Estimation of Groundwater Recharge by The Water Balance Analysis Using DA WAST Model
Estimation of Groundwater Recharge by The Water Balance Analysis Using DA WAST Model
Estimation of Groundwater Recharge by The Water Balance Analysis Using DA WAST Model
1. INTRODUCTION
Development of substitutable water resources became an impending assignment in efficient
utilization plan of water resources, since surface water development reaches the limit.
Various researches had been continued over artificial rain, seawater desalination, and
groundwater development for the development of substitutable water resources. When
economical efficiency is taken into account, groundwater development can be considered as a
principal plan of substitutable water resources.
Groundwater development preferably as substitutable water resources, however, can cause
natural disaster without solving the following prerequisites or problems.
First, groundwater is difficult to perceive the shortage and its utilization plan should be
established under the annual groundwater recharge rate in order to prevent natural disaster.
Second, long-term utilization policy of groundwater is established on the basis of the
groundwater recharge rate. However, the groundwater recharge rate is produced by the
surface water hydrological analysis based upon weather data and/or foreign statistical
techniques. Therefore, the authoritativeness of national water resources plan is insufficient.
Third, at present, the groundwater recharge rate in Korea is estimated by unverified methods.
Fourth, from the analysis result of future water supply-demand, Korea is classified as the 21th
century water resources insufficiency country and thereafter, the government is promoting the
428
\.. _\ \
\~ -
o
water resources plan. Therefore, groundwater is to be preserved and managed as one of the
main pure water resources.
At present, there are several researches related to infiltration rate for soil layers, however,
there is no reliable research about groundwater recharge. Therefore, there should be some
cornerstones of estimation for groundwater recharge.
This research theoretically establishes the water balance equation for groundwater recharge
from the basic equation of DAWAST model (1992, Tai-cheol Kim), and experimentally
observes the values of model parameters from an experiment site, and then physically verifies
between the experimented and observed results to develop a reliable technique for
groundwater recharge. Furthermore, the research wishes to present suitable estimation
technique for groundwater recharge indirectly by verifying the groundwater recharge of
DAWAST model with observed groundwater recharge.
429
\~ -o
met
3. ESTIMATION OF GROUNDWATER RECHARGE BY DAWAST MODEL
con
3.1 DAWAST Model Thi
run
Since phenomena of rainfall and runoff are very complicated according to watershed fro
characteristics, forecasting natural phenomenon is accomplished by simplifying hydrologic Th
reaction to various kinds mathematical expression in order to apply them to actual watershed. CE
Specially, in practical manners, simplifying weather properties and input data of watershed
(U
characteristics and also simplifying the structures and parameters of model can make it pal
possible to develop a lumped, conceptualized hydrological modeL
wa
This model is based on the conceptualized model and is consisted of 3 sub-models: the
optimized model, the generalized model and the regionalized watershed modeL. 3.3
The optimized model is applied to the watershed that has an observed runoff data by
compensating the model parameters. The generalized model can be applied to the watershed ..
1
So
without the observed runoff data by forecasting parameters from watershed characteristics.
by
The regionalized watershed model can estimate daily runoff for the watershed that has no Ia)
investigation about the watershed characteristics or has difficulty to derive them by converting
the parameters from the optimized modeL
wI
3.2 Conceptualization of DAWAST Model
As shown in Fig.2, under the supposition of a closed system that is constant with time, the
DAWAST model simplifies the system of the watershed in 3 water layers such as ground
surface, unsaturated layer and saturated layers in order to conceptualize hydrological reactions
SI
of rainfall-runoff in the watershed and executes hydrological water-balance analysis in daily
d!
basis.
(CE') ~ ~P
ETS_J-. . ,; l,:'--.----~--.Js~.
. t
T
Uma~~~~~~____~____~=============-~~I
Distributed
Surface Flow (QSD) 3
Lma~ :.l
~
~
.
• , f L
. .' ~,
430
method mentioned above and then the daily runoff is estimated by summing up the each
component.
This model calculates surface runoff, interflow and base runoff in rainy season and base
runoff only in non-rainy season, respectively, and hence calculates total runoff consecutively
from the beginning day to the closing day of the simulation.
The model is consisted of 5 water balance parameters that are UMAX, LMAX, FC, CP, and
CE shown on the above figure and 3 tracing parameters that are daily runoff distribution rate
(Ui), interflow and base runoff recession curve coefficients (kl and k2), respectively. These
parameters are corrected by optimization technique and inputted. Input data for the model are
watershed area, daily rainfall, and daily evaporation and output data is daily runoff at the site.
(3)
431
\~ a
..,.
In this model, SCS retention parameter S has been changed the concept and newly defined
from maximum storage capacity to available storage capacity of unsaturated layer (Sa) so that There
the parameter may be suitable for continuous processing of watershed soil moisture and andn
continuous simulation of the model. Hence, in equation (4), parameter S is replaced into from
aquation (5) to equation (6)
when
QS=(P-0.2S)2/(P+0.8S) (4)
Then, watershed evapotranspiration coefficient (CE) can be derived from the above equations when
(7),(8) and (9).
