Study Guides, Notes, & Quizzes

Ecological, Groundwater, and Human Health Risk Assessment in a Mining Region of Nicaragua

Description
Ecological, Groundwater, and Human Health Risk Assessment in a Mining Region of Nicaragua
Published
of 18
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Risk Analysis, Vol. 30, No. 6, 2010  DOI: 10.1111/j.1539-6924.2010.01387.x Ecological, Groundwater, and Human Health RiskAssessment in a Mining Region of Nicaragua Francisco Picado, 1 , 2 Alfredo Mendoza, 3 , 4 Steven Cuadra, 5 , 6 Gerhard Barmen, 4 Kristina Jakobsson, 6 and G ¨ oran Bengtsson 2 , ∗ The objective of the present study was to integrate the relative risk from mercury exposure tostream biota, groundwater, and humans in the R´ ıo Artiguas (Sucio) river basin, Nicaragua,where local gold mining occurs. A hazard quotient was used as a common exchange rate inprobabilistic estimations of exposure and effects by means of Monte Carlo simulations. Theendpoint for stream organisms was the lethal no-observed-effect concentration ( NOECs ), forgroundwater the WHO guideline and the inhibitory Hg concentrations in bacteria (  IC  ), andforhumansthetolerabledailyintake( TDI  )andthebenchmarkdoselevelwithanuncertaintyfactor of 10 ( BMDLs 0 . 1 ).Macroinvertebrates and fish in the contaminated river are faced with a higher risk to suf-fer from exposure to Hg than humans eating contaminated fish and bacteria living in thegroundwater. The river sediment is the most hazardous source for the macroinvertebrates,and macroinvertebrates make up the highest risk for fish. The distribution of body concentra-tionsofHgin fishinthe miningareasofthe basinmay exceed thedistributionofendpoint val-ues with close to 100% probability. Similarly, the Hg concentration in cord blood of humansfeeding on fish from the river was predicted to exceed the  BMDLs 0 . 1  with about 10% proba-bility. Most of the risk to the groundwater quality is confined to the vicinity of the gold refin-ing plants and along the river, with a probability of about 20% to exceed the guideline value. KEY WORDS:  Gold mining; hazard quotient; mercury; risk; river 1. INTRODUCTION Riskassessmentistheprocessbywhichtheprob-ability and magnitude of adverse effects is evaluated 1 Centro para la Investigaci´ on en Recursos Acu´ aticos deNicaragua, Universidad Nacional Aut´ onoma de Nicaragua, Man-agua, Nicaragua. 2 Department of Ecology, Lund University, Lund, Sweden. 3 Centro de Investigaciones Geocient´ ıficas, Universidad NacionalAut´ onoma de Nicaragua, Managua, Nicaragua. 4 Department of Engineering Geology, Lund University, Lund,Sweden. 5 Facultad de Medicina, Universidad Nacional Aut´ onoma deNicaragua, Managua, Nicaragua. 6 Department of Occupational and Environmental Medicine, Uni-versity Hospital, Lund, Sweden. ∗ Address correspondence to G ¨ oran Bengtsson, Department of Ecology, Lund University, S ¨ olvegatan 37, SE-223 62 Lund, Swe-den; tel: 46 46 222 3 777; goran.bengtsson@ekol.lu.se. as a result of exposure to one or more stress fac-tors. Risk assessment can be used to predict, com-pare, and manage environmental risk and providea quantitative basis for preventive or remedial ac-tion under uncertainty. One of the objectives of riskassessment is to compare and evaluate the realiza-tion of different sources of hazards, and weigh thebenefits of reducing or eliminating the risk versusthose of accepting it. For instance, environmentaland human exposure to toxic chemicals can oftenbe characterized by low doses of multiple chemi-cals, and a major challenge is to assess the potentialrisk by using largely single-chemical databases. Sim-ilarly, risk assessment in different disciplines, suchas economics, engineering, environmental, and hu-man health risk, is practically conducted indepen-dently, but some efforts have been made to integratethem. (1 − 4) 916  0272-4332/10/0100-0916$22.00/1  C   2010 Society for Risk Analysis  Ecological, Groundwater, and Human Health Risk Assessment 917 One