CLIMATE CHANGE IMPACTS ON HYDROELECTRIC POWER

G.P. Harrison(1), H.W. Whittington(1) and S.W. Gundry(2)

(1) Energy Systems Group, Department of Electronics and Electrical Engineering, University of Edinburgh, UK.(2) Institute of Ecology and Resource Management, University of Edinburgh, UK.

ABSTRACT

Anthropogenic emissions of greenhouse gases are expected to lead to significant changes in climate over the nextcentury. One of the many potential effects is that river catchment runoff may be altered. This could haveimplications for the design, operation and viability of hydroelectric power stations. This describes attempts to predictand quantify these impacts. It details a methodology for computer based modelling of hydroelectric resources andproposes analysis of the impacts on the electrical system and on the investment performance of hydro.

INTRODUCTION

Climate change or global warming is the expectedoutcome of increases in atmospheric concentrations of“greenhouse” gases resulting from human activities.Many greenhouse gases, including carbon dioxide(CO2), occur naturally and keep the earth warm bytrapping heat in the atmosphere. However, since theIndustrial Revolution, anthropogenic sources of CO2

have added greatly to the atmospheric concentrations,and in particular, transportation and the burning offossil fuels for electricity generation are frequentlycited as major sources. Other man-made greenhousegases, such as CFC’s, are believed to exacerbate theprocess. Enhanced levels of greenhouse gasconcentrations are predicted to cause a significant risein temperature over the next century. The rates ofincrease are anticipated to be greater than at any timein the past. The current scientific consensus is thatunder present rates of economic and population growthglobal mean temperatures will rise by 3°C by the endof the next century. This is expected to beaccompanied by increases in global precipitation levelsof 15% [1].

Predictions of future climate are based on the output ofcomplex numerical Global Circulation Models(GCM’s) which simulate physical processes in theatmosphere and oceans. A number of research groupshave developed GCM’s including the UKMeteorological Office (UKMO), the GeophysicalFluid Dynamics Laboratory (GFDL) at PrincetonUniversity, and NASA’s Goddard Institute for SpaceStudies (GISS). Although different GCM’s simulatecurrent climate with varying degrees of precision mostagree on the general trends.

The main technique for avoiding the worst extremes ofclimate change is to limit the increase in greenhousegas concentrations by reducing emissions. Aselectricity production is responsible for a significantportion of the emissions, much of the burden will fallon the energy sector. Possible measures includetransferring to lower carbon fuels like natural gas,

together with increased use of renewable energysources including hydropower [1].

Hydropower is an attractive energy source as it isrenewable with minimal operational emissions ofgreenhouse gases. In addition, there are no fuelcharges and the civil works have a long useful life.However, large dams may necessitate populationdisplacement and can impact on the ecology of thebasin [2, 3]. Exploitation of hydropower potential isconsidered by many governments and internationalbodies to be a key feature in economic development,especially in less developed countries (LDC’s).

At first glance, increased global precipitation wouldappear to suggest more water available forhydroelectric power production. However, highertemperatures will lead to increased evapotranspirationlevels. Whether increased global precipitation is seenas increased river runoff depends on the regionalclimate and hydrology. In the past, feasibility studieshave relied on historical rainfall and river flow data forthe assessment of hydroelectric potential at a proposedsite. However, climatic change means that these canno longer be relied on to indicate future potential [2].It is perhaps ironic to consider that attempts to reduceclimatic change by switching to non-fossil fuels couldbe hampered by the legacy of their use.

CLIMATE CHANGE IMPACTS ONHYDROPOWER AND ENERGY SECTORS

Changes in the quantity and timing of river runoff,together with increased reservoir evaporation will havea number of effects on the production of hydroelectricpower. These include impacts upon system operation,financial effects and impacts on other energy sectors.

System Operation and DevelopmentChanges in the availability of existing hydroelectricplant, together with system constraints will affect theability of the electricity supply system to meet averageand peak demands. In the longer term, as demandlevels increase, system planning may have to address

any predicted shortfall in hydro output by constructingadditional generating plant [4]. The likelihood is thatfossil fuels will be used, further enhancing radiativeforcing [2]. Climate change may also result in someplanned projects being cancelled or adapted.

Financial EffectsHydroelectric stations are characterised by lowoperational costs but high capital costs. Generally,revenue from electricity sales is the only way ofservicing the capital debt. Thus reductions inelectricity sales will affect the return on investment andhence the viability of the plant [2]. The loss ofhydroelectric generating capacity will requireadditional plant to be constructed to meet demand,requiring additional capital and thus reducing overallsystem returns.

Many large hydropower developments in LDC’s arebuilt with the intention of stimulating economicdevelopment. Generally, this requires internationalfinancing with a requirement for the loan repaymentsto be in hard currency. Reductions in revenue mayaffect the ability to repay the hard currency debt andthis may severely stress a weak economy. In addition,the fall in electricity availability will hamperGovernments’ attempts to aid economic development.

