ML20198D060

Kerr-McGee Corp Cushing Site Surface Water Quality Evaluation

ML20198D060
Person / Time
Site:	07003073
Issue date:	04/03/1992
From:	STOVER & ASSOCIATES, INC.
To:
Shared Package
ML20198D058	List: ML20198D053 ML20198D060
References
NUDOCS 9205130200
Download: ML20198D060 (19)

Text

_ - _ - _ _ _.. -. _ _ _ _ _ _ _.. _ _ _ _. _ _ _.. - _ _. _ _ _. _ _ _. _ _ _ _.... _ _ _. _.. _.

p' s.

y y..,,,

a -..

1'

/ '

b i.

~

r I

b i-4 KERR-licGEE CORPORATION CUSilING SITE SURFACE WATER QUALITY EVALUATION Preparod For:

+

KERR-McGEE CORPORATION P.O.

Box 89 Cushing, oklahoma 74023 Prepared By:

STOVER & ASSOCIATES, INC.

.P.O.

Box 2056 Stillwater, Oklchoma 74076 April 3, 1992 9205130200 920504 PDR ADOCK 07003073 C

PDR

TABLE OF CollTE!4TS S 0C't1911 P M Le_

Il1TRODUCTIOli..........................,................

1 PRELIMIllARY SAMPLIliG PROGRAM AllD PROBLEM DEFIt1ITIO!1....

1 RECOMME11DED SOLUTioli...................................

7 FOLLOW-UP SAMPLIl1G PROGRAM.............................

9 C o ll C LU S I O !1 S............................................ 15 E

e m

1 e

d

+

LIST OF TABLES Table JiuM2Cr DR.0_QIlp11RD Eaag 1

Sampling Locations 3

2 Proliminary Sampling Program Test 4

Results 3

Follow 'sp Sampling Program Test 10 Result.a M

-m_--

<s <. :

LIST OF FIGURES i

i Figure llumb3I Descrintion page i

sampling Points 2

2 Modified Sampling Points 8

3 Mixing-Zone 11 4-Immediately Downstream of Mixing Zone 12 5

Upstream of Low Water Bridge 13 a

3

)

9 p '.

s I

l.r l'

l I

_ _ _. ~. _ _... _ _

i INTRODUCTION The Kerr-McGee site near Cushing is presently undergoing investigations and remediation activities.

Skull Crook flows through the Kerr-McGee site.

The original channel of Skull Creek entered the site from tho west and meandered near a large acid pit prior to flowing off-site.

Due to concerns for low pH in Skull Creek, part of the site remediation activities consisted of diverting Sku11' Creek away from the acid pit.

A no" creek channel was excavated to divert the creek away from t he acid pit.

Following diversion of the creek, a situation developed creating concern relative to the water chemistry reactions occurring downstream of the intersection of the old channel, the new channel, and a small tributary flowing into Skull Creek.

The intersection of all three of these channels occurred within a very close distance.

Seepage from the acid pit continued _to flow through the old stream channel.

The small tributary _also-flowed near a smaller acid pit.

Following diversion of-Skull Creek into the new channel, a predominantly white precipitate-began to form immediately downstream of the mixing zone where all three streams converged.

Further downstream a predominantly red precipitate' began to form in skull Creek.

Due to the.

formation of these precipitates and concerns for potential environmental impacts on the Skull Creek water chemistry, Kerr-McGee contracted STOVER & ASSOCIATES to perform an environmental evaluation of the surface waters on the site, define'the water chemistry reactions and develop a recommendation for resolving the situation.

PRELIMINARY SAMPLING PROGRAM AND PROBLEM DE/INITION On February 7, 1992 Dr. Enos Stover and Mr. Marty Matlock of STOVER & ASSOCIATES visited-the site to perform a site reconnaissance and to~ collect samples of the surface waters, the precipitates which were forming,_and of the geological material which had been exposed during the construction of the new creek channel.

A simple schematic of the site with the respective sampling points is presented in Figure 1._

The_ sampling points are represented by numbers _ enclosed with circles.- The sampling' locations and the types of samples L

taken are further described in the sampling log (Table -1).

The test _results of this initial analytical testing program are presented in Table 2~.

I

+

l.

