Modelling the effects of climate change on water resources in central Sweden
178CHONG-YU XU relationships between climate and water resources.Various methodologies for sim-ulating hydrological responses to global climate change by using hydrologic mod-els have been reported,which may be described using three categories:(1)Coup-ling high-resolution regional climate models(RCM)with hydrologic models(e.g. Hostetler and Giorgi,1993;Nash and Gleick,1993);(2)Coupling GCMs with hy-drologic models through statistical downscaling techniques(e.g.Wilby and Wigley, 1997);(3)Using hypothetical scenarios as input to hydrologic models(e.g.Arnell, 1992).
Ideally,the climate simulations from the GCMs could be used directly to drive hydrologic models,which in turn could be used to evaluate the hydrologic and water resources effects of climate change.However,the performance of GCMs in the control simulation and the magnitude of the predicated climate change signal is not certain.Moreover,different GCMs are still giving different values of climate variable changes and so do not provide a single reliable estimate that could be ad-vanced as a deterministic forecast for hydrological planning.Accordingly,methods of simple alteration of the present conditions,i.e.hypothetical scenarios methods, are often used.Many published works were done in this way(e.g.Nemec and Schaake,1982;Gleick,1986,1987;McCabe and Ayers,1989;Schaake and Liu, 1989;Lettenmaier and Gan,1990;Vehvil?inen and Lohvansuu,1991;Panagoulia, 1991;Arnell,1992;Ng and Marsalek,1992).Various scenarios have been used and climate predictions for‘double CO2’conditions have become a standard(e.g. Loaiciga et al.,1996).
This article reports some of the results of a larger investigation into the implic-ations of climatic variability and change for river?ow regimes in Sweden.The boreal forest zone has several characteristics that differ from other regions of the world and which make studies of it imperative in the global context(Thomas and Rowntree,1992;Halldin et al.,1998).The earlier stage of the study was presented in a number of publications.Xu et al.(1996)developed and applied a monthly water balance model for water balance calculations for Nordic regions.More re-cently,Xu(1999b)tested the applicability of the model to simulate the hydrological responses of climate changes.Xu(1999a)reviewed the current state of methodo-logies for simulating hydrological responses to global climate change.The current study used the model with a number of hypothetical scenarios of climate change to estimate impacts in river?ow regimes and snow cover in central Sweden.
2.The Model,Study Region and Data
More detailed information about the model’s structure and performance can be found in Xu et al.(1996).A brief description is given as follows with the help of Figure1.The monthly water balance model requires as inputs monthly values of areal precipitation,potential evapotranspiration and air temperature.The model outputs are monthly river?ow and other water balance components,such as actual evapotranspiration,slow and fast components of river?ow,soil-moisture storage
MODELLING THE EFFECTS OF CLIMATE CHANGE ON WATER RESOURCES179
Figure1.Schematic computational?ow chart of the monthly water balance model.
and accumulation of snowpack,etc.The model works as follows:precipitation p t is?rst split into rainfall r t and snowfall s t by using a temperature-index function, snowfall is added to the snowpack sp t(the?rst storage)at the end of the month, of which a fraction m t melts and contributes to the soil-moisture storage sm t. Snowmelt is calculated by using a temperature-index method.Before the rainfall contributes to the soil storage as‘active’rainfall,a part is subtracted and added to interception evaporation loss.The soil storage contributes to evapotranspiration e t, to a fast component of?ow f t and to base?ow b t.
The boundary of the study area is de?ned as the central part of Sweden(about 40000km2,Figure2).Within this area necessary meteorological data and land-use data are available for25gauged catchments ranging in size from6to1293km2.
The landscape of the region is dominated by large lakes and plains separated from each other by high undulating ridges and rich in faults.The geology is char-acterised by oldest granites in the northeastern part while sedimentary gneisses characterise the south.Leptites and h?lle?intas are found in the northwestern side together with some small granite-dominated areas.Forest and agriculture are the dominant landuse.Forest is a dominant factor in the northwest and agriculture is concentrated in the south,with meadow and grain cultivation predominant agricul-ture use.
