1 Genre-Hybridization a Key to Hyper-Bestsellers? The use and function of different fiction genres in The Da Vinci Code and The Millennium Trilogy Ph....
1 J Musculoskel Neuron Interact 2004; 4(4): Tribute to Harold M. Frost M.D. Hylonome Harold M. Frost, M.D. His Contributions Harold M. Frost was a tru...
1 by Harold & Meredith Sears We ronde in almost all of the round dance rhythms, but there have been some especially nice examples in three of our ...
1 Eric Bergman Professional Contact Information: Wildlife Researcher - Colorado Parks and Wildlife 317 W. Prospect Avenue Fort Collins, CO Office: Edu...
1 Lupus Dr. John Bergman2 What is Lupus? The immune system is designed to attack foreign substances in the body. If you have lupus, something goes wro...
1 Briefing IDLS Cpt Nicklas Bergman2 Agenda Aim & scenario Exercise area Participating units Air activity Command and Control3 Aim & Scenario ...
1 Rheumatoid Arthritis Dr. John Bergman2 What is Rheumatoid Arthritis? Rheumatoid arthritis is a chronic inflammatory disorder that can affect more th...
1 HAROLD L. MOSELEY SUBDIVISION, RESUBDIVISION OF LOT 2 Engineering Comments: Must comply with the Mobile County Flood Damage Prevention Ordinance. De...
1 Lyme Disease Dr. John Bergman2 What is Lyme Disease? Lyme disease is caused by the bacterium Borrelia burgdorferi and is transmitted to humans throu...
1 Issues and Challenges in Reducing Health Disparities Among Hispanics/Latinos in the U.S. Aida L. Giachello, Ph.D. Associate Professor and Director M...
ENWR0"TAL EFFECTS OF SALINE 0ILFlEI-D DISCHARGES ON SURFACE WATERS Fred N. Lamming Ann M.Boelter Harold L Bergman Aida M. Farag
In Environmental Toxicology and Chemistry Volume 11
Submitted by Ann M. Boelter Fred N. Lamming Aida M. Farag Harold L Bergman Department of Zoology and Physiology University of Wyoming Laramie, Wyoming
0730-7268192 $5.00 + .00 Copyright 0 1992 SETAC
Environmental Toxicology and Chemistry, Vol. 1 I , pp. 1187-1 195, 1992 Printed in the USA. Pergamon Press Ltd.
ENVIRONMENTAL EFFECTS OF SALINE OIL-FIELD DISCHARGES ON SURFACE WATERS ANNM. BOELTER,*FREDN. LAMMING, ADA M. FARAG and HAROLD L. BERGMAN
Department of Zoology and Physiology, University of Wyoming, Laramie, Wyoming 8207 1 (Received 9 April 1991; Accepted 25 October 1991)
Abstract-Ambient water samples from streams receiving discharges of coproduced brines from the Salt Creek oil field in Wyoming were collected and tested for toxicity from 1988 to 1990. Exposure to waters downstream from the oil field significantly (P< 0.05) reduced survival and reproduction of Ceriodaphnia dubia compared to the upstream control, but fathead minnows (Pimephalespromelas) did not appear to be sensitive to the test conditions. Toxicity to Ceriodaphnia dubia increased as stream flow decreased. Toxicity could not be attributed to H2S, trace metals, or nonpolar organic compounds based on toxicity identification evaluation (TIE) methods. Major inorganic ions ma+, K+,C l , HCO;, and C O f - ) appeared to account for the observed toxicity. Keywords-Oil-field brines
L I I
SO$-,K+,Ca2+, Mg2+,and C0:- . Petroleum
Concerns about the environmental effects of discharging saline, coproduced waters from oil fields have been reported from as early as 1924 [1,2]. Recent implementation of toxicity-testingrequirements for effluent discharges in the United States has led to a reevaluation of saline discharges from oil and gas wells . During oil and gas production, a mix of petroleum and water that is trapped in the petroleum-bearing rocks is pumped to the surface. Coproduced waters are then separated from the oil and either injected into a well or discharged to surface waters. Typically high in total dissolved solids, coproduced waters can also contain petroleum contaminants, sulfide (from the bacterial reduction of sulfate), and trace metals, including Al,Ba, B, Cd, Cu, Fe, Hg, Li, Mn, Pb, Se, Sr, and Zn . Some of these elements are found naturally in subsurface waters, and some are added during production and separation of oil and water. Many of these components, including high total dissolved solids, can be toxic to freshwater biota in receiving waters. To assess this potential problem, this study evaluated the ambient toxicity of saline oil-field discharge waters in Salt Creek and downstream in the Powder River, both in northeastern Wyoming. Subsurface waters in the Salt Creek oil field typically contain high concentrations of Na+, C 1 , and HCO; and sometimes
production operations in the drainage release large quantities of produced water to Salt Creek, which derives almost all of its continuous flow from these discharges  and contributes a significant proportion of the flow in the Powder River. Our objectives were to (a) estimate the ambient toxicity of Salt Creek waters and the persistence of this toxicity downstream and (b) determine the contribution of salts, organic compounds, and other materials to the toxicity of Salt Creek and Powder River waters.