U

FIGURE 1 KERR-McGEE CUSHING, OKLAHOMA SAMPLING POINTS SKULL CREEK PROPERTY LINE OLD (T)

D 7

x p CHANNEL (ACID

/

3 s PIT fACID NEW L/

PIT CHANNEL

\\

INTERMITTENT Ns' CREEK F

WAT ER O" 'O O

L T RE AT M EN T O

POND ROAD W

O

,r

(

(

POOL

\\

'n J' PROPERTY LINE p

LOW WATER v

BRIDGE NONAME CREEK @

O BRiOGE NOTE: MAP IS NOT TO SCALE.

TABLE 1 KERR htCGEE CUSillNG, OKLAllOh1 A SAhlPLING LOCATIONS SAh1PLE NO.

SAh1PLE DESCRilrrlON SAhtPLE TYPE

1. 6 Skull Creek, upstream at property line Water

1. 5A Brown shale mixed with opaque crystals, top layer in new Skull Creek channel Sediment

1. 5B Gray shale beneath #5A Sediment

1. 5C Red clay, below shale layer #5B Sediment

1. 2 New channel of Skull Creek Water

1. 1 Old channel of Skull Creek Water

1. 1A Scum, near pipeline Sediment

1. 1B French drain efDuent Water

1. 3 Intermittent stream, run off from small acid pit Water

1. 3A Red sediment from #3 Sediment

1. 4 Skull Creek, down stream from acid pits, mixing area for #1, #2, #3 Water

1. 4A Precipitate of mixing area Slurry

1. 9 Skull Creek, down stream of property line Water

1. 7 Skull Creek, at low water bridge Water

1. 7A Red precipitate from #7 Sediment

1. 10 Noname Creek, above convergence with Skull Creek W ater

1. 8 Skull Creek, below convergence with Noname creek, at bridge W ater Note:

All results reported in parts per million (ppm) which is approximately equivalent to mg/l for wr'er sample and mg/kg for sediment samples.

)

2l3;E!****jEA:!***EjEER

":jj!!!Hjaneagg

)

O!!* Ra :' ES E E* ** s g g 33

r.an.es;s.e.n.sgg s.

gg di 9i5 8 E8 8 !

.e 24 8

i -

39:R$4*7~'E,S!.#***4$kga

: : ; 4 e s,s a r. R.E s s e s 8

-x e.g JE a

"g5".ggna:-- ggg:na=*sgggg a :; ; a e s s : s..a s, c. o e s

a.

g..

q, rs t G

R : 2 8 E R R R G

~

{

g

!$ 8298 **HRA'n***cB.g2A

• 338aerS.8 cran 3

ng..::--

g a

f8 k8 8 b O

m E,

I!

I~EE

,e

" 3, 2 : R G " *, :' !, t t : H * *

• R B I S A

;

• s a. r. *s :.: s e. s. s. a a g

f 1.

w u

-ea

=

. srsa a

~$3o E

! {it

~

e a ::

Wise 3 Ni 4fj

Asiggas: 8 gg8 ***. gpg a g ;. ;. e g s. 3 s a s s n a 1

{

gg argg We, g g 8<

v ye O

E 1

" g g *. 3 3. 2 S : :' S g s a g * *

• a a g g s

" ;;ssa;s *t.sa.us e

~

y oo Ips y

wg 4

a;es88ygE s

E

.F.j f 4

~

s EE $8 08 8 '

n o{K g

B5 s

t l ~" g B!g8Ae%

8 jH 4

gsger :: gn;s-a npu

inunnuses a

I 130.a 3

eE 0

3 00g641 13 {. e

]:

)

[a" g.

I

,N ft * $Nii}i,f,<",t 1

i.i.i t'!!..

,g

-- a q 3

gg gjp t

s G.u.,,,,,,n!s.,,exa!..a.f....a n.....fl EE a.lo-

~.:.

a

. ~..

u x-va

-+a a.-.~

-.a.

