Get documents about "Estimation of recharge to the fresh-water lens of Tongatapu
Document Sample
Hydroloey of Warm Humid Regions {Proceedings of the Yokohama Symposium, July 1993). ^„
IAHS MM. no. 216, 1993. 317
Estimation of recharge to the fresh-water lens of
Tongatapu, Kingdom of Tonga
L. J. FURNESS
Ministry of Lands, Survey and Natural Resources, Nuku 'alofa, Kingdom of Tonga
S. GINGERICH
Department of Geology and Geophysics, University of Hawaii, Honolulu, Hawaii
Abstract Management of the groundwater resource of Tongatapu
requires an assessment of the annual recharge. Previous studies have
used assumed data, a water balance approach, a model, chloride ion
ratio or a percentage applied elsewhere in similar circumstances. These
steady state estimates are inadequate to describe the high variability of
water quality experienced seasonally and in droughts. This study is
based on monitoring data from well hydrographs, pluviographs, tide
and lagoon gauging and a barometer. The hourly data have enabled an
assessment of the factors influencing water levels and water quality in
wells. The water level and salinity are affected by the superposition of
rainfall, tides, winds, sea level change, barometric pressure and local
water extraction. By removing the tidal, barometric and sea level
changes from the hydrograph, the recharge can be directly assessed
with a knowledge of the specific yield of the aquifer.
INTRODUCTION
Tongatapu is the largest island in the Kingdom of Tonga with a surface area of
approximately 257 km2. It is located to the southeast of Fiji and to the southwest of
Western Samoa in the South Pacific at 175°12'W 21°08'S. The population of 60 000
live in the capital city Nuku'alofa and 57 villages, mainly spread around the coast
(Fig. 1). There are no fresh surface water bodies so public water supplies are obtained
from wells in the fresh-water lens of a highly porous limestone aquifer.
GEOLOGICAL SETTING
Tongatapu is a raised coral island on the Tonga Ridge. It is in an archipelago at the
tectonic boundary between the A»stralia-India plate and the subducting Pacific plate.
Between 170 and 230 m of Pleistocene coral limestone underlie the island, which in
turn rests on a base of volcanoclastic and pelagic sediments. A series of limestone
terraces make up the island which tilts from a high of 65 m in the southeast to the
low-lying north coast.
A volcanic ash (tephra) soil covers the island in two distinct layers up to 5 m
thick. It has a high degree of cracking and readily accepts heavy recharge without
ponding.
318 L. J. Furness & S. Gingerich
Fig. 1 Hydrogeology of Tongatapu.
HYDROGEOLOGY
Aquifer
Tongatapu has a fresh-water lens with a water table less than one metre above mean
sea level. The lens reaches a maximum thickness of about 12 m at the three widest
parts of the island. The limestone aquifer is very porous and modelling indicates that
the hydraulic conductivity is of the order of 1500 m day"1.
Discharge is visible on the fringing reefs and at the lagoon at low tide when fresh
water is seen in trickles from pores in the limestone.
Rainfall
The mean annual rainfall for Nukualofa is 1770 mm (1947-1990) (Falkland, 1991).
Rainfall has been monitored at six other locations on Tongatapu which indicate an
increase of 9% in the southeast, possibly due to orographic effects (Thompson, 1986).
There is a distinct wet season from October to April and a dry season from April to
September. Tropical cyclones are experienced with an average frequency of 1.3 per
year. Droughts are a feature of the El Nino events.
Estimation of recharge to the fresh-water lens of Tongatapu 319
Previous recharge estimates
Estimates of recharge have been produced to give guidelines for the management of
the water resource to prevent overuse and saline intrusion. Pfeiffer & Stach (1972)
assumed a recharge between 5 and 15 % of the mean annual rainfall based on previous
experience and adopted a figure of 10% for the mean annual recharge. Lao (1978)
used the water balance approach with an assumed 5% of rainfall as runoff, and
evaporation losses of 70% of rainfall. He derived an estimate of recharge as 25% of
mean annual rainfall.
