B.2 Mayer Ranch/Commerce Site Case Study

Ottawa County, Oklahoma

The mining team would like to acknowledge David Cates, Oklahoma DEQ, and Dr Robert Nairn, University of Oklahoma, for submitting this Biochemical Reactors Case Study.

Contacts

Dr. Robert W. Nairn

The University of Oklahoma

School of Civil Engineering and Environmental Science

202 West Boyd Street, Room 334

Norman, OK 73019

Telephone: 405-325-3354

E-mail: [email protected]

 

David Cates

Oklahoma Department of Environmental Quality

707 N. Robinson

Oklahoma City, OK  73101

Telephone: 405-702-5124

E-mail: [email protected]

 

Name, Location, and Site Description

The site is located in the Tar Creek Superfund Site, part of the Picher Mining Field, and the northeastern Oklahoma portion of the former lead (Pb) and zinc (Zn) mining area known as the Tri-State Mining District. In this area, the ore deposit consists of Pb and Zn sulfides associated with cherty carbonate host rock. The principal ore host stratum is the Boone Formation, composed of fossiliferous dolomite, limestone and nodular chert. Principal ore minerals are sphalerite and galena, with secondary concentrations of chalcopyrite, enargite, luzonite, marcasite, pyrite, and barite. Significant quantities of Pb and Zn were produced from the Tri-State District from the 1890s through the 1960s. By the late 1950s, depressed global markets resulted in the suspension of most mining operations. By the early 1970s when mining ceased, almost 2 million tons of Pb and 9 million tons of Zn had been produced. The Tar Creek site (Figure 1) was proposed for the Comprehensive Environmental Response, Compensation, and Liability Act (Superfund) National Priorities List (NPL) in 1981 and received final listing in 1983. Nearby are the Cherokee County Superfund Site (Kansas) and the Oronogo-Duenweg Mining Belt and Newton County Mines Superfund Sites (Missouri). The following information contains text taken directly from Nairn et al. (2009).

In the Oklahoma portion of the district, approximately 1,000 hectares (ha) are underlain by underground mines in all or part of 47 sections of the Tar Creek Superfund site. During mining, large-capacity dewatering operations pumped approximately 50,000 m3 of water per day from the mines. Upon decline and cessation of mining, groundwater began to accumulate in the mine voids. Approximately 94 million m3 of contaminated water exist in underground voids. By late 1979, metal-rich waters began to discharge into Tar Creek and its tributaries. The first documented discharge of mine drainage was at a location near southeast Commerce, Oklahoma, denoted by the Oklahoma Water Resources Board as Site 14.

Figure B.2-1. Tar Creek Superfund Site and location of Mayer Ranch PTS.

B.2.1 MIW Chemistry

Artesian groundwater discharges from abandoned underground lead and zinc mines into a tributary of Tar Creek for over 30 years. Sampling of seeps from two old mining boreholes showed concentrations, as high a 192,000 μg/l iron, 11,000 μg/l zinc, 970 μg/l nickel, 64 μg/l arsenic; 60 μg/l lead, and 17 μg/l cadmium from 2004 to 2008. The seeps discharged at a rate (as measured with bucket and stop watch) ranging from 570 to 950 L/min (150 to 250 gpm) with a pH of 5.95. Other trace metals of concern are:  aluminum, cobalt, and manganese. The water is net alkaline with CO₂ degassing at the surface. The influent carbonate, acidity, and sulfate concentrations are: total alkalinity 393 mg/L CaCO₃, total acidity 364 mg/L CaCO₃, and sulfate 2,264 mg/l.

Figure B.2-2. Picture of Mayer Ranch Mine water discharge (OWRB Site 14).

B.2.2 System Design

The U.S. Environmental Protection agency (USEPA), University of Oklahoma, and other federal, state, local and tribal partners collaborated in 2004 to design a PTS on part of the 40-square mile Tar Creek Superfund site in northeast Oklahoma. Through a competitive bidding process, a design/build engineering contract was awarded to CH2M-Hill using funds provided by the U.S. Environmental Protection Agency and U.S. Geological Survey. The intent of this project was to develop engineering plans and specifications for the design and construction of the passive treatment system, build the system, and provide as-built construction documents. The goal was to decrease receiving stream concentrations of iron, zinc, lead, and cadmium to the criterion continuous concentrations suggested by USEPA’s national recommended water quality criteria for fresh water aquatic life. Engineering design, based on the conceptual designs and taking into account on-site conditions, was undertaken as an iterative process between the design firm, the University of Oklahoma, and USEPA. To accommodate site constraints, multiple adjustments were made to the system layout. Most notable of these included the configuration of the system as two parallel flow paths to allow estimation of variance in performance and to perform maintenance, consolidation of the layout into a more compact “footprint” area to comply with construction requirements adjacent to an existing utility corridor, and minimize impacts to adjacent landowners, thereby creating a flatter hydraulic profile. Rotosonic overdrilling of both mine drainage seeps was completed 18 months prior to passive treatment system construction, thus providing hydraulic control and allowing additional investigations to further quantify flow rates and variability.

