Analysis of pumping tests, with regard to tectonics, hydrothermal effects and weathering, for fractured Dalha and stratiform basalts, Republic of Djibouti


Jalludin, M.; Razack, M.

Journal of Hydrology 155(1-2): 237-250

1994


The Republic of Djibouti is the location of an exceptional geodynamic situation, the Afar Depression, which is an emerged triple junction of the Red Sea, the Gulf of Aden and the East African rifts, where a fault network is particularly well developed and controls the permeability of volcanic aquifers. Pumping test data are analysed for two major basaltic series, the stratiform series (3.4-1Ma) and the Dalha series (9-3.4Ma), which are distinguished mainly by their geodynamic characteristics and geological history although they are both recognized as trap rocks. Drawdown and recovery were studied by Jacob's logarithmic approximation of the Theis equation and by the Theis recovery method. Elsewhere, step drawdown data were interpreted with the Cooper-Jacob graphical solution and Rorabough's method. The results describe aquifers in the stratiform and Dalha basalts which are identified by different hydrodynamic characteristics. Stratiform and Dalha aquifer transmissivities range from 1.5X10 (super -4) to 5.7X10 (super -1) m (super 2) s (super -1) and from 2.9X10 (super -6) to 1.6X10 (super -2) m (super 2) s (super -1) . Storage coefficients determined from piezometer observations and computer simulations describe the aquifers in the Gulf basalts, a particular type of the stratiform series, as semiconfined, and those in the Dalha basalts as confined or unconfined depending on the area. Thus, the calculated specific capacity demonstrated that the stratiform basalt aquifers have more favourable hydrodynamic characteristics than the Dalha basalt aquifers. Specific capacities vary between 0.372 and 510.01m (super 3) h (super -1) m (super -1) within the stratiform basalts and between 0.008 and 35.2m (super 3) h (super -1) m (super -1) in the Dalha basalts. As they were originally composed of the same type of trap rocks, the stratiform and Dalha basalts should have a similar magnitude of permeability, and this was confirmed by some wells. The lower specific capacities exhibited by the Dalha basalts is explained by a longer period of alteration and greater hydrothermal activity, as has been observed in well cuttings and field studies.

Journal
of
Hydrology
ELSEVIER
[2]
Journal
of
Hydrology
155
(1994)
237-250
Analysis
of
pumping
tests,
with
regard
to.
tectonics,
hydrothermal
effects
and
weathering,
for
fractured
Dalha
and
stratiform
basalts,
Republic
of
Djibouti
M.
Jalludin*'a,
M.
Razack
b
a
lSERST,
Laboratoire
d'Hydrogeologie,
PB
486,
Djibouti,
Republic
of
Djibouti
b
Universite
de
Poitiers,
Laboratoire
d'Hydrogeologie,
40,
Avenue
du
Recteur
Pineau,
86022
Poitiers
Cedex,
France
(Received
10
June
1992;
revision
accepted
18
July
1993)
Abstract
The
Republic
of
Djibouti
is
the
location
of
an
exceptional
geodynamic
situation,
the
Afar
Depression,
which
is
an
emerged
triple
junction
of
the
Red
Sea,
the
Gulf
of
Aden
and
the
East
African
rifts,
where
a
fault
network
is
particularly
well
developed
and
controls
the
permeability
of
volcanic
aquifers.
Pumping
test
data
are
analysed
for
two
major
basaltic
series,
the
stratiform
series
(3.4-1
Ma)
and
the
Dalha
series
(9-3.4
Ma),
which
are
distinguished
mainly
by
their
geodynamic
characteristics
and
geological
history
although
they
are
both
recognized
as
trap
rocks.
Drawdown
and
recovery
were
studied
by
Jacob's
logarithmic
approximation
of
the
Theis
equation
and
by
the
Theis
recovery
method.
Elsewhere,
step
drawdown
data
were
interpreted
with
the
Cooper-Jacob
graphical
solution
and
Rorabough's
method.
The
results
describe
aquifers
in
the
stratiform
and
Dalha
basalts
which
are
identified
by
different
hydro-
dynamic
characteristics.
Stratiform
and
Dalha
aquifer
transmissivities
range
from
1.5
x
10
-4
to
5.7
x
10
-1
m
2
s
-1
and
from
2.9
x
10
-6
to
1.6
x
10
-2
m
2
s
-1
.
Storage
coefficients
determined
from
piezometer
observations
and
computer
simulations
describe
the
aquifers
in
the
Gulf
basalts,
a
particular
type
of
the
stratiform
series,
as
semiconfined,
and
those
in
the
Dalha
basalts
as
confined
or
unconfined
depending
on
the
area.
Thus,
the
calculated
specific
capacity
demonstrated
that
the
stratiform
basalt
aquifers
have
more
favourable
hydrodynamic
characteristics
than
the
Dalha
basalt
aquifers.
Specific
capacities
vary
between
0.372
and
510.01
m
3
h
-1
m
-1
within
the
stratiform
basalts
and
between
0.008
and
35.2
m
3
h
-1
m
-1
in
the
Dalha
basalts.
As
they
were
originally
composed
of
the
same
type
of
trap
rocks,
the
stratiform
and
Dalha
basalts
should
have
a
similar
magnitude
of
permeability,
and
this
was
confirmed
by
some
wells.
The
lower
specific
capacities
exhibited
by
the
Dalha
basalts
is
explained
by
a
longer
period
of
alteration
and
greater
hydrothermal
activity,
as
has
been
observed
in
well
cuttings
and
field
studies.
