Hydrological Sciences Journal
ISSN: 0262-6667 (Print) 2150-3435 (Online) Journal homepage: www.tandfonline.com/journals/thsj20
The estimation of groundwater recharge from
spring hydrographs
NURI KORKMAZ
To cite this article: NURI KORKMAZ (1990) The estimation of groundwater recharge
from spring hydrographs, Hydrological Sciences Journal, 35:2, 209-217, DOI:
10.1080/02626669009492419
To link to this article: https://doi.org/10.1080/02626669009492419
Published online: 29 Dec 2009.
Submit your article to this journal
Article views: 2314
View related articles
Citing articles: 3 View citing articles
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=thsj20
Hydmlogieal Sciences - Journal - des Sciences Hydrologiques, 35,2, 4/1990
The estimation of groundwater recharge
from spring hydrographs
NURIKORKMAZ
Geotechnical Services and Groundwater Division, General Directorate
of State Hydraulic Works (DSI), YUcetepe, Ankara, Turkey
Abstract The amount of groundwater recharge to an aquifer
depends on the amount of precipitation and its seasonal distribu-
tion, the air temperature, the land use, and other factors; direct
measurements of the recharge components are not possible.
Because of these factors, determining the recharge to aquifers is
a difficult problem in all water resource studies. A project for
determining quantitative groundwater recharge is illustrated by the
computation of the groundwater balance of the Kirkgôz springs
area in the Mediterranean region of Turkey.
Estimation de la recharge des nappes aquifères a partir
dTrydrogrammes de sources
Résumé La recharge d'un aquifère dépend du montant des
précipitations et de sa distribution saisonnière, de la température
de l'air, de l'occupation des sols et d'autres facteurs; les mesures
directes de ces composantes de la recharge ne sont pas possibles.
Comme ces facteurs, déterminant la recharge des aquifères posent
de difficiles problèmes dans l'étude des ressources en eau. Un
projet pour déterminer quantitativement la recharge des aquifères
est caractérisé par le calcul des bilans aquifères des sources de la
region de Kirkgôz du domaine Méditerranéen de la Turquie.
INTRODUCTION
Successful long-term development of groundwater resources requires a
quantitative understanding of flow mechanisms and boundary conditions in
aquifers. However, quantification of aquifer recharge implies that predictions
from groundwater models may also be unreliable (Kitching et al, 1980).
A common indirect method of evaluating recharge involves consideration
of the difference between rainfall and evaporation estimates, taking into
account any surface runoff. Unfortunately, estimates of rainfall, and especially
of evaporation, are subject to considerable errors. Where rainfall and
evaporation are of the same order, as in many temperate and arid areas, the
difference between the two estimates, both with appreciable potential errors,
leads to much uncertainty in the calculation of recharge. In order to try to
overcome the uncertainties inherent in indirect methods, a programme of
measurements on aquifer samples has been undertaken to obtain a more
direct determination of recharge (Kitching et al, 1980).
Open for discussion until 1 October 1990 209
Nuri Korkmaz 210
The amount of moisture which a land area loses by évapotranspiration
depends primarily on the incidence of precipitation, secondly on the climatic
factors of temperature, humidity etc. and thirdly on the type, manner of
cultivation and extent of vegetation. The amount may be increased, for
example, by large trees whose roots penetrate deeply into the soil, thus
transpiring water which would otherwise be far beyond the influence of
surface evaporation (Wilson, 1974).
In a dry period, when évapotranspiration rates exceed available moisture
from precipitation, recharge to the water table is negligible and groundwater
levels decline (Figs 1 and 2).
Explanation
Land surface
r*~ Spring
—*- Flow line
"••— Potential line
\2Zà Impermeable bedrock
Fig. 1 Schematic depiction of an aquifer.
to t) t 2 tn
-Fig. 2 Typical hydrograph of an aquifer.
In 1905 Maillet derived the general equation of the retardation curve
expressing the discharge rate of a river under the exclusive influence of an
aquifer (Mijatovic, 1970).
A baseflow hydrograph is shown in Fig 2. The hydrograph of baseflow
211 Estimation of groundwater rechargefromspring hydrographs
may be approximated by a simple exponential relationship of the from :
QrQ0eM (i)
where Qo and Qt are the discharge at the beginning of the measurement
period and at time t, respectively, and a is the coefficient of recession or
discharge coefficient. The coefficient, a, will be a function of the aquifer
transmissivity, T, storage coefficient, S (or specific yield (Sy)), and the
catchment geometry (Foster, 1974). Equation (1) describes the discharge
characteristic of a linear reservoir, i.e. one in which the discharge, Q, is
proportional to the storage, V, with Q = aV, Rearranging and allowing for
units gives (Castany, 1968; Mijatovic, 1970):
86 400 Q
V = — (2)
a
where V = storage capacity or dynamic reserve (m3);
Q = groundwater discharge (m3 s"1); and
a = coefficient of recession (discharge coefficient) (day"1).