Design and construction tasks included capture and control of the two known artesian mine drainage discharges, diversion of stormwater flows from a 470-ha upgradient watershed, implementation of all passive treatment process units including water conveyance structures, and provision of as-built documents. Design performance was estimated using inflow concentrations of 192 mg/L (Iron) Fe, 11 mg/L Zn, 17 µg/L (Cadmium) Cd and 60 µ/L Pb, and Fe removal rate of 20 grams per square meter per day with target Fe effluent of 1 mg/L and assuming net alkaline conditions, a discharge rate of 1000 L/minute, available site area of approximately 3.6 ha, and treatment area of 2 ha.

Construction began in July 2008 and was completed in late November 2008. Issues requiring resolution during construction and start-up included the incorporation of a third seep (Seep D) into the inflow oxidation pond, removal of debris and existing iron oxide muck, delays throughout the course of construction caused by a record rainfall during the summer and related site water management issues, and changes in organic substrateEither (a) a chemical which reacts or (b) a solid surface or (c) an electron donor. requirements. The third seep was isolated as a distinct inflow into the first process unit and did not result in inflow discharge rates or mass loadings substantially different from initial designs. Based on additional laboratory work, the vertical-flow bioreactors received a mixture of 45% spent mushroom compost, 45% hardwood chips, and 10% manufactured limestone sand in the organic substrate layer. The total cost of design and construction totaled $1,196,000. The system is expected to last 25 years, with an estimated $20,000 annual operation and maintenance (O&M) cost. Total costs (including research) are estimated at $4 million.

The completed system includes 10 distinct process units with a single initial oxidation pond (cellAn individual unit in a treatment system. 1) followed by parallel surface-flow aerobic wetlands/ponds (cells 2N and 2S), vertical-flow bioreactors (cells 3N and 3S), re-aeration ponds (Cells 4N and 4S) and horizontal flow limestone beds (Cells 5N and 5S), and a single polishing pond/wetland (Cell 6). Mine water was diverted into the passive treatment system for the first time on December 2, 2008.

Figure B.2-3. Oblique aerial photograph of passive treatment system taken showing mine drainage discharges and individual process unit designations.

B.2.2.1 BCR Design

Water flows from the aerobic wetlands / ponds to the third process unit comprising parallel vertical flow BCRs designed to provide a reducing environment to remove cadmium, lead and zinc. The BCRs were constructed with a 45-cm (18-inch) layer of organic substrate (a mixture of 45% spent mushroom compost, 45% hardwood chips, and 10% manufactured limestone sand) overlying a 30-cm limestone gravel drainage layer containing perforated drainage pipes. The substrate composition was determined through mainly laboratory testing and was modified slightly prior to installation to increase its hydraulic conductivity. A low permeability HDPEhigh density polyethylene liner beneath the gravel maintains design integrity.

The sizing was designed using the 0.3 moles/m3-d volumetric sulfate reductionThe stripping of oxygen atoms from sulfate (SO₄²⁻), most often yielding sulfide (S²⁻) as an ultimate product. rate and the sum of the system influent molar loadingMass of something per time entering a volume (volumetric loading rate) or flowing into an area (areal loading rate). of zinc, cadmium, nickel, and lead. As it turns out the system may be over sized designed somewhat since the influent concentration of metals to the BCR is somewhat reducedIn chemistry, having gained electrons. Often gaining electrons is accompanied with gaining protons (hydrogen). As an example, when O₂ reacts with H₂, the oxygen is reduced, forming H₂O. by pretreatment due to co-precipitationWhen a chemical is precipitated due to inclusion in a solid made from a different chemical. with iron in the oxidation pond and aerobic wetlands. to the  Sulfate concentration changes in the vertical-flow bioreactors, although demonstrating a great deal of temporal variability, provided mean calculated sulfate reduction rates of 0.53 and 0.40 mol / m3 - d for cells 3N and 3S, respectively.

The system and BCR became operation in December 2008. There was no inoculation period as the mine discharges were turned into the system at this time. High H₂S odors were apparent initially but have reduced over time to the point that they are only noticeable during the summer months. Water samples for biological oxygen demand (BOD) and sulfide analyses are currently being collected at the inlet and outlet of Unit 4 to evaluate the re-aeration process.