Finall
3.5 Estimation of Groundwater Recharge Rate by DAWAST Model
Groundwater recharge consists of unconfined and confined groundwater recharge, and it
In th(
depends on topographical factor, meteorological factor, land utilization factor, geological grout
factor of the watershed, and characteristics of aquifer. grout
However, most geological structure in Korea including Chojeong watershed can not be clearly
distinguished between unconfined and confined groundwater. Therefore, there used to no
classification of groundwater type for groundwater recharge estimation. On the other hand, for 4.R1
the case of no observed data, watershed hydrological model such as DAWAST can estimate
4.1 I
the variation and recharge amount of groundwater indirectly from water balance analysis.
Where, groundwater recharge (RG) is to say in 2 aspects: the groundwater recharge in narrow 4.1
sense that is groundwater storage (ASg ) stored by deep percolation and the groundwater To IT
recharge in broad sense that includes base runoff (QB) returning to rivers from infiltration. 50 c
This relation is shown in equation (10) mea~
432
\~ o..,
RG=QB+ASg (10)
Therefore, the water balance including groundwater in a watershed is defined in equation (11)
and represented in Fig. 3.
ASg P-Q ET+Q g ASs (11)
where, ASg : variation of groundwater storage, P : Precipitation,
ET : watershed evapotranspiration, Q : runoff (direct runoff + base runofl),
ASs : soil moisture change in watershed,
Qg: groundwater inflow (or outflow) from (to) other watersheds.
ET p
;ed
ent
Qg2
Fig. 3 Watershed Map of Water Balance
Supposing that the groundwater inflow and outflow are the same, then Qg = O. If soil moisture
change in watershed (ASs) and variation of groundwater storage (ASg ) are replaced with
8WSU and 8WSS in DAWAST model, the groundwater recharge in narrow sense (ASg ) can
be defined as:
8WSS =P-QD-QB-ET 8WSU (12)
tions where, 8 WSS : variation of groundwater storage, QD : direct runoff,
8WSU: soil moisture change in watershed, QB : Base runoff.
Finally, the groundwater recharge in broad sense is estimated by adding base runoff to 8WSS.
ld it RG - 8WSS + QB = P QB ET - 8WSU (13)
~ical
In the research, daily precipitation, runoff, soil moisture content and the variation of
groundwater level in Chojeong station have been observed and measured to estimate
groundwater recharge using DAWAST, one of the watershed hydrological models.
arly
I no
433
\~ o
Also, since the measured data were given for soil moisture content (%v) and they should be
Table
converted to WSU values in mm unit in order to apply to the modified Beken's equation (7),
Class
the model could find the correlation between soil moisture content and SMe. The correlated
equation is shown in equation (14) with r=0.984. From the equation of SMC(%v) = 0.448 +
0.086 x WSU(mm),
WSU (SMC - 0.448)/0.086 (14)
The transformed soil moisture content, WSU, is shown in Fig. 4. 2001
270 .~~---------------------------------------------------------_
230
210 200:
190
170
1-Jan 51-Jan 2-1 ar 1-Apr 1-1 a, 51-I a, SO-Jun SO-Jul 29-AUO 28-Sep 28-0ot V-Noy 27-0eo
20C
Fig. 4 Transformation of SMC into WSU in 2002
434
.. .i
\~ a
be
(7), Table 1. Water Balance Analysis with Observed Data. (unit:I111!!L_---,-_~
.ted Classific~ Jan ! Feb Mar. A~~a~~~i ! Au~~sep i Oct iN()~1 Dec !TotaIJ
g+ Precip.~ - : 1 12 .7 i 8.0 I 8.9 ,233.5,122.8: 70.8· 7.0 146 .6 1 6.8 8.4 l525.5!
Runoff' - 1 I 5.3 : 6.0 6.2 '120.0167.0 i 24.6 6.2·25.61 6.0 I 7.3 ;274.2
I J I I - i I ! - - I!
20011~ WS I -
II
6.2 h
31.2 :785.3
:er
:er The monthly CE was estimated from the calculated ET by using the modified Beken's
equation (7) and is shown in Table 2 and Fig.6.
5.
::>r
In
tn
IS
E,
435
\~
Whe
0.035 CE(
obse
0.030 than
0.025
Tabl
0.020
•••
0.015 •
• •• •
0.010
0.005
• ..
I------~~-~~ .--~- ~*-:-------.......~~~-~~
••
•• Th
sea
0.000·."
1-Jan 31-Jan 2-Mar 2-Apr 2-May 2-Jun 2-Jul 1-Aug 1-Sep 1-Oct 1-Nov 1-Dec 31·Dec
gel
4.2
Fig. 5. Daily CE calculated from the water balance analysis in 2002 OJ
Kc
COl
tht
0.012 100
•
0.010 o
2Ql
0.008
,.
