Fish Consumption Risk Analysis for Tonto Creek, ArizonaJuly 2011

BACKGROUND INFORMATIONTonto Creek is a perennial stream located in Gila

County, in central Arizona. From its origin at TontoSprings, just below the Mogollon Rim, Tonto Creekflows through dense Ponderosa pine forest, deepcanyons and desert chaparral to its confluence withthe Salt River at Roosevelt Lake (Figure 1). Along itscourse, it receives the flows of several smallerstreams including Horton, Spring, Christopher andHaigler Creeks and runs past summer cabins in theKohl’s Ranch area and Bear Flats and through therural areas known as Gisela, Punkin Center andTonto Basin.

At 72.5 miles in length, Tonto Creek is a Strahlerfifth order stream and has a 955 sq. mi. watershedat the confluence with the Salt River. Starting justbelow the Mogollon Rim at an altitude of 7,400feet, the creek then descends through a steepcanyon, in places almost 1,000 feet deep within theHellsgate Wilderness. At the lower wilderness

boundary at Gisela, the creek flows into the valleybetween the Mazatzal and Sierra Ancha mountainsending at an altitude of 2,119 feet where it flowsinto Roosevelt Lake. Geology

Tonto Creek Spring issues from a limestoneaquifer in an area of Permian to Pennsylvanian sed-imentary material and then flows throughMississippian, Devonian, and Cambrian sedimenta-ry rocks. As it flows through the canyon of theHellsgate Wilderness it meets Early Proterozoicquartzite, granitic and metavolcanic rocks. Belowthe wilderness at Gisela, Tonto Creek runs throughPliocene to middle Miocene deposits and earlyPleistocene to latest Pliocene surficial deposits.Within the Tonto basin, there are several occur-rences of barite (barium sulfate), a mineral thatoften co-occurs with cinnabar (mercuric sulfide), aswell as mineable quantities of cinnabar itself.

Publication Number: OFR 11-051

Figure 1. Tonto Creek, Arizona

2

SAMPLING, STANDARDS AND ANALYSISMercury speciation and analysis

All fish tissue samples referenced in this reportwere analyzed for, and the results reported as, totalmercury (THg). The total mercury data was thenused as a surrogate for methylmercury in the riskanalysis. While methylmercury is generally consid-ered the more toxic fraction in consumable fishtissue, total mercury and methylmercury in fishtissue are well correlated (Bloom, N.S. 1992,Maršálek et al. 2006, Kuwabara et al. 2007, Storelliet al. 2003, Lasorsa and Allen-Gil 1995) withmethylmercury comprising ~95-99 percent of thetotal mercury analyzed. Total mercury analysis isalso less complex, less expensive and allows longterm comparison with national and State data sets.These data sets are influenced by the condition(relative weight (Wr) or “fatness”) of the fish.Sampling

In 2002, the U.S. Environmental ProtectionAgency (USEPA) sampled fish from Tonto Creek justbelow the Hellsgate Wilderness Area as a part of theRegional Environmental Monitoring and AssessmentProgram (REMAP). Five fish were taken withoutrecording species and categorized as either small orlarge. The mean mercury concentration for all fishfrom this sample was 1.11 mg/Kg with a range of0.652 to 2.275 milligrams per kilogram (mg/Kg)(N=5). Due to flooding in Tonto Creek, fish werenot again sampled until 2006.

Currently, the Arizona Department ofEnvironmental Quality (ADEQ) and the ArizonaGame and Fish Department (AGFD) work coopera-tively in the collection of fish for the fish consump-tion advisory program. Since 2006 fish in TontoCreek were sampled on five occasions along threesections of the creek by ADEQ and AGFD. A totalof eight different species were taken: smallmouthbass (Micropterus dolomieu), common carp(Cyprinus carpio), green sunfish (Lepomis cyanellus),black bullhead catfish (Ameiurus melas), desertsucker (Catostomus clarki) and largemouth bass(Micropterus salmoides) were taken in the sectionfrom Tonto Narrows upstream of Gisela toRoosevelt Lake. Rainbow trout (Oncorhynchusmykiss), desert sucker and one brown trout (Salmotrutta) were taken in the area of Bear Flats justupstream of the wilderness boundary. An extensivesurvey for piscivorous fish was conducted in theBear Flats and Camp Tontozona portions of TontoCreek to determine the presence or absence of

piscivorous fish. No fish other than rainbow trout,brown trout and desert suckers were found. Resultsare summarized in Table 1 and Figure 2 below.

