Water-Level Declines in the Woodbine, Paluxy, and Trinity Aquifers of North-Central Texas

Abstract

Groundwater mining of the Woodbine, Paluxy, and Trinity aquifers has led to substantial water-level declines in North-Central Texas since the turn of the century. Water-level maps constructed from R. T. Hill's 1901 well survey data show that water levels were initially above land surface before development. Numerous wells were drilled for water supply because the wells flowed at land surface. Water levels declined rapidly, and many of the wells around Fort Worth stopped flowing by 1914. Many of these wells were then abandoned, which slowed the rate of water-level decline. Since the turn of the century, water levels have declined nearly 850 ft in the Trinity aquifer in the Fort Worth area. As of 1990, water levels had declined about 400 ft in the Woodbine aquifer near Dallas and 450 ft in the Paluxy aquifer near Fort Worth. Maps drawn on the basis of water-level measurements in 1935, 1955, 1960, 1970, 1980, and 1990 show how the shape of potentiometric surfaces has evolved during the century. This great drawdown in water levels has increased pumping costs, reversed groundwater flow directions in the Dallas-Fort Worth and Sherman areas, and may have affected water quality. Land subsidence from water-level decline has not been observed in North-Central Texas, perhaps because of the structural stability of the geologic units or a consolidation time lag. Pumping costs and water-quality problems have caused many groundwater users to switch to surface sources of water. Consequently, the rate of water-level decline has decreased in some parts of the aquifers, and in the case of the Paluxy aquifer, this may have caused recent water-level recovery.

Full text

Water-Level
Declines
in
the
Woodbine,
Paluxy,
and
Trinity
Aquifers
of
North-Central
Texas
Robert E. Mace, Alan R. Dutton,
and
H.
Seay
Nance
Bureau
of
Economic Geology,
The
University
of
Texas
at
Austin,
University
Station,
Box
X,
Austin, TX 78713-7508
Abstract
Ground-water
mining
of
the
Woodbine, Paluxy,
and
Trinity aquifers
has
led to
substantial
water-level
declines
in
North-Central
Texas since
the
turn
of
the
century. Water-level
maps
constructed
from
R.
T.
Hill's 1901 well survey
data
show
that
water
levels were initially above
land
surface before development.
Numerous
wells were drilled for
water
supply
because
the
wells flowed
at
land
surface.
Water
levels
declined rapidly,
and
many
of
the
wells
around
Fort
Worth stopped flowing
by
1914.
Many
of
these
wells
were
then
abandoned, which slowed
the
rate
of water-level decline. Since
the
turn
of
the
century,
water
levels
have
declined
nearly
850 ft
in
the
Trinity
aquifer
in
the
Fort
Worth
area.
As
of
1990,
water
levels
had
declined
about
400 ft
in
the
Woodbine
aquifer
near
Dallas
and
450 ft
in
the
Paluxy
aquifer
near
Fort
Worth.
Maps
drawn
on
the
basis
of
water-level
measurements
in
1935, 1955, 1960, 1970, 1980,
and
1990 show how
the
shape
of
potentiometric surfaces
has
evolved
during
the
century.
This
great
drawdown
in
water
levels
has
increased
pumping
costs,
reversed
ground-water
flow
directions
in
the
Dallas-Fort
Worth
and
Sherman
areas,
and
may
have
affected
water
quality.
Land
subsidence from water-level decline
has
not
been
observed
in
North-Central
Texas,
perhaps
because
of
the
structural
stability of
the
geologic
units
or
a consolidation
time
lag.
Pumping
costs
and
water-quality
problems
have
caused
many
ground-water
users
to switch to surface sources
of
water
. Consequently,
the
rate
of
water-level decline
has
decreased
in
some
parts
of
the
aquifers,
and
in
the
case
of
the
Paluxy
aquifer,
this
may
have
caused
recent
water-level recovery.
Introduction
Ground-water mining occurs
when
more
water
is
pumped
from
an
aquifer
than
is replenished
by
recharge
or
cross-
formational
flow.
This causes
water
levels to decline
in
the
aquifer
and
increases
the
costs
of
pumping
water
from
aquifers. The removal of large
amounts
of
ground
water
can
also change regional ground-water
flow
directions, affect
water
chemistry,
and
cause
land
subsidence.
Many
aquifers
have
undergone ground-water mining such
as
the
Ogallala aquifer
in
northwest
Texas (Gutentag
et
al., 1984),
the
Gulf
Coast
region
in
Texas (Winslow and
Wood,
1959),
and
the
sediments
beneath
Mexico City, Mexico (Poland
and
Davis, 1969).
Early
in
the
century
well
drillers
discovered
that
Cretaceous aquifers of North-Central Texas
had
water
levels
that
reached
above
land
surface
in
many
areas.
For
instance,
Waco
was
known
as
the
"City
of
Geysers" because
of
its
many
flowing wells
that
produced
between
500,000
and
1,500,000 gallons
per
day (Hill, 1901).
The
Fort
Worth
area
was also
known
for
numerous
flowing wells. Because
no
pumping
was
required
for
production,
these
aquifers
were
easily
exploited
and
numerous
wells
were
drilled
.
Because
of
this
wide-scale
release
of
artesian
water
pressure,
water
levels declined
rapidly
until
pumping
was
required
to produce
water
from wells
in
every portion
of
the
aquifers.
Over
the
past
century,
these
aquifers
have
undergone
ground-water
mining
as
evidenced by
steadily
decreasing
water
levels.
Decreasing
water
levels
in
the
Cretaceous
aquifers
of
North-Central
Texas
have
been
documented
with
water-
level
maps
by
Klemt
et
al. (1976),
Nordstrom
(1982a),
Nordstrom
(1987),
and
Baker
et
al. (1990). However,
the
complete
history
of
water-level declines
in
the
aquifers
has
not
been
represented
because extensive water-level
maps
have
not
been
made
for
the
first
half
of
this
century.
