of exploiting newly deglacierizedareas has not often been addressed, although it wassuggested earlier that new proglacial lakes, likely toform in such environments, could take over importantroles in hydropower production or the managementof both water resources and natural hazards (Freyet al 2010, NELAK 2013). In support of such a possibleutilization, the supposed ‘neutrality’ in terms of ecolo-gical value of the surfaces becoming ice-free is some-times mentioned: Since the exposed forefields arebare, non-vegetated and relatively simple in their eco-system structure (e.g. Nemergut et al 2007, Hodsonet al 2015), a potential exploitation could result incomparatively limited ecological impacts. Similarly, itcould be argued that an intervention in the upper-most part of a given river stretch would have onlyminimal consequences on river continuity.
This opens the admittedly provocative question onwhether newly deglacierized areas could be used tomitigate, at least partly, the hydrological effects causedby the shrinkage of glaciers and the snow cover. Postu-lating that preserving the current natural runoffregime is a desirable target, the additional water expec-ted during spring time under changed climate condi-tions could in fact be temporarily stored andseasonally transferred in order to compensate for thereduction in summer water yields (figure 1). In such ascenario, reservoirs would replace part of the hydro-logical effect currently provided by glaciers and theseasonal snow pack.
In this study we focus on the European Alps, aregion expected to be significantly affected by futurechanges in snow- and glacier-fed water resources (e.g.Beniston et al 2011, Huss 2011), and use a first-orderapproach in order to assess whether ‘replacing glacierswith dams’ is an option theoretically worth
considering. For doing so, we use the output of a glo-bal glacier model (GloGEM) (Huss and Hock 2015) incombination with latest climate projections (Tayloret al 2012) to assess the evolution of both glacier extentand runoff until the end of the century. We then com-pute storage capacities required for achieving themax-imal possible mitigation, and compare these results toan estimate of retention volumes that could hypotheti-cally be installed in newly deglacierized forefields.While we don’t discuss the technical feasibility or anyof the actual ecological implications of such a hypothe-tical way of action, we provide a first assessment of themaximal effect that such a strategy could potentiallyachieve.
2.Methods
2.1. Runoff frompresently glacierized surfacesWe define glacier runoff Qgl as the sum of liquidprecipitation (rain) that does not refreeze within aglacier, and snow- and ice-melt that stems fromcurrently glacierized areas (glacier runoff as defined byconcept 1 in Radić and Hock (2014) neglectingevaporation). The individual components of Qgl aretaken from theGloGEMbyHuss andHock (2015) (seesection 2.2) and evaluated within individual catch-ments defined by present glacier extent. Presentlyglacierized areas are discerned with the glacier outlinesprovided by the Randolph Glacier Inventory (Pfefferand The Randolph Consortium 2014) version 4.0(RGIv4.0; Arendt et al 2014), which refers to the year2003 for Central Europe (Paul et al 2011). For allanalyses, the area contributing to Qgl is fixed to theglacier extent given by the RGI. This is to avoidsuggesting misleadingly large reductions in futurewater yields in case of shrinking catchment area due toglacier retreat.
For quantifying the relative importance of glaciercontributions to total runoff, we consider thepercental runoff share that stems from presently
Figure 1. Illustration of the potentialmitigation throughwatermanagement.Volume surplus (light blue) is the runoff volume that,according to the projection, will be in excess to the runoff in the reference period during late spring and early summer. This volumecould potentially be transferred into late summer and early autumn (arrow) in order to partially compensate (light green) the runoffreduction caused by glacier depletion.Volume deficit (light red) is the net reduction in annual runoff. The example refers toGlacier dela Girose, France.
2
Environ. Res. Lett. 11 (2016) 054022
glacierized surfaces, that is
= · ( )RS Q Q100 . 1gl gl tot
Here, Qtot is the total runoff at the consideredlocation, and is estimated through a weighted flowaccumulation operation (e.g. Pistocchi and Penning-ton 2006, Pistocchi 2014) of the climatologicalmonthly composite runoff fields provided by the Uni-versity of NewHampshire/Global Runoff Data Center(Fekete et al 2002). The analysis is performed on theGTOPO30 digital elevationmodel (DEM) provided bythe HYDRO1k data base (https://lta.cr.usgs.gov/HYDRO1K). The DEM has a resolution of 30 arc-sec-onds (about 1× 1 km), which is maintained in all ourresults.
In principle, RSgl is defined for any time aggrega-tion, but only summer (July, August, September) andannual values are considered in the following. Melt-water transit-time along the considered streams (in theorder of some days to a few weeks; Huss 2011) aretherefore neglected.