*To whom correspondence may be addressed.
Salt Creek originatesapproximately40 km north of Casper, Wyoming, and flows north through the Salt Creek oil field for approximately 56 km before joining the Powder River (Fig. 1). Three study sites were selected on Salt Creek. The first (USC) was located upstream from any oil treater facility (SE1/4 S26 T39N R78W); the second &IN) was located 26 km downstream from USC (SE1/4 S8 T42N R79W) in the Linch, Wyoming area;and the third (CSC) was about 9.1 m above the confluence of Salt Creek and the Powder River (SE1/4 S15 T43N R79W) (Fig. 1). USC served as the upstream control. All the oil companies releasing discharges that reach Salt Creek are located between USC and LIN. Three sample sites were chosen on the Powder River. The first (CPR) was approximately 0.6 km
A.M. BOELTERET AL.
Fig. 1. Location of Salt Creek and Power River study area. Sample sites: USC = upstream Salt Creek; LIN = Linch; CSC = Salt Creek just upstream from confluence; CPR = Powder River just upstream from confluence; HBR = Powder River 6 km downstream from confluence at Highway 192 bridge; IRY = Powder River 68 km downstream from confluence at Irigary Bridge.
above the confluence of Salt Creek and the Powder River (SE1/4 S15 T43N R79W); the second (HBR) was located near Highway 192 bridge, 6 km downstream from the confluence (NW1/4 S13 T43N R79W); and the third (IRY) was 68 km downstream from the confluence at the Irigary Bridge (SW1/4 S19 T46N R77W) (Fig. 1).
Sample collection Samples were collected from all study sites three times in year 1 (1988), during high (May), medium
(October), and low (August) flow regimes. Stream flow data from 1986  (gaging stations shown in Fig. 1) were used to estimate these flow regimes. During year 2 we collected samples from the LIN and HBR sites in September 1989 (USC was dry at this time) and from the USC and LIN sites in February 1990. Water quality measurements in the field included temperature (degrees Celsius), dissolved oxygen (milligrams per liter) using a YSI model 64A dissolved oxygen meter (Yellow Springs Instrument Co., Yellow Springs, OH), and stream flow (cubic feet per second; 1 cubic foot per second = 0.028 cubic meters per second) using a Teledyne Gurley Pygmy (No. 625) current meter (Teledyne Gurley, Troy, NY). At each sample location, a 19-L grab sample was collected in a polyethylene jug and stored on ice for transport to the Red Buttes Environmental Biology Laboratory at the University of Wyoming, Laramie. There, samples were refrigerated at 4°C for subsequent tests. A 500-ml aliquot was drawn from each sample and filtered with a 0.45-pm cellulose acetate filter, acidified with redistilled HN03 (1 ml/L), and stored at 4°C for later chemical analyses.
Sediments Sediment samples were collected in February 1990 at the USC and LIN sites for pore water and sediment extract analyses. Sediments were scraped from approximately the top 5 cm of stream bottom and stored in 1-L polyethylene bottles for transport to Laramie. There, excess water was decanted and the contents of each bottle thoroughly homogenized by hand. For pore water analyses, four 50-ml centrifuge tubes were each filled with 40 g of homogenized sediment from one sample location. Samples were centrifuged for 10 min at 7,000 rpm in a model B-20A centrifuge (International Equipment Co. , Needham Heights, MD). Supernatants from all four tubes were combined and filtered through a plankton net before toxicity testing to remove visible invertebrates. Sediment extracts were prepared by combining homogenized sediment samples with hard reconstituted water in a 1:4 (v/v) ratio of sediment to water . Each mixture was placed in a 2-L polyethylene container with a lid and shaken vigorously on a shaker table at 150 rpm for 30 min. After settling for 1 hour, the supernatant from each container was decanted and centrifuged for 15 min at 10,OOO rpm. The supernatant was again decanted for toxicity tests.
