fliE DENVER AREA EARTHQUAKES AND

r ilE

ROCKY MOUNTAIN ARSENAL DISPOSAL WELL

i V l D M. EVANS: Consulting Geologist, Denver. Colorado

ual'RACT: During 1961, a deep well was drilled at the Rocky Mountain Arsenal northeast of
V n v e r , Colorado, to dispose of contaminated waste w a t e r . The well is bottomed in 75 f e e t of
,,ly f r a c t u r e d Precambrian gneiss. P r e s s u r e injection of waste water into the f r a c t u r e d Pre.ibrian rock was begun in March 1962. Since the start of fluid injection, 710 Denver-area earth ikes have been r e c o r d e d . The majority of these earthquakes had epicenters within a five-mile
.'uiius of the Arsenal well. The volume of fluid and p r e s s u r e of fluid injection appear to be directly
-.-ated to the frequency of earthquakes. Evidence also suggests that rock movement is due to the
- c r e a s e of fluid pressure within the fractured reservoir and that open fractures may exist at depths
than previously c o n s i d e r e d possible.

-, raoDUCTiON
Products for chemical w a r f a r e have been
...inufactured on a large scale under the direc jn of the Chemical Corps of the U. S. Army
.; -.he Rocky Mountain Arsenal since 1942. A
; . - p r o d u c t of this operation is contaminated
i i = te water and, until 1961, t h i s ' w a s t e water
. 1 5 disposed of by evaporation from dirt
r e s e r v o i r s (Scopel, 1964).
When it was determined that Arsenal
j..idte water was contaminating the local ground a t e r supply and endangering crops (Gahr,
, :ol; Walker, 1961), the Chemical Corps tried
-.aporation of the contaminated waste f r o m
l i t e r - t i g h t r e s e r v o i r s . This proved unsuccess:.::. The Chemical Corps and the Corps of
Engineers then decided to drill an injection disjosal wel! for the purpose of disposing of the
jontaminated waste water (Scopel, 1964).
The U. S. Army Corps of Engineers,
Jmaha. District, commissioned the firm of
.. A. Polumbus, J r . , and Associates, I n c . ,
~) design the well, s u p e r v i s e the drilling and
^Jmpletion, provide the necessary engineering
.eological services, and manage the project.
-ouis J. Scopel, as an associate, was the
Project Geologist and was responsible for all
.?ological aspects of the operation. Another
.cological associate was George R. Downs,

- Mountain G e o l o g i s t , v. 3, no.

1, p. 2 3 - 3 6

Figure 1. Structural map of a portion of the

Denver- Julesburg Basin (after Anderman and
Ackman, 1963), showing the location of the
Rocky Mountain Arsenal well.

23

ROCKY MTN.
ARSENAL
WELL (-8000'

8000
6000

^'^'^^n^^^s^^rW^-^'r^HV'/^.V^.^

4000

^GRANIJET'^'/IDAHOJ'/
:;-V^fiMFTqS^'-r'qpR<5_ pM.

20001

6000'
TERTIARY-

4000'
2000'

PRECAMBRIAN

0
-2000']

-2000'

-4000'

-4000'
PERMIAN

-6000'

PENN.

-6000'

CAM B.-8000'
GRANITE GNEISS;

-8000'

3 MILES

Figure 2.
C r o s s - s e c t i o n showing the subsurface geology f r o m the Arsenal well to the outcrop of
Precambrian granite gneiss west of Denver (after M. F. and C. M. Boos, and H. H. Odiorne).
The line of cross-section is shown in f i g u r e 1.

who contributed materially to the intial design

of the project and acted in an a d v i s o r y capacity
throughout the the operation.
The well was located and drilled in the
N W / 4 N E / 4 sec. 26, T. 2 S., R. 67 W. (395 1 . 5 ' N. , 104-51' W . ) , Adams County, Colorado. It was spudded 10 March 1961 and c o m pleted at a total depth of 12, 045 f e e t 11 September 1961.

P r e c a m b r i a n granite gneiss w e s t of Denver.

The granite gneiss is identified as the
Mount M o r r i s o n Formation by C. M. and
M. F. Boos ( 1 9 5 7 ) , who describe typical
Mount M o r r i s o n granite as medium to fine
grained, pink to tan, and delicately gneissic.
Parts of the granitic gneiss are p e r m e a t e d
with ill-defined pegmatite.
A p p r o x i m a t e l y 1 3 , 0 0 0 f e e t of structural
r e l i e f exists between the top of the Precamb r i a n in the Arsenal well and the Precambrian
o u t c r o p w e s t of Denver.

REGIONAL GEOLOGY
The Rocky Mountain Arsenal Disposal well
is located on the gently dipping east flank of
the D e n v e r - J u l e s b u r g Basin, just a few miles
w e s t of the axis of the basin. As indicated in
f i g u r e 1, the Arsenal well is located in a region
of the s u b c r o p of Cambro-Ordovician rocks,
near the area w h e r e these rocks are truncated
by Pennsylvanian sediments.
Figure 1 is a structural map of a portion
of the D e n v e r - J u l e s b u r g Basin in the vicinity
of the Rocky Mountain Arsenal well after
Anderman and Ackman (1963). Figure 2 is a
c r o s s section a f t e r M. F. and C. M. Boos and
H. H. Odiorne which shows the subsurface
geology f r o m the Arsenal well to the outcrop of

STRATIGRAPHIC SECTION IN INJECTION

DISPOSAL W E L L
A d i a g r a m and log of the well are shown in
f i g u r e 3 (Scopel, 1964). F i g u r e 4 is a log of
the p r e - P e n n s y l v a n i a n portion of the well by
Scopel ( 1 9 6 4 ) . Scopel (1964) described the
P r e c a m b r i a n section cut by the well as follows:
Precambrian
The a b o v e - d e s c r i b e d sediments [Ordovician or Cambrian] overlie 20 f e e t of
b r i g h t g r e e n weathered schist which con-

24

t,

4" FLOVLMV

rtWM HUCCT1M

-^1_
^

PUMPS

JOtTM VU.V*

INJECTION
2=^2_T *>M.yf. TOM-CUW
mittun DISPOSAL WELL ~-<|TS||P3^TO
^H
"* OITICTO* o
q

>/

ROCKY

MOUNTAIN

ARSENAL

WHULUi

ALAMH

^
f"

KB

y.

en
"""

//
//

tf-*"

Vi

\\
\\

13 W CSI6

r^i-8ULA

24 'MOLE "*^i
zo'CAsmo ^^
IT n' MOLE

/^
'

'

1.250' FOX HILLS SS.