Derived values of both Qtot and RSgl, are validatedagainst independent estimates based on station mea-surements (supplementary figures S.1 and S.2). Thegeneral agreement is satisfactory although a tendencyto underestimate Qtot in relatively small catchments(i.e. catchments with small runoff) emerges. This cau-ses RSgl to be overestimated in such cases. However,since RSgl is only used for quantifying the relativeimportance of glacier runoff during the reference per-iod 1980–2009 (and thus is irrelevant for the con-siderations on potential mitigation and storagevolume), this has no effect on the presentedinterpretations.
2.2. Glacier and climate evolutionGloGEM (Huss and Hock 2015) computes glaciermass balance and associated glacier geometry changeson a glacier-by-glacier basis. Glacier outlines areobtained from the RGI v4.0 and are intersected withthe DEM provided by the shuttle radar topographymission (Jarvis et al 2008) in order to derive relevanttopographical information (glacier hypsometry,aspect, and slope in particular). For each glacier, theinformation is aggregated into 10 m elevation bands,which is the resolution at which the model operates.The initial ice thickness distribution for every glacier iscalculated with the method presented by Huss andFarinotti (2012), updated to the RGIv4.0 (Huss andHock 2015). The climatic glaciermass balance—that isthe balance of snow accumulation, snow- and ice-melt, and refreezing (Cogley et al 2011)—is calculatedat monthly resolution based on air temperature andprecipitation time series (Huss and Hock 2015). Thesetime series are taken from ERA-Interim reanalysis(Dee et al 2011) for the past (period 1979 to present)and from the 5th phase of the coupled modelintercomparison project (CMIP5; Taylor et al 2012)for the future (present to 2099). For CMIP5, results
from 14 different Global Circulation Models (GCMs)and 3 different representative concentration pathways(RCP 2.6, 4.5, and 8.5; Meinshausen et al 2011) areconsidered (supplementary figure S.3). The so-com-puted glaciermass balance is then used in combinationwith the parametric approach by Huss et al (2010) foradjusting glacier surface elevation and extent on ayearly basis.
The model is calibrated at the regional scale withthe consensus glacier mass-change estimates by Gard-ner et al (2013), and is validated against observations ofboth glacier mass balances (in situ and geodetic mea-surements; WGMS 2012) and glacier area changes(Fischer et al 2014). For Central Europe,model perfor-mance is high, with average deviations in the calcu-lated glacier-wide (point) mass balance of±0.04(± 0.16)m water equivalent a−1, and average area-change rates reproduced within±0.04% a−1
(see supplementary tables 5–7 in Huss andHock 2015).
For additional details on the glaciological model-ling, including details on the downscaling of the clima-tological data, and the calibration and validationprocedures, refer toHuss andHock (2015).
2.3. Computation ofmitigable volume changeWe compute the maximal potentially mitigablevolume change for a given scenario period as the sumof all positive volume differences with respect to themean runoff volume during the reference period1980–2009 (figure 1). All calculations are performedusing monthly values. We present results for a near-term (2010–2039), a medium-term (2040–2069), anda long-term horizon (2070–2099). The monthlydifferences are computed on a glacier-by-glacier basisfor simulations appertaining to the same GCMensemble member, and aggregated to the resolutionof the HYDRO1k DEM (1×1 km). The same DEM isused for deriving flow directions with which theaggregated values are accumulated downstream.Both the derivation of the flow directions and thedownstream flow accumulation follows Jenson andDomingue (1988).
Presented results refer to the mean of all availableGCMs, while individual RCPs are considered sepa-rately. We use the standard deviation between indivi-dual GCM results in order to quantify the uncertaintylinked to climate model choice, and the standarddeviation of the time series of a given GCM in orderto quantify internal year-to-year variability. Pre-sented uncertainties refer to the combination of botheffects (added in quadrature). The uncertainty intro-duced by GloGEM (see previous section) is assumedto be included in this range as typical year-to-yearvariations in glacier mass balance are larger than theassociated uncertainties by about one order ofmagnitude.
3
Environ. Res. Lett. 11 (2016) 054022
3. Results
Presently glacierized surfaces in the EuropeanAlps (ca.2050 km2 on 3800 glaciers) contributed an averagerunoff volume of 5.28±0.48 km3 a−1 during1980–2009 (table 1). About 75% of this volume(3.97± 0.36 km3 a−1) occurred between July andSeptember (summer hereafter). The spatial distribu-tion of the resulting runoff share is shown infigures 2(a) and (b). Regions with very importantglacier contributions are found across the wholeAlpine Ridge, with two main clusters around thecommon borders of France/Italy/Switzerland (MontBlanc Massif and Pennine Alps) and Austria/Italy/Switzerland (Bernina and Adamello–PresanellaRanges, as well asOrtler Alps andÖtztal Alps).