Saline oil-field discharges
years. Larvae and neonates from these stock cultures were used for all toxicity tests conducted in this study. Undiluted stream water (ambient water sample) was tested for each sample location. The USC sample served as the ambient control, and a hard reconstituted water (pH 8.5, conductivity 600 pS/cm, alkalinity 114 mg/L as CaC03; and hardness 168 mg/L as CaC03) was used as the laboratory control. For statistical comparisons, the USC sample was used as the control. Temperature for all tests was maintained at 25 2°C by using recirculating water baths, and tests were run under a 169-h 1ight:dark photoperiod. Fathead minnow tests were conducted in 1-L glass beakers with 250 ml of test water, with four replicate beakers for each test water and the laboratory control. Each beaker contained 10 larval fathead minnows, for a total of 40 minnows per test water. At the start of each test, the minnows were between 3 and 7 d old and were hatched within 48 h of each other. They were fed 0.1 ml newly hatched brine shrimp (Artemia salina) two times per day during the test. Test waters were renewed daily, and Toxicity identification evaluation (TIE) after 7 d surviving minnows in each replicate beaWe characterized the toxicity of the sample col- ker were dried and weighed as a group and average lected in September 1989 at the LIN site by using dry weight per fish was calculated. phase I toxicity identification evaluation (TIE) proCeriodaphnia dubia tests were started with necedures developed by the U.S.Environmental Pro- onates <24 h old and released within 4 h of each tection Agency (EPA) [111, Manipulations included other. One neonate was placed in each of 10 replipH adjustment, filtration, aeration, CI8Sep-pak cate beakers for each test water and the laboratory (J.T. Baker Inc., Phillipsburg, NJ), oxidant reduc- control. Test beakers were 30-ml plastic cups, each tion, EDTA chelation, and graduated pH tests. The containing 15 ml of test water. Animals were transLIN sample collected in February 1990 was not ferred to fresh test water daily and fed 100 p1 per toxic to the test organisms; therefore, TIE proce- beaker with a mixture of algae (Selenastrumcapdures could not be performed on this sample. In- ricornuturn) and No. 1 trout starter food (U.S. stead, a second TIE characterization of the LIN Fish and Wildlife Service Specification Diet SD9water that was still toxic after being refrigerated 30, obtained from A.J. Balshi Inc., Catawissa, PA) since September 1989 was performed. Also tested slurry. The mean number of young per female was was a reconstituted salt solution containing major calculated at the end of the test by dividing the tocations and anions in concentrations approximat- tal number of offspring produced in each test waing those of toxic stream waters. ter by the number of females that started the test in that water. If a female died before reproducing, Toxicity tests her offspring total was recorded as zero. To test instream toxicity during year 1, we conDuring year 2, toxicity tests were conducted on ducted 7-d survival and growth tests with fathead instream, ambient waters with C. dubia only, beminnows (Pirnephalespromelas) and 7-d survival cause it was the more sensitive species. The test for and reproduction tests with Ceriodaphnia dubia, the water samples collected in September 1989 was using standard protocols developed by the EPA started as a chronic test (10 replicates with one . Stock cultures of fathead minnows and C. organism per replicate), but because all C. dubia dubia were originally obtained from the EPA’s En- in stream water samples died in 24 h, the test was vironmental Research Laboratory in Duluth, Min- discontinued after 48 h and considered an acute nesota, and have been cultured at the Red Buttes test. Thereafter, for the instream water samples, Environmental Biology Laboratory for many 48-h static acute tests were conducted using four
Laboratory chemical analyses Standard methods  were used to analyze routine chemical parameters of temperature, pH (Corning model 10 pH meter, Corning Scientific Instruments, Medfield, MA), conductivity (Extech model 440 conductivity meter, Extech, Boston, MA), alkalinity and hardness, and total ammonia (Orion Analyzer model 407A with a selective-ion ammonia probe, Orion Research Inc., Boston, MA). Major cations, Na+, K+,Ca2+, and Mg2+, and SO:- were analyzed by using a Perkin-Elmer model 5500 inductively coupled argon plasma spectrophotometer (ICP) (Perkin-Elmer Corp., Norwalk, CT). Ten trace elements (Al,As, Fe, Hg, Li, Mn, Ni, Pb, Se, and Sr) were analyzed by ICP during year 1. In year 2, samples were analyzed for these elements plus Cd, Cr, Cu, and Zn. C1- was measured with a Buchler-Cotlove model 4-2000 chloridometer (Buchler Instruments, Fort Lee, NJ) [lo]. Concentrations of HCOT and C0:- were calculated from total alkalinity and phenolphthalein alkalinity titration values.