dimension for integration is the spatialscale. Ecological and human health risk assessmentcan be accommodated on a small, local-scale level,for example, when growth inhibition of a soil bac-terium and blood levels of benzene in workers at abenzene contaminated gasoline station are used asassessment endpoints. One can also apply a larger,for example, watershed, scale, when ecological andhuman health risk assessment are integrated withgroundwater vulnerability assessment using popula-tion and community-level endpoints. A watershedapproach unifies the evaluation of the impacts of in-dustrial discharges and agricultural activities on wa-ter quality, biota, and human welfare, and may beused to identify habitats, subareas, and communi-ties within a region most at risk. Different mod-els have been used to evaluate and compare risk atlarger geographical areas, for example, the relativeriskmodel (5) combinedwithtieredprocedures, (6) andweight-of-evidence approaches. (7) Here we calculate the relative risk of mercury(Hg), released by gold mining in a Nicaraguan water-shed, to stream biota, local inhabitants, and ground-water using hazard quotients (HQ) approaches.Those are essentially applicable to large as well assmall areas, they assume that toxicity can be assessedrelative to a reference chemical, and they allow thecomparison of risk from exposure to different stressfactors. (8) The quotients are useful for screening pur-poses and can be evaluated by probabilistic meth-ods as the probability of exceeding 1.0 (which rep-resents exceeding a benchmark value, for example,a reference or guideline value). The values are, how-ever, not measures of risk in a statistical sense, do notreflect effects on population-based metrics, and areusually nonlinear above 1.0 (9 , 10) (but how far above avalue is has no meaning to the interpretation of risk).In many developing countries around the world,Hg is used for amalgamation in mining of gold andother metals. Although gold mining plays an impor-tant role in the economic development, rural ecosys-tems in which mining activity has taken place haveundergone dramatic deterioration. (11 , 12) Discharge of inorganic Hg into river water by amalgamation of gold particles in crushed ores reduces biodiversityof macroinvertebrates and bioaccumulation of Hg infish, (13) and exposure to inorganic Hg from amalga-mation and burning of the amalgam represents a po-tential risk, not only to people directly handling theHg but also to their families and other persons liv-ing in the surroundings. (14 , 15) The human populationin these areas is also potentially exposed to methylHg through the diet, mainly by consumption of localfish. Local cattle milk and meat may represent othersources of methyl Hg. (16) The aims of this work were to (a) develop anapproach to estimate on a watershed basis the riskto groundwater, stream biota, and local inhabitantsfrom exposure to Hg in a mining area in centralNicaragua,(b)calculatetheprobabilitybywhichpre-dicted exposure concentrations or internal concen-trations would exceed hazard endpoints. The selec-tion of endpoints was based on the precautionaryprinciple, that is, reference or guideline values as-sumed to represent absence of effects, or tolerablelevels, and (c) identify a template by which the riskscan be compared. 2. AREA DESCRIPTION The study area is a 28 km 2 basin located 177 kmeast of the capital of Nicaragua, Managua (Fig. 1).The Rio Artiguas, also called Rio Sucio, drains fromnorth to south, meandering along the steep topogra-phy of the basin. The altitude ranges from 400 m a.s.lin the south to 800 m a.s.l in the north. The tropicalsavannahclimatemeetswithatropicalhumidclimatein the basin. The rainy season lasts from May to De-cember and the dry season for the rest of the year.The average precipitation is 2,400 mm/year, and theyearly temperature varies from 15 ◦ C to 34 ◦ C, withhumidity up to 80%.The land use in the basin is primarily for cattlefarming and crops for domestic