In terms of relative importance, the picture is simi-lar for both annual and summer values (figures 2(a)and (b) respectively), although the actual magnitudecan largely differ. For the Mont Blanc region, forexample, the annual runoff-share from glacierizedsurfaces is estimated to be around 20%, whereas thisshare increases to more than 60% during the summermonths.
Also for the major low-lying rivers, the runoff-share from presently glacierized surfaces can be sig-nificant. Summer (annual) values for the Rhône atChancy, the Inn at Innsbruck, the Po at Pontela-goscuro and the Rhine at Basel (see figures 2(a) and (b)for locations) are in the order of 53 (15), 38 (14), 11 (2)and 6% (2%), respectively, with uncertainties in therange of 25% of the estimated values. At its confluencewith the Danube at Ingling, the Inn is estimate to stillhave an annual (summer) glacier runoff share of 3%(10%). These values compare well with the results byHuss (2011), although the observed tendency of over-estimating the share of glacier runoff in smaller basinsseems to re-emerge (see section 2.1).
At the Alpine scale, the total runoff yield from pre-sently glacierized surfaces is expected to increase sig-nificantly in the short term (+1.05± 0.75 km3 a−1 by2010–2039 according to RCP 4.5) and to decrease inthe long term (−0.73± 0.67 km3 a−1 by 2070–2099,RCP 4.5). The general evolution is similar for otherRCPs although differences occur as a result of differingtemperature and precipitation projections (see table 1and supplementary figure S.3).
Without any intervention, the temporal evolutionof the summer runoff contributions (Qgl
JAS) is similarto the annual ones, although appreciable reductions inthe total water yield are expected in the mid-termalready (table 1). It is important to note, however, thatat the local scale such effects are visible even in theshort term: figures 2(a)–(c) presents projected changesin Qgl
JAS for RCP 4.5 (results for RCPs 2.6 and 8.5 aregiven in supplementary figures S.4 and S.5, respec-tively) and shows that significant reductions are expec-ted in the easternmost and westernmost parts of theAlps in particular (figure 2(a)). Such a decrease inwater availability becomes widespread in themid term(figure 2(b)) and is virtually ubiquitous for the longterm (figure 2(c)).
Reductions in QglJAS are substantially less impor-
tant when surplus volumes occurring in the formerpart of the year are virtually transferred to the summermonths through temporary storage in hypotheticalreservoirs (figures 2(d)–(f)). On the Alpine scale, theexpected decrease in summer runoff could even be off-set entirely in the mid term for all considered RCPs(table 1), although the statement does not hold true atthe local scale (figure 2(e)). On average for the period2070–2099, the results suggest that about two thirds(65%) of the change in summer runoff from presentlyglacierized surfaces could be mitigated through activewater management. Also at the local scale the impactsof such a strategy are significant. Figures 2(g)–(i) showthe differences between runoff reductions in the case
Table 1.Projected changes in total runoff frompresently glacierized surfaces for the EuropeanAlps. Changes aregivenwith respect to the reference period 1980–2009 and refer to yearly totals (first block), summer (July–September)totals without intervention (second block), and summer totals with implementation of themaximal potentialmitiga-tion (third block). Values are in km3.Mean runoff volumes for the reference period are given at the end of the table.Confidence intervals refer to the 95% level.