An
exception is for a
small
area
in
the
Coryell
and
McLennan
County
area
where
Klemt
et
al. (1975)
generated
a pre-
development water-level
map
using
Hill's (1901)
data
to aid
in
calibrating
a
numerical
ground-water
flow model.
This
paper
presents
a
predevelopment
water-level
map
and
a
1930's water-level
map,
and
it
discusses
implications for
ground-water
withdrawal
for
the
Woodbine, Paluxy,
and
Trinity
aquifers
of
North-Central
Texas
.
This
work
is
important
because
predevelopment
water-level
maps
are
useful
for
modeling
of
regional-scale
ground-water
flow,
understanding
the
regional
distribution
of total drawdowns,
and
quantifying
the
potential
for
land
subsidence.
Hydrogeologic
Setting
The Cretaceous of
North-Central
Texas hosts
three
major
aquifers
in
the
sandstones
of
the
Woodbine
Formation,
Paluxy
Formation,
and
Trinity Group. Water-bearing
strata
of
the
Trinity
Group
consist
of
the
Hensell
and
Hosston
Formations
(Fig.
1).
As confining
units
pinch
out
between
the
Hensell, Hosston,
and
Paluxy, formation
names
change.
For
instance,
west
of
where
the
limestone-
and
shale-rich
Pearsall
and
Sligo Members of
the
Trinity
Group pinch out,
the
Hensell
and
Hosston
Members
are
called
the
Travis

414 TRANSACTIONS OF
THE
GULF COAST ASSOCIATION
OF
GEOLOGICAL SOCIETIES, VOL. XLIV, 1994
West
East
0Aa5779c
Figure
1.
Stratigraphy
and
formation
nomenclature
of
the
Lower
Cretaceous
aquifers
of
North-Central
Texas.
Peak
or
Twin
Mountains
Formation,
depending
on
locality.
Where
the
Glen Rose
Formation
pinches
out
between
the
Paluxy
and
the
Hensell
and
Hosston
Formations
(Twin
Mountains
Formation),
the
remaining
interval
is called
the
Antlers
Formation. Some
stratigraphers
include
the
Paluxy
Formation
in
the
definition
of
the
Trinity
Group
(for
example,
the
Texas
Water
Development Board); however,
most
stratigraphers
do
not
incorporate
the
Paluxy
with
the
Trinity
Group
on
the
basis
of
work by Hill ( 1937)
and
Lozo
(1949).
Stratigraphic
descriptions
of
the
Woodbine, Paluxy,
and
Trinity
sandstones
are
presented
in
Oliver
(1971),
Caughey
( 1977),
and
Hall
(1976), respectively.
This
paper
refers to
the
Hensell
and
Hosston
Formations
(Twin Moun-
tains
and
Travis
Peak
Formations)
as
the
Trinity
aquifer.
Lower
Cretaceous
strata
form a wedge
that
thickens
toward
the
East
Texas
Basin
and
are
about
7,400 ft
thick
at
the
Mexia
Fault
Zone.
The
regional dip
ranges
from
about
0.3°
near
the
outcrop
to
nearly
1.5°
at
the
Mexia
Fault
Zone. The 250- to 375-ft-thick Woodbine is a medium-
to coarse-grained iron-rich
sandstone
with
some clay
and
lignite
seams
. Most wells
are
completed
in
the
lower
part
of
the
formation, which yields
better
quality
ground
water
.
Transmissivities
in
the
Woodbine
aquifer
range
from
101.
7 to 103 6 ft2
/day
with
a
mean
of 10
26
ft2/day.
The
100-
to
400-ft-thick
Paluxy
aquifer
is a fine
sand
with
shale
and
sandy
shale
interbeds
. Transmissivities
in
the
Paluxy
aquifer
range
from 102·2 to 103·3 ft2/day,
with
a
mean
of
102·8
ft
2/day.
The
200- to
1,000-ft-thick
Trinity
aquifer
consists of fine to coarse
sandstone
with
some
shale
and
clay seams.
Transmissivities
in
the
Trinity
aquifer
range
from
101.
4 to 103·6 ft2/
day
,
with
a
mean
of
102·9 ft2/day.
Confining
layers
consist
of
the
Eagle
Ford
Shale,
Austin
Chalk,
and
Ozan
Formation
overlying
the
Woodbine,
Washita,
and
upper
and
middle
Fredericksburg
Groups
between
the
Woodbine
and
Paluxy,
and
the
Glen
Rose
between
the
Paluxy
and
Trinity
aquifer. Paleozoic rocks
consisting
of
sandstone,
limestone,
and
shale
underlie
the
Trinity
aquifer
(Hill, 1901;
Henningsen,
1962).
The
Eagle
Ford
Shale
includes
thin
beds
of
sandstone
and
limestone.
The
Washita
and
upper
and
middle
Fredericksburg
consist
of
limestone,
marl,
and
clay.
The
Glen
Rose
consists
of
limestone, marl,
and
shale.
Recharge occurs
by
precipitation
on
the
outcrops of
the
aquifers,
and
discharge occurs by cross-formational flow
in
the
subsurface
and
by
ground-water
pumping,
which
has
been active for
the
last
100 yr. Ground-
water
flow
rates
have
been
estimated
to be 10 to 40 ft/yr for
the
Woodbine aquifer
(Thompson, 1967)
and
1 to 2 ft/yr for
the
Trinity
aquifer
in
the
northern
part
of
the
study
area
(Antlers
Formation)
(Baker, 1960).
Baker
(1960)
estimated
the
flow
rate
in
the
Woodbine
aquifer
to be 15 ft/yr.
