Proceedings of International Conference on Finite Element Models, MODFLOW, and More: Solving Groundwater Problems. 1316 September 2004, Karlovy Vary, Czech Republic, Eds. K. Kovar, Z. Hrkal and J. Bruthans.
1
Effects of Climate Change on the Groundwater
Systems of the Grote Nete Catchment, Belgium
S.T. WOLDEAMLAK(1), O. BATELAAN(1) & F. DE SMEDT(1)
(1) Department of Hydrology and Hydraulic Engineering, Vrije Universiteit Brussel, Pleinlaan
2, 1050 Brussels, Belgium
Email: swoldeam@vub.ac.be
Abstract Modeling the impact of climate change on the hydrology of river
basins has been the center of attention in the past decade and is essential to
adopting integrated water management strategies for water supply and flood
mitigation. However, the effect of climate change on groundwater systems has
only recently gained some interest. In this paper the effects of climate change
on the groundwater systems of the Grote Nete catchment, Belgium, covering
an area of 525 km2, was modeled using incremental climate scenario approach.
A wet (greenhouse), cold (NACCT) and a dry climate scenario were used for
the analysis. Classical, low, central and high estimates with three wind-speed
variants were adopted for each scenario to cover model and emission
uncertainties at 80% confidence interval. Water balance components, such as
groundwater recharge, evapotranspiration, and runoff were simulated using the
WetSpass model in conjunction with a MODFLOW groundwater model.
Groundwater discharge intensities and areas were obtained from the
groundwater model. Results for Green house scenario show that, for the
overall year, as well as the winter season, there is an increase in the total
discharge (combined groundwater and surface runoff). In the summer the slow
discharge coefficients showed a decrease. Both discharge areas and intensities
as well as recharge intensities showed an increase. In this scenario wet winters
and drier summers are expected relative to the present situation. Results
obtained for NATCC (cold) scenarios depicted an opposite picture of the
greenhouse scenario, thus relatively drier winters and wetter summers are
expected. The dry scenario showed a decrease in slow and fast discharge
coefficients, discharge intensities and areas, and recharge intensities. As a
result drier conditions for the whole year are expected.
Key words groundwater sytems, recharge, discharge, climate change, GIS,
INTRODUCTION
Water is indispensable for life, but its availability in a desirable quality and quantity is
threatened by many factors, of which climate plays a leading role (IPCC, 1997).
According to IPCC (2001a), Climate is defined as "the average weather in terms of the
mean and its variability over a certain time-span and a certain area". Thus statistically
significant variation of the mean state of the climate or of its variability, lasting for
decades or longer, is referred to as climate change.
There has been scientific evidence that human activities may already be
influencing the climate, rapid changes being recorded for the second half of the 18th
century mostly owing to the industrial revolution (Miller and Tyler, 1990; IPCC,
1996). This is primarily attributed to burning of fossil fuels and changes in land use
and land cover that eventually increase the atmospheric concentrations of greenhouse
gases, which tend to warm the atmosphere (Miller and Tyler, 1990). These changes in
turn are expected to bring about change in global temperature and precipitation and
other climatic variables. Thus threat of floods, change in global soil moisture, increase
Proceedings of International Conference on Finite Element Models, MODFLOW, and More: Solving Groundwater Problems. 1316 September 2004, Karlovy Vary, Czech Republic, Eds. K. Kovar, Z. Hrkal and J. Bruthans.
2
in global mean sea level, change in availability of groundwater, and associated
damages, loom at large, in conjunction with the high water demand and pollution
levels associated with the ever-increasing world population (IPCC, 1997).
Compounding this, there are increasing concerns being raised on the need to predict
future climate conditions and their effect on water resources, so as to ensure constant
water availability and mitigate foreseeable damages caused by water.
There have been many studies relating the effect of climate changes on surface
water bodies (Allen et al. 2003; Gebremeskel, 2003). But very little research exists on
the potential effects of climate change on groundwater, although groundwater is the
major source of drinking water across much of the world and plays a vital role in
maintaining the ecological value of an area (Batelaan et al., 2003; IPCC, 2001b).
Available studies show that groundwater discharge/recharge conditions are a reflection
of the precipitation regime, climatic variables, landscape characteristics and human
impacts such as agricultural drainage and flow regulation (Allen et al., 2003; De Wit,
2001). Thus predicting the behavior of recharge and discharge conditions under future
climatic and other changes is of great importance for integrated water management.
Climatic changes that bring about a change in the amount of effective rainfall and
duration of the recharge season will induce a change in recharge. Substantial
reductions in groundwater recharge near Grenoble, France, was simulated by Bouraoui
et al. (1999), which was almost entirely attributed to increases in evapotranspiration
during the recharge season. According to Sandstrom (1995) a 15% reduction in
rainfall, with no change in temperature, resulted in a 40–50% reduction in recharge. He
infers that small changes in rainfall could lead to large changes in recharge and hence
groundwater resources.
The main objective of this study is to analyze the sensitivity of the water balance
components of the Grote Nete basin towards climate changes, especially the recharge
and groundwater discharge using different climate scenario outputs at a regional scale.