I
0.006
0.004
0.002
_..,//t"
0'
0.000
01 02 03 04 05 06 07 08 09 10 11 12
436
... .\
\~ o
Whereas the existing model used the only CE value of 0.006, this study applied the monthly
CE calculated in Table 2 to the model and the result is shown in Table 3. Compared with the
observed runoff, the estimated runoff with mean monthly CE's presented the improved result
than that with single CEo
. I CE value and Mean Monthly CE's applied to the Model
t ompanson 0 f Smgle
T able 3. Resu IC
~~ the observed data of Chojeong watershed including precipitation, runoff, evaporation and soil
moisture content, and compared the result with the simulated one from DAWAST model.
The simulated result is shown in Fig. 7 and Table 4. The data in Table 4 was calculated in the
-I
~
12
Fig. 7 Daily Runoff and Hydrologic Response with Measured and Calculated Data in 2002
Table 4. Simulated Results ofDAWAST Model for Chojeong Station (unit: mm)
437
. . i'
\~ -
n
Estimated runoff (EQ) RG
I D. .• Ob . . . . Evapo- A
This
)C reclpl served.'· Ll ~ Ws,,,,------,---.,-----i Kore;
Year • tation runoff. Direct Base i ~ra~s- WSU 10 I
Rate
2001 525.5 i 274.2 175.5 72.7 248.2 303.6 -51.30 -3.4 -3.4 59.3 0.138
......
loun
•
2002 1,400.5 785.3 594.6 220.6 815.2 476.2 50.26 4.6
I 4.6 225.2 ! 0.161 Arno
F
2003 1,297.0 792.7 604.5 i 196.1 800.6 396.3 . 30.28 -4.6 -4.6 189.6 i 0.151 I
f
~ Observation period: 2001.3.08-2003.8.31
Kim,
The estimated rates of runoff for the test period were about 47.2% in 2001, 58.2% in 2002, J;
and 61.7% in 2003, respectively. The errors between observed and simulated runoffs were Kim,
9.5%,3.8%, and 1.0% in order. The negative values in ~ WSU and ~ WSS were resulted from b
the subtraction between the first and last days of each year. 3
The groundwater recharge in narrow sense can be represented as ~ WSs. The amount of ~ Lee,
WSS was -3.4mm in 2001, 4.6mm in 2002, and -4.6mm in 2003, respectively. The F
groundwater recharge in narrow sense showed near zero percent to the annual precipitation.
Lee,
With the result of water balance analysis, it is inferred that the deep percolation mostly
depleted as the form of base runoff and therefore the groundwater recharge in narrow sense l.
(~ WSS) can be negligible. F
The soil moisture content expressed by ~ WSU is about 50mm in a year. This amount is Rust
almost 4% of the mean annual precipitation and can not be ignored as compared to the I
groundwater recharge rate. Suo
The groundwater recharges in broad sense were 59.3mm in 2001, 225.2 mm in 2002, and I
189.6 mm in 2003, respectively. These values are converted to 13.8%, 16.1%, and 15.1% as Wit
the ratio of groundwater recharge to the annual precipitation respectively. It also indicated that
the recharge is about 24%-28% of total runoff
438
-
o
ACKNOWLEDGEMENTS
This work was performed with the fund supported by the basic research grants program of
Korea Science and Engineering Foundation. Reference No.: ROO-2000-00053-0
te
REFERENCES
l8 Journal:
51 Arnold, IG., RS. Muttiah, R Srinivasan, and P.M. Allen, 2000, Regional Estimation of Base
)J Flow and Groundwater Recharge in the Upper Mississippi River Basin, Journal of
Hydrology, 227, pp. 21-40.
Kim, Tai-cheol, 1991, Daily Streamflow Model Based on the Soil Storage in the Watershed,
)2, Journal of Korean Society of Agricultural Engineering, 33-4.
~re Kim, Seong-Joon and Chae Hyo-Seok, 2000, Groundwater Recharge Assessment via Grid
)m based Soil Moisture Route Modeling, Journal of Korea Water Resources Association,
33(1), pp.61-72.
/l Lee, Dong-ryul and Yong-Nam Yoon, 1996, Estimation and Analysis of Groundwater
he
Recharge in Korea, Journal of Korean Society of Civil Engineers, 16(II-4), pp.321-334.
m.
:ly Lee, Dong-ryul and Ho-Bon Goo, 2000, Analysis of Groundwater Recharge Characteristics
Be Using Relationship between Rainfall and Groundwater Level, Journal of Korea Water
Resources Association, 33(1), pp.51-60.
IS Rushton, K.R, and C. Ward. 1979. The estimation of groundwater recharge, Journal of
he Hydrology, 41, pp. 345-361.
Suo N., 1994, A formula for computation of time-varying of groundwater, Journal of
ld Hydrology, 160, pp.123-135.
as Wittenberg, H., and M. Sivapalan 1999, Watershed groundwater balance estimation using
lat
Streamflow recession analysis and baseflow separation, Journal of Hydrology, 219,
pp.20-33.
Thesis:
Han, Young-Min, 2002, Estimation of the Quantity of Watershed Evapotranspiration
considering Soil Moisture Contents, Master's Thesis, Chungnam National University,
Korea.
~f
le
le
lS
Y
l'
e
439
\~ -n