Period RCP 2.6 RCP 4.5 RCP 8.5
ANNUAL TOTAL S
2010–2039 +1.00±0.74 (+18%) +1.05±0.75 (+19%) +1.14±0.76 (+21%)2040–2069 −0.10±0.69 (−1%) +0.05±0.74 (0%) +0.32±0.80 (+6%)2070–2099 −0.74±0.62 (−14%) −0.73±0.67 (−13%) −0.93±0.89 (−17%)SUMMER—WITHOUT INTERVENT ION
2010–2039 +0.55±0.56 (+13%) +0.61±0.58 (+15%) +0.60±0.59 (+15%)2040–2069 −0.61±0.53 (−15%) −0.64±0.58 (−16%) −0.62±0.63 (−15%)2070–2099 −1.16±0.48 (−29%) −1.48±0.53 (−37%) −2.21±0.71 (−55%)SUMMER—WITH MAX IMAL MIT IGAT ION
2010–2039 +1.68±0.75 (+42%) +1.83±0.77 (+46%) +1.90±0.77 (+47%)2040–2069 +0.21±0.69 (+5%) +0.44±0.74 (+11%) +0.84±0.81 (+21%)2070–2099 −0.57±0.57 (−14%) −0.52±0.65 (−13%) −0.75±0.84 (−18%)
1980–2009 ANNUALTOTAL: 5.28±0.48 SUMMERTOTAL: 3.97±0.36
4
Environ. Res. Lett. 11 (2016) 054022
without intervention and the case with maximalpotential mitigation (that is the effectively mitigablechange), and highlight that the largest potential isavailable in regions that currently have high degrees ofglacierization. This observation, which might seem
contra-intuitive at first, can be explained by the factthat these regions are also expected to experience thelargest changes in annual runoff regime, which is thepre-requisite for the mitigation strategy defined here(see figure 1).
Figure 2.Past and future runoff contribution frompresently glacierized surfaces. Themean annual (a) and summer (b) runoff shareformpresently glacierized surfaces (see section 2.1 for definition) is given for the reference period 1980–2009. The expected change inannual runoff from the same surfaces for 2010–2039 (c), 2040–2069 (d), and 2070–2099 (e) is expressed as a relative changewithrespect to the reference period. Changes (means over all consideredGCMs) refer to RCP 4.5. Results for RCPs 2.6 and 8.5 are shownin supplementary figure S.6. In (a) and (b), river-width is proportional to the annual and July–August–September (JAS) runoffestimated from theUNH/GRDCdata set, respectively. In (c)–(e) thewidth is proportional to the estimated annual runoffcontribution fromglacierized surfaces.
5
Environ. Res. Lett. 11 (2016) 054022
4.Discussion
The potential for mitigating the runoff reduction inthe summer months is significant, and possibly ofimportance for individual regions. An estimate on theactual feasibility of such a water management strategyis thus of interest. This opens the question about howmuch of the required storage volume—in the order of1 km3 over the European Alps by the end of thecentury—could potentially be installed. In order tominimize the impact of such important technicalinfrastructure, one strategy could be to focus on areasthat are becoming ice-free due to glacier retreat. Thiswould be in line with the general idea of artificiallysustaining the effect that glaciers have in the hydro-logical cycle, and possibly result in smaller ecologicaland socio-economic impacts when compared to pla-cing storage volumes in lower lying—and thus typi-cally vegetated and inhabited—valleys.
We perform such a first-order assessment by vir-tually placing dams at the locations of current glacierterminus (determined through the RGI v4.0), and cal-culate the water volume of each individual lake beingformed. We do this by making use of the subglacialbedrock topography provided by Huss and Farinotti(2012) and using watershed delineation as describedby Jenson and Domingue (1988). Dams are initiallyplaced perpendicularly to the glacier centre lines pro-vided byMachguth andHuss (2014) and are iterativelyrotated in 5° steps in order to minimize the ratiobetween required dam-wall area (product of crestlength and average damheight) and obtained reservoirvolume. Dam heights and widths are limited to amax-imum of 280 m and 800 m, respectively (the dimen-sions of the largest dam currently installed in theAlpine area) and ‘grown’ to the maximal size allowedby the local topographical conditions. Note that thisprocedure aims at estimating the maximal installabletotal reservoir volume, and is not intended to designedreservoirs capable of accommodating the volume sur-plus of individual glaciers only.
According to our analysis, the potentially install-able volume over the entire Alps exceeds the requiredvolume by more than one order of magnitude(figure 4(a)). In general, the hypothetically availablevolume is largely in excess to the locally required onealso at the sub-regional scale. Figure 4(b) shows thespatial distribution of the storage volume hypotheti-cally available as the percentage of the volume requiredfor achieving the maximal potential mitigation at agiven location (see figure 3). As expected, the excess inpotentially available storage increases when movingdownstreamof the river network.
Clearly, such a widespread installation is neitherdesirable, nor technically or economically viable.Figure 4(a) highlights, however, how a relatively smallset of the virtually created installations could alreadyprovide the required total storage volume—assumingthese installation being fed by the surplus water
volumes of all surrounding glaciers. According to theresults, less than a dozen of the largest virtual reser-voirs would be sufficient for providing the totalvolume required for achieving the maximal potentialmitigation for the period 2070–2099 under theRCP 4.5 scenario. This is true even when consideringthe relatively large uncertainties introduced by theGCM spread, the RCP choice and the year-to-yearvariability (grey band in figure 4(a)). The total uncer-tainty, in fact, influences the required total storagevolume by a factor of 2 atmost (table 1).