In logarithmic form to the base 10, the exponential formula (1) is
transformed into:
l o g ô , = log Q0 - 0.4343 at (3)
The discharge coefficient, a, may be obtained directly from equation (3),
via the following equation:
log Qo - log Qt
« = (4)
0.4343 t
Equation (2) applies at each instant, so that Q in equation (2) is the
same as Qt in equation (1). Thus :
V = 86400 (Q0/a) exp (-at) (5)
For a given time period, At = (t - t ) , quantitative data about the
dynamic reserve at the end of the period (f ) may be calculated from the
following equation:
v v + R
m= o -Q (6)
where Vm = dynamic reserve at the end of the period (t ) (m );
Vo = dynamic reserve at the beginning of the period (r ) (m3);
R = groundwater recharge volume during the time period At (m3);
and
Nuri Korkmaz 212
Q = groundwater discharge volume during the time period At (m3).
The volume of groundwater discharge during a water year period may
be estimated from the spring hydrograph (Fig. 3). The difference between the
dynamic reserve (Vm) at the end of a water year (tm) and the dynamic
reserve at the beginning of a water year (tj is the dynamic reserve change
(AV). The volume of groundwater recharge during a water year will be:
i? = g l A F (7)
where R = groundwater recharge volume during the water year (m3);
Q = groundwater discharge volume during the water year (m3); and
AV = dynamic reserve change during the water year (m3).
Water years
Fig. 3 Hydrograph of Kirkgôz springs.
ESTIMATION OF GROUNDWATER RECHARGE
This approach for the estimation of groundwater recharge was applied in the
Kirkgôz springs area in the Mediterranean region in Turkey (Fig. 4).
The investigation area which includes the catchment area of the Kirkgôz
springs has an intensive karstic character. It is located between 36° 50' - 37°
10' N and 30° 51' - 30° 55' E.
The Taurus Mountain of Southern Turkey has several large aquifers and
springs of the type of the Kirkgôz Springs, with large and well regulated
spring flow and significant low flows. Tracer techniques have not been feasible
for the investigation of such aquifers. The classical geophysical and hydrogeo-
logical methods of establishing karst water connections (swallow holes-to-
springs dye test, determination by hydrogeological methods, finding quantities
of leakage water from karst lakes etc.), are not applicable to very large karst
213 Estimation of groundwater rechargefromspring hydrographs
Fig. 4 Location of the study area in Turkey.
aquifers (Yevjevich, 1986).
The investigation of catchment areas in the Taurus Mountain region and
its foothills by several Turkish geologists and hydrogeologists in the last 10-20
years has revealed important karst features. Tectonic movements of pervious
limestones and of associated impervious formations may have "compartmenta-
lized" the limestone into mosaic-type aquifers. These follow one another,
often separated by thin impervious formations. A mosaic type aquifer has
such compartments connected to produce large springs (Yevjevich, 1986).
Because of these factors, an estimate of the area of the catchment
contributing to a karst spring is a difficult problem. The catchment area of
the Kirkgôz springs is about 1800 km2.
The geological formations from the oldest to the youngest are Mesozoic
karstic limestones, Cataltepe and Ispartaçay impervious formations, Paleogene
and Miocene formations. Mesozoic karstic limestones covering the catchment
area of the Kirkgôz springs constitute the main recharge areas of the
groundwater and there are a lot of karst structures such as sinkholes, dolines,
caves, poljes and solution cavities.
The Mediterranean climate, which is warm and rainy in the winter and
hot and dry in the summer, prevails in the catchment area of the Kirkgôz
springs.
The Kirkgôz springs, discharging from karstified limestones about 30 km
north of Antalya city, are the most important water resource. The discharge
Nuri Korkmaz 214
of these springs has been measured since January 1963 at a gauging station.
The average monthly discharges of the Kirkgôz springs for the water years
1973-1983 are given in Table 1, and the discharge data for the water years
1973-1982 are shown graphically in Fig. 3.
Table 1 Average monthly discharge (mJ s'1) of the Kirkgôz springs
Water
year Oct. Nov. Dec. Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep.