A.M. BOELTERET AL.
replicate beakers of five C. dubia (<24 h old) for each test water. Each beaker of C. dubia was fed 100 p1 algae (S. capricornutum) and 100 pl of a yeast-CerophyP (Agri-Tech, Kansas City, M0)trout starter food mixture (YCT) at the start of the test. For the TIE samples, sediment pore waters, sediment extracts, and salt solution, we conducted 48-h static acute tests with four replicate beakers of 15 ml test water, five C. dubia (<24 h old), and 125 pl of a food mix of algae and YCT.
For year 1 survival tests, we used stepwise logistic regression for fathead minnow and C. dubia survival vs. chemical and physical variables. We also used multiple linear regression for fathead minnow weights and C. dubia number of young vs. chemical and physical variables. For these regressions, we used the BMDP statistical program [15). Pearson product-moment correlation coefficients were computed using SPSS-X to evaluate the covariance among physical and chemical variables, including Na+, Ca2+,Mg2+,K+, C 1 , SO:, alkaStatistical analyses linity, pH, conductivity, and stream flow. BMDP Data for all 7-d toxicity tests were analyzed by and SPSS-X statistical programs were run on the using the computer program TOXSTAT Version VAX mainframe computer at the University of 2.1 [131, which applies all the statistical procedures Wyoming. recommended in the EPA's protocol . For the RESULTS 48-h acute tests, we calculated LCSO values with the Flow program described in the EPA's manual for acute In year 1, flows in the Powder River ranged tests , which uses the probit, binomial, and from 3.5 cubic feet per second (cfs) at the CPR site moving-average met hods.
Table 1. Water chemistry parameters and flow in Salt Creek and the Powder River during 1988, 1989, and 1990* ~
*Dissolved oxygen, temperature, and flow were measured at sample sites in the field; pH, conductivity, alkalinity, hardness, and ammonia were measured in the laboratory. bSee Figure 1 for sample site locations and codes. 'Not measured.
Saline oil-field discharges
after Salt Creek entered the Powder River (HBR and IRY). In contrast, hardness decreased with the addition of coproduced waters. Dissolved oxygen measurements for all sampling locations were similar to the upstream control and remained consistent throughout sampling periods (Table 1). Concentrations of major inorganic ions (Table 2) followed a trend that would be expected from conductivity, alkalinity, and hardness measurements. Concentrations of Na+, K+, C1-, HCO,, and C0:- increased with the addition of oil-field discharges at the LIN site and then generally decreased downstream, whereas SO:-, Ca2+, and Mg2+ decreased at the LIN site and downstream. Chemical analyses Concentrations of Fe, Mn, Ni, and Sr were relDue to the nature of the surrounding soils and atively low, with few exceeding the upstream Salt bedrock, waters at the USC site have naturally high Creek values, except for the 3.39-mg/L value for hardness and conductivity  (Table 1). Conduc- Fe in the February 1990 LIN sample (Table 3). Valtivity, alkalinity, and pH increased within the area ues for Li increased at the LIN site and then deof oil-field discharges &IN site) and then decreased creased downstream. Concentrations of As, Hg,
in August to 389 cfs at IRY, the site furthest downstream, in May (Table 1). Flows in Salt Creek were 0.04 cfs at the USC site in August and 53 cfs at the CSC site in October. Salt Creek contributed l6Or0 of the water in the Powder River during high flow, 67% during medium flow, and 89% during low flow. Flows at the LIN site in year 2 were 24.8 cfs in September 1989 and 60.0 cfs in February 1990 (Table 1). In September 1989, the streambed at the USC site was essentially dry; the flow was low (approximately 0.03 cfs) in February 1990, but there was enough water to collect a sample.