Size, intensities and volume of recharge and discharge are identified using three
dimensional groundwater flow model, MODFLOW, in conjunction with physically
distributed water balance model, WetSpass, integrated in GIS environment.
METHODOLOGY
The methodology consisted of three steps. In the first step climate scenarios were
formulated for the years 2050 and 2100 (Table 1). Based on these scenarios recharge
was simulated with the WetSpass model. Consequently, groundwater system
conditions were simulated by the coupled MODFLOW model for all the developed
scenarios.
Climate scenarios
A climate scenario is defined here as, a plausible future climate that has been
constructed for explicit use in investigating the potential consequences of
anthropogenic climate change (IPCC, 2001a).
The climate scenarios used have been developed by the KNMI. They have been
used in the 4th Dutch National Policy Document on water management (4e Nota
Waterhuishouding NW4) and a study on Dutch Water Management in the 21st Century
(WB21) (Kors et al., 2000). They are based on an incremental scenario approach
where by particular climatic elements are changed by realistic but arbitrary amounts,
and are commonly applied to study the sensitivity of an exposure unit to a wide range
Proceedings of International Conference on Finite Element Models, MODFLOW, and More: Solving Groundwater Problems. 1316 September 2004, Karlovy Vary, Czech Republic, Eds. K. Kovar, Z. Hrkal and J. Bruthans.
3
of variations in climate. In most studies constant changes throughout the year have
been adopted (e.g., Terjung et al., 1984; Rosenzweig et al., 1996). Six climate
scenarios categorized under three general types were used for this study namely:
Greenhouse (wet) scenarios, North Atlantic Thermohaline Circulation Change
(NATCC, cold), and dry scenarios (Table 1). These three scenarios are assumed to be
realistic representations of the controversial views regarding the change in climate for
the future and are discussed in the following paragraphs.
Table 1. Description of modeled climate scenarios.
Modeled Climate
Scenarios
present
Wet low 1
Wet low 2
Wet low 3
Wet central 1
Wet central 2
Wet central 3
Wet high 1
Wet high 2
Wet high 3
Cold (NATTC)
Wet 1 & Cold
Wet 2 & Cold
Wet 3 & Cold
Dry 1
Dry 2
Dry 3
Description
Actual situation at 2000
Greenhouse (wet), low estimate of temperature change, -5% wind-speed
Greenhouse (wet), low estimate of temperature change, no wind-speed change
Greenhouse (wet), low estimate of temperature change, +5% wind-speed
Greenhouse (wet), central estimate of temperature change, -5% wind-speed
Greenhouse (wet), central estimate of temperature change, no wind-speed change
Greenhouse (wet), central estimate of temperature change, +5% wind-speed
Greenhouse (wet), high estimate of temperature change, -5% wind-speed
Greenhouse (wet), high estimate of temperature change, no wind-speed change
Greenhouse (wet), high estimate of temperature change, +5% wind-speed
Sudden changes in the North Atlantic circulation, no wind-speed change
combination of Wet central 1 and Cold
combination of Wet central 2 and Cold
combination of Wet central 3 and Cold
dry scenario, high estimate of temperature change, -10% wind-speed change
dry scenario, high estimate of temperature change, -5% wind-speed change
dry scenario, high estimate of temperature change, no wind-speed change
Greenhouse scenarios (Wet)
The first three scenarios fall under this type of scenario as described by Können et al.
(1997) and Kors et al. (2000). They are results of the global warming and are based on
three classical estimates of temperature change namely, low, central and high
estimates. Model and emission uncertainties are assumed to be covered by the range
between the low and high estimates. Further, it is assumed that the range between low
and high temperature estimates represents 80% confidence interval in the case of
Europe (Van Deursen et al., 2001). The temperature is expected to rise by 1, 2 and 4 oC
for the low, central and high estimates respectively until 2100 with respect to the
baseline period of 1990. Since the temperature increase is assumed to evolve linearly
with time, the increase after 50 years (by 2050) will be half of that of the year 2100
(Table 2). These conditions are assumed to be applicable to the Grote Nete basin (Van
Rossum et al., 2001).
Empirical relations between observed mean temperature and precipitation in the
past show that the precipitation changes are directly proportional to the temperature
changes. Accordingly, the average annual precipitation increases for the year 2100 will
be 3 %, 6 %, and 12 % for the low, central and high estimates respectively. Likewise,
precipitation changes for the year 2050 will be half that of 2100. But there is a seasonal
variation of precipitation increase (Table 1 and 2) unlike temperature change.
Potential evapotranspiration will increase linearly with increase in temperature and
the changes in evapotranspiration for the year 2100 are expected to be 4 %, 8 % and 16
% respectively while that of 2050 being half of the changes in year 2100 (Haasnoot et
al., 1999) (Table 2). The change in evapotranspiration is assumed to be independent of
the season (Van Rossum et al., 2001).
Proceedings of International Conference on Finite Element Models, MODFLOW, and More: Solving Groundwater Problems. 1316 September 2004, Karlovy Vary, Czech Republic, Eds. K. Kovar, Z. Hrkal and J. Bruthans.
4
Table 2. Climate changes for the year 2050 according to proposed climate scenarios.