1,480' TRANSITION
TO PIERRE SHALE

IO4L
CCMCNT WITH 2%
CALCIUy CHLONIOC

5TAOC SI 2O20 SACKS

/or 90-90 poz*a,

/ SeHTOWITC.

2 OOO1 -
II" HOLE
8 5/B" CASING

3S

~~*' .
^

=--=^=

5 1/2" TUBIS
SECOND STAC COCLA*

4307'

fj

5,448'
ZONE

i?^

HYGIENE

SS.

^- STA0E 11 2590 SACKS

Of 90-9O POZHtX,*%
2 CCNTKALIZER3

k!

8k.

i*

* 0
Q <j
t
1-2
1-3

7,710' NIOBRARA
C H A L K Y SH.

^\8078' CARLILE SHALE^l

v

;',,v_y;.;,

8120' GREENHORN

PKKEH LD. 2.3^

N

LSVB

9 000'

^-FlftST STAflt COLLAH

AMD CEMENT ftASKET.

^S'CC
S ,
n OOO'
>011'
'/
^SAKE* MOOCLV
RETAINCH
PflOOUCTIOM
JTINOCH
-^
\ PACKER.
ift W
^STINflER.

''V^aess' "DAKOTA" SS.

Jj\873ff "LAKOTA" S3.

8-13

> I

%S345'6RANEROS SH.J
\'\\\848S' "J'OR "MUDDY" SS. ""

III 8786' MORRISON FM,

^-STAttC 1 WM SACKS Of
(40 fOZMIX A, a3* NH4
AN04K) SACKS Of
OILSONITE MO fOZMIX.
0.3* HM, 3 (XNTMALIZtMS.

" 1^8972' LYKINS FM.

'*" 1 9582' LYONS SS.

7V

?-;
_*li
,

9772' FOUNTAIN FM.

It.OOT .
X
, ,.LIlie,-!il^-> 1M'^ --AAM MSS LIHEK

J 1.880' REGOLITH
f\ 1,895' ORDOVICIAN?

'" ""i v4"Mou

yL-HI950'-f PRE ' CAMBRIAN>

i.II.95U ^(jR^NITE ETC

MO A 0.9* HIM, 10

M.975k

J
*

Figure 3.

POLUMBUS.

"""" ooa

^~^T.tt 12,045'

T.D. 12,045'

E.

HAMCM.
ISO SACKS Or MZMIX

JR.

AND

ASSOCIATES.

INC.

Diagram and log of the Rocky Mountain Arsenal Injection Disposal Well (Scopel, 1964).

25

FOUNTAIN

REGOLITH
MAROON-RED BROWN SHALE
MAROON QUARTZITE

f r a c t u r e s o b s e r v e d w e r e partially to completely
cemented with quartz.

ORDOVICIAN ? CAMBRIAN ?
ORANGE QUARTZ

CONGLOMERATE

A drill stem test was taken of the basal

Fountain Formation, the pre-Pennsylvanian
rocks and Precambrian rocks f r o m the bottom
of the 8 - 5 / 8 - i n c h casing at 11, 171 f e e t to the
total d e p t h of 11, 985 f e e t . R e c o v e r y was
5,400 f e e t of salt w a t e r , in addition to 2, 000
feet of water cushion, in 156 minutes. Ninetyt h r e e - m i n u t e final shut-in p r e s s u r e was 4, 128
pounds, m e a s u r e d at 1 1 , 0 0 2 f e e t . Density of
the water was 1.05 g m . / c c .
The well was drilled ahead to 12, 045 f e e t
w h e r e it was completed in Precambrian gneiss.
Considerable lost circulation was experienced
while coring, testing, and drilling the Precambrian gneiss f r o m 11,970 to 12,045 f e e t .
A 5 - 1 / 2 - i n c h liner was cemented five feet
into the Precambrian gneiss f r o m the bottom
64 feet of the 8-5/8-inch casing. Five-andone- half -inch tubing was run to 9 , 0 1 1 f e e t to
complete the well.
During November and December 1961 a
conventional oil field pump was run in the well,
and pumping tests w e r e conducted. After
pumping 1, 100 b a r r e l s of w a t e r , a quantity in
excess of the amount of fluid that had been lost
into the formation during drilling operations,
the well pumped down and fluid r e c o v e r y became negligible. It was concluded, at the time
of t e s t i n g , that fluid r e c o v e r y was f r o m f r a c t u r e s . It was f u r t h e r believed that as fluid
was w i t h d r a w n f r o m these f r a c t u r e s , they
w e r e squeezed shut by c o m p r e s s i v e f o r c e s
which r e s t r i c t e d fluid entry into the well bore.
P r e s s u r e injection tests w e r e conducted on
the well during January 1962 to determine the
rates and injection p r e s s u r e s at w h i c h the Precambrian would take the fluid. As a result of
t h e s e t e s t s , it was noticed that calculations of
the drainage radius and formation capacity inc r e a s e d as fluid was injected ( s e e Calhoun,
1953, for m o r e on r e s e r v o i r calculations).
As a result of the testing p r o g r a m , it was
concluded that the formation would take f r e s h
w a t e r at 400 gallons per minute under 650
pounds p r e s s u r e , and that the r e s e r v o i r consisted of f r a c t u r e s which expanded as additional
volumes of fluid w e r e injected.