B.2.2.2 Pre- and Posttreatment Requirements

The first process unit (Table B.2-1) is an oxidation pond that receives the artesian MIW and oxidizes and hydrolyzes the Fe2+. Oxidized iron then settles out as iron oxyhydroxideA compound containing an unspecified arrangement of oxygen and/or hydroxides. solids. Effluent water flows by gravity feed to the two parallel surface water wetlands. Emergent vegetation in the shallow zones (<0.3 m) increases retention time of solids, while the deeper pools facilitate settling of oxidized and hydroloyzed iron solids from solution.

In the fourth process unit submerged aeration systems in two parallel aeration ponds re-oxygenate water from the BCRs. Aeration reduces odors caused by the emission of hydrogen sulfide gas during bacterial sulfate reduction. Water in one pond is aerated by a vertical displacement pump powered by a 20-foot-tall windmill; as a result, water is aerated only when the wind is blowing. Water in the adjacent pond is aerated by a high- volume compressor powered by a 120-watt photovoltaic (PV) system that charges a deep-cycle battery. The PV system operates the compressor on a 20-hours-on, 4-hours-off cycle. The passive use of a solar panel and a windmill requires minimal maintenance and uses a natural process – not fossil fuels.

Horizontal-flow limestone beds in ponds of the fifth process unit further improve water quality by removing additional zinc and manganese and increasing water hardness to offset the bioavailability of any remaining trace metals. The limestone beds are greater than 1 meter thick to provide a minimum 14-hour retention time.

The final aerobic polishing pond / wetland facilitates settling of residual solids through vegetative filtration and re-aerates water through passive photosynthesis. Water from the pond discharges via an 8-inch exit pipe with flow meter to a limestone boulder cascade for aeration, then to a small channel containing a weir for monitoring flow that feeds into the Tar Creek tributary. The total retention time for the complete system is about 3 weeks.

Table B.2-1. Summary of process units, primary targeted water quality parameters and design function for passive treatment system.

Process unit

Targeted parameter

Function

Oxidation pond

Fe

Oxidation, hydrolysis and settling of iron oxyhydroxide solids; Trace metal sorption

 

Surface-flow wetlands/ponds

Fe

Solids settling

Vertical-flow bioreactors

Zn, Pb, Ni and Cd

Retention of trace metal sulfides via reducing mechanisms

 

Re-aeration ponds

Oxygen demand and odor

Wind- and solar-powered re-aeration; Oxygen demand and H₂S stripping; Addition of O₂

 

Horizontal-flow limestone beds

Zn, Mn and hardness

Final polishing of Zn as ZnCO₃ and of Mn as MnO₂; Hardness addition to offset bioavailability of remaining trace metals

 

Polishing pond/wetland

Residual solids

Solids settling: Photosynthetic oxygenation; Ecological buffering

B.2.3 BCR Performance

In the vertical-flow bioreactors, Zn concentrations decreased from 5.76 ± 0.81 mg/L to 0.24 ± 0.25 mg/L in Cell 3N and from 5.97 ± 0.87 mg/L to 0.28 ± 0.25 mg/L in Cell 3S. These cells were designed specifically to remove Zn and other trace metals through reductive mechanisms, converting free Zn and SO₄2- to ZnS. Removal rates were much higher in these cells (1.46 ± 0.27 and 1.45 ± 0.21 g / m2- d in cells 3N and 3S, respectively) than in cell 1 (1.46 ± 0.27 and 1.45 ± 0.21 g m-2d-1 in cells 3N and 3S, respectively). Biological sulfate reduction in vertical-flow bioreactors is well-documented as a metal removal mechanism (e.g., Jong and Perry 2006; Neculita et al. 2007) and was the primary Zn retention processes for this passive treatment system.

Although the vertical-flow bioreactors were designed to remove Cd and Pb as well as Zn, Cd and Pb rarely remained in measureable concentrations at this stage of the treatment system. On two instances (January and August 2009), Cd was detected in the inflow to the vertical-flow bioreactors at a concentration of approximately 1 μg/L, but was removed through reductive processes occurring within the cells.

A small percentage (<10%) of Ni was removed through co-precipitation and sorption in Cell 1. However, the majority of Ni (~95%) was removed via reductive mechanisms in the vertical-flow bioreactors. The concentration decreased from 0.81 ± 0.03 mg/L to 0.05 ± 0.04 mg/L in Cell 3N and from 0.82 ± 0.04 mg/L to 0.04 ± 0.03 mg/L in Cell 3S. Ni concentrations did not decrease significantly in the remainder of the system, with final system effluent concentrations of 0.035 ± 0.01 mg/L.