The necessity of collecting the surplus watervolumes of surrounding glaciers into centralizedreservoirs would clearly be a major technical challengewhen adopting the described strategy. Figure 4(b),however, suggests that sufficient storage volume ispotentially installable also at the sub-regional scale,which would reduce the difficulty in centralizing thewater from individual glaciers. Obviously, this decen-tralization of reservoirs would significantly increasethe total number of required installations, and—on apar—the ecological and socio-economic impacts.
Four visualization examples (one for each majorAlpine country) of the virtually installed reservoirs areshown in figures 4(c)–(f). In this respect we would liketo emphasise that it is not our intention to suggest thatthese particular locations could indicate a priority forinstallation. Our analysis, in fact, clearly neglects awhole series of factors—linked to ecological, environ-mental, economical, and technical considerationsamongst others—that would need to be consideredwhen compiling an actual priority list. Similarly, ouranalysis does not take into account already-installedinfrastructure—the exploitation of which should befirst choice when considering an actual implementa-tion of the strategy—and does not consider potentialcompeting interests, that would likely arise in case of awidespread installation of artificial reservoirs. In thisrespect, in fact, it should be noted that the typical sea-sonal production cycle of current hydropower infra-structure installed in the Alpine region, is exactlyanticyclical with the mitigation strategy proposed here(current hydropower operators with storage typicallyaccumulate water during summer in order to produceenergy duringwinter).
Besides the limitations above, it has obviously tobe noted that seasonally transferring a given runoffvolume cannot compensate a reduction in the totalannual runoff (see figure 1). Even assuming unchan-ged annual precipitation in the future, and neglectingincreased evapotranspiration losses due to higher airtemperatures, such a runoff decrease has to be expec-ted for high alpine catchments in response to glacierretreat. This is because glaciers losing large parts oftheir ice volume will largely reduce their contributionto runoff by ice melt. The water that contributes torunoff during the glacier recession phase can, hence,be considered as a non-renewable water contributionto the water cycle: once depleted, the water volume
6
Environ. Res. Lett. 11 (2016) 054022
Figure 3.Projected changes inmean summer (JAS) runoff formpresently glacierized surfaces. The changes are expressed as relativedifferences with respect to (w.r.t) 1980–2009 and refer to the situationwithout any intervention (a)–(c), and the situation inwhich themaximal potentiallymitigable volume (seefigure 1 for definition) is realized (d)–(f). The difference between the two cases, that is thepotentiallymitigable change, is presented separately (g)–(i). Results are given for three different time periods (rows) and refer toRCP 4.5. Results for RCPs 2.6 and 8.5 are given in supplementary figures S.4 and S.5, respectively. River width is proportional to theestimated summer runoff.
Figure 4.Comparison of required and hypothetically available storage volume. (a)Cumulative number of dams (red) andwall-damarea (blue) required for achieving a given total storage volume according to the virtual set of constructed reservoirs. The storagevolume required for achieving themaximal potentialmitigation during 2070–2099 is shown (green). The grey band represents theuncertainty due toGCMspread, considered RCP, and year-to-year variability. (b)Ratio between (1) themaximal hypotheticallyavailable storage volume along a given river stretch and (2) the storage volume required for achieving themaximal potentialmitigation(seefigure 1 for definition) in summer runoff change at a given location (figures 3(d)–(f)). The results refer to the period 2070–2099andRCP 4.5. Thewidths of the rivers are proportional to the hypothetical storage volumes cumulated along a given river stretch. Fourhypothetical reservoirs are visualized for examples in the (c) French, (d) Italian, (e) Swiss, and (f)AustrianAlps. For each example, thearea (Alake) and volume (Vlake) of the generated lake, as well as themaximal height (Hdam) and the length of the crest of the requireddam (Ldam) is given (for comparison: GrandeDixence, the largest dam currently installed in the EuropeanAlps, has Alake = 0.40 km3,Vlake = 3.7 km2, Hdam = 285 m, Ldam = 700 m) . Topographical imagery is fromGoogleTM Earth.
7
Environ. Res. Lett. 11 (2016) 054022
previously stored as ice will require decades to cen-turies, and drastically changed climatic conditions, tore-form.