consumption. Thereare sparse zones with rainforest, mainly along thestream valleys and near the springs. More than half of the population in the basin (ca. 13,000 inhabitants)lives in the countryside areas, using spring water orstream water as the most common source of domesticwater. For more than one century, the Rio Artiguashas received wastes containing Hg, lead, and cyanidefrom the gold mining industry in a small town, SantoDomingo, but also from artisanal activity. (17) About40 tons of Hg and 10 tons of lead have been releasedinto the environment during the past 100 years of mining activity, practiced with amalgamation by sin-gle individuals in their homes, and in low-technologymills, located at the riverside.Previous investigations have found almost oneorder of magnitude higher Hg concentrations in theRio Artiguas water than the permissible concentra-tions for human consumption, (18) and sediments con-taminated as far as 50 km downstream from a mill inSanto Domingo, La Estrella. (19)  918 Picado  et al. Fig. 1.  Map of the R´ ıo Artiguas (Sucio) basin with sampling sites and pollution sources indicated. 3. METHODS3.1. Groundwater Risk Assessment The groundwater risk is often regarded as a com-bination of the intrinsic vulnerability, the pollutionhazards, and the socioeconomic value that water hasforagiven population oreconomic activity.Acombi-nation of three maps, the vulnerability map, the haz-ard map, and the socioeconomic value map, was usedto construct a risk map of groundwater pollution. Weadopted a combination of the index method (20 , 21) anda method based on hazards quotients. (22 , 23) The vulnerability assessment method was a mod-ified version (24) of the DRASTIC method. It usedthe srcinal seven parameters (25) (depth to ground-water ( D ), net recharge ( R ), aquifer media ( A ),soil media ( S ), topography ( T ), influence of the va-dose zone ( I ), hydraulic conductivity ( C )) plus anadditional parameter for the degree of fracturing( M ), which may facilitate the transport of contami-nants. Maps connected to each parameter were over-laid to produce a vulnerability map as describedby Aller  et al.  (1987) (25) and Mendoza and Barmen(2006): (24)  Ecological, Groundwater, and Human Health Risk Assessment 919 Table I.  Danger of Contamination Index (DCI) Assigned toDifferent Methods for Gold Refining, Adapted from Ducci (21) DCI Pollution Source1 Areas of the basin without mining and polluted sourcesof water4 Tailings disposed at areas with groundwater discharge5 Former mining areas6 Current mining areas where Hg is occasionally handled7 Manual mortar, commonly located by a stream8 Water-driven mill9 Electric mill and polluted river water 5 D R + 4 R R + 3 A R + 2 S R + T R + 5 I R + 3 C R + 5 M R = DRASTIC index , where subscript  R  is the rating, and the coefficient of each term in the equation is the weight assigned toevery parameter. The resulting map with numericalvalues was normalized and classified as high, moder-ate (medium), and low vulnerability. (26 , 27) The pollution hazard is associated with themining mills, where Hg is used in the amalga-mation process. Different procedures for gold re-finement are used in the area, and each of themuses different amounts of Hg. Each activity has adifferent magnitude of hazard, since it handles or re-leases different amounts of Hg. Therefore, each haz-ard was given different values following a danger of contamination index ( DCI  ), modified after Ducci (21) (Table I).The basin has 99 perennial springs and a few dugwells. A socioeconomic value, (21) based on the sizeof the population using the water source and its eco-nomic value (Table II), was assigned to each catch-ment that supplies water to a given well or spring.The groundwater contamination risk was obtainedby linking vulnerability, hazards (Table I), and so-cioeconomic values (Table II) in 50  ×  50 m squaresby a cross-table (Table III). (20) Table II.  