4 g-

)

.y-*-

- + - -.

w%q~g-

,--y--,ym,9,y-9 T=r u

1Mt.'tW*

"v-i1*-

4'N rv i-**"w-tNWW-'-v'tN'f

-'Tm-t

'.9%r pr-

t--Tw-+---r

~

9F"t'T--T

'T-

Q&[Q&QNA WW

... ~

5 The water entering the site in Skull Creek, sample location 6,

was clear and relatively sediment free, while the water in the old creek channel, sample location 1,

was slightly reddish in color and was relatively high in suspended material.

When the two waters mixed with the tributary stream at sample location 4, a white precipitate formed causing an increase in the amount of suspended material.

At sample location 7 the white precipitate had given way to a red precipitate and the water still had a slightly elevated amount of suspended material.

At sample locations 10 and 8 the amount of suspended material had decreased to approximately the same levels as the water entering the site at location 6.

As observed in Table 2, the creek bank or geologic materials, samples SA, SB and SC, which were exposed during the excavation of the new channel, contain relatively large amounts of several metals (Aluminum (A1), Iron (Fe),

Magnesium (Mg), Manganese (Mn), Calcium (Ca), and Potassium (K)).

Both the new and old creek channels run through a shale layer which ranged in color from brown to gray to red.

At the surface of the upper brown shale layer, there were crystals of an opaque white material.

The crystals were identified through the use of X-ray diffraction data as gypsum (Calcium Sulfate (CaSO4)).

Gypsum is a common mineral in most areas of Oklahoma and is the apparent source of the Ca and SO4 found in the water samples.

The shales and clays commonly found in the area are composed of mixtures of aluminosilicate minerals, which contain a varying mixture of several metals as part of their structure.

The presence of the shales and the common occurrence of clays in the soil layers of the area explain the presence of many of the metals found in the water samples.

Iron is commonly associated with the red color which is so common in Oklahoma soils.

Small black manganese nodules are also communally found in many of these red soils.

The results of the water quality analysis from tne new channel above the property line represents fairly typical surface water quality for this area of Oklahoma.

The water pH was 8.2.

The alkalinity (a measure of the water's ability to resist changes jn pH on the addition of acids or bases) was 276 ppm, expressed as calcium carbonate.

The old channel water analysis represented a distinctly atypical water analysis with a pH of 3.4 and an acidity of 6,840 ppm, expressed as calcium carbonate.

This water sample appeared to be primarily seepage from the large acid pit which had percolated through the subsurtace into the old creek channel.

4.

M a

6 The solubility of most metals and minerals is much higher under acidic conditions than !.n neutral or basic conditions.

Because of the low pH of the rater scoping from the large acid pit it readily dissolves many of the minerals in the clays, shales, soils, and geologic material as it moves through the subsurface on its way to the old creek channel.

As a result the water present in the old creek channel contains relatively large amounts of several metals (A1, Ca, Fe, Mg, K, and Na).

When this metal rich water mixed with the slightly basic, higher pH water in the new channel, the resulting increase in pH of the mixed waters resulted in the formation of solid metal oxides and hydroxides (precipitates).

Many dissolved metal species are not stable at the pH and Redox conditions which were present in the mixture of all the waters.

As a result of the conditions present on mixing, several of the metals present reacted with oxygen or hydroxide (both of which are electron rich species) to form insoluble metal oxides and hydroxides which then precipitated and settled out of solution.

Although the water from the third source, the small tributary creek which runs near the small acid pit, contains alightly clevated concentrations of most of the metals of interest it seems to have had little influence on the formation of the precipitates.

The aluminum present in the old creek channel water reacted with the hydroxide ions present in the new channel water to form aluminum hydroxides (AlOH, AlOH2, t.nd AlOH3), which are initially white noncrystalline solids forming a crystal structure as it ages.

This precipitation reaction takes place relatively rapidity under the conaitions whicn were present in the mixing area. Manganese and magnesium react with oxygen to form oxides (MnO and MgO) relatively quickly also. These metals and their oxides / hydroxides account for the white precipitates which formed immediately at the mixing zone of the waters in the area of sampling point 4 (Figures 3 and 4).

Iron reacts with oxygen much more slowly under the conditions present in the creek.

As a result of this relatively slow reaction the red precipitate of iron oxide (Fe2O3) did not occur until later, after the mixed water had flowed further down stream from the immediatn mixing zone.