180CHONG-YU XU
Figure2.The map of Sweden with the location of the study region.
Within the study region,the following data of at least10-year duration are avail-able:discharge from25stations,precipitation from41stations,temperature data from12stations.The daily data from1981to1991were taken and subsequently integrated to the monthly values for use in the study.Areal precipitation was calcu-lated by Thiessen https://www.sodocs.net/doc/d82689736.html,nd-use data are also available for the corresponding catchments.A general information about the catchment characteristics is presented in Table I.
3.Hypothetical Climate Change Scenarios
The preferred source of data for using in the assessment of impacts of climate change is the general circulation model(GCM).Given the limitations of GCMs grid-point predictions for regional climate change impacts studies,one of the most widely used methods of scenario generation has been to estimate average annual changes in precipitation and temperature for a region,and then apply these estim-ates to adjust historic time series of precipitation and temperature.In the simplest procedure,the generation of climate scenarios consists of two steps:
MODELLING THE EFFECTS OF CLIMATE CHANGE ON WATER RESOURCES181
Table I.General information of the study catchments(1981.1–1991.12)
Station Abbr.Code Area Mean Mean a Mean Lake Forest Open
prec.evap.runoff?eld
(km2)(mm)(%)
?kesta Kv.Ak221672760.140.021.6 4.069.027.0
?kers Krut.Ar224921460.343.317.6 5.266.328.5 Bergsh.Be230021.655.640.216.30.269.530.3 Berg Bg221836.563.945.322.20.071.428.6 Bernsh.Bs157359578.043.434.98.677.314.1
Dalkarlsh.Dl2206118276.442.435.27.574.617.9 Fellingsbr.Fb220529862.639.924.6 6.063.830.2 Finntorp.Ft2242 6.9665.943.922.1 4.795.30.0 Gr?nvad Gr221716759.441.319.90.041.158.9
H?rnevi Ha224831260.238.822.8 1.055.044.0
Hammarby Hb215389173.343.130.99.580.99.7
K?falla Kf153241381.043.636.9 6.280.813.0 Kringlan Kl222929478.344.334.27.687.2 5.2 Karlslund Ks2139129369.743.427.0 6.662.730.7 Lurbo Lu224512260.837.025.60.368.231.5
Odensvibr.Ob222111063.641.723.3 6.371.022.7 Ransta Ra224719759.838.222.40.966.133.0
R?lls?lv Rs220729879.343.138.47.478.813.8
S?vja Sa224372259.740.419.7 2.064.034.0 Skr?ddart.Sd222217.766.741.625.3 2.596.1 1.4
Sk?llnora Sn184358.555.039.916.210.444.545.1
S?rs?tra So222061259.733.228.3 1.161.037.9
T?rnsj?Ta229913.759.739.122.2 1.584.514.0 Ulva Kv.Ul224697661.244.216.7 3.061.036.0 Vattholma Va224429360.641.220.3 4.871.024.2
Mean65.241.324.9 4.370.425.3
a Actual evapotranspiration calculated by the model.
182CHONG-YU XU Table II.Hypothetical climate change scenarios
Scenario no.
123456789101112131415
T(?C)111112222244444 P(%)–20–1001020–20–1001020–20–1001020
1)Estimate average annual changes in precipitation and temperature using either
GCM results or historical measurements of change,or personal estimates(typ-ically, T=+1,+2and+4?C and P=0,±10%,±20%).
2)Adjust the historic temperature series by adding T and,for precipitation,by
multiplying the values by(1+ P/100).
In practice,these annual changes were distributed during the year by various meth-ods.For example,Nemec and Schaake(1982)and Ng and Marsalek(1992)as-sumed constant distributions of climatic changes,and multiply historical precipit-ation records by constant factors and adjusted historical temperatures by constant increments.Sanderson and Smith(1990)used GISS(Goddard Institute of Space Studies)scenarios of predicted monthly changes in temperature and precipitation.