*
Corresponding
author.
0022-1694/94/$07.00
©
1994
Elsevier
Science
B.V.
All
rights
reserved
SSDI
0022-1694(93)02354-Z
238
M.
Jalludin,
M.
Razack
Journal
of
Hydrology
155
(
1994)
237-250
1.
Introduction
The
study
area
is
located
in
Eastern
Africa
between
longitudes
41°40'E
and
43°30'E
and
latitudes
10°50'N
and
12°50'N,
in
the
Republic
of
Djibouti.
The
landscape
is
mainly
composed
of
plains,
plateaux
and
a
few
mountainous
zones.
The
Dalha
basalts
(Fig.
1)
underlie
a
wide
plateau
in
the
northern
part
of
the
country
and
extend
from
Arta
to
Dikhil
where
they
form
the
Bour
Ougoul
and
Arta
uplands
as
well
as
hilly
mountains
towards
the
south.
To
the
west
of
the
Dalha
formation,
stratiform
basalts
are
characterized
by
large-scale
ravines
dissecting
flat
plateaux.
Elsewhere,
Jurassic
and
Cretaceous
sediments
encompassed
by
rhyolites
produce
high-level
reliefs
in
the
Ali
Sabieh
region.
The
harsh
climate,
with
an
average
of
less
than
200
mm
rainfall
per
year,
explains
the
lack
of
permanent
rivers,
and
has
led
the
country
to
exploit
underground
water
resources.
Underground
exploitation
now
focuses
on
volcanic
rocks,
as
they
generally
contain
better
quality
water.
On
the
basis
of
available
pumping
test
data
(Bundesanstalt
fiir
Geowissenschaften
and
Rohstoffe,
unpublished
data,
1982;
Landsystem—Technosol,
unpublished
data,
1990)
and
data
collected
in
the
current
hydrogeological
project,
this
paper's
aims
are
as
follows:
(1)
to
apply
classical
analytical
models
based
on
the
Theis
solution
(1935)
for
pumping
tests
run
on
Dalha
and
stratiform
fissured
basalt
aquifers,
and
determine
and
compare
their
hydrologic
characteristics;
(2)
to
analyse
step
drawdown
tests
and
calculate
well
losses,
to
simulate
drawdown
in
pro-
duction
wells;
(3)
to
discuss
the
difference
in
hydrodynamic
characteristics
between
Dalha
and
stratiform
basalts
with
regard
to
tectonism,
weathering
and
hydro-
thermal
effects.
2.
Geology
The
volcanic
series
overlie
Jurassic
limestones
and
Cretaceous
grainstones
as
a
result
of
the
Red
Sea,
Gulf
of
Aden
and
East
African
rifts
triple
junction
system
(Barberi
et
al.,
1975;
Gaulier
and
Huchon,
1991),
which
has
involved
plate
tectonics
for
25-30
my.
Old
or
`adoler
basalts
characterize the
first
rupture
movement
along
the
Arabo-Nubian
bloc
during
the
later
Miocene
epoch.
This
highly
weathered
basalt
is
intensively
affected
by
hydrothermal
alteration.
Then
followed
a
slow
expansion
rate
period
during
which
the
Mabla
rhyolites
(15
Ma)
erupted.
After
an
erosional
stage
of
these
rhyolites,
the
Dalha
basalts
occur
with
an
angular
discordance
(3.4-9
Ma).
Contemporaneously,
Somali
basalts
erupted
to
the
east
but
remain
in
a
stable
tectonic
area.
Between
3.5
and
1.5
Ma
stratiform
basalts
and
Gulf
basalts
poured
out,
at
the
time
of
the
Gulf
of
Tadjourah
opening.
(Geological
ages
are
after
Barberi
et
al.
(1975),
Black
et
al.
(1975)
and
Chessex
et
al.
(1975).)
Gulf
basalts
occur
in
particular
on
both
sides
of
the
Gulf
of
Tadjourah,
and
are
considered
the
first
fissural
lava
flows
along
the
gulf.
4
2'
43'
1
iiii,
III
2
3
4
DORRA
5
/
/
1r
OBOCK
TADJOURAH
DJIBOUTI
Sito.
,rs
.
1
""4
1
4114.
-
;..7;
6
.71
1
DIKHIL
7
Main
•••••=
8
11-
ALI
SABIEH
9
ART
M.
Jalludin,
M.
Razack
/
Journal
of
Hydrology
155
(1994)
237-250
239
Fig.
1.
Geological
sketch
map
of
Republic
of
Djibouti:
1,
Jurassic
and
Cretaceous
sedimentary
rocks;
2,
Adolei
or
ancient
basalts
(25
Ma);
3,
Mabla
rhyolites
(15
Ma);
4,
Dalha
and
Somali
basalts
(9-3.4
Ma);
5,
stratiform
basalts
(3.4—I
Ma);
6,
stratiform
rhyolites;
7,
Gulf
basalts
(3.4-1
Ma);
8,
recent
volcanic
series
(less
than
1
Ma);
9, Pleistocene
sedimentary
rocks.
3.
Hydrogeology
Dalha
basalts
are
greenish
grey
or
black
and
outcrop
in
the
form
of
traps,
lava
flows
a
few
metres
thick,
with
shrinkage
cracks,
vesicles
and
amygdules,
intercalated
with
ignimbrites,
red
and
green
palaeosoils
and
some
sparse
alluvium
layers.