Precipitation change (%)
Temperature
change
(oC)
Yearly Summer Winter
Wet (low)
+0.5
+1.5
+0.5
+3
Wet (central)
+1
+3
+1
+6
Wet (high)
+2
+6
+2
+12
Cold (NATCC)a
-2
-6
-2
-12
Wet (central) & Coldb
-1
-3
-1
-6
Dry
+2
-10
-10
-10
a. Changes remain the same for the year 2100.
b. Scenario does not exist for 2100.
Climate Scenario
Pot. ET
change
(%)
+2
+4
+8
-8
0-4
+8
Windspeed
change
(%)
+5
+5
+5
0
0
-10 to 0
The observed decadal variability for wind-speed in the past century has been in the
range of +5 % (Van Rossum et al., 2001). Therefore, each of the low, central and high
estimate scenarios will have 3 windspeed variants of –5, 0, and +5 % change in the
average windspeed (Table 1 and 2).
NATCC scenario (Cold)
This scenario is expected when the climate change is induced by a sudden change in
the North Atlantic thermohaline circulation resulting in the cooling of the ocean, and
thus cooling of the atmosphere (Alcamo et al., 1994). Assuming that the present
condition prevails and ignoring the effect of greenhouse scenario, the ocean is
expected to cool by 4 oC for the worst-case scenario, with an associated decrease of
2 oC in the study area (Van Rossum et al., 2001). The cooling is expected to happen
within a range of 5 – 10 years between the years 2000 and 2050, thus its effects should
be superposed to the climate conditions prevailing until the year 2050.
Putting this into consideration two scenarios where analyzed. In the first, the effect
of climate changes induced by the sudden change in the NATC was considered
ignoring the effect of global warming, while in the second case this change was
superposed to the change due to the central estimate of the greenhouse scenario for the
year 2050 (Table 2). It is assumed that temperature and precipitation changes remain
coupled. Thus, decrease in temperature will induce a decrease in precipitation by an
amount inversely proportional to the greenhouse scenarios. The same logic holds true
for evapotranspiration. Lastly, the windspeed is expected to remain the same as the
present situation (Table 2). In the second case, the global warming of the central
estimate of the greenhouse scenario will balance half the effect of the cooling of the
NATCC scenario by the year 2050 and completely balance it by the year 2100. So,
scenario analysis is done only for the year 2050 by considering the average effect of –
1oC decrease in temperature and the associated precipitation and evapotranspiration
changes.
Dry scenarios
The dry scenarios represent climate change where the temperature and precipitation
changes are uncoupled, while the relationship between temperature and
evapotranspiration is preserved. In addition there is a lower probability of having
strong winds, expressed in three wind-speed variants, of 0, 5 and 10% of decrease in
windspeed. On the other hand, the precipitation shows a decrease of 10% for all
summer, winter and yearly cases both for 2050 and 2100.
Proceedings of International Conference on Finite Element Models, MODFLOW, and More: Solving Groundwater Problems. 1316 September 2004, Karlovy Vary, Czech Republic, Eds. K. Kovar, Z. Hrkal and J. Bruthans.
5
Recharge estimation and groundwater modeling
The groundwater system was analyzed by applying the USGS modular threedimensional finite difference groundwater model (MODFLOW) (Harbaugh and
McDonald, 1996). Batelaan et al. (2003) describes the present conditions of the study
area for the groundwater model, combined DRAIN-SEEPAGE (Batelaan and De
Smedt, 2004) for groundwater discharge simulation, and calibration on piezometric
and river discharge data.
The groundwater model is coupled with WetSpass model for groundwater
recharge estimation (Batelaan and De Smedt, 2001). The model is a quasi physically
distributed seasonal water balance model, which takes detailed soil, landuse, slope,
groundwater depth, and hydro-climatological distributed maps with associated
parameter tables into account. The water balance equation for estimating groundwater
recharge is given as:
P = R – S – ET
Where the groundwater recharge, R, is calculated indirectly from the precipitation (P),
evapotranspiration (ET), and surface runoff (S), each having [L/T] dimensions. The
model uses seasonal (summer and winter) GIS input grids of the previously mentioned
inputs that are used in calculating recharge values. Since one of the inputs required for
WetSpass model is the groundwater depth data from MODFLOW, an interface has
been developed in an ArcView GIS platform to loosely couple the two models,
facilitating exchange of data between the two models (Kassa, 2001). Initially, the
coupled WetSpass-MODFLOW model was run for the present situation, yielding a
seasonal as well as yearly groundwater recharge output. Consequently, the same
procedure was followed for each of the constructed scenarios.
APPLICATION AND DISCUSSION
Study area
The study area is located about 60 km North-East of Brussels and covers the GroteNete basin with a size of 525 km2 (Fig. 1). It is part of the Central Campine region
with a moderate rolling landscape cut by the Grote-Nete River and its many tributaries,
resulting in long stretched hills, very slightly elevated interfluves and broad swampy
valleys. The base of the sandy aquifer system is formed by heavy clays of the Boom
Formation. The basin is generally composed of marine sediments of the Tertiary and
Quaternary age, with a thickness of up to 90 m (Batelaan et al., 2003).