The general procedure for estimating the impacts of hypothetical climate change on hydrological behaviour has the following stages:
1)Determine the parameters of a hydrological model in the study catchment using
current climatic inputs and observed river?ows for model validation.
2)Perturb the historical time series of climatic data according to some climate
change scenarios.
3)Simulate the hydrological characteristics of the catchment under the perturbed
climate using the calibrated hydrological model.
4)Compare the model simulations of the current and possible future hydrological
characteristics.
In this study,the long-term hydrological response was simulated for climate change scenarios associated with a base case(nominally,present climate conditions),as well as15hypothetical climate change scenarios.The15scenarios are combin-ations of plus1,2and4?C,and minus and plus0,10and20%precipitation (Table II).The changes were applied uniformly to monthly values of the historical time series.This set of scenarios is in general in agreement with the values provided for the Nordic region in other studies(e.g.Kaas,1993a,b).
MODELLING THE EFFECTS OF CLIMATE CHANGE ON WATER RESOURCES183 4.Hydrological Response Analysis
It is very important for water resources managers to be aware of and prepared to deal with the effects of climatic change on stream?ow and related variables. Analysing different hydrologic variables will indicate which hydrologic variable is most affected by changes in the climate.Obviously,stream?ow is essential in order to provide an indication of the extent of impacts of climatic change on water resources.Stream?ow represents an integrated response to hydrologic inputs on the surrounding drainage basin area and therefore affords good spatial coverage. Since it is expected that climatic change will result in a diversity of environmental responses,hydrologic variables other than stream?ow,are also included in this study.The hydrologic variables selected to describe the alternative hydrologies are (a)annual and monthly average catchment runoff,(b)annual and monthly average snow water equivalent over the catchments,(c)annual and monthly average catch-ment evapotranspiration.The results are not discussed for each individual catch-ment,instead,the average values calculated from the25catchments are presented and discussed.The results represent the regional hydrological response to climate change scenarios.
4.1.SNOW WATER EQUIVALENT
The long-term mean monthly snow water equivalent over the region of central Sweden for all the alternative climates is shown in Figure3.A marked reduction in average snow water equivalent for all the alternative scenarios was presented.The combined scenarios of temperature increase by4?C( T=4?C, P=±20,±10, 0%)produced the maximum reduction in snow water equivalent.The maximum value in February reduced from98mm for the present condition to34mm for the scenario T=4?C, P=–20%,and the snow-cover free period increased from?ve months(May to October)to seven months(April to November).These changes re?ect the fact that for temperature increase by4?C,the temperature is the basic factor of snow storage control in relation to precipitation.The other combined scenarios of temperature increase by1and2?C,and for all precipitation changes caused progressive reduction in average snow water equivalent from the wetter to the drier climate.
For the mean annual values,the minimum reduction of snow water equivalent (13%)was produced with the scenario T=1?C and P=20%,and the maximum reduction(76%)with the scenario T=4?C and P=–20%(Figure4).
4.2.EVAPOTRANSPIRATION
The actual evapotranspiration,as calculated by the water balance model,depends on soil moisture and potential evapotranspiration.During the cold and wetter period (October–April)the actual evapotranspiration remained completely unaffected by the precipitation changes(Figure5).During the warm and drier period(May–
184CHONG-YU XU
Figure3.Monthly mean snow water equivalent for the study region calculated from15 climate change scenarios and25catchments.
September)the actual evapotranspiration went up for precipitation increase and dropped for precipitation reduction.
Despite the changes in seasonal distribution of evapotranspiration,the change in annual total evapotranspiration was relatively small with the maximum change of23%compared with the76%for mean annual snow water equivalent changes and52%for mean annual runoff changes(Figure4).