Flow-weighted mean SO₄2- concentrations were elevated in the discharges (2,239 ± 26 mg/L) and decreased to 2,047 ± 72 mg/L in the final outflow. Sulfate concentration changes in the vertical-flow bioreactors, although demonstrating a great deal of temporal variability, provided mean calculated sulfate reduction rates of 0.53 and 0.40 mol / m3 - d for cells 3N and 3S, respectively. These values are slightly greater than reported values for field-scale systems of approximately 0.3 mol m-3d-1 as reported earlier (e.g., Dvorak et al. 1992; Neculita et al. 2007) but were not unexpected during this initial operational period.

B.2.3.1 Overall Water Quality Changes

The passive treatment system described here demonstrated that the quality of these artesian hard-rock mine discharges may be effectively addressed through an appropriate combination of ecologically engineered process units. Final effluent concentrations of Cd, Pb and As were below detection limits and percentage decreases for Fe, Zn and Ni in the entire system were 99.7, 98.8 and 96.3%, respectively (Table B.2-2). Conservative ion (e.g., Mg and Li) concentration changes were between 0.3 and 1.2% indicating that design-targeted biogeochemical mechanisms affected contaminant concentrations and dilution played little if any role.

Table B.2-2. Summary total metals and sulfate concentration data for flow-weighted influent and final system effluent. Data are presented in mg/L as mean ± standard error; BDL = below detection limits.

Parameter

Influent

Final Effluent

Removal

Al

0.094 ± 0.009

0.071 ± 0.030

24.47

As

0.063 ± 0.002

BDL (0.022)

65.08

Ca

742 ± 9.0

740 ± 22.3

 

Cd

0.016 ± 0.002

BDL(0.00064)

96.0

Co

0.066 ± 0.008

0.007 ± 0.0004

89.39

Cr

0.001 ± 0.0002

0.002 ± 0.0006

 

Cu

0.002 ± 0.0003

0.003 ± 0.0003

 

Fe

177 ± 2.33

0.57 ± 0.207

99.7

K

26.0 ± 0.286

31.1 ± 4.82

 

Li

0.366 ± 0.010

0.365 ± 0.018

 

Mg

200 ± 2.53

198 ± 7.49

 

Mn

1.51 ± 0.016

1.38 ± 0.197

8.61

Na

94.9 ± 1.63

96.6 ± 4.23

 

Ni

0.945 ± 0.015

0.035 ± 0.007

96.30

Pb

0.068 ± 0.003

BDL(0.019)

72.06

Zn

8.29 ± 0.078

0.096 ± 0.037

98.84

SO₄

2239 ± 26

2047 ± 72

 

 

The PTS retains approximately 57,000 kg (>62 tons) iron, 3,300 kg (4 tons) zinc, 300 kg (660 lbs) nickel, 100 kg (200 lbs) aluminum, 18 kg (40 lbs) arsenic, 17 kg (37.5 lbs) lead, and 5 kg (11 lbs) of cadmium per year. Beneficial reuse of the accumulated solids and substrates retained by the system is being explored as researchers continue to study receiving stream biogeochemistry, fish and macro-invertebrate communities, microbial community activity and degree of bioaccumulation potential.

B.2.4 System Performance

In the first two years of operation, the passive treatment system performed as designed from a water quality perspective. The net alkaline nature of the mine waters was maintained throughout the system as target metals were removed from solution. Due in part to degassing of elevated CO₂ concentrations in the artesian discharges, pH increased from less than 6 to >7 at final discharge. Total Fe concentrations decreased to 0.57 ± 0.59 mg/L at the final outflow with the great majority of total iron removal in the initial oxidation pond. Total Zn concentrations decreased to 0.10 ± 0.10 mg/L for the entire system, with removal occurring via sorption/co-precipitation in cell 1, sulfide precipitation in cells 3N and 3S, and ZnCO₃ formation in cells 5N and 5S. Total Cd, Pb, and As concentrations were decreased to below detection limits (0.64, 19.5 and 22 µg/L, respectively) before the outflow of the second process units, presumably through sorptive mechanisms. The majority of Ni was removed via reductive mechanisms in the vertical-flow bioreactors, with final system effluent concentrations of 0.035 ± 0.007 mg/L.

Coupled oxidative and reductive processes lead to fluctuating dissolved oxygen (DO) and ORP values throughout the system and a demonstrated seasonality as the rates of biological processes varied. The remaining discussion focuses on constituents of interest.