Table 2. Concentrations of major inorganic ions in Salt Creek and the Powder River during 1988, 1989, and 1990a Cations Siteb
May 1988, high flow USC 845 LIN 1,255
1,410 268 239 73
CPR Oct. 1988, medium flow USC 805 LIN 1.390
1;395 845 IRY 685 CPR 200 Aug. 1988, low flow USC 895 LIN 1,420 csc 1,430 1,290 HBR IRY 1,205 CPR 254 Sept. 1989 LIN 1,015
HBR Feb. 1990 USC LIN
219 55 47 71 70 70
109 52 62 32 31 24
9 18 17 4 4 2
8 1,042 1,066 171 139 8
229 30 25 51 74 161
112 30 33 42 53
9 16 16 9 11 5
8 1,294 1,299 777 552 144
804 242 260 282 273 320
235 24 20 24 37 208
132 24 26 33 53 93
11 14 15 14 15 8
8 1,123 1,238 1,118 1,006 82
914 204 215 249 322 419
‘Values expressed as milligrams per liter.
bSee Figure 1 for sample site locations and codes. ‘Not measured.
466 152 149 85
323 712 878 268 242 157
343 934 876
6.0 7.2 1.8 2.4 0.0
0.0 10.8 7.2 1.8 1.8 0.0
307 927 859 83 1 599 211
0.0 31.8 42.0 15.0 16.8 0.0
A.M. BOELTERET AL.
Table 3. Concentrations of selected trace elements in Salt Creek and the Powder River during the studya Elementb Site'
Fig. 2. (A) Fathead minnow (Pimephalespromelas)survival and weight and (B) Ceriodaphnia dubia survival and reproduction in toxicity tests conducted in waters collected during high (H), medium (M), and low (L) flow periods in 1988. USC = upstream Salt Creek; LIN = Linch; CSC = Salt Creek just upstream from confluence; HBR = Powder River 6 km downstream from confluence at Highway 192 bridge; IRY = Powder River 68 km downstream fromconfluence at Irigary Bridge; CPR = Pb, and Se were below detection limits ( ~ 0 . 2 0 Powder River just upstream from confluence. * = Survival, weight, or number of young significantly (P< mg/L), as were values of Cd, Cr, Cu, and Zn mea- 0.05) lower than upstream control during that flow sured in year 2 samples (cO.01 mg/L). period.
aAdditional analyzed elements were below detection limits: As, Hg, Pb, Se (~0.20 mg/L) and Cd, Cu, Cr, Zn (<0.01 mg/L). bValues expressed as milligrams per liter. 'See Figure 1 for sample site locations and codes.
Hydrogen sulfide had been a suspected toxicant, but we did not detect the characteristic odor of H2S in any of the stream samples. In year 2, two samples from the LIN site were analyzed for sulfide and found to contain <1 mg/L.
Toxicity tests There were no significant effects on fathead minnow survival in any of the water samples collected from Salt Creek and the Powder River during this study. Fathead minnow weights at all sites were significantly lower than weights at the USC site during the low flow period in 1988 (Fig. 2A). No C. dubia survived in ambient water samples collected during all three flow regimes in year 1 from the LIN site or from the CSC site (Fig. 2B). In
Powder River waters, survival was 90 to 100% in water collected at the CPR site but significantlyreduced at the HBR site during medium flow. No C. dubia survived at this site and the IRY site during low flow (Fig. 2B). Sites where all organisms died (LIN and CSC during all flow regimes and HBR and IRY during low flow) were not included in statistical comparisons of C. dubia mean number of young. No significant effects on the number of young were seen in the Powder River sites during high flow (Fig. 2B). But during the medium flow period, the mean number of young produced was significantly reduced at the HBR site and the IRY site.