The
area
of
interest
for
this
study
extends
from
Montague,
Cooke,
and
Grayson
Counties
in
the
north
to
McLennan
County
in
the
south
and
from
Erath
and
Comanche Counties
in
the
west
to
the
Mexia
Fault
Zone
in
the
east
(Fig. 2).
OKLAHOMA I
Wtchtta Upltft
-:/
..
,..
Arbuckle/Ttshomtngo-
Belton Upltft
/.
_.
.
..<..
Ouachtta
Upltft
eq
R1ve~
· ·
:·.
· ···
/
'lJ I
II
~o~
J"
Austin
't-7>~
'I
~
salcone;_
?U~
J.'
--/
~
~
'<>'
San
Antonio 7 .
0
~
X e
'-'-;}'t-o<' 0
:t-'G
N
~<v
~
d<
.;3v~
0
0
100
0
100
200
30 0 km
Counties
a Montague n Hood
b Cooke 0 Johnson
c Grayson p Ellis
d Fannin q Erath
e Wise r Somervell
f Denton s Hill
g Collin t Navarro
h Hunt u Comache
Parker v Hamilton
j Tarrant w Bosque
k Dallas X Coryell
I Rockwall y
Mclennan
m Kaufman z Limestone
0Aa5778c
Figure
2.
Location
of
the
study
area.

MACE, DUTTON, NANCE
Methods
Water-Level
Data
Water-level
measurements
record changes in
the
hydrau-
lic
head
or
fluid
pressure
in
the
aquifer
over time.
Water
levels were
obtained
from two
data
sources: (
1)
the
Texas
Water
Development Board (TWDB)
and
(2)
R.
T.
Hill (1901).
Computerized
water-level
data
files from
the
TWDB
included
other
well
information
and
latitude/longitude
coordinates.
These
files
were
formatted
and
the
desired
portions
merged
in
a file
with
well
name,
well location,
water-level
measurement,
date
of
measurement,
formation,
and
depth
of
well.
In
this
file
are
22,250
measurements
that
were
taken
from 1899 to 1993. However,
data
coverage
over
that
time
span
is
nonuniform.
Most
of
the
data
(94 percent)
were
collected since 1960,
with
very poor
data
control (27 points) from 1899 to 1930.
Numerous
quantitative
and
qualitative
measurements
of
water
levels
in
the
Woodbine, Paluxy,
and
Trinity
aquifers
were
recorded
at
the
turn
of
the
century
by Hill (1901).
Besides
measuring
water
levels
in
wells,
he
contacted
officials
at
various
cities
and
small
towns
in
the
area
to
obtain
information
about
local
water-supply
wells.
Hill
recorded information on approximately 750 wells
in
Central
and
North-Central
Texas. A
large
portion
of
these
data
is
anecdotal
and
qualitative,
typically
only
reporting
the
formation,
approximate
location,
and
whether
or
not
the
well flowed
at
land
surface.
Of
the
750 wells
in
Hill's report,
it
was
determined
that
149 wells
had
the
needed informa-
tion
concerning water-level
measurement,
indication
of
flow
at
surface,
and
well location for construction
of
a water-level
map.
Of
the
149 wells, 46
were
located
in
the
Woodbine
aquifer, 40 were located
in
the
Paluxy
aquifer
,
and
63 were
located
in
the
Trinity aquifer.
Water-Level Map
Construction
Water-level
maps
were constructed for 1900, 1935, 1960,
1970, 1980,
and
1990,
and
a 1955 water-level
map
developed
by Nordstrom (1982a) extends into
the
study
area.
The
1900
water-level
maps
represent
approximate
positions
of
water
levels
in
the
aquifers.
This
is because
many
flowing wells
did
not
have
their
artesian
pressures
measured
at
the
time
of
Hill's
(1901)
study
.
Water
levels
at
these
wells
were
·
assigned
at
least
as
high
as
the
land
surface
elevation
at
the
well location.
The
1935 water-level
map
was
made
by pooling
water
levels collected from 1930 to 1939 from
the
TWDB because
data
were insufficient
in
any one year.
For
the
period
of
1930
to 1939, 18 water-level
measurements
were available for
the
Woodbine aquifer, 8 for
the
Paluxy
aquifer,
and
13 for
the
Trinity
aquifer.
Creating
a water-level
map
in
this
manner
assumes
that
water
levels did
not
change
or
changed
very
little
over
the
decade, a faulty
assumption
because
water
levels declined
throughout
the
aquifer
over
this
period
of
time
. However,
this
composite
map
is
the
best
available
estimate
of
water
levels for
the
midpoint
of
the
decade.
Nordstrom
(1982a) pooled water-level
data
from 1950 to
1959 to
estimate
the
distribution
of
hydraulic
head
in
1955.
Similarly, pooled water-level
data
from 1950 to 1959
were
used
to
extend
Nordstrom's (1982a)
map
to
the
south
in
order
to cover
the
entire
study
area.
Water-levels
maps
for
415
1960, 1970, 1980,
and
1990 were
based
on
data
collected for
single
years
and
thus
are
the
most
precise
estimates
of
hydraulic-head ·distribution.
Because
of
space
limitations,
only
the
1900
and
1990
water-level
maps
for
the
aquifers
(Figs. 3
through
5)
are
presented
in
this
paper.
Determination
of
Land
Subsidence
Land
subsidence
is
defined
as
the
lowering
of
land
surface
elevation
relative
to
sea
level.
Land
subsidence
in
some
areas
is
caused
by
pumping
ground
water
from
aquifers (Winslow
and
Wood, 1959; Poland
and
Davis, 1969).
Pumping
causes
water
levels
and
thus
pressures
in
the
aquifer
to decline
and
leads
to
increases
in
the
effective
stress
in
the
aquifer.