The red iron precipitate also continued to form for a much longer period of time than the other precipitates, resulting in a larger area of the stream being impacted by this reaction (Figure 5).

NG.%2n1 Ml W@ M S. <

p

... ~

7 There was also observed significant bacterial activity, especially in the Fe precipitates.

These appear to be iron bacteria which can tolerate a very low pH (3.0 to 3.5) and oxidize ferrous Fe to ferric Fe which also aids o

precipitating the Fe out of solution.

These iren bacteria then become heavily encrusted with ferric oxide.

Even though the bacteria contribute to the Fe precipitation, they are not the real problem.

The real source of the problem in the heavy metals leached out of the natural formations from the acid sludge pit scopage.

RECOMMENDED SOLUTION

~

The best solution to the precipitation of heavy metals in Skull Creek was determined to be capturing the acid sludge pit scepage.

In order to eliminate the metal precipitation problems, the old creek channel seepage was isolated, collected and treated prior to entering the new creek channel.

This was accomplished by the installation of a french drain in the old channel and the collection of the effluent from the drain.

This seepage runs into a small collection sump (Figure 2).

The french drain was constructed with limestone, which reacts in acidic conditions in such a way that it both neutralizes the acid and increases the alkalinity of the water.

The contents of the collection sump are transferred to a large treatment pond until enough water is obtained to treat it in a batch treatment mode.

This treatment consists of increasirJ the pH of the water to approximately 6.0 to 8.0 with the addition of sodium hydroxide (NaOH).

This increase in pH causes the same reactions which were taking place in the mixing area in the stream to occur in the treatment pond.

The treated water is allowed to stand.

The precipitates settle, resulting in a clear, near neutral water which is then discharged into the new creek channel in the same location that the old channel originally entered the new channel.

The precipitated solids are retained in the treatment pond and prevented from entering the creek.

FIGURE 2 KERR-McGEE CUSHING, OKLAHOMA MODIFIED SAMPLING POINTS SKULL CREEK PROPERTY LINE

'M

,eX ACID PIT FRENCH ACID NEW d

i DR AIN CHANNEL PIT INTERMITTENT cottacno,,

s

'v/

SUMP CREEK

^

L TR ATfAENT O

POND ROAD w

b rh

's POOL J

PROPERTY LINE LOW WATER BRIDGE NONAME CREEK BRIDGE N OTE: MAP IG NOT TO SCALE.

l

s.

9 FOLLOW-UP SAMPLING PROGRAM On March 23, 1992 Dr. Enos Stover and Mr. Ray Powers, of STOVER & ASSOCIATES, revisited the site to perform a follow-up sampling program and to evaluate the treatment procedures in relation to surface water quality in tho stream.

The results of this follow-up sampling program are presented in Table 3.

The old creek channel had been completely filled during the construction of the french drain, several trees had been removed from the old channel area, and the entire area had been leveled and landscaped.

The new channel upstream from the mixing area, the mixing area and the old channel below the mixing area were all completely clear of any metal precipitates.

In addition to the absence of precipitates, these areas also had significant amounts of algae and plant growth on the bottom of the channels.

There was only a slight decrease in pH over the site, from 8.2 at point 6 to a low of 7.4 at point 3 then back up to 7.9 at point 7, which is off-site downutream of the property line.

The Redox potential decreased over the site from a high of 0.207 at sampling point 6, down to a relative constant value of just above 0.000 over the rest of the site.

Th, dissolved oxygen levels in the creek on the sampling day wer2 all considerably above the theoretical saturation poi nt as a result of the algae and aquatic plants.which were present.

These high levels of oxygen also affected the Redox potentials measured in the field, as compared to the original values measured in the laboratory.

1 e Redox potentials of the original samples were a3so affected by the presence of the precipitates.

The alkalinity of the creek during the second sampling program was also relatively constant, indicating that the collection and treatment of the acid pit. seepage was effectively maintaining the original integrity of the Skull Creek water chemistry.

Visual comparisons of the stream before and after completion of the remediation activities dramatically illustrate the positive _ impacts on Skull Creek.

A pictorial comparison of Skull Creek before and after remediation is presented in Figures 3, 4,

and 5.