B.2.4.1 Iron Retention

During this initial period of performance, flow-weighted mean total Fe concentrations decreased from 177 ± 2.33 mg/L to 0.57 ± 0.21 mg/L at the final outflow (Table 3). The great majority of total iron removal occurred in the initial oxidation pond and parallel surface-flow aerobic wetlands/ponds, with mean removal rates of 21 ± 2.8 and 3.7 ± 3.4 g / m2-d, with the latter rate being load-limited. The initial oxidation pond was designed specifically for iron oxidation, hydrolysis, precipitation, and settling. A statistically significant relationship between area-adjusted iron removal rate and effluent water temperature did exist for this process unit (r = -0.83, p <0.01) indicating greater removal under colder conditions.

The parallel surface-flow aerobic wetlands/ponds were designed for iron oxide solids retention and include influent and effluent deep water areas with an interior shallow wetland shelf. Although planted, vegetation establishment on this wetland shelf was limited in this first growing season (<10% coverage), perhaps inhibiting solids retention.

B.2.4.2 Zinc Retention

Flow-weighted mean total Zn concentrations decreased from 8.29 ± 0.08 mg/L to 0.10 ± 0.04 mg/L for the entire system. Zinc was targeted for retention via several mechanisms in individual process units of the treatment system, including sorption (Cell 1), ZnS precipitation (Cells 3N and 3S), and ZnCO₃ formation (Cells 5N and 5S). Over the course of initial operation, Zn was removed in all of these cells. In the initial oxidation pond, Zn concentrations decreased to 6.71 ± 0.37 mg/L, representing an area-adjusted removal rate of 0.28 ± 0.05 g/m2d. Retention likely occurred through co-precipitation and sorption to iron oxyhydroxide precipitates.

In the vertical-flow bioreactors, Zn concentrations decreased from 5.76 ± 0.81 mg/L to 0.24 ± 0.25 mg/L in Cell 3N and from 5.97 ± 0.87 mg/L to 0.28 ± 0.25 mg/L in Cell 3S. These cells were designed specifically to remove Zn and other trace metals through reductive mechanisms, converting free Zn and SO₄2 - to ZnS. Removal rates were much higher in these cells (1.46 ± 0.27 and 1.45 ± 0.21 g / m2d in cells 3N and 3S, respectively) than in cell 1. Biological sulfate reduction in vertical-flow bioreactors is well-documented as a metal removal mechanism and was the primary Zn retention processes for this passive treatment system.

Table B.2-3. Selected mean water quality data from artesian discharges and process unit effluents. All data are mg/L except pH (standard units) and alkalinity (mg/L as CaCO₃). BDL = below detection limits.

 

 

pH

DO

Alk

Fe

Zn

Cd

Pb

Mn

As

Ni

SO₄

Discharges

SA

6.01

1.17

402

171

8.34

0.015

0.065

1.47

0.062

0.94

2209

 

SB

5.97

1.04

371

178

8.15

0.016

0.069

1.51

0.064

0.94

2264

 

SD

5.99

1.27

386

181

9.00

0.015

0.070

1.62

0.061

0.96

2343

Cell 1

 

6.10

3.83

161

30

6.71

0.003

0.029

1.48

BDL

0.87

2173

Cell 2

N

6.51

7.94

135

4.4

5.76

BDL

BDL

1.50

BDL

0.81

2199

 

S

6.40

7.00

131

4.4

5.97

BDL

BDL

1.58

BDL

0.82

2216

Cell 3

N

6.85

0.35

262

1.3

0.24

BDL

BDL

1.36

BDL

0.045

2311

 

S

6.84

0.28

242

1.1

0.28

BDL

BDL

1.35

BDL

0.042

2091

Cell 4

N

7.20

3.86

236

0.87

0.16

BDL

BDL

1.38

BDL

0.035

2076

 

S

7.10

3.51

232

0.92

0.23

BDL

BDL

1.44

BDL

0.042

2099

Cell 5

N

6.98

0.65

229

0.57

0.087

BDL

BDL

1.28

BDL

0.033

2064

 

S

6.95

0.42

227

0.63

0.109

BDL

BDL

1.40

BDL

0.033

2195

Cell 6

 

7.11

2.65

224

0.57

0.096

BDL

BDL

1.38

BDL

0.035

2057

 

In the horizontal-flow limestone beds, Zn concentrations decreased from 0.16 ± 0.21 mg/L to 0.09 ± 0.11 mg/L in Cell 5N and from 0.23 ± 0.23 mg/L to 0.11 ± 0.12 mg/L in Cell 5S. Relatively low influent Zn concentrations resulted in load-limited removal rates of 0.03 ± 0.04 and 0.02 ± 0.02 g / m2d for cells 5N and 5S, respectively. These cells were designed to remove Zn through mechanisms specific to proper pE-pH. Although pE values (converted from field ORP readings) were typically within the needed range for amorphousHaving no crystalline form. ZnCO₃ and ZnO formation (-0.1 to +0.9 v), pH values were 6.95-7.20, below the needed pH values of 7.5-8.2.