Characteristics of the Two Socioeconomic ValueClasses in the St. Domingo AreaValue Class Description of the CatchmentsMedium Well or spring supplying 1,000 to 10,000 inhabitantsor an industry with 10 to 99 workers.Low Spring supplying less than 1,000 inhabitants or anindustry with less than 10 workers.The study area has no higher value classes than medium.       T    a      b      l    e      I      I      I  .     C   r   o   s   s  -    T   a    b    l   e   s      (    2    0     )     t   o    E   v   a    l   u   a    t   e    t    h   e    R    i   s    k   o    f    G   r   o   u   n    d   w   a    t   e   r    C   o   n    t   a   m    i   n   a    t    i   o   n    V    →     V   e   r   y    L   o   w    L   o   w    M   e    d    i   u   m    H    i   g    h    V   e   r   y    H    i   g    h    E   x    t   r   e   m   e    l   y    H    i   g    h    V   r    →     D    C    I     ↓     l    M    h   v    h    l   m    h   v    h    l   m    h   v    h    l   m    h   v    h    l   m    h    V    h    l   m    h   v    h    1   v    l   v    l   v    l   v    l    l   v    l   v    l    l   m   v    l   v    l   m    h    2   v    l   v    l   v    l   v    l   v    l   v    l   v    l    l   v    l   v    l    l   m   v    l   v    l   m    h    l    l   m    h    3   v    l    V    l   v    l   v    l   v    l   v    l   v    l    l   v    l   v    l    l    l   v    l   v    l    l   m   v    l    l    h    h    l   m    h   v    h    4   v    l    V    l   v    l   v    l   v    l   v    l    l    l   v    l    l    l   m   v    l    l   m    h    l   m    h   v    h    l   m    h   v    h    5   v    l    V    l    l    l   v    l   v    l    l   m   v    l    l    l   m    l    l   m    h    l   m    h   v    h    l    h   v    h   v    h    6   v    l    V    l    l   m   v    l    l    l   m    l    l   m    h    l   m    h   v    h    l    h   v    h   v    h    l    h   v    h   v    h    7   v    l    l    l   m    l    l   m    h    l   m    h   v    h    l   m    h   v    h    l    h   v    h   e    h   m   v    h   v    h   e    h    8    l    l   m    h    l    l    h   v    h    l   m    h   v    h    l    h   v    h   e    h   m   v    h   v    h   e    h   m   v    h   e    h   e    h    9    l    M    h   v    h    l   m   v    h   e    h   m    h   v    h   e    h   m   v    h   e    h   e    h   m   e    h   e    h   e    h    h   e    h   e    h   e    h    V   :    V   u    l   n   e   r   a    b    i    l    i    t   y    D   e   g   r   e   e   ;    V   r   :    S   o   c    i   o   e   c   o   n   o   m    i   c    V   a    l   u   e   ;    D    C    I   :    D   a   n   g   e   r   o    f    C   o   n    t   a   m    i   n   a    t    i   o   n    I   n    d   e   x   :    V   e   r   y    L   o   w     (   v    l     ) ,    L   o   w     (    l     ) ,    M   e    d    i   u   m     (   m     ) ,    H    i   g    h     (    h     ) ,    V   e   r   y    H    i   g    h     (   v    h     ) ,   a   n    d    E   x    t   r   e   m   e    l   y    H    i   g    h     (   e    h     ) .  920 Picado  et al. In a parallel effort to assess the groundwa-ter risk, concentrations of Hg were determined ina group of 35 spring water (hereafter referred toas groundwater) samples ( C  GW  ). The samples werefiltered through 0.45  µ m cellulose acetate filters(Millipore Corp., Bedford, MA, USA) and imme-diately preserved in 0.1% HNO 3  and kept on iceprior to analyses by inductively coupled plasma massspectroscopy (ICP-MS) (PerkinElmer Sverige AB,Upplands V ¨ asby, Sweden). The instrumental de-tection limit (IDL) was 0.03 ng of total Hg/mL.It was calibrated against a single  202 Hg standard(1 ng/mL). The data were combined with othersources on total Hg concentrations in groundwaterin the area (19 , 24 , 28 − 30) to expand the pool of exposuredata. They were used in three different approachesto calculate the risk, expressed in HQ terms, of expo-sure from Hg contaminated groundwater. The firstapproach addressed the spatial distribution of risk