PURPLE SHALE

PINK

DOLOMITE

PURPLE SHALE

PRE-CAMBRIAN

SCHIST

GREEN, MICACEOUS
11,970

h- PRE-CAMBRIAN

GNEISS

FRACTURES

12.045

TESTING OF THE WELL

TOTAL DEPTH

F i g u r e 4.
Log of pre-Pennsylvanian portion
of disposal well (Scopel, 1964).

tains brown to c o p p e r - c o l o r e d mica and

kaolinite. The pre-Pennsylvanian sediments and the Precambrian w e r e not
cored.
The Precambrian schist is immediately
above highly f r a c t u r e d hornblende granite
gneiss which contains pegmatite intrusions.
The top eight-foot section of the gneiss was
c o r e d . Hedge and Walthall (1963) have
dated the gneiss to be 1, 350 x 10 years
old.
A portion of the core mentioned above was
examined by the present author. The f r a c t u r e s
observed w e r e almost vertical and f r o m onehalf-inch to two inches apart. When taken
f r o m the core barrel, the core was found to be
split apart along one f r a c t u r e plane, and the
lack of cementing material suggested that this
might have been an open f r a c t u r e . The other

THE PRESSURE INJECTION PROGRAM

Contaminated waste f r o m the Arsenal

26

COLUO.

in
CD
CD

o
LU

O
LU
c -o

OOLUQ-

a
-1 2
o

CD

UJ

S "
~ c

.S u
T3 2

0)

(O

"

n)

11

O
UJ

Sa

C/)LUQ_

ro

31

^tm
C

u o
X u

CD
CD

a -S

1!
El

O
O

COUJQ-

IM

C\J
CO
CD

a S
a o

3
SO

Dcor^-cDio^rooj-

'OIAI/SN011V9 SNOI11IIAI

27

T
plants is f i r s t collected and allowed to settle in
a two-hundred-million-gallon w a s t e - s e t t l i n g
basin that is sealed w i t h an asphaltic m e m brane to prevent seepage. It is then flocculated
and clarified. Next it is f i l t e r e d to less than
20 parts per million of suspended solids less
than 5 m i c r o n s in d i a m e t e r . It is sterilized
and m o n i t e r e d for bacteria, then pumped into
the well. Four 1 3 0 - h o r s e p o w e r positivedisplacement electric pumps are available.
Normally, two or t h r e e pumps are used.
The f i r s t contaminated w a s t e was injected
into the well during March 1962, when 4 . 2
million gallons of waste w e r e injected into the
well. The monthly volume of waste injected
into the well is shown in the lower half of
f i g u r e 5. During the first year of operation,
considerable trouble was experienced with the
f i l t e r plant with the result that the injection
well was often shut down for a few days or
weeks at a time. From March 1962 until
September 1963 the maximum injection p r e s s u r e
is r e p o r t e d to have been about 550 pounds, with
a fluid injection rate of 200 gallons per minute.
At the end of September 1963 the injection
well was shut down, and no fluid was injected
until operations were resumed 17 September
1964. During the shut-down period, surface
evaporation, f r o m the settling basin, was
sufficient to handle the plant output.
From 17 September 1964 until the end of
March 1965, injection operations w e r e resumed
by gravity discharge into the well. No wellhead p r e s s u r e was necessary to inject the
maximum of 2 . 4 million gallons of waste per
month into the well.
Beginning in April 1965 larger quantities
of fluid w e r e injected. The filter plant operated
efficiently, and fluid was usually injected 16 to
24 hours daily. During April and May a maximum pump p r e s s u r e of 800 pounds was r e q u i r e d .
From June to the end of September 1965 a
maximum p r e s s u r e of 1,050 pounds was required to inject 300 gallons per minute into the
well.

e p i c e n t e r s w e r e r e c o r d e d in the Denver area

by e i t h e r the U. S. Coast and Geodetic Survey
or by the Regis College Seismological Observat o r y , located ten miles southeast of the Rocky
Mountain Arsenal well ( J o s e p h V. Downey,
personal communication, 1965).
During Phe period f r o m April 1962 to the
end of September 1965, 710 earthquakes w e r e
r e c o r d e d with epicenters in the vicinity of the
Arsenal at the Cecil H. Green Observatory,
Bergen Park, Colorado, operated by the
Colorado School of Mines (Pan, 1964; Wang,
1965; Jones, 1965, Mines Magazine, 1965).
The total number of earthquakes r e p o r t e d
in the Denver area is plotted in the upper half
of f i g u r e 5. The magnitude of the earthquakes
reported range f r o m 0 . 7 to 4. 3 on the Richter
scale. Table 1 lists the earthquakes in Colorado of magnitude 3 and l a r g e r , according to
the U. S. Coast and Geodetic Survey reports
(Wang, 1965). Wang (1965) calculated the epicenters and hypocenters of the 1963-65 Denver
earthquakes, and the results of his calculations
are shown in figure 6.
The majority of the earthquake epicenters
are within a five-mile radius of the well. All
epicenters calculated f r o m four or m o r e rec o r d i n g stations are within seven miles of the
well.
Wang (1965) calculated the best-fitting
plane passing through the zone of hypocenters
calculated f r o m four or m o r e r e c o r d i n g stations.
He concluded that this plane might be a fault
along which m o v e m e n t was taking place. The
plane dips to the east, and passes beneath the
arsenal well at a depth of about six and onehalf miles (fig. 6).
ROCK MOVEMENT AND EARTHQUAKES
An attempt has been made to develop a
method of estimating, directly f r o m seismog r a m s , wave energy radiated during an earthquake. Using the formula favored by Tocher
(1964) and Richter (1958), the elastic wave
energy of a magnitude 3 earthquake could be
provided by dropping a 100 foot cube of rock a
distance of a few f e e t .
Admittedly, the formula applies to distant
earthquakes and is not routinely applicable to
large number of earthquakes, but it does suggest
that the Denver earthquakes may be caused by
relatively minor rock movements.