This
increase
in
effective
stress
will
cause
a
compressible
aquifer
to
compact
and
will
cause
land
subsidence
if
the
compaction
propagates
to
the
land
surface. Compaction is
greater
if
the
aquifer
contains
or
is
bounded by clays
or
shales
because
the
compressibility
of
clay
is 1
to
2
orders
of
magnitude
larger
than
that
of
sand
(Freeze
and
Cherry, 1979). Because
the
permeability
of
clay
and
shale
is
orders
of
magnitude
smaller
than
that
of
sand
and
sandstones,
compaction
and
land
sub-
sidence
may
continue
years
after
the
time
of
ground-water
withdrawal.
The
magnitude
and
timing
of
the
subsidence
can
be
estimated
if
the
amount
of
drawdown
and
the
elastic
and
hydraulic
properties
of
the
confining
layer
are
known
(Domenico
and
Mifflin, 1965; Domenico
and
Schwartz
,
1990). The conceptual model involves
an
aquitard
bounded
on
the
top
and
bottom
by
aquifers
that
have
undergone
drawdowns.
An
estimate
of
the
amount
of
land
subsidence,
1;,
can
be calculated
with
s=SsLc(~hi
;~hz)
(1)
where
S5 is
the
specific
storage
of
the
aquitard,
Lc
is
the
thickness of
the
aquitard,
t.h
1 is
the
amount
the
water
level
has
been
lowered
in
the
overlying aquifer,
and
t.h
2 is
the
amount
the
water
level
has
been
lowered
in
the
underlying
aquifer
.
The
specific
storage
term
accounts
for
the
compressibility
of
the
aquitard.
The
time
for
half
of
the
subsidence to occur, t50,
can
be
estimated
with
(2)
Kv
where
~
is
the
vertical
hydraulic
conductivity
of
the
aquitard
(Domenico
and
Schwartz,
1990).
These
calculated
values were compared
to
geodetic
survey
data
for
the
region.
Effects
of
Water-Level
Declines
on
Water
Chemistry
The
effects
of
water-level declines on
water
chemistry
were
determined
by
inspecting
results
of
water
sampling
over long periods
of
time.
These
records
were
found
in
the
TWDB open well
report
files. A
subset
of
data
is published
in
Nordstrom (1982b).

416 TRANSACTIONS OF THE GULF COAST ASSOCIATION OF GEOLOGICAL SOCIETIES,
VOL.
XLIV,
1994
0
40
mi
0 60 km 0 60 km
Contour interval 1
00
ft
Contour
interval 100 ft
-600-
Water-level elevation (ft) -
6oo-
Water-level elevation (It)
0Aa5782
c
Figure
3.
Water-level
maps
for
the
Woodbine
aquifer
in
(a)
1900
and
(b) 1990.
Water-level
elevations
shown
in
feet
above
mean
sea
level.
(a)
Study
area
0
40
mi
0 60 km 0 60 km
Contour
interval 1 00 It
Contour
interval 100 ft
-600-
Water-level elevation (ft) -
600-
Water-level elevation (It)
0Aa5781c
Figure
4.
Water-level
maps
for
the
Paluxy
aquifer
in
(a)
1900
and
(b) 1990.
Water-level
elevations
shown
in
feet
above
mean
sea
level.

MACE, DUTTON, NANCE 417
0
40
mi
0
0 60 km 0 60 km
Contour interval 100 It Contour interval 100
It
-
600-
Water-level elevation (It)
-600-
Water-level elevation (It) QAa5783c
Figure
5.
Water-level
maps
for
the
Trinity
aquifer
in
(a) 1900
and
(b) 1990.
Water-level
elevations
shown
in
feet
above
mean
sea
level.
Results
and
Discussion
Water-Level
Maps
and
Water-Level
Declines
Water
levels
have
declined
steadily
in
the
regional
aquifers
since
the
late
1800's.
In
many
places
at
the
beginning
of
the
century,
the
aquifers
had
hydraulic
heads
that
extended over
the
land
surface.
In
Fort
Worth, a well
drilled into
the
Trinity
aquifer
in
1890
had
a
water
level
90
to
100 ft above
land
surface. After
the
discovery of flowing
wells
in
the
Trinity
aquifer
in
1882,
about
150 to 160 wells
were drilled
in
the
Fort
Worth
area
by
1897 (Hill, 1901). As
water
was produced from
the
aquifer,
water
levels
began
to
fall,
and
by 1914
many
of
the
wells
had
stopped
flowing
(Leggatt, 1957).
At
this
time
the
city
abandoned
many
of
these
wells
and
started
using
impounded
lakes
for
water
supply.
Pressures
declined
at
a
rapid
rate
in
the
aquifers
in
the
early
part
of
the
20th
century
and
slowed
after
World
War I as aquifer development decreased (Leggatt, 1957).
For
instance,
the
Thcker Hill
experimental
well was drilled into
the
Paluxy
aquifer
in
1890
in
Fort
Worth.
The
water
level
in
this
well was 90
ft
below
land
surface
in
1890, 277 ft below
land
surface
in
1942,
and
285 ft below
land
surface
in
1954.
Woodbine
The
1900
water-level
map
for
the
Woodbine
aquifer
shows
water
levels ranging from
about
400
ft
above
sea
level
in
eastern
Dallas
County
to
just
over 700
ft
at
the
outcrop
(Fig. 3a). Although
the
aquifer
extends
downdip to
the
east,
wells
were
drilled
in
a
narrow
band
because
of
the
high
drilling
costs
of
deeper
wells
and
poor
water
quality
downdip. Although
there
is limited well control, water-level
production
in
Grayson
County
had
already
caused
a cone
of
depression.
The
1990
water-level
map
for
the
Woodbine
aquifer
shows
water
levels
ranging
from
just
under
100 ft
in
Dallas
County
to
over
600ft
near
the
outcrop (Fig. 3b). These
water
levels
represent
a 200- to 400-ft decline
in
water
levels over
the
past
90
years
.