In the Bear Flats area south of Camp Tontozonathe 11 rainbow trout taken had a mean Hg con-centration of 0.2 mg/Kg with a range of 0.02 to 0.3mg/Kg. Six brown trout had a mean concentrationof 0.2 mg/Kg with a range of 0.1 to 0.3 mg/Kg. Fourdesert suckers had a mean Hg concentration of 0.6mg/Kg with a range of 0.4 to 0.7 mg/Kg.

From Tonto Narrows to Roosevelt Lake the 17smallmouth bass taken had a mean tissue Hg con-centration of 1.5 mg/Kg with a range of 0.4 to 3.0mg/Kg. 13 common carp had a mean Hg concen-tration of 0.5 mg/Kg with a range of 0.3 to 0.8mg/Kg. Four green sunfish had a mean Hg concen-tration of 1.0 mg/Kg with a range of 0.6 to 1.4mg/Kg. Three black bullhead catfish had a meanconcentration of 1.1 mg/Kg with a range of 0.6 to1.9 mg/Kg. A single largemouth bass and a singledesert sucker had Hg concentrations of 0.4 and 0.6mg/Kg respectively.Standards

In the last surface water quality standards rule-making approved January 2009, the State adopteda fish tissue standard of 0.3 mg/Kg for methylmer-cury for protection of the fish consumption desig-nated use. This fish tissue standard was used in cal-culating the lower safe limit in this advisory. Theconsumption advice contained in this document isbased on a risk analysis derived using the USEPAreference dose for methylmercury (USEPA, 2002).

RISK ANALYSISIn aquatic ecosystems with a conducive water

chemistry, (the right balance of sulfate, dissolvedorganic carbon, a favorable redox potential andbacteria population), methylation; which is theincorporation of mercury into an organic molecule,takes place at a rapid rate. Once methylated, mer-cury moves through and up the food web quicklyand becomes concentrated in high order predatorssuch as largemouth or striped bass, northern pikeand/or walleye. Upon ingestion by humans,methylmercury is readily absorbed through the gas-trointestinal tract and easily penetrates the blood-brain and placental barriers. While methylmercuryhas been linked to a variety of health effects, theprimary risk arises from its toxicity to the nervoussystem, including the brain. Most at risk are infantsand unborn children whose mothers consume fishcontaining methylmercury during pregnancy or

3

while nursing. Chronic exposure to methylmercuryat elevated concentrations can cause developmentaldelays and learning disabilities, and acute exposures

may cause gross cranial defects, cerebral palsy anda decrease in living fetuses.

4

Table 1. Mercury in fish filet tissue from Tonto Creek, Arizona

Methylmercury is considered a Class C (possible)human carcinogen due to increased incidence ofrenal adenomas, adenocarcinomas and carcinomasobserved in male mice. These effects were found inthe presence of profound nephrotoxicity caused byexposure to the toxicant and may be a conse-quence of reparative changes in the cells. Whilemethylmercury may not be a strong mutagen, it hasbeen shown to induce chromosomal aberrations.Neurologic effects include erythism or “mad hattersyndrome”, tremors, paresthesia, slowed brainactivity, memory disturbance and sleep disorders.(USEPA, 2002)

CALCULATION OF RISK FOR CHRONIC EXPOSUREWhile Arizona has a standard for methylmercury

in fish tissue, fish containing mercury can still bepart of a healthy diet if consumed at a rate that willnot cause mercury to accumulate to toxic concen-trations in the human body. To determine this rate,the following formula is used:

Where: CRlim = daily consumption rate limit (Kg/day)RfD = national reference dose (mg/Kg/day)BW = body weight (Kg)Hg(t) = total mean mercury concentration

(mg/Kg per individual species)

TOXICOLOGY ANDREFERENCE DOSE

An RfD is an estimate of adaily oral exposure that is likelyto be without any appreciablerisk of adverse health effects.For determining safe rates ofconsumption, the EPA RfD of0.0001 mg/Kg-day is used.This RfD is based on a studyof women and children whoepisodically ate methylmercurytainted fish and pilot whalemeat in the Faroe Islands andconsiders impairment of chil-dren’s neurological functionthat cause developmental

delays. Mean body weights for all age/sex classifica-tions are calculated using data from the NationalHealth and Nutrition Examination Survey(NHANES) (Portier, et al., 2007). Total mercuryconcentrations used in the risk calculation are themeans for the individual species.