THE DENVER EARTHQUAKES

The U. S. Coast and Geodetic Survey reports that on 7 November 1882 an earthquake
was felt in the Denver, Louisville, Georgetown, and S. E. Wyoming area (Wang, 1965).
From that date until April 1962 no earthquake

28

u
f

En

>;
3i

tf

D U - O
T3 -O <|

0)0)011)
-OTI-O-O

a s
|
-" c K 2 a
J)

>- S " 3 > - >

> ^ - i n

|H

a a a a
X K S X

D.-...J
T J 3 3

t,

a o o
3JJ
"

"

^ ^ 2 ^ ^ ^.i^.^^ i

u u
a C
D

1)

3 3 3
0 0 0

CU

c a c
D

Q Q Q Q

a a

a a a
D

71

71

01

in
in

r-

71

71

71

0)

1)

0)

s ss
r- r- NO

0)

75 -a

y o

D
T)
3
*j "
*J ^

iT)
- O

CO
- O

IT)

~-t

<M

NO
tN.

[N.

in in m in

c^- c^. % % ^ 5 S pj iC f- ^ Ln

o o o o o o - ^ o

mo

in
in
0

0 0 0

NH m TJ<
O o o o

-H

u
0

ro

!>

0 ~~

O O O O O O ^ O

^"O
IM

in
O

00

ro
fi

ro

ON
ro

0
r-

0 O O
O"N 00 [^
71
L<

ro

3
O

*
*

. NO

m
o

0)

Mon.
ifear

'-'*?

O^ O
ro

ro

in
o

Is-

'

o
Z

u
0

rj

NO

00

ON

IN] Ln TJ<

vO

ON Q

* *

*
* * *
* *
* *
fNj 00
ON
rj

^<?QQ^<?'<i?'S^'^^^t
in

00

o oo

ro

i rq rj

"3

O. Q. Q.
0) D -1)
C/5 to CO

in

in in in
NO NO

NO

ON

.a
-""^

ti U

Q En En

fNl

<^

T r m N O r - o o O N O - ' r M ' - o l ' i n N O r-

#
29

ON 0

-1

1965, w i t h a d d i t i o n s )

(>.

Day

v''
,*'*;

3
J1

a - 2 ^ i u o

t e r m i n e d f r o m local seismograrn data.

r Mountain Standard T i m e , s u b t r a c t 7 ho
r Mountain Daylight Savings Time, subtr

4-)

COLO RADO EARTHQUAK

(Fron i U. S. Coast and G<

I1
i

a c c a

j u a u u S u a a u a a l a u

'I

s< ^.2

Table 1.

}}

aa-ji

c
^
Q so 3 O ? E ^ f f l a Q Q Q Q Q Q Q a Q a S 5

Greenwich Mean Tin

Hour
Min.
Se

1;

Jj"
9

y1
2

s c"-i1 -j - <

n) 3

&
-I*

1
11
=0 -0

Q OH
^ >

j,

e>

Si
*;'
|

^cx

*^

Is"

!
I

, Louisv
, Louisv

lontrose, Opllir,
^durliorn, Ridge-

rgetown and

* **
-*

30

ra

EXPLANATION

HOC*r MOUNTAIN MKMAL *T

TATION AT THC
P T or it4

COLORADO 9CMOOL 0* MUNT.S STA.

LOBMi>(()lTATlO^T|o((

. ,.

P R O J E C T I O N

2 - STATION

ON

S E C T I O N

A - A

OAT*

SEDIMENTARY

'"'ROCK

v-X '

1
4 STATION

Of

,..-1

K VC
,AAX

Jt

J,
""

*
i_

|
_i

_
0

PRECAM8RIAN

METAMORPMIC

ROCK

X^COMPUTEO BEST FITTING PLANE

J,
'

,'. BiSSINU THHOUSH HYPOCENTEHS

\"\

FOU

"-

Sr

TI

"

* *

MO*e-TU.-4-STAT,001TJ
( lt4,,Sl

'
:

U S COAST ft S E O D E T l C SUHVEY 10
POELIMINAHT OETCDHINATION Of
CA1THOUAME KT'OCEMTEIt

'
I

COMPUTED IESTFITTINS "LANE

PISSING THAOU6H MTPOCENTERS

STR.PtE *N ft, 0?P *'

G R A N I T I C - LIKE
7~=
I j

ROCK

BASALTIC - LIKE

ROCK

M A N T L E

S C A L E I 125.000
9

Figure 6.
Earthquake hypocenters during 1963-64 from local seismological stations in the Denver
area (after Wang, 1965). All epicenters calculated from four or m o r e recording stations are within
seven miles of the Arsenal well. All hypocenters calculated from four or m o r e recording stations
are *ithin area indicated on section A - A.

30

PRESSURE INJECTION AND EARTHQUAKE

FREQUENCY

1962 is not used in the a v e r a g e because the

exact day when injection was started is not
known.
Figure 7 indicates that t h e r e is a direct
correlation between average monthly injection
and earthquake f r e q u e n c y when an injection
p r o g r a m is carried out for a period of five
months.
Bardwell (1966) has prepared a statistical
analysis, presented e l s e w h e r e in this issue of
The Mountain Geologist, that suggests that a
mathematical relationship exists between the
Denver earthquakes and the volume of contaminated waste injected into the Arsenal well.

P r e s s u r e injection began in March 1962.