Hunt (1978) used a sharp interface, steady state model based on the chloride ion
flux upward from the interface being balanced by the recharge. A recharge rate of 25
to 30% of rainfall was used to calibrate the model. Kafri (1989) obtained a simple
chloride ion ratio between the freshest recorded well sample and rainfall. This
indicated a recharge of 35 % of the rainfall. He used a more conservative value of 25 %
in his calculation of yield, based on the previous studies.
Hasan (1989) estimated the mean annual recharge to be 20% of rainfall. This was
based on a monthly water balance and monthly crop water requirements. Falkland
(1991) used a computer model of water balance. Using 44 years of data, the mean
annual recharge was 528 mm or 30% of mean annual rainfall. The results show high
monthly variability with little or no recharge coinciding with El Nino events. This was
the only study to show the variability of recharge and as such was useful in
understanding the large fluctuations in water quality.
COLLECTION OF RECHARGE DATA
Assessment of recharge in this study was based on the results of a series of
measurements from data recorders. At well 193 located at Fua'amotu International
Airport near the centre of the widest part of the island, groundwater levels have been
monitored on an hourly interval. A tilting bucket rain gauge totalised rainfall over each
hour. Similarly at well 105 at the Mataki'Eua well-field, which supplies the capital
Nuku'alofa, there is an hourly water level logger with a rain gauge and a barometer.
On the Vuna wharf along the north coast is a tide gauge with a continuous chart
recorder. Within the western reach of the Fanga'Uta Lagoon a water level recorder
has been installed to record the fluctuations of water level. The tide gauge has
operated for two years, and the water level loggers for between one and two years.
Thus a considerable number of data are available for analysis.
During the period of continuous records the Pacific region has been under the
influence of the El Nino phenomenon. In 1991-1992 Tongatapu recorded eight
continuous months of below average rainfall. There has been a persistent soil moisture
deficit and only relatively light rainfall observed. Therefore the hydrographie record
has enabled a long period of calibration of water levels with other external influences
and is mostly unaffected by rainfall.
ANALYSIS OF RESULTS
The hydrograph and rainfall records of well 193 are shown in Fig. 2. Tide gauging
320 L. J. Furness & S. Gingerich
i.. Il i1 J lli li i LII i u H Jl J liL iLl L\. y. II i , . , Â .
JAN FEB MAR APR MAY JUN
1992
Fig. 2 Hydrograph and rainfall records at well 193.
and barometric pressure are shown on Fig. 3. There are several important features in
the records which have a bearing on the analysis of recharge.
The tide gaugings show a semi-diurnal pattern with a range from 0.5 to 2.35 m.
The mean sea level for the period of measurement is 1.39 m on the chart. During the
period of the El Nino the sea level rose up to 0.1 m above the mean. The smoothed
sea level curve was obtained by using a filter for frequencies of one day or less.
The hydrograph from well 193 shows a semi-diurnal response to the tidal
constituents. These were removed by applying a numerical filter such as the Doodson
Filter (Pugh, 1987). The residual is still affected by a signal which is related to the
barometric pressure.
The barometric record shows the tropical response with the chief characteristic of
a 12-h cycle with amplitudes around 2 mbars and maximum pressures around 1000 h
and 2200 h local time. The barometric record also shows the passage of troughs of low
pressure which typically take up to one week. It is frequently during the period of low
pressure that associated clouds bring heavy rain.
The inverse relationship between atmospheric pressure, sea level and the
hydrograph can be seen (Figs 2 and 3). Using standard values of sea water density and
gravity, an increase in atmospheric pressure of 1 mbar will produce a decrease in sea
level of about 1 cm. This ideal relationship is closely approximated in the tide
gauging.