POSSIBILITY OF ACUTE TOXICITYWhile developmental chronic toxicity is the

major concern when considering mercury, it is alsoimportant to address the possibility of acute toxicity.Biokinetic modeling (Ginsberg and Toal, 2000) haspredicted the elevation of certain biomarkers asso-ciated with neurodevelopmental effects as a resultof the consumption, during pregnancy, of a singleeight ounce fish meal at or above 2 mg/Kgmethylmercury. Using this concentration as a deci-sion making tool will allow an analysis of possibleacute effects that may occur if a much larger thanaverage fish were caught and consumed. Becausetissue mercury concentrations often correlate wellwith length and/or weight within a single fishspecies (Gewurtz, S. et al. 2011), ADEQ may beable to predict Hg concentrations in fish larger thanthose taken for analysis by correlating length withHg and extrapolating out to State record length.

Smallmouth bass taken from Tonto Creek had amean Hg concentration of 1.5 mg/Kg. Because thecalculated consumption rate was less than one mealper month for that species, the published consump-tion advice will be “no consumption”. For blackbullhead catfish, no consumption is also advised forall age and sex classifications due to the high maximumconcentration (1.9 mg/Kg) found in a specimen well

5

Figure 2. Mercury in fish filet tissue from Tonto Creek, Arizona

below State record length (420 mm). Green sunfishhad a mean concentration of 1.0 mg/Kg with a max-imum concentration of 1.4 mg/Kg occurring at lessthan 80% of the state record. Given this high max-imum concentration, no consumption is alsoadvised for all age and sex classifications.

The unlimited consumption advice for both troutspecies is based on tissue concentrations below theState standard for methylmercury (0.3 mg/Kg).

Desert suckers are not considered in this analysisfor several reasons. First, because of their subapical

mouth structure and grazing/scraping feedingbehavior, it is rare that they can be successfullytaken on hook and line. Also, due to the prevalenceof hard-to-remove bones, if caught, they are hard toprepare and consume. These factors along with thegenerally lower tissue concentrations of mercuryfound for this species in this study, it is consideredunlikely that they will present significant risk to theangling population.

The following Table 2 contains consumptionadvice for the sport fish taken in this survey:

6

Table 2. Safe consumption limits for sport fish from Tonto Creek, Arizona

was 0.00 cfs at the Gun Creek gauge, a “pinchpoint” on the creek where bedrock pushes subflowto the surface.

Tonto Creek also has three separate and distinctsections that impact use and fishing pressure:Upper Tonto Creek, which is regularly stocked withrainbow trout from the AGFD fish hatchery nearTonto Springs and is considered a prime trout fish-ery by many anglers; Middle Tonto Creek, throughthe Hellsgate Wilderness, where fishing pressure isrestricted by difficult, non-motorized access; andlower Tonto Creek where low (or no) flows, degradedlandscape lacking shade and proximity to betterrecreational fishing in Roosevelt Lake probablyreduces fishing pressure (Marc Dahlberg, pers.com).

SYNTHESIS AND CONCLUSIONSWhen determining whether to issue a fish

consumption advisory, a number of factors need tobe considered:

Factors that may discourage consumption ormitigate the consequences of consumption include:• Relatively large distance between the stream and

any major population center• Low density of persons from ethnic groups with

traditionally high fish consumption• Below average birth rate per 1,000 population• Low average flows in lower Tonto Creek• Open, degraded watercourse not esthetically

pleasing• Proximity to better recreational fishing in

Roosevelt Lake

PROBABILITY OF CONSUMPTIONIt is important that the probability and, by infer-

ence, the possibility of consumption be addressedin any fish consumption risk analysis. Also, whilenational default consumption rates are importantfor the calculation of risk for the general populace,they are often problematic when extrapolating riskto vulnerable or underrepresented portions of thepopulation. For these reasons, this analysis address-es factors that may contribute to the possibility ofconsumption rates in excess of the default and datafrom other studies where consumption rates inexcess of the default were found. These factors willbe considered qualitatively in the risk analysisprocess.

DemographicsRelative to the Arizona

and Maricopa County pop-ulation statistics, GilaCounty has a higher child-hood poverty and overallpoverty rate, a lower medi-an income and birth rateand a lower overall popula-tion density. The lowerpopulation density doesnot necessarily present alower risk to any one per-son or group of people.The only amelioratinginfluence of the lower pop-ulation would be the lowernumber of persons exposedto that risk.