Topography ranges from 13 – 73 m, with an average of 32 m above sea level, the mean
slope being 0.24%. The transmissivity of the aquifer system ranges from an average of
about 500 m2 day-1 on the western part to an average of about 3000 m2 day-1 on the
eastern part of the area. Although the dominant soil type is sand, covering 67 % of the
area, loamy sand (24%), silty loam (7%), and sandy loam (1%) are also found in the
valleys (Fig. 2). The landuse types of the area is comprised of: 28% agriculture, 18%
deciduous forest, 10% coniferous forest, 3% mixed forest, 15% grass lands, 5%
heather, 2% open-water bodies and about 19% of built-up area (Fig. 3). The average
yearly precipitation of the area ranges from 743 to 801 mm with an average of about
764 mm, while the average summer and winter precipitation are respectively 392 and
Proceedings of International Conference on Finite Element Models, MODFLOW, and More: Solving Groundwater Problems. 1316 September 2004, Karlovy Vary, Czech Republic, Eds. K. Kovar, Z. Hrkal and J. Bruthans.
6
372 mm. The average yearly potential evapotranspiration is 670 mm. The area has a
moderate average winter and summer temperatures of 5 and 14.1 oC respectively, with
wind speed of 3.84 and 3.27 m s-1.
80000
120000
160000
200000
40000
80000
120000
160000
200000
240000
160000
160000
200000
200000
240000
240000
40000
240000
N
Legend
Study area
Flanders
10
0
10
20 km
Fig 1. Study area location
195000
200000
205000
210000
215000
190000
195000
200000
205000
210000
215000
220000
200000
200000
205000
205000
210000
210000
215000
215000
190000
Legend
Soil
sand
loamy sand
Project boundary
220000
N
sandy loam
silty loam
loam
2
0
2 km
Fig 2. Soil types
195000
200000
205000
210000
215000
190000
195000
200000
205000
210000
215000
220000
200000
200000
205000
205000
210000
210000
215000
215000
190000
Legend
Project boundary
220000
N
Landuse types
Agriculture
Forest and heather
Fig 3. Landuse types
Builtup area
Open water
2
0
2 km
Proceedings of International Conference on Finite Element Models, MODFLOW, and More: Solving Groundwater Problems. 1316 September 2004, Karlovy Vary, Czech Republic, Eds. K. Kovar, Z. Hrkal and J. Bruthans.
7
Calibration of the groundwater modeling
The groundwater model for the present condition was calibrated on the basis of 38
piezometers in the area. Fig. 4 shows that with a correlation of 0.99 there is a good
agreement between simulated and observed heads. With a mean absolute error (MAE)
of 0.43 m, root means square error (RMSE) of 0.61 m, Nash-Sutcliffe model efficiency
(NS) of 0.98 and a coefficient of determination (CD) of 0.99 it is shown that the model
has a good ability of reproducing measurements and the observed and simulated heads
are proportionally scattered around the mean of observations.
50
Simulated phreatic level (m)
45
R2 = 0.99
40
35
30
25
20
15
10
10
15
20
25
30
35
40
45
50
Measured phreatic level (m )
Fig 4. Comparison of measured and calculated groundwater levels
Climate scenario analysis for present situation
General hydrological conditions
The results of the WetSpass-MODFLOW model for the present hydrological situation
of the Grote Nete catchment were a set of yearly, summer, and winter distributed
evapotranspiration, recharge and runoff grids. The simulated distributed yearly
recharge map is given in Fig. 5. The basin averaged values are given in Fig. 7-12.
195000
190000
195000
200000
205000
200000
205000
210000
215000
220000
200000
200000
205000
205000
210000
210000
215000
215000
190000
Legend
Project boundary
Recharge (mm/y)
<0
0 - 50
50 - 100
100 - 150
210000
150 200 250 300 350 -
215000
200
250
300
350
410
Fig 5. Wetspass calculated yearly recharge values
220000
N
2
0
2 km
Proceedings of International Conference on Finite Element Models, MODFLOW, and More: Solving Groundwater Problems. 1316 September 2004, Karlovy Vary, Czech Republic, Eds. K. Kovar, Z. Hrkal and J. Bruthans.
8
These results show that summer season is characterized by high evapotranspiration,
low recharge and runoff, respectively 85.7, 9.7, and 5.5% of the precipitation. In
winter the runoff remains almost the same at 6%, but the evapotranspiration goes down
to 29.9% owing to the low temperatures, as a result more water is available for
recharge (64.1%). This implies that major part of the recharge to the aquifer (86.2%)
occurs during the winter, while only 13.8% occurs during the summer. It can also be
inferred that 81.4% of the total average yearly discharge is the contribution of winter
groundwater recharge and runoff. But there is a wide range of time lag before the
recharged precipitation is discharged to the stream networks or wetlands as shown by
Batelaan et al. (2003).