The f i r s t two earthquakes with epicenters in
the Arsenal area w e r e r e c o r d e d during April
1962.
The lower half of figure 5 is a graph of the
monthly volume of w a s t e injected into the
Arsenal well. The total number of earthquakes
r e c o r d e d in the Arsenal area is plotted each
month in the upper half of the graph.
During the initial injection period f r o m
March 1962 to the end of September 1963, the
injection program was often shut down for r e - ,
pairs to the filter plant. During this period
t h e r e does not appear to be a direct month by
month correlation. However, the high injection
months of April, May and June 1962 seem to
c o r r e l a t e with the high earthquake f r e q u e n c y
months of June, July and August. The high
injection months of February and March 1963
may correlate with the high earthquake month
of April.
The period of no injection f r o m September
1963 to September 1964 coincides with a period
of minimum earthquake frequency. The period
of low volume injection by gravity flow, from
September 1964 to April 1965, is characterized
by two months (October and February) of
greater earthquake frequency than experienced
d u r i n g the preceeding y e a r .
The most direct correlation of fluid injection with earthquake frequency is during the
months of June through September 1965. This
period was characterized by the pumping of
300 gallons per minute, 16 to 24 hours a day,
at p r e s s u r e s of f r o m 800 to 1,050 pounds.
A review of the injection p r o g r a m reveals
that there have been five characteristic periods
of injection into the well ( f i g . 7):
1. April 1962-April 1963: High injection
at medium p r e s s u r e .
2. May 1963-September 1963: Medium injection at m e d i u m p r e s s u r e .
3. October 1963-September 1964: No injection.
4. September 1964-March 1965: Low injection at zero p r e s s u r e ( g r a v i t y ) .
5. April 1965-September 1965: High in jection at high p r e s s u r e .
The average numbers of earthquakes per
month are plotted in f i g u r e 7 above the average
volumes of fluid injected per month for each of
these five periods. The injection for March

EFFECT OF EARTHQUAKES ON INJECTION

PRESSURE
The w e l l h e a d - p r e s s u r e injection charts
w e r e not available for the years 1962 and 1963.
Only the earthquakes of magnitude 3 or larger
w e r e checked against the p r e s s u r e injection
charts for 1965. These earthquakes are listed
in table 1.
No charts w e r e being r e c o r d e d at the time
of the February and July 1965 earthquakes.
Three pumps were maintaining a pressure of
725 pounds when the September 14 earthquake
occurred. There was no change in injection at
the time of the event.
Allowing for a few minutes time d i s c r e p ancy between the chart time and recorded time
of the earthquakes, the two earthquakes of 29
September may have a f f e c t e d the injection
p r e s s u r e . During the f i r s t earthquake, at
12:59 P.M. MOST, the p r e s s u r e r e c o r d i n g
needle on the p r e s s u r e chart jumped from 970
pounds to 940 pounds and also repeated a ten
minute time interval on the chart. During the
second earthquake, at 1:20 P . M . MOST, the
p r e s s u r e dropped f r o m 960 pounds to 780
pounds. Whether this 180-pound p r e s s u r e d r o p
was due to the earthquake or to the slowing up
of one of the pumps is not known.
FLUID PRESSURE AND THE ARSENAL
EARTHQUAKES
The evidence gained from drilling and
testing the Arsenal disposal well indicates that
the Precambrian r e s e r v o i r is composed of a
highly f r a c t u r e d granite gneiss which is substantially impermeable. The f r a c t u r e s are
almost vertical, and porosity of the r e s e r v o i r
is f o r m e d by these f r a c t u r e s . The evidence

31

60
50
cc
UJ

40

AVERAGE NUMBER OF DENVER EARTHQUAKES

PER MONTH

30

0 20
| 10
10
37
LULL. O
O
co c;

AVERAGE GALLONS OF WASTE INJECTED

PER MONTH INARSENAL DISPOSAL WELL

23
03

-_MAXIMUM INJECTION-PRESSURE 550 LBS-

1962
Figure 7.

-NO WASTE
INJECTED

1963

1964

MAXIMUM
^INJECTED INJECTION}
Y GRAVITY PRESSURE
1050 LBSA

1965

Earthquake frequency - waste injection relationships during five characteristic periods.

slopes. This arises from the circumstance

that the weight of such a block is jointly supported by solid stress and the p r e s s u r e of interstitial fluids. As the fluid p r e s s u r e approaches
the lithostatic p r e s s u r e , corresponding to a
flotation of the overburden, the sheer stress
required to move the block approaches zero.
If high fluid pressures reduce frictional
resistance and permit rocks to slide down very
gentle slopes, it follows that, as fluid pressure
is decreased, frictional resistance between
blocks of rock is increased, thus permitting
them to come to rest on increasingly steep
slopes. The steeper the slope upon which a
block of rock is at rest, the lower the required
raise in fluid pressure necessary to produce
movement.
In the case of the Precambrian reservoir
beneath the Arsenal well, these rocks w e r e at

indicates that as fluid was pumped out of the

reservoir the fractures closed, and as fluid
was injected into the reservoir the fractures
opened. In other w o r d s , the pumping and injection tests indicated that rock movement
o c c u r r e d as fluid was withdrawn or injected at
relatively low p r e s s u r e s .
The p r e s s u r e - d e p t h relations of the Precambrian reservoir, showing hydrostatic and
lithostatic p r e s s u r e variations with depth, are
shown in figure 8. These data w e r e determined from the drill stem test. As shown on
the chart, the observed pressure of the Precambrian reservoir is almost 900 pounds less
than hydrostatic p r e s s u r e .
Hubbert and Rubey (1959) devised a simple
and adequate means of reducing by the required
amount the frictional resistance to the sliding
of large overthrust blocks down very gentle

32

FORMATION PRESSURE (Ib./in2)

2000 4000 6000_8000 10.000 12,000 14,000
OPEN FRACTURES
The hypocenters in the Arsenal area
plotted f r o m data derived f r o m four or more
r e c o r d i n g stations indicate that movement is
taking place beneath the Arsenal well at depths
of f r o m 1 - 1 / 2 to 1Z miles. If the Precambrian
f r a c t u r e system extends to a depth of 12 miles,
then fluid p r e s s u r e could be transmitted to this
depth by m o d e r a t e s u r f a c e injection p r e s s u r e
as long as the f r a c t u r e system was open for the
transmission of this p r e s s u r e .
Secor (1965) concluded that open fractures
can occur to great depths even with only m o d erately high fluid p r e s s u r e - o v e r b u r d e n weight
ratios. It appears possible that high-angle,
open f r a c t u r e s may be present beneath the
Arsenal well at great depths with much lower
fluid p r e s s u r e - o v e r b u r d e n weight ratios than
has f o r m e r l y been c o n s i d e r e d possible.
Almost 150 million gallons of contaminated
w a s t e had been injected into the Arsenal well
by the end of September 1965. Since this
amount of water would be enough to fill four
continuous I / 1 6 - i n c h f r a c t u r e s each seven
miles long (the maximum distance of epicenters
from the well located by four or more recording stations) and five miles deep, it can be
seem that a relatively small area is being
affected by the injection p r o g r a m .