In Figure 3, comparison of Skull Creek at the mixing zone during the preliminary sampling program

. clearly shows-the white: precipitates (Aluminium hydroxide, Manganese oxides, and Magnesium oxides) forming in Skull Creek.

The second picture in Figure 3 shows the exact same stream are; in the mixing zone during the follow-up sampling program.

The flow in Skull Creek was much lower during the l

follow-up period and no precipitation reactions were observed.

At the lower flow rate in Skull Creek, the precipitation reactions would be more apparent if the source of the heavy metals had not been removed.

l l

Q

. - ~

TABLE 3 KERR-McGEE CUSHING, OKLAHOM A FOLLOW-UP SAMPLING PROGRAM TEST RESULTS SAMPLE LOCATION

~

Parameter SC SC SC French Tributary Mixing Property Low Water Up Stream N.Channe! OLhannes Dr ain Creek Area Line Bridge

(#6)

(# 2)

(# 1)

(#1B)

(#3)

(# 4)

(# 9)

(#7) pH, s u 8.2 8

4 4.3 7.4 7.5 7.9 7.9 Redox Potential, Volts 0.207 0.020 0.234 0.001 0.002 0.001 0.003 0.010 16.3 13.3 13 3 10.5 a ygen, ppm 16 9 14.1 Dissolve x

COD, p 100 130 1290 1780 120 120 140 140 BOD, pp.

< 5.0

< 5.0 142 184

< 5.0

<50

< 5.0

< 5.0 Total Dissolved Solids, ppm 900 906 19400 23900 1250 1070 1010 1000 Total Suspended Solids, ppm 2

6 232 304 12 10 2

6 l

Chloride, ppm 216 226 446 486 22C 236 256 216 Su:f ate, ppm 223 205 12500 15500 420 285 212 230 l

Tota! Alkalinity, mgCaCo3/l 318 316 7400# ~

300 306 320 316 Total Hardness,mgCaCo3/1 482 407 5482 5551 724 575 487 506 i

METALS l

S. Aluminum, ppm

<01

<01 799 852

< 0.1

< 0.1

< 0.1 0.2 S. Calcium, ppm 103 96 8 408 472

57 124 104 108 S. It on, ppm 0.29 0.26 1245 1673 0.48 0.30 0.66 1 02 l

S. Magnesium, ppm 54.5 56.5 1983 1061 80 5 64 55 58 S. Manganese, ppm 0.22 0.18 148 165 2 50 0.92 0 44 0 59 S. Potassium, ppm 3.41 3.74 169 305 4.16 3 94 3.79 3.92 S. Socium, ppm 144 155 581 592 146 228 159 149

)

• Could nct determine because ofinterferences.

• • Insuthcient sample.

j

1. Acidity - initial pH <4.5 1

l

~ t, FIGURE 3. MIXING ZONE pu s 1,

e 2:

~,S.y

.e

..fm

.,d g'

6h

• ', f-

• 'st }y

..,k h-A M f:w:$+2 y~

DEFORE

~

...&... :A.

h,f

,g.

e ~. <.4ib w..~ s : p,c.....e. s ; #m k

s.

...c n

t

.* w

!.- = = '

1 ?!

• i.)

i i

l AFTER l

l

's ARROW ltJDICATES SAME LOCATIOld 4

FIGURE 4.

IMMEDIATELY DOWNSTREAM OF MIXING ZONE 8 r

, r I

[

aj

• Y l

f,.

a ;

i l

I

'\\;)a,

h,..

.:: e l

M l

J.

ph.

.y,

.t,,'.{.

'I f?!

.>. p k, -

ir v a!;[ 3' s

.}

j jr

+

k,.i

+ <: 0.1

j

. i 3,,..;>;>,,,

,2,'

.st a

+

..g y <

s.u m_ _

W k J,

; ;l d;

A n.,

l b.

c l

" % l G. n ;f.'.

,,,$,; y

..,p,g.

2:< 3yw;:y

.s

).- - -

,y

. o m...