195000
200000
205000
210000
215000
190000
195000
200000
205000
210000
215000
220000
200000
200000
205000
205000
210000
210000
215000
215000
190000
Legend
Project boundary
220000
N
Discharge (mm/d)
0-2
2-4
4-6
6-8
8 - 10
> 10
2
0
2 km
Fig 6. MODFLOW calculated yearly discharge values
Groundwater systems (discharge and recharge conditions)
Groundwater discharge areas and associated intensities obtained from the groundwater
model are represented by Fig. 6. The discharge areas cover a total of 95.1 km2
(18.6 %) of the total area with an average discharge intensity of 217 mm y-1. In
addition to this yearly, summer and winter fast (surface runoff) and slow
(groundwater) discharge coefficients were calculated as percentages of the average
precipitation of the basin. Slow discharge coefficients were calculated from WetSpass
calculated average groundwater recharge values by assuming all the groundwater
recharge reaches the river or the wetland areas at a time ranging between few days to
hundreds of years as demonstrated by Batelaan et al. (2000).
The average recharge intensity of the area is 277 mm y-1 with a minimum and
maximum groundwater recharges of –375 mm y-1 and 408 mm y-1 respectively (Table
5). In valley areas the recharge is negative or negligibly small due to the shallow
groundwater depth and evapotranspiration from the groundwater by phreatophytes,
while high recharge values are recorded for areas dominated by sandy soils, flat areas,
and grass or forest covered areas. Similar findings were reported by De Smedt and
Batelaan (2001).
Proceedings of International Conference on Finite Element Models, MODFLOW, and More: Solving Groundwater Problems. 1316 September 2004, Karlovy Vary, Czech Republic, Eds. K. Kovar, Z. Hrkal and J. Bruthans.
9
Climate scenario analysis for future situation
10
5
precipitation
ET
recharge
runoff
-15
-20
Climate scenario (2050)
ET
recharge
runoff
Climate scenario (2050)
Dry
Wet &
Cold
Cold
Wet
(central)
Fig 9. Summer recharge changes (2050)
% change relative to present
(yearly)
Dry
Cold
Wet
central
recharge
runoff
-15
-20
-25
Climate scenario (2100)
precipitation
ET
recharge
runoff
Climate scenario (2050)
Fig 11. Yearly recharge changes (2050)
Dry
Cold
Wet
(central)
50
0
precipitation
-50
ET
recharge
-100
runoff
-150
-200
Climate scenario (2100)
Fig 10. Summer recharge changes (2100)
Dry
precipitation
10
5
0
-5
-10
-15
-20
-25
-30
-35
ET
Cold
60
40
20
0
-20
-40
-60
-80
-100
-120
precipitation
5
0
-5
-10
Fig 8. Winter recharge changes (2100)
Dry
Wet &
Cold
Cold
Wet
(central)
% change relative to present
(summer)
Fig 7. Winter recharge changes (2050)
20
15
10
Wet
(central)
-10
% change realtive to present
(summer)
-5
% change relative to present
(yearly)
0
% change relative to present
(winter)
Dry
Wet &
Cold
Cold
Wet
(central)
% change relative to present
(winter)
The results of the hydrological situation for the different scenarios considered are
presented in Fig. 7-12. These figures show the changes of the average yearly and
seasonal precipitation, evapotranspiration, groundwater recharge and surface runoff
with respect to the corresponding actual situations for all climate scenarios.
20
10
precipitation
0
ET
-10
recharge
-20
runoff
-30
-40
Climate scenario (2100)
Fig 12. Yearly recharge changes (2100)
Greenhouse scenarios
General hydrological conditions
For all cases of the greenhouse scenarios an increase in the average runoff,
groundwater recharge, and evapotranspiration was observed for seasonal and yearly
cases, with the exception of summer groundwater recharge, which showed a reduction
of up to 89.5% compared to the present situation. Within each greenhouse scenario
Proceedings of International Conference on Finite Element Models, MODFLOW, and More: Solving Groundwater Problems. 1316 September 2004, Karlovy Vary, Czech Republic, Eds. K. Kovar, Z. Hrkal and J. Bruthans.
10
highest changes (increase or decrease) of hydrological parameters was observed, where
the wind-speed is expected to decrease by 5% from the present situation, though the
difference between the wind-speed variants is not that much significant.
For a closer look at the greenhouse scenarios, the WetSpass results of wet central
scenario with no windspeed change, was compared with the results of the actual
situation (Fig. 4, 6, and 8). For a 6% (+45.8 mm) increase in the yearly precipitation
7% increase in evapotranspiration and 6.2% increase in groundwater recharge were
recorded. However, the runoff had almost a doubled increase (11.2 %). Although the
percentage increase for runoff is higher than the other water balance components most
of the additional rainfall is either evapotranspired or recharged to the aquifer. In
winter, where the precipitation increased by 12% and the temperature increased by
40% relative to the prevailing conditions, the increase in evapotranspiration,
groundwater recharge, and runoff were respectively 8.6%, 13.4% and 13.5%. In this
case, most of the additional water from precipitation is recharged to the aquifer. In
summer, the changes were 6.4%, -39.1%, and 8.8% of the same hydrological
parameters for an increase of 2% in precipitation and 14% for temperature.
It is clear that the effects of temperature increase in the summer outweigh the
effects of relatively smaller increase in precipitation. More water is lost via
evapotranspiration than that added by precipitation or recharge. This implies that
groundwater replenishment of the area is taking place in the winter season.