Hydraulic-head
gradients
(change
in
water
level over distance)
in
the
Woodbine
have
increased
since
the
beginning
of
the
century
.
For
instance
,
in
the
Johnson
and
Ellis
County
area,
gradients
have
increased
from 6.6 to 31.7 ftJmi.
This
indicates
that
the
velocity of
ground
water
has
increased
locally
about
five
times
since
1900.
Grayson
County
has
maintained
a cone of depression
since
the
early
part
of
the
century
.
Paluxy
The 1900 water-level
map
for
the
Paluxy
aquifer
shows
water
levels
ranging
from 500 ft above
sea
level
in
western
Dallas
County
to over 800 ft
at
the
outcrop (Fig. 4a).
There
were no obvious
areas
of
water-level depression,
and
gradi-
ents
appear
to
be uniform over
the
aquifer.
There
is little
or
no well control
in
the
eastern
portion of
the
aquifer because
of
the
greater
depth
of
the
aquifer
and
poor
water
quality
.
The
1990 water-level
map
for
the
Paluxy
aquifer
shows
water
levels
ranging
from
under
100 ft
in
Denton
County
to over 900
ft
at
the
outcrop (Fig. 4b).
These
water
levels
represent
a 100- to 400-ft decline over
the
past
90 years.
As
in
the
Woodbine aquifer,
ground-water
gradients
have

418 TRANSACTIONS OF
THE
GULF COAST ASSOCIATION
OF
GEOLOGICAL SOCIETIES, VOL. XLIV, 1994
~
c
0
-~
>
QJ
Qi
QJ
>
QJ
'
Cii
cr;
s:
800
Woodb
ine
600
400
200
0
-200
-+----.,---.----.-
~
---.--,--,--,..-
-,-
-,-....,---1
1880 1900 1920 1940 1960 1980 2000
Year
QAa5780c
Figure
6.
Water-level
declines
for
the
Woodbine
aquifer
near
Dallas,
the
Paluxy
aquifer
near
Fort
Worth,
and
the
Trinity
aquifer
near
Fort
Worth.
increased since
the
beginning
of
the
century.
For
instance,
in
Tarrant
County
gradients
have
increased
from 9. 7 to
32.0 ft/mi.
This
indicates
that
the
velocity of
ground
water
has
increased
locally
about
three
times
since 1900.
Trinity
The
1900 water-level
map
for
the
Trinity
aquifer
shows
water
levels
ranging
from
just
under
600 ft above
sea
level
in
western
Dallas
County to over 800 ft
near
the
outcrop
(Fig. 5a). No obvious
areas
of
water-level depression existed
at
this
time,
and
gradients
appear
to be smooth over
the
aquifer, although
data
control is sparse
in
the
northern portion
of
the
study
area. As
in
the
other
aquifers,
there
is
little
or
no well control
in
the
eastern
portion
of
the
aquifer
because
of
high
drilling costs
of
deeper
wells
and
poor
water
quality.
The 1990 water-level
map
for
the
Trinity
aquifer
shows
water
levels
ranging
from
-500
ft
in
Denton
County
to over
900ft
at
the
outcrop (Fig. 5b). These
water
levels
represent
a 200- to
1,100-ft
decline
in
water
levels
over
the
past
90
years.
The
largest
amount
of
drawdown
occurred
in
Tarrant
County
where a cone
of
depression formed
in
the
aquifer. Because
of
this
cone
of
depression,
ground-water
flow directions
have
reversed from
east
to
west
in
Dallas
and
eastern
Tarrant
Counties.
As
in
the
Woodbine
and
Paluxy
aquifers,
ground-water
gradients
have
increased
since
the
beginning
of
the
century.
For
instance,
gradients
have increased from 15.8 to 33.0 ft/mi
in
the
Parker
-
Tarrant
County
area
and
from 6.0 to 25.4 ft/mi
in
the
Somervell-
Johnson
County
area.
This
indicates
that
the
flux
of
ground
water
moving
through
the
aquifer
in
this
area
has
increased
between
two
and
four
times
since 1900.
Another
cone
of
depression,
though
smaller, exists
in
Grayson County where
there
have
also
been
large
ground-water
withdrawals.
Rate
of
Water-Level Declines
Hydrographs
of
wells
in
Dallas
for
the
Woodbine
aquifer
and
in
Fort
Worth for
the
Paluxy
and
Trinity
aquifers were
used to gauge
the
rate
of
water-level declines
in
the
study
area.
These
points
approximately
correspond to
the
areas
of
maximum
water-level
declines
in
each
aquifer.
The
Woodbine
aquifer
has
had
a
nearly
steady
decline
in
water
levels,
although
data
are
lacking
between
1900
and
1950
(Fig.
6)
. The
rate
of
decline is
about
4.2 ft/yr. More
variation
in
rates
of
water-level declines is documented for
the
Paluxy
aquifer
(Fig. 6).
In
the
early
part
of
this
century,
between
1900
and
1935,
water
levels
declined
at
a
rate
of
about
8.6 ft/yr. The
rate
of
decline decreased
between
1935
and
1955 to 2.5 ft/yr
and
subsequently
increased
between
1955
and
1976 to 11.4 ft/yr. However, since 1976,
water
levels have
recovered slightly
in
the
Paluxy
aquifer
near
Fort
Worth.
Varied
rates
of
water-level decline
are
also observed
in
the
Trinity aquifer
in
Fort
Worth.
Early
in
the
century
between
1892
and
1911
water
levels
declined
at
a
rate
of
about
20 ftlyr.
From
1911 to 1942
the
rate
of
decline decreased to
3 ft/yr.