2000
4000

.^LITHOSTATIC
PRESSURE

^6000
J 8000
10,000

OBSERVED
^uvnoncTATir*
PRESSURE *\\HYDROSTAT1C^
12,000

oFpc WATER* \ PRESSURE

pC WATER

F i g u r e 8.
P r e s s u r e - d e p t h relations, Prec a m b r i a n r e s e r v o i r , Rocky Mountain Arsenal
Disposal well.

equilibrium on high-angle f r a c t u r e planes w i t h

a fluid p r e s s u r e of 900 pounds less than h y d r o static p r e s s u r e b e f o r e injection began.
As fluid was injected into the Precambrian
r e s e r v o i r , the fluid p r e s s u r e adjacent to the
well bore r o s e , and the frictional resistance
along the f r a c t u r e planes was thereby r e d u c e d .
When, finally, enough fluid pressure was
e x e r t e d over a large enough area m o v e m e n t
took place. The elastic wave energy released
was r e c o r d e d as an earthquake.
Since the formation fluid p r e s s u r e is 900
pounds sub-hydrostatic, m e r e l y filling the hole
w i t h contaminated waste (mostly salt w a t e r )
raises the formation p r e s s u r e 900 pounds, or
to the equivalent of hydrostatic p r e s s u r e . Any
applied injection p r e s s u r e above that of gravity
flow brings about an increase in p r e s s u r e resulting in a total in excess of hydrostatic
p r e s s u r e . For example, an injection p r e s s u r e
of 1, 000 pounds would raise the r e s e r v o i r
p r e s s u r e adjacent to the well bore 1, 900
pounds, or by the amount n e c e s s a r y to bring
the p r e s s u r e to hydrostatic (by filling the hole)
plus 1,000 pounds.
Apparently a rise in fluid p r e s s u r e within
the Precambrian reservoir of from 900 to
1, 900 pounds is sufficient to allow movement
to take place.

TIME LAG BETWEEN FLUID INJECTION AND

EARTHQUAKES
The correlation of fluid injected with earthquake f r e q u e n c y ( f i g . 5) suggests that a time
lag exists between the two. Bardwell (1966)
notes that the f r e q u e n c y of Denver earthquakes
appears to lag injected waste by approximately
one to four months. This phenomenon is probably the same as that d e s c r i b e d by Serafim and
del Carnpo ( 1 9 6 5 ) . They describe the o b s e r v e d
time lag between water levels in r e s e r v o i r s
and the p r e s s u r e s m e a s u r e d in the foundations
of dams, and ascribe this to an unsteady state
of percolation through open joints in the rock
m a s s due to the opening and closing of these
passages resulting f r o m internal and externally
applied p r e s s u r e s .
The time lag between waste injected in the^__
Arsenal well and earthquake frequency is
probably due to an unsteady state of percolation
t h r o u g h f r a c t u r e s in the Precambrian r e s e r v o i r

33

3. The majority of the earthquake epic e n t e r s are located within five miles of
the Arsenal well. All e p i c e n t e r s d e t e r mined f r o m f o u r - o r - m o r e station data
are within seven miles of the well.
4. T h e r e is evidence that the earthquake
activity is taking place along a plane
that dips e a s t w a r d and passes beneath
the Arsenal well at a depth of 6. 5 miles
( W a n g , 1965).
5. When the Arsenal injection p r o g r a m is
c o n s i d e r e d on the basis of high, m e d i u m ,
low, or no injection, t h e r e is a c o r r e l a tion between the fluid injected and earthquake f r e q u e n c y .
6. The best correlation of earthquake f r e quency with fluid injected o c c u r r e d
during July, August and September 1965,
when relatively large amounts of fluid
w e r e injected at higher p r e s s u r e s and
for longer p e r i o d s of time than previously.
7. A statistical analysis (Bardwell, 1966)
is cited that suggests a mathematical
relationship b e t w e e n Arsenal earthquakes and volumes of waste i n j e c t e d
into the Arsenal well.
The volume of fluid injected appears to be
affecting the Precambrian r e s e r v o i r only for a
limited distance f r o m the well bore, and r o u g h
estimates of the e n e r g y r e l e a s e d by a single
earthquake suggest that relatively minor rock
m o v e m e n t is involved.
The Precambrian r e s e r v o i r r e c e i v i n g the
Arsenal w a s t e is highly f r a c t u r e d g r a n i t e
gneiss of v e r y low p e r m e a b i l i t y . The f r a c t u r e s
are nearly vertical. The f r a c t u r e porosity of
the r e s e r v o i r is filled with salt w a t e r . R e s e r voir p r e s s u r e is 900 pounds s u b - h y d r o s t a t i c .
It appears that m o v e m e n t is taking place
in this f r a c t u r e d r e s e r v o i r as a result of the
injection of water at p r e s s u r e s f r o m 900 to
1,950 pounds g r e a t e r than r e s e r v o i r p r e s s u r e .
Hubbert and Rubey ( 1 9 5 9 ) point out that
rock m a s s e s in f l u i d - f i l l e d r e s e r v o i r s are
s u p p o r t e d by solid s t r e s s and the p r e s s u r e of
interstitial f l u i d s . As fluid p r e s s u r e approaches
lithostatic p r e s s u r e , the shear s t r e s s required
to move rock m a s s e s down v e r y gently slopes
approaches z e r o .
It a p p e a r s that t h e s e principles offer an
explanation of the r o c k m o v e m e n t in the
Arsenal r e s e r v o i r . T h e highly f r a c t u r e d rocks
of the r e s e r v o i r are at r e s t on s t e e p slopes
under a condition of s u b - h y d r o s t a t i c fluid