B.2.4.3 Trace Metals

Other metals of specific interest in these waters were Cd, Pb, and As. All three were removed to below detection limits (0.64, 19.5 and 22 μg/L, respectively) before the outflow of the second process units, presumably through sorptive processes. Although the vertical-flow bioreactors were designed to remove Cd and Pb as well as Zn, Cd and Pb rarely remained in measureable concentrations at this stage of the treatment system. On two instances (January and August 2009), Cd was detected in the inflow to the vertical-flow bioreactors at a concentration of approximately 1 μg/L, but was removed through reductive processes occurring within the cells.

The other trace metal found in significant concentrations in these waters was Ni. A small percentage (<10%) of Ni was removed through co-precipitation and sorption in Cell 1. However, the majority of Ni (~95%) was removed via reductive mechanisms in the vertical-flow bioreactors. The concentration decreased from 0.81 ± 0.03 mg/L to 0.05 ± 0.04 mg/L in Cell 3N and from 0.82 ± 0.04 mg/L to 0.04 ± 0.03 mg/L in Cell 3S. Ni concentrations did not decrease significantly in the remainder of the system, with final system effluent concentrations of 0.035 ± 0.01 mg/L.

Figure B.2-4. Clean water exiting the Mayer Ranch Treatment System (Pond 6).

B.2.4.4 Sulfate

Flow-weighted mean SO₄2- concentrations were elevated in the discharges (2,239 ± 26 mg/L) and decreased to 2,047 ± 72 mg/L in the final outflow. Sulfate concentration changes in the vertical-flow bioreactors, although demonstrating a great deal of temporal variability, provided mean calculated sulfate reduction rates of 0.53 and 0.40 mol m-3d-1 for cells 3N and 3S, respectively. These values are slightly greater than reported values for field-scale systems of approximately 0.3 mol/m3 - d as reported earlier (e.g., Dvorak et al. 1992; Neculita et al. 2007) but were not unexpected during this initial operational period.

B.2.4.5 Receiving Stream 

Biological testing in the receiving stream down-stream from the BCR system shows the presence of less tolerant species of fish that have been absent from the Tar Creek tributary for more than thirty years - since the mine water discharge began in 1979. These species include darters, sunfish, and large-mouth bass.

B.2.5 BCR Monitoring

Future research at this site will focus on development of a detailed mechanistic understanding of passive biogeochemical processes leading to water quality changes. Research infrastructure has been installed to examine sedimentationThe process of depositing entrained particles from water. rates in the initial oxidation pond, pore water chemical composition in the vertical-flow bioreactors, oxygenation kineticsThe study of rates of reaction. in the re-aeration cells, dissolution/precipitation dynamics in the horizontal-flow limestone beds, and evaluation of metals transport after high flow rainfall events by means of auto-samplers at the inlet and outlet of the aerobic wetlands and the system outlet. Ancillary ecological benefits provided in the passive treatment system itself and in the receiving waters will be examined, as well as public educational opportunities.

Monthly monitoring for the first three years of system operation has transitioned to quarterly monitoring. This level of monitoring consists primarily of collection of water samples (for analyses of seven metals, alkalinity, and sulfate, as well as field parameters: dissolved oxygen, specific conductance, pH and temperature) at the inflow and outflow of each process unit and measurement of volumetric flows (with bucket and stop watch at the system influent and a weir and flow meter with pressure transducer at the effluent). Some continuous data are collected from YSIs and data loggers installed at the system influent and outfall. This stored data (dissolved oxygen, specific conductance, pH and temperature) is downloaded during these monitoring events. Other research data is collected during these monitoring events. For example, water samples at the inlets and outlets of the re-aeration ponds for BODbiological oxygen demand and sulfide analyses are now collected monthly to evaluate the kinetics of oxygenation in the re-aeration ponds. Also auto-samplers stationed at the outlets of the oxidation pond, the aerobic wetlands, and the system outlet (polishing pond) will be used to collect samples after high flow rainfall events to evaluate metals re-mobilization and transport through the system.

The YSI multi-meters installed at continuous monitoring sites are inspected during periodic monitoring and are serviced annually. An optical dissolved oxygen probe with a wiper is now used on the YSI multi-meter since the membrane probes tended to foul often from iron oxyhydroxide precipitates.

Figure B.2-5. Inspection of a YSI multi-meter.

Routine maintenance was conducted during the monthly monitoring events. Currently this is being accomplished by OU during the quarterly monitoring and weekly by interns from the local college (Northeastern Oklahoma A&M) hired under a cooperative agreement with the City of Commerce, the Quapaw Tribe of Oklahoma, and the University of Oklahoma. The weekly duties involve mowing, keeping the inlets and Agra-drains clear of vegetation, visual inspection for normal operations and reporting problems. They also collect samples from the auto samplers and mail them to OU.