The
rate
of
water-level decline
increased
from 1942
to 1970 to 10 ft/yr
and
decreased since 1970 to 5 ft/yr.
Baker
et
al
..
(1990) documented a well
in
Tarrant
County completed
in
the
Trinity
aquifer
in
which
water
levels
have
recovered
126
ft
between
1976
and
1989.
They
attributed
the
rise
in
water
level to
decreased
pumping
in
the
aquifer
in
the
general
area
of
the
well.
Land
Subsidence
Whenever
there
are
large
decreases
in
water
pressures
in
aquifers
with
or
bounded
by
compressible
materials,
land
subsidence is a potential problem.
Equation
1
estimates
the
potential
amount
of
land
subsidence
given
water-level
declines
in
the
bounding aquifers
and
hydrologic
parameters
of
the
confining
layer
(Fig. 7).
For
the
Glen
Rose confining
layer
between
the
Paluxy
and
Trinity
aquifers
at
the
maximum
drawdown
point
in
the
Trinity
aquifer,
llh
1
is
450 ft
and
llh
2 is 1,100 ft. Therefore, (llh1 +
llh
2
)/2
equals
102·9 ft.
The
specific
storage
for
the
confining
unit
is
an
unknown
parameter,
but
reasonable
values
for specific
storage
for a
unit
such
as
this
might
range
from 10-4 to
10-6 n-1.
Thus,
with
Lc
equal
to 500 ft for
the
Glen Rose,
storativities
range
from 10-
1.
3 to
10-
3·3.
When
these
values
are
used
in
Figure
7,
potential subsidence
might
range
from
25 to 0.3 ft, respectively,
because
of
compaction
in
the
confining
layer
between
the
Paluxy
and
Trinity
aquifers.
Equation
2
can
be
used
to
determine
the
time
for
half
the
subsidence to occur.
For
a vertical hydraulic conductivity
of
the
confining
layer
of
I0
-5 ft/day,
which
is
a
typical
hydraulic conductivity for a
shale
(Freeze
and
Cherry, 1979),
the
time
for
half
the
subsidence
to
occur
for a
total
subsidence
of
0.3 ft
with
a specific
storage
of
10-4 n-1 is
300 yr,
and
the
time
for
half
the
subsidence to occur for a
total
subsidence of 25 ft
with
a specific
storage
of
10-6 n-1
is 3
yr
.

MACE, DUTTON, NANCE
108
10 7
10 6
105
J.1}l
10
4
ai
103
0
c
QJ
10 2
"0
"(ii
.0
10
1
::>
(/)
100
10-1
10-2
10-3
10-4
100
10
1
10
2
10
3
6h1
1 o4
105
+
6h2
2
S=S5
Lc
1
0.1
0.01
0.001
0.0001
106
10
7
108
QAa5777c
Figure
7.
Magnitude
of
subsidence
due
to
dewatering
of
two
aquifers
bounding
a
confining
layer
for
different
stora-
tivities,
S.
85
is
the
specific
storage,
Lc
is
the
confining
layer
thickness,
6h
1
is
the
drawdown
in
the
overlying
aquifer,
and
6h
2
is
the
drawdown
in
the
underlying
aquifer.
Historical
geodetic
information
collected
south
of
the
Dallas-Fort
Worth
area
in
Ellis County indicates
that
there
has
been
no
detectable
land
subsidence
greater
than
measurement
error
between
1957
and
1991 (Dave Goss,
personal communication, 1994).
This
conclusion is
based
on
USGS geodetic survey
data
for
the
region.
If
subsidence
has
occurred
during
this
time,
it
is
masked
within
the
precision
of
the
data
collection,
which
is 0.2
ft
.
The
lack
of
land
subsidence
suggests
that
the
storativity
of
the
confining
layer
is lower
than
10-4 to
10--<i
or
that
the
vertical hydraulic
conductivity is lower
than
I0-
5 ftJday.
Effects
of
Water-Level
Declines
on
Pumping
and
Water
Quality
Water-level declines affect
pumping
because (1)
the
cost
of lifting
water
increases, (2) well yields decrease with
the
hydraulic gradient,
and
(3)
the
pump
may have to be moved
to
remain
below
the
water-level surface.
For
instance,
a
water
supply
well
in
northwestern
Ellis County
has
had
its
pump
lowered several
times
because
of
declining
water
levels
in
the
Trinity
aquifer
(Alton
Adams
, personal com-
munication,
1993).
In
the
vicinity
of
the
large
cone
of
depression
in
the
Trinity
aquifer,
lowering
of
pumps
is
a common
occurrence
and
pumps
are
placed
as
deep
as
1,500 ft below
land
surface (Nordstrom, 1982).
Former
large
-....___/
ground-water
users
such
as
Euless, Bedford,
and
Arlington
have
switched to surface-water supplies (Nordstrom, 1982a).
Ground-water
use
in
Dallas
County
has
decreased
over
50
percent
since
the
mid-1970's, while
water
usage
as
a
whole
has
increased
42
percent.
Ground-water
use
in
Tarrant
County
has
remained
the
same
since
the
mid-1970's
419
although
the
percentage
contribution
of
ground
water
to
total
water
usage
has
dropped from 10 to 7 percent. Coryell,
Hill,
and
Navarro
Counties show decreases
in
ground-water
usage
,
and
Bosque,
Comanche,
Cooke,
Denton,
Grayson,
Hamilton,
Hood,
Johnson,
McLennan,
Parker,
Somervell,
and
Wise
Counties
show
increases
in
ground-water
usage
since
the
mid-1970's.
Ground-water
usage
remained
about
the
same
for Collin,
Erath,
Fannin,
Montague,
and
Tarrant
Counties. Projections for
ground-water
usage
for
this
region
by
the
Texas
Water
Development Board show a 10-percent
increase
in
ground-water
pumping
by
2020
and
then
a
decrease
to
present
rates
of
pumping
by
2040.