For the water level response in well 193 the inverted barometer effect appears to
be reduced in amplitude by about 40-50%. The phase shift is variable with peaks and
Estimation of recharge to the fresh-water lens ofTongatapu 321
1020
J2
e,
LU
CC
r>
w
w
LU
CC
CL
g
ce
H
LU
o
ce
<
m
JAN FEB MAR MAY
1992
Fig. 3 Tide gauging and barometric pressure.
troughs appearing up to 48 h behind the recorded pressure. As the aquifer is
unconfined it is likely that barometric effect on the sea level is subsequently
transmitted to the groundwater levels. There are several factors which may cause the
variable phase shift, recharge is one of them.
There is not a simple one-to-one correlation between barometric changes, sea level
changes and the well hydrograph. The island ofTongatapu is subject to persistent wind
patterns which cause the build up of sea water on one side (up to 0.2 m) and a
corresponding decrease on the other. Wind pattern shifts accompany the passage of
low pressure troughs.
The wind pattern also strongly affects the disposition of water in the shallow
Fanga'Uta Lagoon. Build up of water on one side of the lagoon may be as much as
0.3 m, which significantly affects the groundwater level. The lagoon level which is
damped and up to three hours out of phase with tides is also variable around its
shoreline depending on the wind pattern.
The barometric effect is expressed as inverted changes in the water level, but it
is also damped and retarded. If recharge is to be separated with confidence from the
record it must clearly be larger than the background signal.
During the period of record only a few rainfall events have been observed which
show a slight increase in the hydrographie level. From modelling of the aquifer and
examination of drill-hole cores and exposures of rock in quarries, caves and cliffs it
is apparent that the porosity and specific yield of the aquifer is very high. The latter
probably averages as much as 0.4.
322 L. J. Furness & S. Gingerich
Subtracting the smoothed well signal from the smoothed tide gauging leaves a
residual fluctuation of the order of 0.03 m. Thus an individual recharge event of less
than 13 mm (specific yield 0.4) may pass undetected. However, it is likely that such
magnitude is insignificant compared with recharge which occurs during the frequent
heavy rainfall associated with a normal wet season.
Fine tuning of the process of estimating recharge will rely on developing
an understanding of the relationship between barometric pressure and well water level.
Extreme rainfall events, such as in the passage of a tropical cyclone will give the
largest response in both atmospheric pressure and recharge to the aquifer. The effect
of the pressure is transient, whereas the recharge is retained much longer in the
hydrographie record. Continued monitoring over several years should produce
sufficient data to quantify the annual recharge and pinpoint when it occurs.
REFERENCES
Falkland, A. C. (1991) Tonga Water Supply Master Plan Study. Water Resources Report for PPK Consultants Pty
Ltd, October 1991.
Hasan, M. R. (1989) Report on Preliminary Survey of Rainwater Harvesting. TCP/TON/8952, FAO, Rome.
Hunt, B. (1978) An analysis of the groundwater resources of Tongatapu island. Research Report 78-15. Dept of
Civil Engineering, University of Canterbury, New Zealand.
Kafri, U. (1989) Assessment of groundwater potential in the island of Tongatapu, Kingdom of Tonga. Report
GSI/8/89, Geological Survey of Israel, Ministry of Energy and Infrastructure.
Lao, C. (1978) Assignment Report (25 October to 5 December 1978), Groundwater Resources Study of Tongatapu.
WHO, Regional Office for the Western Pacific.
Pfeiffer, D. & Stach, L. W. (1972) Hydrogeology of the Island of Tongatapu, Kingdom of Tonga, South Pacific.
Geologisches Jahrbuch, Reihe C , Heft 4, Hannover.
Pugh, D. T. (1987) Tides, Surges and Mean Sea-Level. National Environment Research Council. John Wiley.
Thompson, C. S. (1986) The Climate and Weather of Tonga. New Zealand Meteorological Service, Misc. Publ.
188(5), Wellington, New Zealand.