BCR longevity is evaluated through periodic coring of the substrate. So far no reaction front has been observed. The design life of the BCR is for 30 years.

The area is filled with wildlife, including golden winged butterflies, turtles, frogs, muskrats, red-wing blackbirds, ducks and a few geese. A wildlife camera has recorded other animals including deer and bobcat. No vandalism has occurred even though trespassers have been seen on the wildlife camera (including kids fishing in Pond 6).

B.2.6 Regulatory Challenges

This passive treatment system represents the first full-scale mine water treatment system in the mining district. At this time, the fund-balancing waiver regarding surface waters under the OU 1 ROD is still in place and any concerted effort to address water quality has not been undertaken. However, in the Fourth Five Year review of the OU 1 ROD, it was noted that passive treatment "could be an economically feasible engineered remedy for surface water at the site. For these reasons, in this fourth five-year review, the fund balancing ARARsapplicable or relevant and appropriate requirements waiver included in the OU1 ROD may no longer be appropriate and should be reevaluated" (USEPA 2010).

The discharges addressed by the passive treatment system represent approximately 20% of the contaminant mass loading from artesian mine drainage discharges in the Tar Creek watershed. The stream and its tributaries receive additional contaminant mass loading from other artesian discharges, waste pile and pond runoff, and other sources. It is obvious that considerable efforts would be required to satisfactorily address mining-related water quality deterioration in this watershed. In 1994, approximately 15 years after the first discharges began to flow, the State of Oklahoma concluded that “the impacts to Tar Creek… are because of irreversible man-made conditions” and furthermore USEPA “concurs with the State’s conclusion that the surface water conditions are irreversible” (USEPA 1994). However, appropriately designed passive treatment systems appear to be capable of treating these waters effectively and evaluation of future watershed-scale applications is warranted.

From the bioavailability and eco-toxicity perspectives, the untreated artesian discharges exceeded the hardness-adjusted USEPA Criterion Continuous Concentrations (CCC) for Cd, Fe, Ni, and Zn and the Criterion Maximum Concentration (CMC) for Zn (USEPA 2010b). In general, as hardness increases the toxicity of the metal decreases. Despite the fact that there was no significant increase in hardness values in the system, the system effluent no longer exceeds any of these criteria. LaBar et al. (2010) report on stream water quality changes related to operation of this passive treatment system.

The Mayer Ranch pilot project was conducted within a Superfund site under the oversight of the USEPA and consequently a discharge permit was not required. Also mitigation of the volunteer (cattail) wetlands that sprung up after the mine water began to discharge in 1979 was not required. In theory the clean polishing pond at the end of the treatment system could replace (in terms of mitigation credits) the contaminated volunteer wetlands that were destroyed during construction of the treatment system.

B.2.7 Stakeholder Challenges

A century of hard rock mining in the Tri-State Mining District resulted in substantial environmental degradation of land and water resources. Although the region was designated as a suite of USEPA Superfund sites in the early 1980s, streams draining the mining-impacted watersheds still fail to meet designated beneficial uses, in part due to administrative decisions rendering them "irreversibly damaged". However, three decades of remediation work, coupled with changes in mine pool water quality and advances in passive treatment technologies, merit reconsideration of these decisions. Furthermore, given the recent relocation of residents of the most severely impacted communities, unique opportunities exist for large-scale ecological engineering implementation projects that could effectively address water quality. Initial results from a pilot passive treatment system are promising and fit well with possible future land uses being considered by the Quapaw Tribe of Oklahoma (i.e., a return to native prairie and associated wetland and riparian habitats). Successfully addressing water quality higher in the watersheds would also provide substantial downstream benefits to other Native American tribes, residents and fish and wildlife resources.

An ‘attractive nuisance’ for wildlife, is an issue resulting from the creation of the treatment system and potential exposure of wildlife to contaminants in the water and sediment of the unit cells. The most likely exposure would occur in the water of the aerobic wetland and the BCR but this is thought to be much less contaminated with metals than the previous volunteer wetlands.

B.2.8 Other Challenges and Lessons Learned

There have been several instances where the mine water backed up in the treatment system and the emergency bypass had to be activated to divert some of the water to the un-named tributary of Tar Creek to avoid overflowing unit the berms while work was conducted to remedy the problem. In one instance the system effluent pipe had floated out of the water, shutting off the outflow and caused water to back up in the system. This circumstance was identified during one of the routine inspections. It was promptly repaired (with coordination and assistance from the City of Commerce) and the water was re-directed into the treatment system as very short time after discovery.