The change in this non-renewable contribution isshown in figures 2(c)–(e) for the three considered timehorizons and RCP 4.5 (see supplementary figure S.6for RCP 2.6 and 8.5). According to these results, a totalof 0.73±0.67 km3 of water per year could be missingacross the European Alps on average between 2070and 2099 (see table 1). This is equivalent to roughly80% of today’(s) annual freshwater consumption inSwitzerland (SFSO 2014), and highlights the need ofalternative water management strategies in addition tothe hypothetical one presented above. A part from theobvious need of curbing current greenhouse gas emis-sions in order to minimize changes in global temper-ature, concepts aiming at a more efficient use of waterresources are required.
5. Conclusions
Surfaces presently covered by glaciers are importantwater sources across the European Alps. During theperiod 1980–2009, glaciers provided an average5.28±0.48 km3 a−1 of freshwater. About three quar-ters of this volume occurred during July–September,resulting in significant glacier contributions also inlarge low-lying rivers including the Po, the Rhine, andthe Rhône.
Important decreases in both annual and summerglacier runoff contributions are anticipated inresponse to ongoing climatic change. For an averageemission scenario (RCP 4.5), annual runoff contribu-tions from presently glacierized surfaces are expectedto decrease by 16% by 2070–2099—despite of nearlyunchanged contributions from precipitation. Thedecrease is even more pronounced during the latesummer months (JAS), for which a decrease of about37% is projected. This highlights the need of adequatewatermanagement strategies in the future.
Here, we addressed the somewhat provocativequestion of whether replacing glaciers by dams couldbe an option theoretically worth considering in orderto mitigate the projected changes in summer runoff.Postulating that keeping the past runoff regime ofindividual river stretches unaltered is a desirable tar-get, we quantify the water volume that could season-ally be transferred by means of reservoirs in order tocompensate for such runoff reductions. We estimatethat by 2070–2099, roughly 1 km3 a−1 of water couldbe seasonally redistributed, and that this volumewould be sufficient to offset about two thirds of theexpected changes in July–September runoff across theEuropeanAlps.
With a first-order assessment that neglects a seriesof environmental, ecological and economic considera-tions, we compute the storage volume that couldhypothetically be installed in areas becoming ice-free
due to glacier retreat. We estimate that this hypothe-tical volume is in excess of the required one by morethan one order of magnitude, and that less than adozen of large dams could theoretically provide thetotal required storage. Centralizing and redistributingthe waters from the numerous glacierized catchmentswould, however, be amajor technical challenge.
Obviously, the here addressed technocrat solutioncannot compensate for the expected net reduction inannual water yields. For compensating this non-renewable component of the water cycle, alternativestrategies aiming at a more efficient management ofwater resources are required.
Acknowledgments
This work was funded by the Swiss National ScienceFoundation and the European Communities 7thFramework Programme (Grant AgreementNo. 603629-ENV-2013-6.2.1-Globaqua). We thankBettina Schaefli for her helpful feedback on an earlierversion of this manuscript, as well as the anonymousreferees and the editorial team for their constructivecomments.
References
Arendt A et al 2014Randolph glacier inventory [4.0]: a dataset ofglobal glacier outlines, version4.0Global Land IceMeasurements from Space (GLIMS)Boulder, Colorado,USA,digitalMedia (www.glims.org/RGI/)
Barnett T P, Adam JC and LettenmaierDP 2005 Potential impactsof a warming climate onwater availability in snow-dominated regionsNature 438 303–9
BavayM, LehningM, Jonas T and LöweH2009 Simulations offuture snow cover and discharge inAlpine headwatercatchmentsHydrol. Process. 23 95–108