1973 - - - 13.089 16.560 18.260 17.330 14.050 14.470 12.820 11.370 11.050
1974 10.850 10.808 10.620 12.150 18.960 - - 17.953 16.507 14.272 11.733 10.625
1975 9.694 10.508 10.451 - 16.775 16.050 12.327 11.663 9.832
1976 8.350 9.951 12.100 16.400 21.200 19.550 16.564 17.946 16.546 12.854 10.940 9.987
1977 8.276 8.441 10.256 12.810 14.020 14.570 12.960 13.887 13.088 9.213 7.545 8.643
1978 6.437 7.030 7.721 14.641 24.618 24.120 23.213 22.294 18.203 15.554 11.987 10.758
1979 10.165 10.230 11.252 16.730 30.434 26.747 19.770 18.068 20.644 17.050 15.249 13.861
1980 10.027 14.810 16.207 20.107 18.820 - - 17.812 17.000 14.473 13.114 12044
1981 9.220 9.990 11.974 19.520 21.700 25.635 22.000 18.650 18.500 15.087 13.886 13.220
1982 11.313 11.880 15.596 22.128 19.000 21.635 25.220 26.147 23.400 18.836 15.971 14.400
1983 13.660 12.100 15.153 17.250 23.467 - 23.425 19.794 19.320 15.732 14.327 14.132
The discharge rise of the Kirkgôz springs is largest during the late
autumn, winter, and spring months, when plants are dormant and évapotrans-
piration rates are small. In the summer when évapotranspiration rates exceed
the available moisture from precipitation, recharge to the water table is
negligible and the discharge rates of these springs decline (Fig. 3).
The results of the computation of the discharge coefficient (a) for each
water year during the period 1973-1983 by means of equation (4) are given
in Table 2 (see Figs 5, 6 and 7). The average discharge coefficient (a) of
this aquifer during the period 1973-1983 is approximately 0.00405 day"1
(Table 2).
The average monthly discharge data of the Kirkgôz springs in October of
Table 2 Recession hydrograph analysis for Kirkgôz springs, 1973-1983
Water year Selected interval, t Discharge coefficient, a
Qo
(days) °? .7 (nrs1) (day'1)
(mJsl)
1973 210 18.26 10.5 0.00263
1974 150 18.5 9.2 0.00465
1975 150 17.0 8.8 0.00439
1976 180 23.2 11.2 0.00405
1977 210 16.5 6.8 0.00422
1978 180 24.4 11.3 0.00427
1979 210 30.0 12.0 0.00436
1980 180 20.0 9.8 0.00396
1981 180 26.0 12.6 0.00402
1982 180 24.0 12.0 0.00385
1983 120 23.4 14.3 0.00410
Average discharge coefficient 0.00405
215 Estimation ofgroundwater rechargefromspring hydrographs
|=:
50 J——
j1 J_ - Log Q t = L o g Q 0 - ( 3 . 4 3 4 3 « : t
1
| ::
^ 5 ==^i=r t - | L o g 10,5 = Log 1 8 , 2 6 - 0 , 4 3 4 3 . 2 . 0 < *
_£ E||_E|
30 » ^ = - , i____- T &c-0,0026 day '
" Z i-, ^ •—-
-4-
ZT | _| . ,. '
-. •-
~4_û,=:i 3 -- _rnL__
.^r^ ^ ___S_1: 1
^ • ^ 1 1 . ~' -1 -i
I •
1 '""T—t L ' '
i 4» ! r"*-*__.
Ï-« ;
3
E ^ - -+- : r1 •+^r ^ _ _ - - j _
c ' 1 ' | ' t-It
1
1
^ ,; Qt = 10,5 m / s e c
1 1 H-4-
T Tltr " _"__
lii__p|^
•g s ^ H = Î =P _-__-____
ill! _jZ | A
_~~
«
Q 7 —
_______ _.
_j_ 1
— —: ._ ^_i
6 —r_._ =r =
5.
" _li
___-__^
- -ife -M. î
-_4- i -
-
r
Marc Apr. May June July Aug- Sept. Oct.
Fig. 5 Discharge line for Kirkgôz springs during recession of the
aquifer.
/ Ç j t = ll.3ir?/__
0 D F A J A O D F A J A O D F A J A 0 D F A J A 0 D F A J A
1974 | 1975 | 1976 | 1977 1978
Water years I
Fig. 6 Recession hydrograph analysis for Kirkgôz springs.
of each year during the period 1974-1982 are given in Table 3. The results
of the computation of the dynamic reserves at the beginning and end of each
water year during the period 1974-1982 by means of equation (2) are also
given in Table 3. The parameters used for this calculation were the discharge
(Qo and Qm) at the beginning and end of each water year during the period
1974-1982 based on discharge measurements of the Kirkgôz spring (Figs 6
and 7 and Table 3) and on the average discharge coefficient (a) of
Nuri Korkmaz 216
Qt=l4.3n?/sec
3 D F A J A O D F A J A O D F A J A O D F A J A O D F A J A
1979 1 1980 1 1981 I 1982 | 1 983
Water years
Fig. 7 Recession hydrograph analysis for Kirkgbz springs.