tohumans from drinking groundwater, by calculatingHQ for each individual observation of Hg in ground-water and the guideline of 1  µ g/L for total Hg indrinking water. (18) The numerical values of the HQwere transformed into categorical values followingthe classification suggested in other studies, (31 , 32) Inthe second approach, a probability density function(PDF) was calculated by applying BestFit 2.0 d (Pal-isade Inc., Ithaca, NY, USA) to the same Hg data asin the first approach. The probability that HQ of theratio between the PDF and the guideline value of 1 µ g/Lwouldexceed1.0wasevaluatedbyMonteCarloanalysis (see the next section for details). The thirdapproach addressed the risk of Hg contamination togroundwater microorganisms. Data on inhibitory Hgconcentrations in bacteria (  IC  ) were collected fromthe literature, (33 − 39) a PDF was fitted to the data, di-vided by an arbitrary uncertainty factor of 100 to becomparable with the guideline value, and then usedas the denominator in a Monte Carlo analysis to cal-culate the probability by which the HQ would exceed1.0. 3.2. Ecological and Human Health Risk Assessment  3.2.1. Hazard Endpoints The endpoint for stream organisms was the no-observed-effect concentration ( NOECs ). Since few NOEC   data on Hg were available on freshwater or-ganisms, we used the scaling factor of 100 from acomparison of   NOEC   and  LC   50  (the lethal concen-tration that kills 50% of the exposed organisms) forHg in embryonic and larval stages of fish (40 , 41) to es-timate  NOECs  from  LC  50  for Hg in fish and aquaticmacroinvertebrates. (42) For humans, the endpoints were tolerable dailyintake ( TDI  ), (43 , 44) including an uncertainty factorof 10 to the NOAEL, and benchmark dose levelswith an uncertainty factor of 10 ( BMDLs 0 . 1 ), (44) be-ing equivalent to the construct of a reference dose.The  BMDLs 0 . 1  account for toxicokinetic and toxi-codynamic human variability by applying an uncer-tainty factor to the reported benchmark dose level( BMDL ). (44 − 46) The reported  BMDLs  were derivedby benchmark dose analysis on standardized neu-ropsychological endpoints from three longitudinalprospective studies. (44)  3.2.2. Calculation of Predicted Environmental Concentration (PEC) and Predicted BodyConcentration (PBC) for Aquatic Organisms In the absence of accurate and spatially ex-plicit data, Hg concentrations in stream livingorganisms were calculated from data on Hg concen-trations in sediments and stream water. The calcula-tion of the ecological risk of Hg contamination waslimited to two groups of aquatic organisms, benthicmacroinvertebrates and fish. Benthic macroinverte-brates were assumed to become exposed to Hg fromtwo sources, passively via pore water and actively viaingested sediment. Fish were assumed to become ex-posed via river water and via consumption of benthicmacroinvertebrates. The body concentrations of Hgwere calculated in the benthic macroinvertebratesand fish to facilitate a comparison of the ecologi-cal and health risk, as the latter uses expressions of  PBC  .The pore water concentrations were calculatedfrom the PDFs of the total Hg concentration ( C   s )in the top 5 cm of sediment cores (Appendix A)from each of six sampling sites in the river (Fig. 1),and of the sediment partition coefficient,  K   p (47) (Ap-pendix A). The  PBC   was calculated from the porewater concentrations and bioconcentration factors( BCFs ), (42) defined as the ratio of Hg concentrationin the organism to its concentration in pore water.The uptake of Hg from ingested sediment wascalculated from the PDFs of the total Hg concen-trations in the sediments, the estimated fraction of organic Hg to total Hg in a sediment (48 , 49) and dataon the number of times that the concentration of or-ganic Hg in macroinvertebrates, such as the poly-chaete worm  Nereis diversicolor  , may exceed theconcentration in the sediment. (48) The latter two
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x