BenistonM2003Climatic change inmountain regions: a review ofpossible impactsClimate Variability andChange inHighElevation Regions: Past, Present& Future (Advances inGlobalChange Research vol 15) edHFDiaz (Netherlands: Springer)pp 5–31
BenistonM, StoffelM andHillM2011 Impacts of climatic changeonwater and natural hazards in theAlps: can current watergovernance copewith future challenges? Examples from theEuropeanACQWAprojectEnviron. Sci. Policy 14 734–43
BerghuijsWR,Woods RA andHrachowitzM2014Aprecipitationshift from snow towards rain leads to a decrease in streamflowNat. Clim. Change 4 583–6
Bliss A,Hock R andRadićV2014Global response of glacier runoffto twenty-first century climate change J. Geophys. Res. 119717–30
Brutel-Vuilmet C,MénégozMandKrinnerG 2013An analysis ofpresent and future seasonalNorthernHemisphere land snowcover simulated byCMIP5 coupled climatemodelsCryosphere 7 67–80
CareyM2005 Living and dyingwith glaciers: people’s historicalvulnerability to avalanches and outburst floods in PeruGlob.Planet. Change 47 122–34
Cogley J G 2012The future of theWorld’s glaciersThe Future of theWorld’s Climate edAHenderson-Sellers andKMcGuffie(Amsterdam: Elsevier) ch 8, pp 197–222
Cogley J G et al 2011Glossary of glaciermass balance and relatedterms IHP-VII Technical Documents inHydrologyNo86 IACSContributionNo. 2, UNESCO-IHP, Paris (http://unesdoc.unesco.org/images/0019/001925/192525E.pdf)
8
Environ. Res. Lett. 11 (2016) 054022
DeeDP et al 2011The ERA-Interim reanalysis: configuration andperformance of the data assimilation systemQ. J. R.Meteorol.Soc. 137 553–97
Dussaillant A, BenitoG, BuytaertW,Carling P,Meier C andEspinoza F 2010Repeated glacial-lake outburstfloods inPatagonia: an increasing hazard?Nat.Hazards 54 469–81
Farinotti D, Usselmann S,HussM, Bauder A and FunkM2012Runoff evolution in the Swiss Alps: projections for selectedhigh-Alpine catchments based onENSEMBLES scenariosHydrol. Process. 26 1909–24
Fekete BM,Vörösmarty C J andGrabsW2002High-resolutionfields of global runoff combining observed river dischargeand simulatedwater balancesGlob. Biogeochem. Cycles 1615-1–15-10
Fischer L, KääbA,Huggel C andNoetzli J 2006Geology, glacierretreat and permafrost degradation as controlling factors ofslope instabilities in a high-mountain rockwall: theMonteRosa east faceNat.Hazards. Earth Syst. Sci. 6 761–72
FischerM,HussM, BarbouxC andHoelzleM2014The new Swissglacier inventory SGI2010: relevance of using high-resolutionsource data in areas dominated by very small glaciersArcticAntarct. Alpine Res. 46 933–45
FreyH,HaeberliW, Linsbauer A,Huggel C and Paul F 2010Amulti-level strategy for anticipating future glacier lakeformation and associated hazard potentialsNat.HazardsEarth Syst. Sci. 10 339–52
Gardner A S et al 2013A reconciled estimate of glacier contributionsto sea level rise: 2003–2009 Science 340 852–7
HaggW J andBraun LN2005The influence of glacier retreat onwater yield fromhighmountain areas: comparison of AlpsandCentral AsiaClimate andHydrology ofMountain Areas edCde Jong et al (NewYork:Wiley)pp 263–75
HockR, Jansson P andBraun L 2005Modelling the response ofmountain glacier discharge to climate warmingGlobalChange andMountain Regions—AState of KnowlegdeOverview edUMHuber et al (Dordrecht: Springer)pp 243–52
HodsonA, Brock B, PearceD, Laybourn-Parry J andTranterM2015Cryospheric ecosystems: a synthesis of snowpack andglacial researchEnviron. Res. Lett. 10 110201
HussM2011 Present and future contribution of glacier storagechange to runoff frommacroscale drainage basins in EuropeWater Resour. Res. 47W07511
HussMand Farinotti D 2012Distributed ice thickness and volumeof all glaciers around the globe J. Geophys. Res. 117 F04010
HussMandHockR 2015Anewmodel for global glacier change andsea-level rise Frontiers Earth Sci. 3 54
HussM, Jouvet G, Farinotti D andBauder A 2010 Future high-mountain hydrology: a new parameterization of glacierretreatHydrol. Earth Syst. Sci. 14 815–29
ImmerzeelWW, van Beek L PHandBierkensMFP 2010Climatechangewill affect the Asianwater towers Science 328 1382–5
Jarvis J, ReuterH,NelsonA andGuevara E 2008Hole-filled SRTMfor the globe, CGIARConsortium for Spatial Information,Montpellier, France, CGIAR-CSI SRTM90 mDatabaseVersion 4 (http://srtm.csi.cgiar.org/)