Groundwater systems (discharge and recharge conditions)
recharge flux
-5
-10
-15
-20
gw. dis. flux
-25
-30
-35
ET
gw. dis. area
precipitation
Climate scenario (2050)
Fig 11. Yearly discharge changes (2050)
Dry
Cold
Wet
(central)
% change relative to present
10
5
0
Dry
Wet &
Cold
Cold
Wet
(central)
% change relative to present
As stated briefly in the opening parts of this section there is a general increase in the
discharge areas, intensity and volume for all greenhouse scenarios investigated.
Summarized results pertaining to discharge conditions are presented in and Fig. 13 and
14. The results show that the discharge areas are expected to increase by 1.1% at the
end of 2050 (for wet central scenario) and then goes on to increase to 11.1% by the end
of 2100. Simultaneously, there is an associated increase in the volume and intensity of
the groundwater discharge, which could create an imbalance in the fragile riverine and
wetland ecosystems as well as increase the possibility for flooding.
15
10
5
0
-5
-10
-15
-20
-25
-30
-35
-40
recharge flux
gw. dis. flux
gw. dis. area
precipitation
ET
Climate scenario (2100)
Fig 12. Yearly discharge changes (2100)
In contrast to the discharge areas, there is no significant change (increase or
decrease) for groundwater recharge areas. On the other hand, there is appreciable
increase in the recharge intensity and volume as a result of the increase in
precipitation. The lowest and highest average annual increase in the recharge volume
to the aquifers for all the greenhouse scenarios ranges from 0.2% to 7.5%. As a result
of this the phreatic water level of the basin is expected to rise by an average of 7.5 cm
increasing the risk of possible flooding.
Proceedings of International Conference on Finite Element Models, MODFLOW, and More: Solving Groundwater Problems. 1316 September 2004, Karlovy Vary, Czech Republic, Eds. K. Kovar, Z. Hrkal and J. Bruthans.
11
Cold (NATCC) and Dry scenarios
General hydrological conditions
Results of WetSpass run for the identical cold scenarios (NATCC) of 2050 and 2100
show that the relative changes of hydrological parameters with respect to the present
situation have the same magnitude as the central estimate of the greenhouse scenarios
but with the opposite sign (negative). This is obvious since the input parameters of the
former are just the negative of the latter one (Table 1). The same holds true for the
combined wet central and cold scenario, except that the magnitudes for this scenario
are half that of the cold scenario because of the effects of greenhouse scenario.
The situation with the dry scenarios was more dramatic and the changes in
hydrological parameters were more pronounced. Only evapotranspiration shows an
increase, while both groundwater recharge and surface runoff showed a decrease for all
seasons (Fig. 7-12). For a closer look at the dry scenarios, the WetSpass results of dry
scenario with -5% windspeed change, was compared with the results of the actual
situation. For a 10% decrease in the yearly precipitation, 9% decrease of
evapotranspiration, 38.6% groundwater recharge and 16.5% decreases of surface
runoff were recorded (Fig. 3 and 4). In winter, where the precipitation increased by
10% and the temperature by 80%, the increase in evapotranspiration was 12.5%, while
decrease in groundwater recharge, and runoff were 19.7% and 18% respectively (Fig. 5
and 6). In summer, the changes were 7.9%, -157%, and –14.9% f the same
hydrological parameters for a decrease of 10% in precipitation and 28 % increase in
temperature (Fig. 7 and 8). The extremely high decrease in groundwater recharge in
summer shows that a net evapotranspiration of groundwater will occur starting before
the year 2050. Since the average precipitation for this scenario decreases, the increase
in evapotranspiration takes place at the expense of groundwater recharge and surface
runoff.
Groundwater systems (discharge and recharge conditions)
Both cold and dry scenarios depict opposite picture of that of the greenhouse scenarios
in terms of discharge and recharge conditions. The discharge areas will decrease by
2.8% for the combined wet and cold scenarios climate scenario in 2050 and reaches an
alarming decrease of 34.8% for dry scenario with -10% windspeed change (Fig. 13 and
14). This, in association with a huge decrease in volume of discharged water ranging
between –2.8% and –34.7% could be detrimental for the hydrologic cycle of the basin
in general and wetland and river ecosystems in particular.
There is an appreciable decrease in the recharge intensity and volume as a result of
decrease in precipitation. The average annual decrease in the recharge volume to the
aquifers ranges from 1.8% for the combined cold and dry scenario in 2050 to a
massive 40% for dry scenario with -10% windspeed change in 2100. As a result of this
the phreatic water level of the basin is expected to decrease by an average of 49 cm
and could lead to a possible substantial water shortage.
CONCLUSION
Results of present situation
The results obtained for the general hydrologic situation, discharge and recharge
conditions show a good agreement with what have been obtained from previous works
Proceedings of International Conference on Finite Element Models, MODFLOW, and More: Solving Groundwater Problems. 1316 September 2004, Karlovy Vary, Czech Republic, Eds. K. Kovar, Z. Hrkal and J. Bruthans.
12
on the basin by Batelaan et al. (2003). The groundwater model for the present situation
showed a good agreement between the simulated and observed heads.