Major
decreases
in
ground-water
usage
are
predicted for
Tarrant
County
(-290
percent)
and
Bosque
County
(-100
percent).
Water
quality
can
be affected
by
water-level
declines
because,
as
pressures
in
the
aquifer
decrease, lower
quality
water
might
move from
shale
interbeds
or
from
bounding
aquitards
into
the
aquifer. Because
shales
and
aquitards
in
general
have
lower permeability,
they
generally
have
poorer
water
quality
owing to
the
presence
of
modified
connate
fluid
or
longer water-rock residence times. An inspection
of
water
chemistry
data
from
the
study
area
does
not
con-
clusively show
any
effect
of
pumping
on
water
quality
in
the
aquifers.
Most
long-term
records
of
water
quality
do
not
exist
before 1940.
Of
these
long-term records,
most
show
little
change
in
water
quality, a few show
total
dissolved
solids
and
concentrations
of
sodium, chlorine,
and
sulfate
increasing,
and
several
show
total
dissolved
solids
and
concentrations
of
sodium, chlorine,
and
sulfate decreasing
with time.
Baker
et
al. (1990) plotted
the
concentration
of
total
dissolved solids
areally
and
noted
higher
than
normal
concentrations
in
the
Trinity aquifer of
northern
Tarrant,
southeastern
Wise,
and
southwestern
Denton
Counties.
They
suggested
that
this
anomaly of
higher
concentrations
might
be due to
the
large water-level declines
in
that
area.
It
is theoretically possible
that
water
quality
might
also
be affected by
the
reversal
of
ground-water
flow directions
in
the
Trinity
aquifer
in
Tarrant
and
Dallas Counties.
The
reversal
of
ground-water
flow directions will move lower
quality
water
updip
into locations of
fresher
water
.
Baker
et
al. (1990) noted
that
evidence does
not
exist
to show
that
water
quality
along
the
slightly
saline
water
line
has
been
affected.
This
is due to
the
slow
movement
of ground
water
through
the
aquifer
. Because
gradients
are
presently
32 ft/
mi
in
the
updip
direction,
and
assuming
a
hydraulic
conductivity of 0.498 ftlday for
this
area
(Macpherson, 1983)
and
a porosity
of
20 percent,
the
ground
water
moves
about
5.5 ftlyear. Therefore,
the
slightly saline line
might
move
55 ft to
the
west
in
10
yr
and
550 ft
in
100
yr
at
the
present
hydraulic
gradient.
Within
the
n
ea
r
future
it
appears
that
movement
of
the
slightly
saline
water
line will
not
greatly
affect regional
water
quality
.
Conclusions
Ground-water
development of
the
Woodbine, Paluxy,
and
Trinity
aquifers
has
caused
large
water-level declines over
North-Central
Texas since
the
turn
of
the
century
. A
water-
level
map
for 1900 constructed from
R.
T.
Hill's well
survey
data
shows
that
water
levels
were
initially
above
land
surface
in
many
parts
of
the
area
before development.
When

420
TRANSACTIONS
OF
THE
GULF
COAST
ASSOCIATION
OF
GEOLOGICAL
SOCIETIES,
VOL. XLIV, 1994
these
1900
water-level
maps
are
compared
to
1990
water
levels,
the
regional
magnitude
of
water-level
declines
can
be
observed.
Between
1900
and
1990,
water
levels
have
declined
nearly
850
ft
in
the
Trinity
aquifer
in
the
Fort
Worth
area,
about
400
ft
in
the
Woodbine
aquifer
near
Dallas,
and
450
ft
in
the
Paluxy
aquifer
near
Fort
Worth.
Ground-water
flow
gradients
have
increased
in
all
the
aquifers
because
of
pumping.
Areas
of
the
Woodbine, Paluxy,
and
Trinity
aquifers
have
seen
a 480-, 330-,
and
200-percent
increase
in
gradients
and
thus
ground-water
flow
rates,
respectively,
since
the
beginning
of
the
century
.
In
the
case
of
the
Trinity
aquifer
in
Tarrant
and
Dallas
Counties,
ground-water
flow
has
reversed
flow
direction
as
well
as
increased
the
magnitude
of
flow.
Large
water-level
declines
cause
land
subsidence,
affect
pumping,
and
influence
water
quality.
Simple
calculations
show
that
only
a
small
amount
of
subsidence
would
be
caused
by
the
largest
water-level
declines
in
the
study
area
and any
measurable
subsidence
that
occurred
would
take
a
long
time
to
manifest
itself.
Pumping
has
been
affected
by
water-level
declines
because
(1)
the
cost
of
pumping
water
increases,
(2) well yields decrease,
and
(3)
pumps
have
to
be
lowered.
Clear
evidence
does
not
exist
to
show
that
water-level
declines
have
affected
water
quality,
although
Baker
et
al. (1990)
suggested
that
elevated
concentrations
of
total
dissolved
solids
in
the
Trinity
aquifer
of
northern
Tarrant,
southeastern
Wise,
and
southwestern
Denton
Counties
may
be
due
to
water-level
declines.
Because
ground-water
flow
directions
have
reversed
in
the
Trinity
aquifer
in
Tarrant
and
Dallas
Counties,
the
slightly
saline
water
line
will
move
updip
into
regions
of
fresher
water.
However,
the
calculated
movement
of
the
slightly
saline
water
line is
very
slow
at
5.5 ft·'year
and
poses no
immediate
threat
to
ground-water
quality
farther
to
the
west.
Acknowledgments
This
work
was
conducted
as
part
of
a
study
to
characterize
hydrogeologic
properties
and
water
resources
near
the
Superconducting
Super
Collider
site
for
the
Texas
National
Research
Laboratory
Commission
under
Contract
No. IAC(92-93)-0301
and
IAC 94-0108.