Often
240
M.
Jalludin,
M.
Razack
/
Journal
of
Hydrology
155
(1994)
237-250
intertrap
layers
pinch
out.
The
thicknesses
of
these
basalts
vary
between
250
m
in
the
south
and
400
m
in
the
north.
Red
and
green
shales
with
rare
detrital
quartz
constitute
the
substratum
of
this
geological
formation.
Gulf
basalts
are
generally
black.
The
sequence
looks
like
the
Dalha
basalts
but
with
lava
flows
and
frequent
alluvium
layers
that
pinch
out,
thus
making
geological
correlation
between
wells
difficult.
Their
thickness
can
exceed
100
m
but
the
bottom
of
these
basalts
has
not
yet
been
recognized.
The
total
thickness
of
stratiform
basalts
may
reach
1000
m.
Although
the
surface
of
the
younger
stratiform
basalts
is
characterized
by
spheroidal
weathering,
the
complete
sequence
remains
in
a
fresh
condition.
Hydro-
thermal
effects
and
weathering
are
almost
absent
and
do
not
appreciably
affect
shrinkage
cracks,
faults
and
interbedded
layers.
In
contrast,
weathering
affects
the
Dalha
basalts
intensively,
and
has
smoothed
off
rugged
mountains.
Scoriaceous
layers
show
varying
degrees
of
alteration
depending
on
the
quantity
of
vesicles.
It
follows
that
such
layers
are
partly
altered
to
clay.
Field
observations
show
that
plugging
of
fissures
and
faults
as
a
result
of
hydrothermal
action
is
in
the
form
of
calcite
veins
that
can
be
several
centimetres
thick.
Calcite
veins
are
made
up
of
one
or
more
generations
of
crystallization
that
have
developed symmetrically
from
the
walls
of
open
fissures.
Different
generations
of
crystallization
could
signify
tectonic
events
or
changes
in
thermodynamic
conditions
of
the
water
or
both.
Various
types
of
crystallization,
at
millimetre
and
centimetre
scales,
or
microscopic
mineral
crystals,
may
explain
the
speed
of
crystallization.
An
important
fault
network
was
generated
during
the
various
tectonic
stages
(Arthaud
et
al.,
1980).
Within
the
study
area,
faults
affecting
the
Dalha
basalts
form
five
groups
(Arthaud
and
Jalludin,
1993),
composed
of
ancient
faults
and
recurrent
faulting;
the
main
fault
directions
are
110°,
east—west,
north—south
and
N130°—N140°.
Faults
affecting
the
stratiform
basalts
trend
east—west,
north—south,
N150°
and
N130°.
Fault-line
scarps
are
less
apparent
within
the
Gulf
basalts
because
of
recent
lava
flows
that
mask
the
underlying
topography.
Tectonic
events
may
be
accompanied
by
volcanic
eruption.
Those
faults
injected
with
lava
are
usually
recognizable
through
alignment
of
volcanic
cones.
Taking
into
account
stratigraphy,
geological
logs,
ancient
and
recent
volcano-
tectonic
and
tectonic
activity,
weathering
and
hydrothermal
effects,
the
basaltic
reservoir
structures
appear
to
be
heterogeneous
and
anisotropic.
Hence,
under-
ground
flow
within
basaltic
aquifers
might
not
respond
to
the
conditions
required
for
Darcy
flow.
Effectively,
basaltic
aquifers
include
horizontal
and
vertical
permeabilities.
Horizontal
permeability
is
located
within
the
horizontal
water-bearing
layers,
where
underground
flow
occurs
through
fissured
media
(with
shrinkage
cracks)
and
porous
media
(ignimbrites
or
alluvium).
Laterally,
these
layers
are
interrupted
owing
to
their
pinching
out
or
because
of
tectonic
activity,
with
or
without
fissure
eruption.
Vertical
permeability
occurs
in
tectonic
faults,
which
play
an
essential
role
in
the
aquifer
permeability
as
they
can
bring
into
contact
laterally
limited
horizontal
pervious
and
impervious
layers.
When
tectonic
faults
are
not
reactivated,
they
lose
their
basalt
alluvium
pb
„y
e
e.
.
• .
••
.
.•
41
.
.
. . .
.
.
.
.
• •
"
"
••
••
••
••
••
.•
e.o
o.e••
•••.e..e.
. .
.e
.
e
. .
e
.
o
.
•1:•
%•/..•
••
••••• ••••• ::•••• •••••
:
41';
i
t
Ke:'c'
s
ireZN
....1•
•••
'4Z*
Val
M.
Jalludin,
M.
Razack
/
Journal
of
Hydrology
155
(1994)
237-250
241
Fig.
2.
Schematic
model
for
basalt
aquifer
recharge:
pb,
piezometric
level
in
basalt
aquifer;
pa,
piezometric
level
in
wadi
underflow.
permeability
as
a
result
of
hydrothermal
action
and
weathering,
and
become
impervious
zones
in
the
aquifer.
Within
an
ancient
geological
formation,
a
basaltic
reservoir
will
have
been
sub-
jected
to
a
high
degree
of
weathering
as
well
as
intense
hydrothermal
action,
which
act
to
decrease
the
types
of
permeability.
In
fact,
there
are
no
water
wells
within
the
Adold
basalts
(25
Ma),
as
they
are
completely
hydrothermalized
and
weathered.
The
main
process
of
porosity
and
permeability
decrease
has
been
described
by
Davis
(1974).