Greenhouse (Wet) scenarios
Increase in the average precipitation and temperature induced increase in the water
balance components of groundwater recharge, surface runoff, and evapotranspiration
for the yearly and seasonal cases, except for the average summer groundwater
recharge, which showed a decrease relative to the present situation. This decrease is
attributed to increase in temperature, which in turn induces higher evapotranspiration
rates that exceed the recharge induced by the increase in precipitation.
There is relatively small increase in the phreatic level. However, the total
groundwater discharge increased substantially, with the exception of summer season.
Higher difference between summer and winter recharge causes higher seasonal
fluctuation of the groundwater system. Discharge areas, yearly discharge intensities
and volume, and yearly recharge intensities also increased substantially especially for
the high estimates. Based on these scenarios, there is an increased risk of flooding for
the catchment. The fragile wetland and riverbank ecosystems could also be in danger,
as the possibility of inundation looms larger.
Cold (NATCC) scenarios
The results obtained for this scenario are just the reverse of the greenhouse scenarios
(central estimate). Accordingly, evapotranspiration, surface runoff and recharge show
a yearly, and winter decrease and only summer recharge shows an increase. Contrary
to that of the wet scenarios, this happens because the decrease in temperature
overwhelms the decrease in precipitation and thus inducing colder summers that are
characterized by having higher recharge than the evapotranspiration. Results of
discharge and recharge conditions indicate that there is a general lowering of the
groundwater level, and the discharge in the rivers is expected to decrease. There is a
possibility of water shortage, and also the fragile wetland and riverbank ecosystems
would be affected from lack of water.
Dry scenarios
The result of dry scenarios follows the same trend as that of cold scenario, with the
exception that the evapotranspiration increases. This fact, added to the decrease of
precipitation, reduces the summer recharge to values of more than 100 % with respect
to the actual situation. As a result the phreatic level is expected to decrease by as much
as half a meter. In addition, both slow and fast discharges in all seasons showed a
decrease, specially the slow discharge coefficients in the summer case. This indicates
that a net evapotranspiration is occurring in the area, way before the prediction years
(2050 and 2100). The consequences could be drastic, as this could reduce the water
availability (especially in the summer) to a crucially low level for both aquatic life and
human uses.
Generally, although hydrological models have uncertainties, the greatest
uncertainties in the effects of climate on hydrological outputs arise from uncertainties
in climate change scenarios, as long as a conceptually sound hydrological model is
used (IPCC, 2001b).
Feasible integrated water management strategies could only be adopted if these
climate scenario analyses are complimented with landuse scenarios, since the land use
is changing owing to different economical, social and environmental factors. In
addition, as climate change models are characterized by high uncertainty, stochastic
approaches should be made for a refined delineation of discharge areas that can be
Proceedings of International Conference on Finite Element Models, MODFLOW, and More: Solving Groundwater Problems. 1316 September 2004, Karlovy Vary, Czech Republic, Eds. K. Kovar, Z. Hrkal and J. Bruthans.
13
expressed in terms of probability. The results of such work could yield a range of
options for making decisions of appropriate landuse development planning for the
catchment.
References
Alcamo, J., G.J. van den Born, A.F. Bouwman, B.J. de Haan, K. Klein, Goldewijk, O. Klepper, J. Krabec, R. Leemans,
J.G.J. Olivier, A.M.C. Toet, H.J.M. de Vries, and H.J. van der Woerd, 1994. Modeling the global society-biosphereclimate system, part 2: computed scenarios. Water, Air, and Soil Pollution, 76, 37–78.
Allen, D. M., D. C. Mackie, and M. Wie, 2003. Groundwater and climate change: a sensitivity analysis for the Grand
Forks aquifer, southern British Columbia, Canada, Hydrogeology Journal, pp. 1 – 47.
Batelaan, O. and De Smedt, F., 2004. SEEPAGE, a new MODFLOW DRAIN Package. Ground Water, vol. 42, no. 4, p.
576-588.
Batelaan, O., F. De Smedt, and L. Triest, 2003. Regional Groundwater Discharge: Phreatophyte Mapping, Groundwater
Modeling and Impact Analysis of Land-Use Change, Journal of Hydrology 275, pp 86-108.
Batelaan, O. and De Smedt, F., 2001. “WetSpass: a flexible, GIS based, distributed recharge methodology for regional
groundwater modelling”, Gehrels, H., Peters, J., Hoehn, E., Jensen, K., Leibundgut, C., Griffioen, J., Webb, B. and
Zaadnoordijk, W-J. (Eds.), Impact of Human Activity on Groundwater Dynamics, IAHS Publ. No. 269, pp. 11-17.
Batelaan, O., Asefa, T., Van Campenhout, A., and De Smedt, F., 2000. Studying the impact of land-use changes on
discharge and recharge areas. In: Verhoest, N.E.C., Van Herpe, Y.J.P. and De Troch, F.P. (eds), Book of abstracts of
European Network of Experimental and Representative Basins (ERB) Conference: Monitoring and Modeling
Catchment Water Quantity and Quality, Ghent, Belgium, 27-29 September, 2000, pp. 215-218.
Bouraoui, F., G. Vachaud, L.Z.X. Li, H. LeTreut, and T. Chen, 1999. Evaluation of the impact of climate changes on water
storage and groundwater recharge at the watershed scale. Climate Dynamics, 15, 153–161.