due to the opening and closing of t h e s e f r a c t u r e s

resulting f r o m the applied fluid p r e s s u r e of the
injected w a s t e . The delayed application of this
p r e s s u r e at a distance f r o m the well bore is
believed to t r i g g e r the m o v e m e n t r e c o r d e d as
an e a r t h q u a k e .
EARTHQUAKES DURING S H U T - D O W N PERIOD
In c o n s i d e r i n g the earthquake f r e q u e n c y
d u r i n g the year the injection well was shut d o w n ,
it is unfortunate that n e i t h e r periodic bottom hole p r e s s u r e t e s t s nor checks of the fluid
level in the hole w e r e run. Had these m e a s u r e m e n t s been made, then speculation as to how
long it took for the bottom-hole p r e s s u r e to
decline would have been unnecessary.
By the end of September 1963, 102.3
million gallons of fluid had been injected into
the well. It is believed that this injection had
raised the fluid p r e s s u r e in the r e s e r v o i r for
some distance s u r r o u n d i n g the well bore. During the s h u t - d o w n period this elevated p r e s s u r e
was equalizing t h r o u g h o u t the r e s e r v o i r and at
i n c r e a s i n g distances f r o m the well b o r e . As
this fluid p r e s s u r e r e d u c e d the frictional resistance in f r a c t u r e s f a r t h e r away f r o m the
well, m o v e m e n t o c c u r r e d , and small earthquakes w e r e the result.
SUMMARY AND CONCLUSION
The Rocky Mountain A r s e n a l P r e s s u r e Injection Disposal well was d r i l l e d for the purpose of d i s p o s i n g of contaminated w a s t e w a t e r ,
w h i c h is a b y - p r o d u c t of chemical w a r f a r e
p r o d u c t s m a n u f a c t u r e d at the A r s e n a l .
During the month following the initiation of
injection of waste water into the a l m o s t v e r t i cally f r a c t u r e d Precambrian rocks t h e r e w e r e
two earthquakes with e p i c e n t e r s in the Arsenal
area.
A s u m m a r y of the evidence relating the
A r s e n a l injection p r o g r a m w i t h the earthquakes
is:
1. The f i r s t earthquakes o b s e r v e d d u r i n g
the p r e s e n t century w i t h e p i c e n t e r s in
t h e Denver area w e r e r e c o r d e d d u r i n g
the m o n t h following the initiation of the
A r s e n a l injection p r o g r a m .
2. Since the initiation of the injection p r o g r a m in M a r c h 1962, 150 million gallons
of w a s t e have been injected into the
A r s e n a l well, and t h e r e have been 710
e a r t h q u a k e s (to 1 October 1965).

34

r
p r e s s u r e . As the fluid p r e s s u r e is raised
within the r e s e r v o i r , frictional resistance
along fracture planes is reduced and, eventually, movement takes place. The elastic
wave energy released is r e c o r d e d as an earthquake.
If earthquake hypocenters indicate the
point at which movement is taking place and
injected fluid is triggering this m o v e m e n t , then
t h e r e is evidence that open f r a c t u r e s exist at
depths of 12 miles under conditions of lower
fluid p r e s s u r e - o v e r b u r d e n weight ratios than
has f o r m e r l y been considered possible (Secor,
1965). It is believed that the high angle of this
f r a c t u r e system is an important factor. Because the f r a c t u r e s are almost vertical, only a
small part of the lithostatic p r e s s u r e is acting
to f o r c e the f r a c t u r e s closed, and they can remain open under conditions of lower fluid
p r e s s u r e , and at greater depth than if they
w e r e horizontal or inclined at a lower angle.
The time lag between fluid injection and
earthquake f r e q u e n c y is believed to be due to
the unsteady state of percolation of fluid through
the f r a c t u r e s in the r e s e r v o i r due to the opening and closing of these passages resultin'g
f r o m the applied p r e s s u r e of the injected waste.
It is believed that as fluid continues to be
injected into this r e s e r v o i r fluid p r e s s u r e will
be increased at greater distances f r o m the well
bore, and rock movement will be occurring at
ever increasing distances.
In the present case it is believed that a
stable situation in this Precambrian reservoir
is being made unstable by the application of
fluid p r e s s u r e . However, it is interesting to
speculate that the principle of increasing fluid

p r e s s u r e to release elastic wave energy might

have an application in the subject of earthquake
modification. That is, it might someday be
possible to relieve the stresses along some
fault zones in urban areas by increasing the
fluid p r e s s u r e s along the zone using a series
of injection wells. The accumulated s t r e s s
might thus be released at will in a series of
non-damaging earthquakes instead of eventually
resulting in one large event that might cause a
major d i s a s t e r .
ACKNOWLEDGEMENTS
The author is especially obligated to Ben
H. P a r k e r , who critically read the m a n u s c r i p t
and made many valuable suggestions. Thanks
are also due to Lt. Col. Martin J. Burke, J r . ,
Commanding O f f i c e r of the Rocky Mountain
A r s e n a l , Lt. Col. William J. Tisdale, Director of Industrial Operations, and to the Arsenal
Engineering Department, for aid in compiling
the well and injection data; John C. Hollister
and Maurice W. Major of the Colorado School
of Mines Geophysical Engineering Department,
for help in compiling the earthquake data;
Joseph V. Downey, S. J. , Director of the
Regis College Seismological Observatory, for
earthquake data prior to 1962; George E.
B a r d w e l l , for undertaking the statistical
analysis, p r e s e n t e d e l s e w h e r e in this i s s u e ;
L. T r o w b r i d g e Grose and David T. Snow for
information concerning the role of fluid p r e s sure in jointing; Harry C. Kent and David A.
Moore for critically reading the m a n u s c r i p t ;
and to John A. Rathbone and Charles C. Works
for suggestions concerning the statistical
analysis .