0.00405 day"1 (Table 2). Also, the dynamic reserve changes and the yearly
discharge data (Figs 3 and 8) of this aquifer are given in Table 3. The results
of the computation of the recharge (R) by means of equations (7) are given
both in Table 3 and in Fig. 8. The dynamic reserve change during the period
1974-1982 is 59.9 x 106 m3 (Table 3). This value is the difference between
the total recharge volume and the total discharge volume for the same
period. The average annual recharge volume and discharge volume are 493.2
x 106 m3 and 486.5 x io 6 m3, respectively (Table 3).
Table 3 Kirkgôz springs hydrograph analysis, 1974-1982
Water October Dynamic reserve Dynamic Discharge Recharge
year discharges reserve change
Q0 Q0 V
o V
m AV Qy R
3 1 3 1 3 3
(m *' ) (m *- ) (10 6 3
m) (lé m ) (106 3
m) (lém ) (106 m3)
1974 10.850 9.694 231.5 206.8 -24.7 451.5 426.8
1975 9.694 8.350 206.8 178.1 -28.7 378.0 349.3
1976 8.350 8.276 178.1 176.6 -1.5 441.0 439.5
1977 8.276 6437 176.6 137.3 -39.3 357.0 317.7
1978 6.437 10.165 137.3 216.9 + 79.6 525.0 604.6
1979 10.165 10.027 216.9 213.9 -3.0 567.0 564.0
1980 10.027 9.220 213.9 196.7 -17.2 504.0 486. S
1981 9.220 11.313 196.7 241.3 +44.6 546.0 590.6
1982 11.313 13.660 241.3 291.4 + 50.1 609.0 659.1
Total +59.9 4378.5 4438.4
Average 486.5 493.2
Average discharge coefficient, a = 0.0040571
day
217 Estimation ofgroundwater recharge from spring hydrographs
800 T 1 , , J r
(ft
v 700
»
E
,9 600
J 500
c 400
S 300
S
° 200
E
| 100
0
1974 75 76 77 78 79 80 8! 1982
Fig. 8 Recharge and discharge for Kirkgôz springs.
CONCLUSION
The aim of this study has been a synthesis of the results which can be
obtained by an analysis of spring discharge. The accuracy of the results
depends exclusively on the basic hydrometric observation data.
The present work suggests that the Kirkgôz springs discharging from
karstified limestones about 30 km north of Antalya city are the most
important water resource. Mesozoic karstic limestones covering the catchment
area of the Kirkgôz springs constitute the essential recharge area of the
groundwater. Finally, the average discharge coefficient of this aquifer is
approximately 0.00405 day"1 while the dynamic reserve change during the
period 1974-1982 is 59.9 x 106 m3. The average annual recharge volume and
discharge volume for the same period are 493.9 x 106 m3 and 486.5 x 106
m3, respectively.
REFERENCES
Castany, G. (1968) Prospection et Exploitation des Eaux Souterraines. DUNOT, Paris.
Coskun, N., Nazik, M. & Altug, A. (1985) Antalya-Kirkgoz springs and travertine plateau karst
area investigation report. Geotechnical Services and Groundwater Division (DSI), Ankara.
Foster. S. S. D (1974) Groundwater storage-river flow relations in a chalk catchments. /.
Hydro!. 23(3/4), 299-311.
Kitching, R., Edmunds, W. M, Shearer, T. R., Walton, N. R G. & Jacovides, J. (1980) Assess-
ment of recharge to aquifers. Hydrol. Set Bull. 25, 217-235.
Mijatovic, B. (1970) A method of studying the hydrodynamic regime of karst aquifers by
analysis of the discharge curve and level fluctuation during recession. Institute for
Geological and Geophysical research, Beograd.
Wilson, E. M. (1974) Engineering Hydrology. The Macmillan Press Ltd. London.
Yevjevich, V. (1986) Specificities of karst water resource. In: Karst Water Resources. Proc.
Ankara-Antalya Symp. July 1985 (ed. Gunay, G. & Johnson, A. I.), IAHS Publ. no. 161,
3-26.
Received 26 April 1988; accepted 16 July 1989