Effects on other Energy SectorsClimate change will have impacts on both electricitydemand and supply. Higher air temperatures will tendto lower winter heating demands but increase summercooling demand. Thermal generating stations requiringrivers for cooling water may suffer operationalconstraints due to reduced river flows [2, 5, 6].Warmer river and sea water will reduce the efficiencyof steam cycles, resulting in lost output or increasedfuel consumption. Predicted sea level rise may alsothreaten coastal stations; climate change may lead tomore extreme weather patterns causing increasedsystem damage costs. Climate change may affect otherrenewable technologies: wind patterns may change as aresult of changed temperature gradients, and changesin cloud cover may affect the performance of solarpanels [7].

CLIMATE IMPACTS ASSESSMENT

The potential impact of climate change on waterresources has been suggested since the 1980s, as workprogressed on predicting climate change [5]. AlthoughGCM’s can be used to predict runoff directly, thecoarse scale used means that this information is onlyuseful for the most general studies. As a result, manystudies have been carried out on individual basins,showing that river basins display a range ofsensitivities to climate change [8]. Figure 1 shows theresponse of a typical river basin to variations inprecipitation and temperature. It can be seen that

increased temperature results in non-linear variationsin runoff due to changes in precipitation.

-80

-60

-40

-20

0

20

40

60

80

100

-20 -10 0 10 20

percent change in precipitation

per

cen

t ch

nag

e in

ru

no

ff

1°C temperature rise 2°C temperature rise

Figure 1: River Basin Response To Climate Change

Later studies have considered not only the effect onriver flows but also the impact on generation fromhydroelectric stations [9]. In particular, one studyexamined a number of international river basins [4].The study drew upon existing hydrological anddedicated basin models and the experience ofinternational experts. For example, for one GCMscenario (GFDL), hydroelectric production on theIndus River would fall by 22%. Another study [2]qualitatively examined the effects of reducedhydroelectric output on sub-Saharan Africa and centralEurope. However, to date, studies have failed toquantify the impacts in terms of the investmentperformance of plant or on the electrical network.

Modelling ImpactsClimate impact assessment requires scenarios of futureclimate to be translated into potential changes onnatural and human systems. To assess climate impactson hydropower production a number of key steps mustbe taken [4]:

1. A river basin is selected and its rainfall-runoffprocesses are modelled and calibrated;

2. Climate data emanating from different GCM orarbitrary climate scenarios is applied to the modeland the runoff computed;

3. River runoff values are converted into estimates ofhydroelectric power production.

The first step involves the accurate modelling of thehydrology of the chosen river basin. A wide variety ofmodelling techniques have been applied to simulatingrunoff processes [10]. Three basic approaches exist:

• Empirical• Conceptual

• Deterministic The first type requires a relationship to be establishedbetween climate inputs (e.g. rainfall) and hydrologicaloutputs (i.e. runoff). The second type uses a simplifiedrepresentation of the physical processes to mimic thestorage and flow of water. The technique requires suchmodels to be calibrated for each catchment usingrelevant climate and river flow data. The finalapproach is based on complex physical theory andmost examples are spatially distributed in two or threedimensions. Such models claim to give a more explicitrepresentation of hydrological processes, but sufferfrom the requirement for significant quantities ofinformation for operation. Despite the sophistication of current hydrologicalmodelling techniques, particular difficulties exist intranslating the GCM’s large spatial and temporalpredictions into a form that can be used by ahydrological model. Techniques to overcome this aretermed downscaling methods [5], and attempt togenerate local values for precipitation (say) from thelarge-scale GCM values. Other methods includenesting smaller-scale regional climate models insidethe GCM’s, creating weather stochastically and usinganalysis of different weather types. Given that climate data from GCM’s can be convertedinto a form suitable for use, the output from a suitablycalibrated hydrological model would be similar to thatshown in figure 2 for present and assumed futureclimates. In this hypothetical case it can be seen thatclimate change affects the magnitude and timing ofriver runoff. Winter runoff is higher as precipitationfalls as rain rather than snow, the winter thaw occursearlier, and summer runoff is also lower.

0

10

20

30

40

50

60

70

80

90

J F M A M J J A S O N D

Month

Ru

no

ff

Present Climate change

Figure 2: Effect Of Climate Change On Runoff The potential for hydroelectric generationapproximately follows runoff, so here it can be seenthat hydroelectric potential would also be affected. Amore accurate estimate of climate impacts onhydropower would involve assessment of the relativeimportance and cost of hydro, the economic

development of the country and the policies ofgovernments and international organisations.