FISHING PRESSURE ON TONTO CREEKTonto Creek is unregulated over its full length

and has a highly variable hydrograph. BetweenTonto Spring and the point where the creek crossesthe northern boundary of the Hellsgate Wilderness(upper Tonto Creek), annual flow is largely influ-enced by groundwater discharged into the water-course by multiple springs. However, at its southernboundary, Tonto Creek enters Tonto Basin whereflows can downwell into the wide alluvial deposits.At the USGS gauging station on lower Tonto Creek,monthly mean flows fluctuate between a high of4,272 cubic feet per second (cfs) for a single month(January, 1993) to a low of less than 1 cfs for 15separate months between January, 1990 andSeptember, 2010. For six of those months the flow

7

Table 2. Safe consumption limits for sport fish from Tonto Creek, Arizona

1Arizona Department of Health Services, 2009

Factors that may promote consumption or exac-erbate the consequences of consumption include:• There are several areas consisting both of vaca-

tion/winter homes and year round residenceswithin the vicinity of the creek.

• Tonto Creek contains resident populations oflarge, predatory fish species with a higher potentialto bioaccumulate methylmercury.

• Concentrations of methylmercury found incertain species present a reasonable potential risk(if consumed) to all persons in general and childrenand women of child bearing age in particular.Considering the rates of unemployment and the

low median income found in Gila County, theremay be economic pressures encouraging theutilization of fish caught from Tonto Creek as anepisodic supplemental protein source. Given thatdefault consumption assumptions may underesti-mate the risk to sensitive populations and that theconcentrations of methylmercury found in fishtissue samples taken from Tonto Creek are in excessof the calculated lower safe limit and present apotential risk to consumers, it is recommended thata fish consumption advisory be instituted on TontoCreek as detailed below.

ADDITIONAL PROGRAM INFORMATIONArizona Department of Environmental QualityWater Quality Division – Fish Advisory Program1110 W. Washington St. Phoenix, AZ 85007Contact: Sam Rector(602) 771-4536 or (800) 234-5677 Ext: 771-4536e-mail: smr@azdeq.gov Web site: www.azdeq.gov

ReferencesBloom, N.S. 1992. On the Chemical Form of Mercury inEdible Fish and Marine Invertebrate Tissue. Canadian Journalof Fisheries and Aquatic Science. 49: 1010-1017

Gewurtz, Sarah B., Satyendra, P. Bhavsar and RachaelFletcher, 2011. Influence of fish size and sex on mercury/PCBconcentration: Importance for fish consumption advisoriesEnvironment International 37: 2, 425-434.

Ginsberg, G.L. and B. F. Toal. 2000. Development of a single-meal fish consumption advisory for methylmercury. RiskAnalysis. 20:41-47.

Kuwabara, J.S., Arai, Yuji, Topping, B.R., Pickering, I. J., andGeorge, G.N., 2007. Mercury speciation in piscivorous fishfrom mining-impacted reservoirs: Environmental Science andTechnology 41, 2745-2749.

Lasorsa, B.K. and S. M. Allen-Gil. 1995. The methylmercury tototal mercury ratio in marine, freshwater, and terrestrial organ-isms: an overview. Water, Air and Soil Pollution. 80: 905-913.

Marc Dahlberg, 2011. Personal communication, AZ Gameand Fish Department

Maršálek, P., Z. Svobodová and T. Randák, 2006. TotalMercury and Methylmercury Contamination in Fish from Sitesalong the Elbe River. Acta Veterinaria Brno. Journal of theUniversity of Veterinary and Pharmaceutical Sciences in Brno,Czech Republic 75: 579-585

Portier K.; Tolson, J.; Roberts, S., 2007 Body weight distribu-tions for risk assessment. Risk Analysis 27(1)11-26.

Storelli M.M.; Giacominelli-Stuffler R.; Storelli A.; D'AddabboR.; Palermo C.; Marcotrigiano G.O., 2003. Survey of totalmercury and methylmercury levels in edible fish from theAdriatic Sea. Food Additives and Contaminants, Volume 20,Number 12, pp. 1114-1119(6)

U.S. Census Bureau, 2009. Census Data for the State ofArizona. www.census.gov/census2000/states/az.html

U.S. EPA, 2002. Integrated Risk Information System (IRIS)toxicological review for methylmercury. USEPA, Washington,D.C. www.epa.gov/iris/

8

For rainbow and brown trout caught in TontoCreek upstream of the northern boundary ofHellsgate Wilderness:• Unlimited consumption for all groups

For smallmouth bass, green sunfish and blackbullhead catfish from Tonto Creek downstream ofthe northern boundary of Hellsgate Wilderness:• No consumption

For common carp from Tonto Creek downstreamof the northern boundary of Hellsgate Wilderness:

• Children younger than six years old: No consumption

• Children between six and sixteen years of age: One 8 oz. (uncooked weight) fish meal per month

• All Adults (16 years or older): One 8 oz. (uncooked weight) fish meal per month