4.3.RUNOFF
The signi?cant changes in seasonal distribution of regional runoff for all15hy-pothetical scenarios are shown in Figure6(a)( T=1?C, P=±20,±10,0%), Figure6(b)( T=2?C, P=±20,±10,0%),and Figure6(c)( T=4?C, P=±20,±10,0%),respectively.
Summer(June,July,August)runoff in14of the15cases dropped considerably in relation to the observed summer runoff.The summer runoff that resulted from
MODELLING THE EFFECTS OF CLIMATE CHANGE ON WATER RESOURCES185
Figure4.Percent change of annual mean snow water equivalent,runoff and evapotranspir-ation for the study region calculated from15climate change scenarios and25catchments.
Refer to Table II for the corresponding temperature change and precipitation change for each scenario number.
T=1?C and P=20%went up a little,re?ecting the fact that the temperature increase was too small to cause early snowmelt and signi?cant increase of evapo-transpiration.The summer runoff drop is more obvious for the driest simulated climates.For a temperature increase by4?C and a precipitation decrease by20%, the summer runoff dropped by about80%.
Winter(December,January,February,March)runoff increased in10of the15 cases combined with any temperature increase and positive or no precipitation change.The maximum winter runoff increase reached80%above the base case runoff was caused by a temperature increase of4?C and a precipitation increase of20%.These two high-value variables raised up the winter runoff through earlier snowmelt and rainfall increase.The maximum winter runoff reduction resulted from temperature increase by1?C and precipitation decrease by20%re?ecting the fact that the snowmelt mechanism could not operate due to the low value of the temperature increase,while the precipitation dropped at the maximum percentage.
In the base case and10of the15combined scenarios of temperature increase by1and2?C and for all precipitation changes,the spring?ow peaked in April, while for the other?ve scenarios( T=4?C),the peak shifted one month earlier (to March).
On the annual base,the temperature increase by1,2and4?C combined with no precipitation change produced annual runoff down by7,12and21%,respectively
186CHONG-YU XU
Figure5.Monthly mean evapotranspiration for the study region calculated from15climate change scenarios and25catchments.
(Figure4).The maximum reduction and increasing of annual runoff were–51and 35%caused by T=4?C, P=–20%and T=1?C, P=20%,respectively.
5.Conclusions
The following conclusions can be reached by this study:(1)The three temperature increases,associated with all precipitation changes,could result in substantial de-
MODELLING THE EFFECTS OF CLIMATE CHANGE ON WATER RESOURCES187
Figure6.Monthly mean runoff for the study region calculated from15climate change scenarios and25catchments.
188CHONG-YU XU creases in average snow accumulation in the study region.(2)The climatic input changes led to a signi?cant redistribution of the stream?ow within a year.The changes caused by scenarios T=4?C combined with precipitation changes showed a strong increase in discharge over the whole winter period and especially at the beginning and the end of the winter.(3)Increased temperature could increase spring and summer actual evapotranspiration,this could counterbalance the effect of a precipitation increase during summer and the change in discharge was the smallest in summer.
It is necessary to make clear at this juncture that the climate change scenarios used in this study should not necessarily be seen as the future climates in the region: They are primarily designed to show the sensitivity to change within a reasonable interval.
Acknowledgements
This research is a part of my work within SWECLIM(Swedish Regional Cli-mate Modelling Programme).I also received research funding from NFR(Swedish Natural Science Research Council).The author appreciates the helpful comments made by the reviewer.
References
Arnell,N.W.:1992,Factors controlling the effects of climate change on river?ow regimes in a humid temperate environment,J.Hydrol.132,321–342.
Gleick,P.H.:1986,Methods for evaluating the regional hydrologic impacts of global climatic changes,J.Hydrol.88,97–116.
Gleick,P.H.:1987,Regional hydrologic consequenses of increases in atmospheric CO2and other trace gases.Climatic Change10,137–161.