Saline oil-field discharges
Table 4. Year 2 Ceriodaphnia survival and LC50sa in ambient stream samples and in sediment pore waters, sediment extracts, and a salt solution reconstituted to match major ion concentrations at the LIN site
Yo Survival in 100% water
Sept. 1989/LIN Sept. 1989/HBR Feb. 1990/LIN Feb. 199O/USC Feb. 199O/USC sediment pore water Feb. 199O/USC sediment extract Feb. 1990/LIN sediment pore water Feb. 1990/LIN sediment extract Reconstituted salt solution
0 0 100 100 100 100 100 100 0
0 100 100 100 100 75 100
aLC50 = median lethal concentration. b95%C.1.s shown in parentheses. "No dilutions were tested, so LC50 could not be determined.
1 I *
In year 2, samples collected at the LIN site and the HBR site in September 1989 were toxic to C. dubia (Table 4). No test organisms survived in undiluted water samples, and we calculated a 48-h LC50 of 66% for the LIN sample (Table 4). All C. dubia survived in the USC and LIN samples collected in February 1990 (Table 4). Sediment pore waters and extracts from the USC sediments were not toxic to C. dubia nor were extracts from sediments at the LIN site (Table 4). However, 48-h survival in the sediment pore water from the LIN site was reduced to 75% (Table 4). In all tests, survival of laboratory control organisms was 90 to 100%, and in 7-d tests C. dubia young production averaged 19 young per female and fathead minnow weights averaged 0.51 mg per fish. Statistical regressions that separately regressed cations, anions, alkalinity, pH, conductivity, and flow (cubic feet per second) against year 1 toxicity test results revealed that none of these parameters were significant in accounting for variance in fathead minnow weights. Several parameters did significantly account for the variance in C. dubia survival in year 1 toxicity tests (Table 5). Alkalinity and pH, for example, accounted for 94 and 81% of the variance, respectively. The inorganic ions Cl-, Na+, and K+ accounted for 94, 90,and 82% of the variance, respectively. Two of these ions, Na+ and K+,significantly contributed to the variance in the mean number of C. dubia young (55 and 45%, respectively), but not as much as alkalinity, which accounted for 70% (Table 5). Chemical variables that significantly accounted for toxicity (Na+, K+,C1-, alkalinity, and pH) were highly correlated with each other and with other chemical and physical parameters (Table 6).
None of the standard TIE manipulations reduced the toxicity of the LIN sample collected in September 1989, suggesting that nonpolar organics, cationic metals, or sulfide was not the cause of toxicity. To test whether major ions accounted for the toxicity, we used average values of major cations and anions from the toxic LIN samples (Na+, 1,272 mg/L; Cat+, 3 1 mg/L; Mg2+, 32 mg/L; K+, 22 mg/L; C1-, 1,195 mg/L; SO",, 424 mg/L; and alkalinity, 746 mg/L as CaC03) to prepare a reconstituted salt solution. Analyzed concentrations in the reconstituted solution were Na+, 1,270 mg/L; Ca2+, 15 mg/L; Mg2+, 24 mg/L; Cl-, 1,740 mg/L; SO:-, 427 mg/L; and alkalinity, 678 Table 5. Significant (P < 0.05) regression analyses for Ceriodaphnia dubia survival and reproduction during 7-d toxicity tests in Salt Creek and Powder River waters conducted in year 1 Toxicity end point
Table 6. Correlations among chemical and physical variables measured in Salt Creek and Powder River waters during 1988 Bivariate correlation coefficient ( r ) between variable 1 and variable 2a Variable 2 Variable 1
mg/L as CaC03. The 48-h LC50 for the reconstituted solution (58%) was essentially the same as the 48-h LC50 for the September 1989 LIN sample (66'70) (Table 4). DISCUSSION
Several researchers have noted detrimental effects on biota from oil-field brines [17-191. Birge et al.  reported a strong negative relationship between stream chloride concentrations and fish and macroinvertebrate community structure downstream from a brine outfall. The Salt Creek oil field has been a suspected source of pollution to surface waters since at least 1%8 . Researchers reported free and emulsified oil concentrations up to 11 mg/L in Salt Creek and Powder River waters but concluded these levels did not detrimentally affect fish populations [21,22]. Taraldsen et al.  tested five effluents from Salt Creek oil treater facilities and ambient water samples from Salt Creek above and below one of the dischargers for toxicity to fathead minnows and C. dubia. They also conducted TIESon the effluents. One effluent was not acutely toxic, H2S was the suspected toxicant in three effluents, and nonpolar organic compounds accounted for the toxicity of the fifth effluent. Ambient Salt Creek water collected above one of the dischargers was not toxic, but water immediately below the discharge was toxic to both C. dubia and fathead minnows . Woodward and Riley  reported petroleum