Alan
Cherepon
and
Erika
Boghici
assisted
in
data
collection
and
entry.
Susan
Lloyd did
the
word processing.
Figures
were
drafted
by
Tari
Weaver
under
the
direction
of
Richard
L. Dillon.
Jeannette
Miether edited
the
paper. Design by
Jamie
H. Coggin.
Publication
authorized
by
the
Director,
Bureau
of
Economic
Geology,
The
University
ofTexas
at
Austin.
References
Baker, Bernard, Duffin, Gail, Flores, Robert, and Lynch, Tad, 1990,
Evaluation of water resources in
part
of North-Central Texas:
Austin, Texas Water Development Board, Report 318,
67
p.
Baker,
E.
T.,
1960,
Geology
and ground-water resources of Grayson
County, Texas: Austin, Texas Board of Water Engineers,
152
p.
Caughey,
C.
A.,
1977,
Depositional systems in the Paluxy Formation
(Lower Cretaceous) northeast Texas-oil, gas, and groundwater
resources:
The
University of Texas
at
Austin,
Bureau
of
Economic Geology Geological Circular 77-8, 59
p.
Domenico, P. A.,
and
Mifflin,
M.
D., 1965, Water from
low
permeability sediments and land subsidence: Water Resources
Research, v.
4,
p.
563-576.
Domenico,
P.
A.,
and Schwartz,
F.
W.,
1990, Physical and chemical
hydrogeology: New
York,
John
Wiley and Sons, Inc., 824
p.
Freeze, R.
A.,
and Cherry,
J.
A., 1979, Groundwater: Englewood
Cliffs, New Jersey, Prentice Hall, Inc., 604
p.
Gutentag, E. D., Heimes,
F.
J., Krothe,
N.
C.,
Luckey,
R.
R.,
and
Weeks,
J.
B.,
1984, Geohydrology of the High Plains aquifer in
parts of Colorado, Kansas, Nebraska, New
Mexico,
Oklahoma,
South Dakota, Texas,
and
Wyoming: Alexandria, Virginia,
U.S. Geological Survey Professional Paper 1400-B,
63
p.
Hall,
W.
D., 1976, Hydrogeologic significance of depositional
systems and facies in Lower Cretaceous sandstones, North-
Central Texas: The University of Texas
at
Austin, Bureau of
Economic Geology Geological Circular 76-1,
29
p.
Henningsen,
E.
R.,
1962, Water diagenesis in Lower Cretaceous
Trinity aquifers of Central Texas: Baylor Geological Studies,
Bulletin No.3, 37
p.
Hill,
R.
T.,
1901, Geography and geology of the Black and Grand
Prairies, Texas, with detailed descriptions of
the
Cretaceous
formations
and
special reference to
artesian
waters:
Washington, D.C., Twenty-First Annual Report of the United
States Geological Survey,
part
VII, 666
p.
Hill,
R.
T.,
1937, Paluxy sands, with further notes
on
the Comanche
Series: Geological Society of America Proceedings,
p.
79--80.
Klemt,
W.
B.,
Perkins,
R.
D.,
and Alvarez,
H.
J., 1975, Ground water
resources of
part
of Central Texas with emphasis
on
the Antlers
and Travis Peak Formations: Austin, Texas Water Development
Board,
v.
1,
63
p.
Klemt,
W.
B.,
Perkins,
R.
D.,
and Alvarez,
H.
J., 1976, Ground water
resources of
part
of Central Texas with emphasis
on
the Antlers
and Travis Peak Formations: Austin, Texas Water Development
Board,
v.
2,
528
p.
Leggatt, E.
R.,
1957, Geology and ground-water resources of
Tarrant
County, Texas: Texas Board of Water
Eng
ineers
Bulletin 5709,
187
p.
Lozo,
F.
E., 1949,
Stratigraphic
relations of Fredericksburg
Limestone, North-Central Texas: Shreveport Geological Society
Guidebook
17,
p.
85
-
95.
Macpherson,
G.
L.,
1983, Regional trends in transmissivity and
hydraulic conductivity, Lower Cretaceous Sands, North-Central
Texas: Ground Water,
v.
21,
no.
5,
p.
577-583.
Nordstrom,
P.
L.,
1982a, Occurrence, availability, and chemical
quality of ground water in the Cretaceous aquifers of North-
Central
Texas:
Volume
1:
Texas Department
of
Water Resources,
Report 269,
61
p.
Nordstrom,
P.
L., 1982b, Occurrence, availability, and chemical
quality of ground water in the Cretaceous aquifers
of
North-
Central Texas:
Volume
2:
Texas Department
of
Water Resources,
Report 269, 386
p.
Nordstrom,
P.
L., 1987, Ground-water resources of the Antlers and
Travis Peak Formations in the outcrop area of North-Central
Texas: Texas Water Development Board, Report
No.
298, 280 p.
Oliver,
W.
B., 1971, Depositional systems in
the
Woodbine
Formation (Upper Cretaceous), Northeast Texas: The University
of Texas
at
Austin,
Bureau
of Economic Geology Report of
Investigations
No.
73,
28
p.
Poland,
J.
F.,
and Davis,
G.
H., 1969, Land subsidence due to
withdrawal of fluids: Geological Society of America Reviews of
Engineering
Geology,
v.
2,
p.
187-269.
Thompson, G. L., 1967, Ground-water resources of Ellis County,
._____/
Texas: Texas Water Development Board, Report
No.
62,
115
p.
Winslow,
A.
G.,
and
Wood,
L.A., 1959, Relation ofland subsidence
to groundwater withdrawals in the upper Gulf Coastal region
of Texas: American
Institute
of Mining Engineers Mining
Division, Mining Engineering,
p.
1030-1034.