Halldin,S.,Gottschalk,L.,van de Griend,A.A.,Gryning,S.-E.,Heikinheimo,M.,H?gstr?m,U., Jochum,A.and Lundin,L.-C.:1998,NOPEX–A northern hemisphere climate processes land, J.Hydrol.212–213(1–4),171–187.
Hostetler,S.W.and Giorgi,F.:1993,Use of output from high-resolution atmopsheric models in landscape-scale hydrologic models:an assessment.Water Resour.Res.29(6),1685–1695. Houghton,J.T.,Meira Filho,L.G.,Callander,B.A.,Harris,N.,Kattenberg,A.and Maskell,K.
(eds):1995,Climate Change1995.The Science of Climate Change:Contribution of Working Group I to the Second Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press.
Kaas,E.:1993a,Greenhouse Induced Climate Change in the Nordic Countries as Simulated with the Hamburg Climate Model.Part1:Direct Model Output,Danish Meteorological Institute, Scienti?c report No.93–2,Copenhagen.
Kaas,E.:1993b,Greenhouse Induced Climate Change in the Nordic Countries as Simulated with the Hamburg Climate Model.Part2:Statistical Interpretation,Danish Meteorological Institute, Scienti?c report No.93–3,Copenhagen.
Lettenmaier,D.P.and Gan,T.Y.:1990,Hydrologic sensitivity of the Sacremento-San Joaquin River Basin,California,to global warming,Water Resour.Res.26,69–86.
Loaiciga,H.A.,Valdes,J.B.,V ogel,R.,Garvey,J.and Schwarz,H.:1996,Global warming and hydrologic cycle,J.Hydrol.174,83–127.
MODELLING THE EFFECTS OF CLIMATE CHANGE ON WATER RESOURCES189 McCabe,G.J.and Ayers,M.A.:1989,Hydrologic effects of climate change in the Delaware River basin,Water Resour.Bull.25(6),1231–1241.
Nash,L.and Gleick,P.:1993,The Colorado River basin and climate change,Rep.EPA230-R-93-009,United States Environmental Agency,Washington,DC.
Nemec,J.and Schaake,J.:1982,Sensitivity of water resources system to climate variation,Hydrol.
Sci.J.27,327–343.
Ng,H.Y.F.and Marsalek,J.:1992,Sensitivity of stream?ow simulation to changes in climatic inputs,Nordic Hydrol.23,257–272.
Panagoulia,D.:1991,Hydrological response of a medium-sized mountainous catchment to climate changes,Hydrol.Sci.J.36(6),525–547.
Sanderson,M.and Smith,J.:1990,Climate change and water in the Grand River,Ontario.Proc.
Of the43rd Annual Conf.Of the Canadian Water Resources Association,May16–18,Penticton, British Columbia,pp.243–261.
Schaake,J.C.and Liu,C.:1989,Development and applications of simple water balance models to understand the relationship between climate and water resources.New Directions for Surface Water Modelling(Proceedings of the Baltimore Symposium,May1989)IAHS Publ.No.181, 343–352.
Thomas,G.and Rowntree,P.R.:1992,The boreal forests and climate.Quarterly Journal of the Royal Meteorological Society118,469–497.
Vehvil?inen,B.and Lohvansuu,J.:1991,The effects of climate change on discharges and snow cover in Finland,Hydrol.Sci.J.36,109–121.
Wilby,R.L.and Wigley,T.M.L.:1997,Downscaling general circulation model output:A review of methods and limitation.Progress in Physical Geography21(4),530–548.
Xu,C.-Y.,Seibert,J.and Halldin,S.:1996,Regional water balance modelling in the NOPEX area: Development and application of monthly water balance models,J.Hydrol.180,211–236.
Xu,C.-Y.:1999a,From GCMs to river?ow:A review of downscaling methods and hydrologic modelling approaches,Progress in Physical Geography23(2),229–249.
Xu,C.-Y.:1999b,Operational testing of a water balance model for predicting climate change impacts.Agricultural and Forest Meteorology98–99,295–304.