hydrocarbons and Cd, Cr, Cu, Pb, and Zn in water and sediments downstream from an oil-field discharge to the Little Pop0 Agie River in Wyoming. They noted a reduction in species diversity of macrobenthos in downstream sampling stations. In our study, toxicity tests showed adverse effects on C. dubia from exposure to surface waters from the Linch site and from sitesfar downstream in the Powder River. In August 1988, when Salt Creek comprised about 89% of the Powder River flow, toxicity was observed at the IRY site, 68 km downstream from the confluence of Salt Creek and the Powder River. Survival and reproduction were significantlyreduced at the Linch site and at one or more downstream sites at all sampling times except February 1990, when stream flow was the highest recorded at that site during our study. Toxicity increased as stream flow decreased. At low flow, no C. dubia survived in ambient stream waters from the Linch site and all downstream sites. Exposure to sediment pore water from the Linch site reduced survival of C. dubia, but the limited volume of this sample precluded further analysis. None of the trace elements analyzed in this study appeared to contribute to the toxicity of the stream waters. Toxicity due to Hg, Pb, or Se can occur at concentrations much lower than the detection limits used in this study, but evidence from the EDTA chelation tests performed in the TIE and from the reconstituted salt solution tests indicated that these elements did not appear to be present at toxic concentrations. And although Taraldsen et al.  found that H2S and nonpolar organics were suspected toxicants in the effluents, these components had apparently dissipated, decayed, or were diluted to nontoxic concentrations at sites further downstream where we collected our samples. This study illustrates the importance of using more than one test organism whenever possible, as C. dubia was more sensitive than fathead minnows to the test waters. Similarly, a study of produced waters from a degasification field in Alabama revealed that Ceriodaphnia spp. were more sensitive to the waters than fathead minnows and that chloride accounted for most of the toxicity . At the produced water 48-h LC50 for Ceriodaphniu (71%), O'Neil et al.  reported a chloride concentration of 994 mg/L, similar to the chloride concentration of 859 mg/L we found at the LC50 for the September 1989 Linch sample (66%). Birge et al.  reported LC50 values for chloride (as NaCI) of 6,570 mg/L for fathead minnows (96 h) and 1,470 mg/L for Daphnia pulex (48 h), again demonstrating the greater sensitivity of inverte-
Saline oil-field discharges
brates to chlorides. Our study found significant correlations between toxicity and individual parameters, especially C1-, Na+, K+,and alkalinity. However, singling out any one parameter would be misleading due to covariance among the variables. The toxicity of the reconstituted solution was very similar to toxic stream waters from the Linch site. Also, none of the standard TIE manipulations, which do not address toxicity due to salinity, reduced toxicity. Therefore, it appears that major ions accounted for the toxicity of these stream waters. Acknowledgement- We thank Fred Carr, Amoco Oil Co., and the U.S. Naval Petroleum Reserve for providing access to study sites. Sampling locations were suggested by John Wagner and Kurt King of the Wyoming Department of Environmental Quality. Joe Fernandez, Dave Gulley, Connie Boese, Mike Szumski, and Sheryl Hill assisted with sample collection, toxicity tests, and statistical and chemical analyses at the University of Wyoming. Funding was provided by the U.S. Geological Survey through the Wyoming Water Research Center at the University of Wyoming, Laramie.