Porosity
and
permeability
have
not
been
studied
in
Djibouti;
however,
some
values
have
been
presented
for
porosities
within
Deccan
trap
basalts
by
Deolankar
(1980)
and
in
tuffaceous
rocks
by
Foster
et
al.
(1985).
Hydrothermal
interaction
experiments
on
the
tholeitic
basalt
from
the
Umtanum
flow
of
the
Columbia
River
Plateau
showed
several
reaction
products—silica,
illite,
scapolite,
wairakite,
heulandite,
K-feldspar
and
smectite
(Moore
et
al.,
1985).
The
extremely
weathered
surface
of
basalts
and
the
high
degree
of
argillization
(indicated
by
red
soils
with
carbonate
concretions)
in
some
areas
make
infiltration
negligible.
Infiltration
operates
essentially
in
wadi
alluvial
aquifers
during
flows
generated
by
rainfall.
Alluvium
along
wadis
directly
overlies
the
basalts
and
gener-
ally
follows
tectonic
fractures.
In
these
circumstances,
wadi
underflows
act
as
inter-
mediates
in
the
basaltic
aquifer
recharge,
mainly
through
tectonic
faults
but
also
through
horizontal
pervious
layers
(Fig.
2).
4.
Pumping
tests
Volcanic
aquifers
in
the
Republic
of
Djibouti
are
exploited
by
more
than
100
wells.
Nevertheless,
pumping
test
data
are
available
for
only
a
few
of
them.
These
data,
after
a
preliminary
analysis
of
their
reliability,
were
included
in
the
study.
Wells
are
located
mainly
in
the
southern
part
of
the
Republic
of
Djibouti
(Fig.
3);
17
wells,
including
242
M.
Jalludin,
M.
Razack
/
Journal
of
Hydrology
155
(
1994)
237-250
1
42°
43
01
,
i
12°
t
1
Fig.
3.
Studied
well
locations
with
main
geological
contacts.
piezometers,
tap
stratiform
basalt
aquifers
and
23
tap
Dalha
basalt
aquifers.
The
wells
can
be
considered
complete
as
they
penetrate
most
of
the
aquifer.
As
indicated
above,
the
system
can
be
described
as
a
multilayered
aquifer
dissected
by
tectonic
faults
acting
as
drains
or
as
impervious
walls.
A
first
analysis
of
drawdown
data
has
been
processed
using
arithmetical
graphs
to
characterize
aquifer
behaviour
as
far
as
possible.
Drawdowns
related
to
aquifer
losses
and
well
losses
are
significantly
higher
in
the
Dalha
basalts
(Fig.
4).
Typical
loga-
rithmic
curves
as
predicted
by
theoretical
models
for
homogeneous,
isotropic
and
infinite
aquifers
occur
in
most
wells
despite
the
particular
reservoir
structure
of
the
basalts.
Several
stabilizations
and
slope
changes
appear
in
the
water-table
evolution
of
the
stratiform
basalts,
whereas
this
evolution
seems
to
be
more
regular
for
the
Dalha
basalts.
Similar
behaviour
of
volcanic
formations
during
pumping
has
been
encountered
in
other
parts
of
the
world
(Versey
and
Singh,
1982;
Dheilly-Carn,
1983;
Uhl
and
Joshi,
1986).
Pumping
tests
help
in
delineating
quantitative
hydraulic
characteristics
of
aquifers
and
determine
boundary
conditions.
Thus
several
drawdown
and
recovery
data
sets
were
analysed
or
re-interpreted
for
the
purpose
of
studying
the
stratiform
and
Dalha
basalts.
Before
processing
hydrodynamic
interpretations,
the
wellbore
storage
effect
has
been
evaluated
(Plegat,
1987),
as
it
is
not
clearly
apparent
in
some
cases.
The
Cooper—Jacob
method
(1947),
which
results
from
a
logarithmic
approxi-
mation
of
the
Theis
equation
(1935),
has
also
been
applied.
Drawdowns
vs.
time
are
plotted
on
a
semilog
graph
where
the
slope
for
one
logarithmic
cycle
(c)
of
the
0.6
Zi3
0.5
E
0.4
0.3
0.2
E
0.1
0
D
RA
WDO
WN,
me
te
r
M.
Jalludin,
M.
Razack
/
Journal
of
Hydrology
155
(
1994)
237-250
243
A
••
A
A
A•••••••
GABLAGALAN
0
PK20-
7
A
HINDI
MEM
AWRAWSA
3
0 0
co3SC3
39
C
DADIN
3
0
50000
100000
150000
200000
TIME,
seconds
Fig,
4.
Drawdowns
on
arithmetic
plot:
A,
stratiform
basalts;
B,
Dalha
basalts.
straight
line
and
the
pumping
rate
(Q)
are
used
to
calculate
transmissivity
using
T
=
0.183Q/c.
The
Theis
method
was
applied
for
recovery
data.
Residual
drawdown
vs.
t/t'
is
reported,
where
t
is
the
time
from
the
start
of
the
pumping
test
and
t'
is
the
time
since
it
stopped.
The
same
calculation
as
above
gives
the
transmissivity.
After
the
wellbore
storage
effect
period,
an
ideal
drawdown
on
a
semilog
graph
follows
a
straight
line.
Depending
on
whether
the
limits
of
the
aquifer
are
permeable
or
form
a
barrier,
the
drawdown
evolves
towards
stabilization
or
an
increased
slope,
respectively.