Jenson SKandDomingue JO1988Extracting topographic structurefromdigital elevationdata for geographic information systemanalysisPhotogramm. Eng.Remote Sens.541593–600
de JongC,Collins D andRanzi R 2005Climate andHydrology inMountain Areas (West Sussex, England:Wiley) (doi:10.1002/0470858249)
KääbA, Reynolds JM andHaeberliW2005Glacier and permafrosthazards in highmountainsGlobal Change andMountainRegions (Advances inGlobal Change Research vol 23) edUMHuber et al (Netherlands: Springer)pp 225–34
KaserG,GroßhauserM andMarzeion B 2010Contributionpotential of glaciers towater availability in different climateregimes Proc. Natl Acad. Sci. USA 107 20223–7
MachguthH andHussM2014The length of theworld’s glaciers—anew approach for the global calculation of center linesCryosphere 8 1741–55
Mankin J S, Viviroli D, SinghD,Hoekstra A Y andDiffenbaughN S 2015 The potential for snow to supplyhumanwater demand in the present and future Environ.Res. Lett. 10 114016
MeinshausenM et al 2011TheRCP greenhouse gas concentrationsand their extensions from1765 to 2300Clim. Change 109213–41
NELAK2013New lakes as a consequence of glacier retreat in highmountains—challanges and risks (original title: neue seen alsfolge des gletscherschwundes imhochgebirge—chancen undrisiken) Scientific Report of theNational Research ProgrammeNRP61 edWHaeberli et al vdfHochschulverlag AG anderETHZürich (doi:10.3218/3534-6)
NemergutDR, Anderson S P,ClevelandCC,Martin AP,Miller A E, SeimonA and Schmidt SK 2007Microbialcommunity succession in an unvegetated, recentlydeglaciated soilMicrobial Ecol. 53 110–22
Orlove B 2009Glacier retreat: reviewing the limits of humanadaptation to climate changeEnvironment 51 22–34
Paul F, FreyH andBris R L 2011Anew glacier inventory for theEuropeanAlps fromLandsat TM scenes of 2003: challengesand resultsAnn.Glaciol. 52 144–52
PfefferWTand (TheRandolphConsortium) 2014TheRandolph glacier inventory: a globally complete inventory ofglaciers J. Glaciol. 60 537–52
Piao S et al 2010The impacts of climate change onwater resourcesand agriculture in ChinaNature 467 43–51
Pistocchi A 2014GIS BasedChemical FateModeling: Principles andApplications (Hoboken:Wiley) p 520
Pistocchi A and PenningtonD2006 European hydraulic geometriesfor continental scale environmentalmodelling J. Hydrol. 329553–67
RadićVandHockR 2011Regionally differentiated contribution ofmountain glaciers and ice caps to future sea-level riseNat.Geosci. 4 91–4
RadićVandHockR 2014Glaciers in the Earth’s hydrological cycle:assessments of glaciermass and runoff changes on global andregional scales Surv. Geophys. 35 813–37
Richardson SD andReynolds JM2000An overview of glacialhazards in theHimalayasQuat. Int. 65–66 31–47
SchanerN, VoisinN,Nijssen B and LettenmaierDP 2012Thecontribution of glaciermelt to streamflow Environ. Res. Lett. 7034029
ScottD 2006Global environmental change andmountain tourismTourism andGlobal Environmental Change: Ecological,Economic, Social and Political Interrelationships edSGössling andMCHall (London: Taylor and Francis)pp 54–75
ScottD, Jones B andKonopek J 2007 Implications of climate andenvironmental change for nature-based tourism in theCanadianRockyMountains: a case study ofWaterton LakesNational ParkTourismManage. 28 570–9
SFSO Swiss Federal StatisticalOffice 2014 Freshwater consumption(http://bfs.admin.ch/bfs/portal/de/index/themen/02/06/ind17.indicator.1300101.13001.html) (accessed 17May 2016)
Steger C, Kotlarski S, Jonas T and Schär C 2013Alpine snow cover ina changing climate: a regional climatemodel perspectiveClim.Dyn. 41 735–54
Taylor K E, Stouffer R J andMeehl GA 2012 An overview ofCMIP5 and the experiment design Bull. Am.Meteorol. Soc.93 485–98
Viviroli D et al 2011Climate change andmountainwater resources:overview and recommendations for research,managementand policyHydrol. Earth Syst. Sci. 15 471–504
WGMS 2012 Fluctuations of Glaciers 2005–2010 (Vol. X) edMZemp et al (Zurich, Switzerland: ICSU (WDS)/IUGG(IACS)/UNEP/UNESCO/WM,WorldGlacierMonitoringService) based on database version (doi:10.5904/wgms-fog-2012-11)
ZempM,HaeberliW,HoelzleM and Paul F 2006Alpine glaciers todisappearwithin decades?Geophys. Res. Lett. 33 L13504
9
Environ. Res. Lett. 11 (2016) 054022