The problem of stream pollution in Texas with special reference to salt water from the oil fields. Trans. Am. Fish. SOC. 64~81-85. 2. Clemens, H.P. and W.H. Jones. 1954. Toxicity of brine water from oil wells. Trans. Am. Fish. SOC. 84:97-109. 3. Taraldsen, J.E., J.R. Amato and D.I. Mount. 1988. Toxicity testing and characterization of toxicants from effluents of the Powder River Basin, Wyoming. NETAC Technical Report B-88. U.S. Environmental Protection Agency, Duluth, MN. 4. Collins, A.G. 1975. Geochemistryof Oirfierd Watem Elsevier, Amsterdam, The Netherlands. 5. Crawford, J.G. 1964. Rocky Mountain oil field waters. Section 5. Chemical and Geological Laboratories, Casper, WY. 6. Wyoming Department of Environmental Quality. 1986. Wyoming 1986 water quality assessment. Water Quality Division, Cheyenne, WY. 7. U.S. Department of the Interior. 1986. Water resource data for Wyoming. USGS Water-Data Report WY-86-1. U.S. Geological Survey, Cheyenne, WY. 8. Nebeker, A.V., M.A. Cairns, J.H. Gakstatter, K.W. Malueg, G.S. Schuytema and D.F. Krawczyk. 1984. Biological methods for determining toxicity of contaminated freshwater sediments to invertebrates. Environ. Toxicol. Chem. 3:617-630. 9. American Pubtic Health Association, American Water Works Association and Water Pollution Control Federation. 1980. Standard Methods for Examination of Water and Wastewater, 15th ed. American Public Health Association, Washington, DC. 10. Cotlove, E., H.V. Trantham and R.L. Bowman. 1958. An instrument for and method for automatic,
rapid, accurate, and sensitive titration of chloride in biological samples. J. Lab. Clin. Med. 50:358-371. U.S. Environmental Protection Agency. 1988. Methods for aquatic toxicity identification evaluations. Phase I toxicity characterization procedures. EPA 600/3-88-034. Environmental Research Laboratory, Duluth, MN. U.S. EnvironmentalProtection Agency. 1989. Shortterm methods for estimating the chronic toxicity of effluents and receiving waters to freshwater organisms. EPA 600/4-89-001. Environmental Monitoring and Support Laboratory, Cincinnati, OH. Gulley, D.D., A.M. Boelter and H.L. Bergman. 1988. TOXSTAT Version 2.1. University of Wyoming, Laramie, WY. U.S. Environmental Protection Agency. 1985. Methods for measuring the acute toxicity of effluents to freshwater and marine organisms. EPA 600/4-85013. Environmental Monitoring Systems Laboratory, Cincinnati, OH. Dixon, W.J., M.B. Brown, L. Engleman, J.W. Frane, M.A. Hill, R.I. Jennrich and J.D. Toporek. 1985. BMDP Statistical Software. University of California Press, Berkeley, CA. SPSS. 1988. SFSS-X User’s Guide, 3rd ed. Chicago, IL. Charles, J.R. 1967. Effects of oilfield brines. Proceedings, 18th Annual Conference Southeast Association Game Fish Commission, Clearwater, n, Oct. 18-21, 1964, pp. 371-403. Harrel, R.C. and T.C. Dorris. 1968. Stream order, morphometry, physico-chemical conditions, and community structure of benthic macroinvertebrates in an intermittent stream system. Am. Midi. Nat. 80~220-251. Van Hassel, J.H., D.S. Cherry, J.C. Hendricks and W.L. Specht. 1988. Discrimination of factors influencing biota of a stream receiving multiple-source perturbations. Environ. Pollut. 55:271-287. Birge, W.J., J.A. Black, A.G. Westerman, T.M. Short, S.B. Taylor, D.M. Bruser and E.D. Wallingford. 1985. Recommendations on numerical values for regulating iron and chloride concentrations for the purpose of protecting warmwater species of aquatic life in the Commonwealth of Kentucky. Memorandum of Agreement 5429. Kentucky Natural Resources and EnvironmentalProtection Cabinet, Lexington, KY. Muell&, J.W. 1968. Oil pollution investigations on Powder River, middle and south forks of Powder River and Salt Creek, 1967. Administrative Report, Project 0367-07-6503.Wyoming Game and Fish Department, Cheyenne, WY. Millis, R. 1972. Salt Creek oil investigations. Administrative Report, Project 0371-07-7101. Wyoming Game and Fish Department, Cheyenne, WY. Woodward, D.F. and R.G. Riley. 1983. Petroleum hydrocarbon concentrations in a salmonid stream contaminated by oil field discharge water and effects on macrobenthos. Arch. Environ. Contam. Toxicol. 12:327-334. O’Neil, P.E., S.C. Harris, M.F. Mettee, K.R. Drottar, D.R. Mount and J.P. Fillo. 1989. Biomonitoring of a produced water discharge from the Cedar Grove Degasification Field, Alabama. GRI-8910073. Final Report. Gas Research Institute, Chicago, IL.