The
stratiform
basalts
and
the
Dalha
basalts
display
a
linear
evolution
intercepted
by
slope
changes
(Fig.
5).
Stabilization
occurs,
for
example,
on
the
graph
for
the
well
AWRAWSA
3
after
a
straight
line
with
increasing
slope
demonstrating
a
highly
permeable
boundary.
In
an
intermediate
state
between
a
straight
line
and
stabilization,
the
graph
for
the
well
HINDI
shows
a
decrease
of
slope
owing
to
the
presence
of
a
more
permeable
area
surrounding
the
well.
A
barrier
boundary
is
shown
for
the
wells
DADIN
3
and
E31
where
the
slope
increases.
In
other
respect,
the
wells
GABLAGALAN
and
E31
display
alternatively
several
phases
of
stabilization
and
straight
line
evolution,
which
seem
to
take
place
exclusively
in
stratiform
basalts.
Few
wells
tap
basaltic
aquifers
overlain
by
superficial
alluvium
with
which
they
have
a
direct
hydraulic
relationship.
These
alluvial
aquifers
are
found
in
very
narrow
valleys
formed
by
wadis
cutting
the
basalts.
Hence
pumping
tests
behave
initially
as
for
an
unconfined
aquifer
in
a
porous
medium;
the
graphs
for
DADIN
6
and
OUEAH
8
(Fig.
6)
show
drawdowns
studied
by
Boulton
(1963).
In
this
case,
after
a
complete
or
incomplete
stabilization,
drawdowns
increase
regularly
and
pro-
gressively,
as
and
when
drainage
attenuates
and
disappears
in
the
aquifer.
Curves
for
recovery
observations
(Fig.
7)
show
classical
evolution
of
residual
drawdowns
vs.
t/t'.
These
data
are
very
helpful
when
drawdown
data
sets
cannot
E
31
D
RA
WDO
WN,
me
te
r
0.6
E
31
0.5
0.4
g
0.3
GABLAGALAN
A
A4
A
A
,
A&
0.2
AA
A-
1
AA"
A
PK20-
7
aCCCC
0
7
5
3
A
AWRAWSA
3
liP
l
scts
5C
1C
00
0 °
'
3-j
DADIN
3
HINDI
dico
DGEP
00,11
0.111.
too
t
EstE
naminil
i
1
DADIN
6
Ursa
i
NE
••
OUEAH
8
gaime
meNalla
244
M.
Jalludin,
M.
Razack
/
Journal
of
Hydrology
155
(
1994
)
237-250
1
100
1000
10000
100000
1000000
TIME,
seconds
Fig.
5.
Drawdowns
on
semilog
plot:
A,
stratiform
basalts;
B,
Dalha
basalts.
be
interpreted
because
of
aquifer
heterogeneities.
At
the
end
of
the
recovery,
the
aquifer
limit
effects
described
by
Forkasiewicz
(1972)
can
be
observed.
5.
Results
and
discussion
Results
are
presented
in
Tables
1
and
2.
The
transmissivity
values
obtained
reflect
significant
differences
between
the
stratiform
and
Dalha
basalts.
Transmissivity
in
stratiform
basalts
ranges
from
1.5
x
10
-4
to
5.7
x
10-1
m
2
s
-1
and
in
Dalha
basalts
from
2.9
x
10
-6
to
1.6
x
10
-2
m
2
s
-1
.
The
well
MIDGAOUNE
shows
a
low
transmissivity
that
could
be
explained
by
its
location
over
a
highly
fissured
area
in
southern
Djibouti,
but
injections
of
lava
plug
the
fissures.
10
100
1000
10000
100000
Time,
sec
Fig.
6.
Drawdowns
in
wells
where
basaltic
aquifer
is
overlain
by
local
alluvium
underflow.
10
0
0
0.1
GABLAGAIAN
ae
on
OPIII
Me
0
PK20
-
7
PK20
-
1
E••
DIKHIL
8
VVRAVVSA
3
0
OUEAH
10
° °
DABADERE
2
B
0.5
0.4
0.3
0.2
c)
0.1
D
RA
WDO
WN,
me
te
r
0
M.
Jalludin,
M.
Razack
/
Journal
of
Hydrology
155
(
1994)
237-250
245
1
10
100
1000
10000
t
/
Fig.
7.
Recovery:
A,
stratiform
basalts;
B,
Dalha
basalts.
There
is
a
systematic
difference
between
transmissivity
readings
calculated
by
drawdown
and
recovery
methods.
This
variation
seems
to
be
higher
in
the
stratiform
basalts,
where
aquifer
limit
effects
are
more
evident
during
pumping.
Only
a
few
storage
coefficient
values
were
calculated,
as
piezometers
near
pumping
wells
are
rare.
At
first
the
values
suggest
a
confined
aquifer
for
stratiform
basalts,
and
parti-
cularly
for
Gulf
basalts
near
Djibouti,
and
a
confined
or
semiconfined
aquifer
for
the
Dalha
basalt
aquifers,
depending
on
the
area.
Furthermore,
step
drawdown
data
interpretation
based
on
the
Cooper—Jacob
graphical
method,
s
=
BQ
+
CQ
2
,
and
Rorabaugh's
method,
s
=
BQ
+
CQ",
led
to
the
aquifer
loss
coefficient
B
and
well
loss
coefficient
C.
Four
step
drawdown
data
sets
within
stratiform
basalts
and
10
within
Dalha
basalts
were
analysed.
Generally,
aquifer
and
well
loss
coefficients
are
lower
for
stratiform
basalts.