4

'e'

p '

9-D k e ;

R%"<.,9.. l.

f b &.'

pD

{

l

kh.%

s mnlff.'S) lN a i

W

.D.079 h,4M 9[, %, b,,,

I 1

..., m BEFORE AFTER 1

i J

l ARROW ltJDlCATES SAME LOCATIOf4 i

gr tv er:

y-ee----.+-,.-,-

.,+----w,--w w-msw rw

---

g-w.+.,w

---

+--4

,--s-,

suis---

+

es-vy---

.--in,-i,.

l FIGURE 5. UPSTREAM OF LOW WATER DRIDGE j

4 "~'

]gh.e h i

~ Y,., j

'.T

.[+

I 7,, r+

1

~

9

.T

'? -

n, y. _.

.. ",,gQ BEFORE e,.

i.n

+u i,

}4

,m k.

,.s'

- ' 3E F

i

..,3 l

. s r.,

l c '.

I 1

i w

l l

i s '

1

,~1. ;.#4M'g m:y 5;'..

g.7. y :..:u.nu#w. m.n.,

2 M e' o.D

,,r :s

.=.

py

pfI@m+DO.g?'. (g &qt[ L!,.$4 : y;+p! t,>..

n

,,g g.m*' y p. ns.

- _ P; U

I:-

., g:n. s'.-

a :. ' ;. '

%e : e

. f.

+ 1P p, pg,. " '., ', y + fg y'.,')%..

g.

s

~4 t

e.

+

s.,

I s

e l

i d

g l

AFTER

~

^

y...,'.

n k'

p -'

8%l

.e p

I b,

.s

.e o

-?-

ws..

~:

i i

ARROW INDICATES SAME LOCATION s

p h

m.....

,o=

j 14

\\

In Figure 4, a visual comparison of Skull. reek downstream of the mixing zone is presented during the preliminary and follow-up sampling programs.

As in Figurn 3,

the difference in water characteristics is readily apparent.

During the follow-up sampling program, the water was crystal clear, and the bottom of the stream bed can be clearly observed.

Bubbles can be seen on the water surface.

As previously indicated, the water is supersaturated due to the extensive aquatic plant growth in the stream bed.

Under supersaturated conditions, oxygen bubbles can be released from the water as the water travels downstream due to the turbulence from the creek bed.

Extensive algae and aquatic plant growth were observed throughout the entire reach of Skull Creek during the follow-up sampling program.

During the preliminary sampling program no aquatic plant growth was observed in the creek bed downstream of the mixing zone.

As described previously, during the preliminary sampling program there was a significant amount of iron precipitate observed in the creek bed at the low water bridge.

Figure 5 shows a comparison of the water quality just upstream of the low water bridge during both sampling programs.

During the follow-up sampling program there were no signs of the iron precipitation reactions observed earlier at the low water bridge.

During the follow-up sampling program, the water is clear and both rocks and aquatic plant growth can be observed on the creek bed.

The analyses of the water from the collection sump and the effluent from the french drain indicate that the high concentrations of meta', which caused the crigina) precipitates are now being captured and retained in the treatment pond.

The actual concentrations of several of the metals in the collection sump were higher than the original concentrations in the old creek channe'..

The continued high concentrations of Al, Fe, Mg, and Mn in the effluent from the french drain and the water in the collection sump indicate that the leaching of these metal is continuing, but since they are no longer reaching the creek channel, tne environmental concerns associate 2 with the metal precipitates in the stream have been addressed in an adequate and reasonable manner.

I

m< wr, ww. ~

15 CONCLUSIONS A detailed sampling and surface water quality testing program at the Kerr-McGee Cushing site indicated that leaching of metals from the subsurface formations was the source of deteriorated water quality in Skull Creek.

The low pH acid pit water leaching through these metal laden formations solubilized the metals which then precipitated back out of solution when this water mixed with the Skull Crock water.

The recommended solution to this problem was therefore to remove the source of the metals to Skull Creek.

This solution was accomplished by installing a french drain and collection sump to collect the acid pit seepage and treat it to remove the metals prior to discharge to Skull Creek.

A follow-up sampling program after start-up of operation of the remediation system proved that the implemented program had resolved the water quality situation described in this report.

The source of the heavy metals from the large acid pit had been removed and the water quality in Skull Creek was essentially the sar.e both upstream and downstream of the Kerr-McGeo site.

I

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ~ ' - - - - - - " - - - - - - - - - ~ - ' ~ ~ - - ' - ' ^