De Smedt, F. and Batelaan, O., 2001. The impact of land-use changes on the groundwater in the Grote Nete river basin,
Belgium. Proceedings of the 3rd International Conference on ‘Future Groundwater Resources at Risk’, Ed. L.
Ribeiro, Lisbon, Portugal, June, 25-27, 2001: 151-158 pp.
De Wit, M.J.M (ed), 2001. Effect of Climate change on the Hydrology of the River Meuse, Environmental Sciences report
104, Wageningen University, The Netherlands.
Gebremeskel, S. 2003. Modelling the effect of climate and land-use changes on the hydrological processes: an integrated
GIS and distributed modeling approach. PhD thesis, Vrije Universiteit Brussel.
Haasnoot, M., Vermulst, J.A.P.H. and Middelkoop, H., 1999. Impacts of Climate Change and Land Subsidence on the
Water Systems in the Netherlands. Terrestrial areas, NRP project 952210, RIZA report 99.049, RIZA.
Harbaugh, A.W. and McDonald, M.G., 1996. User's documentation for MODFLOW-96, an update to the U.S. Geological
Survey modular finite-difference ground-water flow model. Open-File Report 96-485: 56.
IPCC, 1996. Houghton, J.T., L.G. Meira Filho, B.A. Callander, N. Harris, A. Kattenberg, and K. Maskell (eds.) Climate
Change 1995: 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, Cambridge, United Kingdom and
New York, NY, USA, 572 pp.
IPCC, 1997. Watson, R. T., M. C. Zinyowera, R.H. Moss, and D.J. Dokken (eds.) The regional Impact of Climate Change:
An Assessment
of Vulnerability –
A Special Report of IPCC Working Group II.
http://www.grida.no/climate/ipcc/regional/ index.htm.
IPCC, 2001a. Climate Change 2001: The Scientific Basis - Contribution of Working Group I to the Third Assessment
Report of the Intergovernmental Panel on Climate Change (IPCC), Cambridge University Press, Cambridge, United
Kingdom and New York, NY, USA, 944 pp. http://www.grida.no/climate/ipcc_tar/wg / index.htm
IPCC, 2001b. Climate Change 2001: Impacts, Adaptation and Vulnerability - Contribution of Working Group II to the
Third Assessment Report of the Intergovernmental Panel on Climate Change (IPCC), Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 517 pp. http://www.grida.no/ climate/ipcc_tar/wg2/
index.htm
Kassa, Y., 2001. Development and Application of a WetSpass-MODFLOW model for ArcView. Masters Thesis, InterUniversity Program in Water Resources Engineering, KUL-VUB, Belgium.
Können, G.P., Frasnse, W. and Mureau, R., 1997. Meteorology for the 'Fourth Note on Water Management (NW4), (in
Dutch), KNMI report, KNMI.
Kors, A.G., Claessen, F.A.M., Wesseling, J.W. and Können, G.P., 2000. Scenario's externe krachten voor WB21
[Scenarios concerning external forces for WB21, in Dutch], RIZA/WL and KNMI publication, KNMI.
Miller, Jr., and G. Tyler, 1990. Living in the environment: An introduction to environmental science, 6th ed., Wadsworth
Publishing Company, Belmont, CA. 620 pp.
Rosenzweig, C., J. Phillips, R. Goldberg, J. Carroll, and T. Hodges, 1996. Potential impacts of climate change on citrus
and potato production in the U.S. Agricultural Systems, 52, 455-479.
Sandstrom, K., 1995. Modeling the effects of rainfall variability on groundwater recharge in semi-arid Tanzania. Nordic
Hydrology, 26, 313–330.
Terjung, W.H., D.M. Liverman, J.T. Hayes, P.A. O'Rourke, and P.E. Todhunter, 1984. Climatic change and water
requirements for grain corn in the North American Great Plains. Climatic Change, 6, 193-220.
Van Deursen, W.P.A., Middelkoop, H., Kwadijk, J.C.J., Haasnoot, M., Buiteveld, H., van Asselt, M.B.A., van ‘t Klooster,
S.A., Rotmans, J., van Gemert, N., Können, G.P., Bronstert, A., Fritsch, U., Niehoff, D., De Smedt, F., Batelaan, O.,
van Rossum, P., Bárdossy, A. and Zehe, E., 2001. Development of flood management strategies for the Rhine and
Meuse basins in the context of integrated river management. Executive summary of the IRMA-SPONGE project
3/NL/1/164 / 99 15 183 01, Rotterdam, Utrecht, 38 pp.
Proceedings of International Conference on Finite Element Models, MODFLOW, and More: Solving Groundwater Problems. 1316 September 2004, Karlovy Vary, Czech Republic, Eds. K. Kovar, Z. Hrkal and J. Bruthans.
14
Van Rossum, P., Batelaan, O., Gebremeskel, S., and De Smedt, F., 2001. Discharge Coefficients for the Kikbeek Sub-basin,
a Brook Sub-basin at the Belgian Side of the River Border Meuse (Grensmaas), VUB Contribution to the Final
Report of the IRMA-SPONGE Subproject 2: Development of Flood Management Strategies for Rhine and Meuse
Basins in the Context of Integrated River Management. 58 pp. + app.