REFERENCES
Calhoun, J. C . , 1953, R e s e r v o i r rocks and
rock-fluid s y s t e m s , Part II, Fundamentals
of Reservoir Engineering, Norman,
University of Okla. P r e s s .
Gahr, W. M. , 1961, Contamination of ground
w a t e r , vicinity of Denver (abstract):
Symposium on Water I m p r o v e m e n t , Am.
Assoc. Adv. Sci. , p. 9 - 2 0 .
Hedge, C. E., and Walthall, F. G. , 1963,
Radiogenic Strontium-87 as an index of
geologic p r o c e s s e s :
Science, v. 140,
p. 1Z14-1217.
Hubbert, M . K . , a n d Rubey, W . W . , 1959,
Role of fluid p r e s s u r e in mechanics of
overthrust faulting: Pt. I, Mechanics of
fluid-filled porous solids and its application to o v e r t h r u s t faulting: Geol. Soc.
A m e r i c a ' B u l l . , v. 70, p. 115-166.

Anderman, G. G. , and Ackman, E. J. , 1963,

Structure of the D e n v e r - J u l e s b u r g Basin
and surrounding areas: in Rocky Mtn.
Assoc. Geologists Guidebook to the Geology
of the northern Denver Basin and adjacent
uplifts.
Anonymous, 1965, Geophysical observatory.
Mines M a g . , v. 55, no. 5, p. 2 4 - 2 6 .
Bardwell, G. E. , 1966, Some statistical
f e a t u r e of the relationship between Rocky
Mountain Arsenal waste disposal and
f r e q u e n c y of earthquakes:
The Mountain
Geologist, v. 3, no. 1, p. 37-42.
Boos, C. M. , and Boos, M. F., 1957,
Tectonics of eatern flank and foothills of
the Front Range, Colorado: Am. Assoc.
Petroleum Geologists Bull., v. 41, p.
7.603-2676.
35

?
''',

, 1961, Role of fluid pressure

in m e c h a n i c s of o v e r t h r u s t faulting, a
reply:
Geol. Soc. A m e r i c a B u l l . , v. 72,
p. 1445-1452.
Jones, P. B., 1965, Derby a r e a e a r t h q u a k e s ,
1964:
Colorado School of Mines, Dept.
of Geophysics, unpublished r e p o r t .
M e c h e m , O. E. , and G a r r e t t , J. H. , 1963,
D e e p injection d i s p o s a l w e l l for liquid
toxic \vaste:
Proc. A m . Soc. Civil Eng. ,
Jour. C o n s t r u c t i o n Division, p. 1 1 1 - 1 2 1 .
Pan, P o h - H s i , 1963, The 1962 e a r t h q u a k e s
and m i c r o - e a r t h q u a k e s n e a r Derby,
Colorado: unpublished M. Sc. t h e s i s ,
Colo. School of Mines.
R i c h t e r , C . F . , 1958, E l e m e n t a r y seismology,
San F r a n c i s c o , W. H. F r e e m a n & Co.
Scopel, L. J. , 1964, P r e s s u r e injection disposal well, Rocky Mountain Arsenal,
D e n v e r , Colorado: The Mountain Geologist,
v. 1, no. 1, p. 35-42.

Secor, D. T., J r . , 1965, Role of fluid

p r e s s u r e in j o i n t i n g : A m . J o u r . Sci. ,
v. 2 6 3 , p. 6 3 3 - 6 4 6 .
S e r a f i m , J . L . , a n d d e l Campo, A . , 1 9 6 5 ,
I n t e r s t i t i a l p r e s s u r e s on rock foundations
of dams:
J o u r , of the Soil M e c h a n i c s and
Foundations D i v . , Proc. Am. Soc. of
Civil E n g r s . , v. 91, no. SM5.
T o c h e r , D. , 1964, Earthquakes and rating
scales:
GeoTimes, v. 8, no. 8, p. 1 5 - 2 0 .
W a l k e r , T. R. , 1 9 6 1 , Ground w a t e r c o n t a m i nation in the Rocky Mountain A r s e n a l a r e a ,
D e n v e r , Colorado:
Geol. Soc. America
B u l l . , v. 72, p. 489-494.
W a n g , Y u n g - H a n g , 1965, Local h y p o c e n t e r
d e t e r m i n a t i o n in linearly v a r y i n g l a y e r s
applied to e a r t h q u a k e s in the D e n v e r a r e a :
unpublished D. Sc. thesis, Colo. School of
Mines.
M a n u s c r i p t r e c e i v e d D e c e m b e r 20, 1965.

me dona Id and childers

LEON J. PANKAU

Consulting. Geologists

CONSULTING PHOTODEO LOGIST

STRUCTURAL 3EO MD H PHD LQGIST

Specialists

in interpreting and mapping

obscure surface Geology

Robert E. McDonald
Milton O. Childers, Ph.D.

212 Executive Air Terminal Bldg.

205 North, 2300 West
Salt Lake City, Utah
Phone (801) 364-5024

PHONE: 7 5 7 - 6 3 2 2

2 3 1 3 SO. ELM DRIVE

AREA COOK: 3 0 3

DENVER, C O L O . B O 2 2 3

POWERS ELEVATION
COMPANY, INC.
Oil Well Elevation Service

INDEPENDENT
EXPLORATION

COMPANY

MAIN OFFICE
Houston5825 Chimney Rock Road713MO 6-2561

IN DENVER

POWERS SURVEYORS, INC.

Oil Well Locations and Surveys

Covering
Eight States in the Rocky Mountain Area
Six District Offices to Serve You

P. O. Box 1156
3600 South Huron. Phone 761-1983, Code 303
J. R. Hunt, Vice-Pres. & Division Mgr.

OFFICES: Dallas, Midland, Wichita Falls, Texas; Denver,

Colorado; New Orleans, Louisiana; Newark, Ohio. OVERSEAS: Algiers, Brisbane, London, Madrid, Paris, Tripoli,
Palmerston North, N. Z.

P. 0. Box 10205

Denver, Colorado 802TO

777-1469