After
drawdown
correction
for
well
losses,
a
series
of
simulations,
run
on
a
PC,
were
processed
for
those
wells
to
estimate
the
storage
coefficient
(Fig.
8).
The
simula-
tion
results
confirm
the
Dalha
basalt
identification,
but
Gulf
basalts
appear
to
con-
stitute
a
semiconfined
aquifer.
Using
a
semiconfined
characteristic
for
stratiform
basalts
and
a
confined
characteristic
for
Dalha
basalts,
specific
capacities
were
gen-
erated
for
wells
for
1
x
10
5
s
of
pumping.
Uhl
and
Joshi
(1986)
calculated
specific
capacities
for
12
h
of
pumping
for
wells
in
Deccan
trap
basalts.
The
results
for
the
stratiform
and
Dalha
basalts
are
illustrated
in
Fig.
9,
where
specific
capacities
are
plotted
against
transmissivities,
taking
into
account
the
error
made
in
applying
a
Table
1
Pumping
tests
results
for
stratiform
basalts
Well
Diameter
Q
B
C/n
T
T
S
S
Specific
(m)
(m
3
h
-1
)
m
(m
3
h
-1
)
10
-1
m
(m
3
h
-1
)10
-
"
a
drawdown
recovery
(piezometric)
(simulated)
capacity
(m
2
s
-1)
(m
2
s
-1
)
(m
3
m
1
)
E
8
0.244
43.5
3.3
x
10
-2
66.85
E
19
0.244
36.0
1.2
x
10
-2
25.46
E25
0.244
65.0
0.0057
0.000011/2.9
5.8
x
10
-2
5.7
x
10
-1
6
x
10
-4
124.33
E
28
0.203
50.0
2.8
x
10
-1
510.01
E
31
0.195
62.2
0.0040
0.0003
8.2
x
10
-2
1.4
x
10
-1
2
x
10
-5
146.63
RG
3
0.244
19.0
1.2
x
10
-2
25.46
PK20-1
0.193
40.8
8.3
x
10
-2
158.28
PK20-7
0.254
45.1
0.0053
0.0012
3.7
x
10
-2
9
x
10
-4
82.99
Piezometer
2.9
x
10
-2
2
x
10
-5
PK20-5
GEDMARREH
0.244
18.0
3.3
x
10
-4
2.6
x
10
-3
0.842
AGADERE
3
0.209
42.0
0.0120
0.0006
5.7
x
10
-2
1
x
10
-4
111.21
GODCHABEL
0.203
64.5
1.4
x
10
-1
262.32
MIDGAOUNE
0.127
11.8
1.5
x
10
-4
0.372
GABLAGALAN
0.168
10.1
5.1
x
10
-3
6.5
x
10
-3
10.88
MOULOUD
3
0.203
29.0
3.9
x
10
-2
77.16
DAWDAWYA
0.250
7.4
1.0
x
10
-2
21.45
KOUTABOUYA
0.244
48.6
1.9
x
10
-2
39.46
CHABELLEY
0.152
29.5
9.0
x
10
-3
2.6
x
10
-2
51.05
M
.
R
azack
/
J
our
nal
of
H
yd
rol
o
g
y
155
(
199
4
)
23
7
-2
50
a
When
not
equal
to
2,
n
is
given
in
the
table.
Table
2
Pumping
tests
results
for
Dalha
basalts
Well
Diameter
Q
B
C/n
T
T
S
S
Specific
(m)
(m
3
h
-1
)
m
(m
3
h
-
')10
-1
m
(m
3
h
-1
)10
- '
a
drawdown
recovery
(piezometric)
(simulated)
capacity
(m
2
s
-i
)
(m
2
s
-i
)
(m
3
h
-1
m
1
)
DADIN
1
0.244
7.0
0.3770
0.0758
3.4
x
10
-4
3.6
x
10
-4
6
x
10
-4
0.964
DADIN
3
0.244
38.8
0.0737 0.0027
2.4
x
10
-3
2
x
10
-6
4.596
DADIN
6
0.244
66.0
1.5
x
10
-3
1.5
x
10
-3
3.15
AWRAWSA
3
0.168
11.6
0.8080
0.0414
3.2
x
10
-4
3.8
x
10
-4
3
x
10
-4
4
x
10
-4
0.849
Piezometer
AWR3
2.9
x
10
-4
3.7
x
10
-4
5
x
10
-6
AWRAWSA
4
0.168
10.8
0.6800
0.1160
1.4
x
10
-3
7.5
x
10
-4
3
x
10
-7
2.462
AWRAWSA
5
0.168
9.3
1.5
x
10
-3
1.6
x
10
-3
3.045
Piezometer
AWR5
1.5
x
10
-4
4
x
10
-3
OUEAH
10
0.244
16.8
0.1140
0.00011/3.6
1.6
x
10
-3
2.9
x
10
-3
4
x
10
-3
3.35
OUEAH
8
0.195
21.7
1.6
x
10
-3
3.502
OUEAH
4
0.178
8.4
1.3
x
10
-3
2.67
DIKHIL
8
0.100
30.5
0.3263
0.0134
9.1
x
10
-4
1.8
x
10
-3
2
x
10
-4
2.076
DIKHIL
2
0.178
19.2
3.5
x
10
-4
0.764
DABADERE
2
0.195
21.3
0.0590
0.0213
5.7
x
10
-3
5.7
x
10
-3
2
x
10
-4
12.684
HINDI
0.152
11.5
0.4050
0.0097
1.1
x
10
-3
2.2
x
10
-3
3
x
10
-5
2.361
BIEYDLE
0.152
13.1
3.5
x
10
-4
3.7
x
10
-4
0.753
DAASBIYO
0.244
6.5
3.9
x
10
-3
1.0
x
10
-4
0.099
HOLL
HOLL
3
0.152
4.0
3.1
x
10
-4
0.671
ASAGAYLA
0.168
12.0
0.4169
0.0220
1.6
x
10
-2
4.3
x
10
-3
6
x
10
-4
35.2
ADAYLOU
0.168
6.8
0.9547
0.2490
1.5
x
10
-4
3
x
10
-5
0.359
Piezometer
ADAYL
2.5
x
10
-4
8.1
x
10
-4
4
x
10
-2
BOLLI
0.168
1.5
2.9
x
10
-6
0.008
DORRA
0.168
6.7
2.9
x
10
-4
9.9
x
10
-5
0.635
R
az
ack
/
J
ou
r
nal
of
H
yd
rol
o
gy
1
55
(199
4
)
2
37
-250
a
When
not
equal
to
2,
n
is
given
in
the
table.