PROPOSED COMPUTER MODEL This work is concerned primarily with the design anddevelopment of a generic hydropower assessment toolsuitable for use with any hydropower scheme andrelevant climate scenario. The assessment tool will beimplemented on a PC and would consist of: A (simple) self-calibrating hydrological model thatwould be able to derive suitable input-responserelationships, given suitable climate and river flowdata. After calibration, the model will convert inputclimate data into estimates of river flows. Theseresults would be processed by the hydropowercomponent, which when given suitable technical andoperational parameters, would compute the electricalpower generated. With this structure, the model will give indications ofthe power generated for desired baseline and predictedclimate scenarios. This will enable projections to bemade concerning the investment performance of thehydroelectric scheme together with assessments of theimpact on the economy. In addition, the generationscenarios may be used for analysis of the electricalsystem, in terms of its generation mix and overallpower requirements. A simplified model may be as effective as a morecomplex one given that at present climate data fromGCM’s is available in terms of mean monthly values.Accordingly, monthly river flows are likely to be moremeaningful than shorter time frames. Although thistime step is too long to allow simulation of hydro-plantscheduling, suitable operational rules can be derived tomimic its output. In general, consideration of hydropower impacts hasbeen concerned with the potential annual production ofelectricity, and not the impact of changes in themonthly availability. Month-to-month changes inelectricity demand are generally greater than year-to-year, so a consideration of the monthly availability islikely to have more relevance. Consideration ofspillage and the likely actual production under climatechange will allow a more realistic representation. It is expected that a number of case studies will becarried out using the software tool. These will tend toconcentrate on large strategically importanthydroelectric power projects both in existence orplanned. The majority of such projects are in LDC’shighlighted as being sensitive to climate change.Among the areas under consideration are sub-SaharanAfrica and East Asia. In particular, the Three GorgesDam on the Yangtze River in China will be examined,

due to its key role in China’s economic and waterresources development, its sheer size and cost, and thealleged potential for environmental and socialcatastrophe. Application of Proposed Software Tool The proposed software tool would have a number ofapplications. It would: • allow planners to make projections of changes in

future availability of water and power resources.• enable water resources and hydropower designers

to compare competing schemes before detailedfeasibility studies commence.

• allow assessment of the impacts of land usechanges.

• interface with existing dedicated basin models toallow accurate water flow projections to be input tothe hydropower element of the software.

The key outcome of this work will be information onhow hydropower may fare as an investmentopportunity, and as a driver of economic developmentin LDC’s.

CONCLUSION

Human activities are expected to lead to substantialchanges in climate. One outcome may be reductions inriver runoff with potentially serious ramifications forthe provision of hydroelectric power. Recent attemptsat quantifying these impacts have been described and amethodology proposed to enable analysis of the impacton the electrical system as well as the investmentperformance of hydroelectric plant.

REFERENCES

1. IPCC (Intergovernmental Panel On ClimateChange): Climate Change. The IPCC ScientificAssessment. Houghton, J.T., Jenkins, G.J. &Ephraums, J.J. (Eds.), Cambridge University Press,pp. 1-40, 93-110, 131-173, 1990.

2. Whittington, H.W. & Gundry, S.W.: ‘GlobalClimate Change and Hydroelectric Resources’, IEEEngineering Science & Technology Journal, March1998.

3. Barber, M. And Ryder, G. (Eds.): Damming TheThree Gorges : What Dam Builders Don’t WantYou To Know, 2nd ed., Probe International,Earthscan Publications Ltd, 1993

4. Reibsame, W.E., Strzepek, K.M., Wescoat Jr., J.L.,Perritt, R., Gaile, G.L., Jacobs, J., Leichenko, R.,Magadza, C., Phien, H., Urbiztondo, B.J.,Restrepo, P., Rose, W.R., Saleh, M., Ti, L.H.,Tucci, C. & Yates, D.: Complex River Basins. InStrzepek, K.M. & Smith, J.B. (Eds.), As ClimateChanges : International Impacts and Implications,Cambridge University Press, pp. 57-91, 1995.

5. Arnell, N: Global Warming, River Flows AndWater Resources, J. Wiley & Sons Ltd, pp. 25-60,151-200, 1996.

6. Gleick, P.H.: ‘Water and energy’. In Gleick, P.H.(ed.) Water In Crisis, Oxford University Press,New York, pp. 67-79, 1993.

7. IPCC (Intergovernmental Panel On ClimateChange): Climate Change. The IPCC ImpactsAssessment. Tegart, W.J.McG., Sheldon, G.W. &Griffiths, D.C. (Eds.), Australian GovernmentPublishing Service, Canberra, pp. 1-25, 1990.

8. IPCC (Intergovernmental Panel On ClimateChange): The Regional Impacts of Climate Change:An Assessment of Vulnerability. Watson, R.T.,Zinyowera, M.C., Moss, R.H. & Dokken, D.J.(Eds.), Cambridge University Press, 1998.

9. Mimikou, M.A. & Baltas, E.A.: ‘Climate changeimpacts on the reliability of hydroelectric energyproduction’, Hydrol. Sci. J., 42 (5), pp. 661-678,1997.

10. Fleming, G.: Computer Simulation Techniques inHydrology, Elselvier Publishing, pp. 18-53, 1975.

AUTHOR’S ADDRESS

Energy Systems Group,Department of Electronics and Electrical Engineering,The University of Edinburgh,King’s Buildings, Mayfield Road,Edinburgh, EH9 3JL,Scotland, UK.

Email : Gareth.Harrison@ee.ed.ac.ukPhone : +44 (0)131 650 5584Fax : +44 (0)131 650 6554