6
E
DRAWD
OWN,
me
te
r
248
M.
Jalludin,
M.
Razack
Journal
of
Hydrology
155
(
1994)
237-250
0.6
0.4
2
cer
rencr.
E
25
0.2
1
0
16
ccc
(
c
°
12
2
Fig.
8.
Pumping
simulation
in
stratiform
and
Dalha
basalts:
1,
measured
data
corrected
for
well
losses;
2,
simulated
drawdowns.
storage
coefficient
value.
This
error
has
been
estimated
to
be
about
10%
for
the
stratiform
basalts
and
approximately
20%
for
the
Dalha
basalts,
as
the
simulated
storage
coefficient
values
for
the
latter
have
a
larger
range.
The
relationship
between
specific
capacity
and
transmissivity
is
linear,
and
both
basalts
appear
on
the
same
line
but
are
separated
clearly
into
two
groups
with
a
transitional
zone:
higher
transmissivity
values
for
the
stratiform
basalts
are
related
to
higher
specific
capacities,
and
lower
values
characterize
the
Dalha
basalts.
Specific
capacity
values
range
between
0.37
and
510.0
m
3
h
-1
m
-1
for
stratiform
basalts
and
between
0.008
and
35.2
m
3
h
-1
m
-1
for
the
Dalha
basalts.
6.
Conclusion
Two
intensively
fissured
trap
series
composed
of
regular
lava
flows,
the
Dalha
basalts
(9-3.4
Ma)
and
the
stratiform
basalts
(3.4-1
Ma),
were
studied
to
identify
their
hydrodynamic
properties.
Analyses
of
pumping
tests
revealed
the
following
major
differences
between
the
series:
(1)
higher
and
regular
drawdowns
were
observed
in
the
Dalha
basalts
for
lower
pumping
rates
than
in
the
stratiform
basalts,
where
drawdowns
typically
show
stabilization
steps.
(2)
Aquifer
and
well
loss
coefficients
deduced
from
step
drawdown
test
studies
were
lower
in
the
stratiform
basalts.
8
i
m
occo
n
or
AWRAWSA
3
4
1
0
100
1000
10000
100000
1000000
TIME,
seconds
GULF
BASALTS
nu
0
DALHA
BASALTS
TRANSMISSIVITY,
m2/h
0
M.
Jalludin,
M.
Razack
Journal
of
Hydrology
155
(
1994)
237-250
249
0.001
0.1
10
1000
Fig.
9.
Specific
capacity
vs.
transmissivity
for
stratiform
and
Dalha
basalts.
(3)
Considering
the
logarithmic
evolution
of
drawdown
vs.
time,
classical
analytical
models
could
be
applied
in
spite
of
the
vertical
and
horizontal
geological
hetero-
geneities
of
these
aquifers.
The
results
show
that
the
transmissivities
of
the
stratiform
basalts
are
two
orders
of
magnitude
higher
than
those
of
the
Dalha
basalts.
Owing
to
the
effects
of
weathering,
hydrothermal
activity
and
tectonism,
which
occur
simultaneously
during
geological
time,
the
permeability
of
the
basaltic
forma-
tions
varies
appreciably.
Within
recent
stratiform
basalts,
the
original permeability
has
remained
practically
unchanged
since
3.4
Ma.
This
results
in
very
high
trans-
missivities
that
range
over
three
orders
of
magnitude
according
to
the
degree
of
fissure
opening
and
the
permeability
of
porous
horizontal
layers.
Hydrothermal
action
and
weathering
over
9
My
have
deeply
degraded
the
primary
permeability
of
the
Dalha
basalts
and
seem
to
be
the
dominant
factor,
as,
despite
an
important
cluster
of
faults,
the
Dalha
basalts
show
lower
transmissivities
than
expected.
Acknowledgements
We
thank
the
Ministry
of
Co-operation
of
France
for
financial
support
of
the
1000
10
0.1
0.001
250
M.
Jalludin,
M.
Razack
Journal
of
Hydrology
155
(
1994)
237-250
hydrogeological
programme.
We
appreciate
the
contribution
of
A.M.
Fida,
A.
Gaffaneh
and
Y.
Osman
in
fieldwork.
References
Arthaud,
F.
and
Jalludin,
M.,
1993.
Cartographie
des
fractures
d'un
reservoir
hydrogeologique
en
milieu
volcanique
a
partir
de
donnees
SPOT
controlees
sur
le
terrain.
(Basaltes
miocenes
de
Djibouti.)
4emes
Journ.
Sci.
Teledetection,
Montreal,
in
press.
Arthaud,
F.,
Choukroune,
P.
and
Robineau,
B.,
1980.
Tectonique,
microtectonique
et
evolution
structurale
du
golfe
de
Tadjourah
et
du
sud
de
la
Depression
Afar
(Republique
de
Djibouti).
Bull.
Soc.
Geol.
Fr.,
XXII(6):
909-915.
Barberi,
F.,
Ferrara,
G.,
Santacroce,
R.
and
Varet,
J.,
1975.
Structural
evolution
of
the
Afar
triple
junction.
In:
A.
Pilger
and
A.
Rosier
(Editors),
Afar
Depression
of
Ethiopia.
Inter-Union
Commission
on
Geodynamics,
Sci.
Rep.
14,
pp.
38-55.
Black,
R.,
Morton,
W.H.
and
Rex,
D.C.,
1975.
Block
tilting
and
volcanism
within
the
Afar
in
the
light
of
recent
K/Ar
age
data.
In:
A.
Pilger
and
A.
Rosier
(Editors),
Afar
Depression
of
Ethiopia.
Inter-Union
Commission
on
Geodynamics,
Sci.
Rep.
14,
pp.
296-300.
Boulton,
N.S.,
1963.
Analysis
of
data
from
non-equilibrium
pumping
tests
allowing
for
delayed
yield
from
storage.
Proc.
Inst.
Civ.
Eng.,
26:
469-482.
Chessex,
R.,
Delaloye,
M.,
Muller,
J.
and
Weidmann,
M.,
1975.
Evolution
of
the
volcanic
region
of
Ali-
Sabieh
in
the
light
of
K/Ar
determinations.
In:
A.
Pilger
and
A.
Rosier
(Editors),
Afar
Depression
of
Ethiopia.
Inter-Union
Commission
on
Geodynamics,
Sci.
Rep.
14,
pp.
221-227.
Cooper,
Jr.,
H.H.
and
Jacob,
C.E.,
1946.
A
generalized
graphical
method
for
evaluating
formation
con-
stants
and
summarizing
well
field
history.
Trans.
Am.
Geophys.
Union,
27:
526-534.
Davis,
S.N.,
1974.
Change
of
porosity
and
permeability
of
basalt
with
geologic
time.
Int.
Symp.
on
Hydrology
of
Volcanic
Rocks,
UNESCO,
Lanzarote.
Deolankar,
S.B.,
1980.
The
Deccan
basalts
of
Maharashtra,
India.
Their
potential
as
aquifers.
Ground
Water,
18(5):
434-437.
Dheilly-Cam,
A.,
1983.
Contribution
a
l'etude
hydrogeologique
des
volcanites
du
Tregor.
These
de
3eme
cycle,
Universite
Montpellier
II,
168
pp.
Forkasiewicz,
J.,
1972.
Interpretation
des
donnees
de
pompages
d'essai
pour
revaluation
des
parametres
des
aquiferes,
2nd
edn.
BRGM,
72
SGN
273
AME.
Foster,
S.D.S.,
Ellis,
T.A.,
Losilla-Penon,
M.
and
Rodriguez-Estrada,
V.H.,
1985.
Role
of
volcanic
tuffs
in
ground-water
regime
of
Valle
Central,
Costa
Rica.
Ground
Water,
23(6):
795-801.
Gaulier,
J.M.
and
Huchon,
Ph.,
1991.
Tectonic
evolution
of
Afar
triple
junction.
Bull.
Soc.
Geol.
Fr.,
162(3):
451-464.
Jalludin,
M.,
Anis,
A.,
Idriss,
G.,
Mohamed,
I.,
San
Juan,
B.,
Gaulier,
J.M.,
Zotskina,
E.,
Gaffaneh,
A.
and
Haga,
A.,
1990.
Alimentation
en
eau
potable
de
la
ville
d'Ali
Sabieh.
Etude
hydrogeologique
des
basaltes
de
Dalha.
Sci.
Tech.,
5:
13-21.
Moore,
E.L.,
Ulma,
G.C.
and
Grandstaff,
D.E.,
1985.
Hydrothermal
interaction
of
Columbia
Plateau
basalt
from
the
Umtanum
flow
(Washington,
USA)
with
its
coexisting
groundwater.
Chem.
Geol.
49:
53-71.
Plegat,
R.,
1987.
Prise
en
compte
de
la
perte
de
charge
a
rentree
lors
de
retude
des
nappes
d'eau
souter-
raines
par
pompage
dans
des
puits
a
effet
de
capacite.
C.R.
Acad.
Sci.,
Ser.
II,
304(11).
Theis,
C.V.,
1935.
The
relation
between
the
lowering
of
the
piezometric
surface
and
the
rate
and
duration
of
discharge
of
a
well
using
ground-water
storage.
Trans.
Am.
Geophys.
Union,
2:
519-524.
Versey,
H.R.
and
Singh,
B.K.,
1982.
Groundwater
in
Deccan
basalts
of
the
Betwa
basin,
India.
J.
Hydrol.,
58:
279-306.
Uhl,
Jr.
and
Joshi,
V.G.,
1986.
Results
of
pumping
tests
in
Deccan
trap
basalts
of
central
India.
J.
Hydrol.,
86:
147-168.