Another instance when the water backed up in the system to the point of nearly overflowing the berms occurred during the first spring of operation. Some of the water was diverted to the emergency bypass to evaluate the cause of the high water levels in ponds 1 and 2. The piping was investigated as a potential cause of the problem. Also the Agra-drain panels were adjusted in an attempt to increase the water flow out of the ponds. This is a longer term iterative process in which the response of the change takes many days to equilibrate to the new flow conditions. The pipes were cleaned out. A muskrat was found partially clogging one of the pipes and accumulation of subaqueous vegetation in the water conveyance structures may have added to the reduced outflow of water. The piping system was modified by adding animal guards (e.g., varmint grates) to the pipe ends and straightening the piping diagram by eliminating a number of elbows and 90 degree joints in the 6 inch pipe to reduce the headA specific measurement of water pressure above a geodetic datum. It is usually measured as a water surface elevation expressed in units of length. loss they contributed. The system was placed back in full service. As a result, inspection and cleaning of vegetation from the inflow and outflow structures (i.e., varmint grates and Agra-drains) is included in the routine monitoring and maintenance.

It was found that the aerators work better if suspended in the water column rather than resting on the bottom of the re-aeration ponds. They were suspended using children’s water noodles of various colors. Apparently turtles like the green ones as they have eaten them to the point of sinking the aerators. Muskrats are also a problem with burrowing in the berms and clogging the piping.

The YSI multi-meters installed at continuous monitoring sites are inspected during periodic monitoring and are serviced annually. An optical dissolved oxygen probe with a wiper is now used on the YSI multi-meter since the membrane probes tended to foul often from iron oxyhydroxide precipitates.     

Additional artesian discharges are found in the upper Tar Creek and Beaver Creek watersheds (Figure B.2-1). However, contaminant concentrations and mass loadings are less than or similar to these untreated discharges (Table B.2-4), making passive treatment an especially attractive option to address surface water quality degradation.

Table B.2-4. Comparison of water quality characteristics for existing passive treatment system design and other potential discharges to be treated (adapted from Nairn et al. 2011)

Parameter

Existing passive system design influent

Representative Beaver Creek artesian discharge

Estimated Tar Creek artesian Discharge at Douthat

pH

5.95

6.62

6.2

Total alkalinity (mg/L)

393

185

100

Net alkalinity (mg/L)

29

90

 

Fe (mg/L)

192

9.28

55.1

Zn (mg/L)

11

1.60

8.05

Ni (mg/L)

0.97

0.02

0.235

Cd (µg/L)

17

1

0.004

Pb (µg/L)

60

15

0.003

As (µg/L)

64

20

0.014

SO₄-2 (mg/L)

2239

244

1231

Discharge rate (LPM)

1000

25

8500

The size of the BCR unit may be reduced compared to this pilot system (which was designed using the 0.3 mole / m3 – d sulfate reduction rate applied to the mine water discharge outflow), if the design were based on the metals loadings after the oxidation pretreatment step. Cadmium and lead and to a lesser extent zinc and nickel loads are reduced due to co-precipitation in the oxidation pond and aerobic wetland. This along with better post treatment re-aeration may reduce the initial high sulfide odors.

B.2.9 References

J.A. LaBar, R.W. Nairn, K.A. Strevett, W.H. Strosnider, D. Morris, C.A. Neely, A.E. Garrido and K. Kauk, 2010, “Stream Water Quality Improvements after Installation of a Passive Treatment System”, this Paper was presented at the 2010 National Meeting of the American Society of Mining and Reclamation, Pittsburgh, PA Bridging Reclamation, Science and the Community June 5 - 11, 2010. R.I. Barnhisel (Ed.) Published by ASMR, 3134 Montavesta Rd., Lexington, KY 40502.

Nairn, R. W., LaBar, J. A., and Strevett, K. A., 2011. “Passive Treatment Opportunities in a Drastically Disturbed Watershed: Reversing the Irreversible?”, Paper presented at the 2011 National Meeting of the American society of Mining and Reclamation, Bismarck, ND, Reclamation: Sciences Leading to Success, June 11 – 16, 2011. R.I. Barnhisel (Ed.) Published by ASMR, 3134 Montavesta Rd, Lexington, KY  40502

Nairn, R.W., J.A. LaBar, K.A. Strevett, W.H. Strosnider, D. Morris, A.E. Garrido, C.A. Neely and K. Kauk. 2010b. Initial evaluation of a large multi-cell passive treatment system for net-alkaline ferruginous lead-zinc mine waters. Proceedings of the 27th National Conference of the American Society of Mining and Reclamation, Pittsburgh, PA, pp. 635-649.

Nairn, R. W., T. Beisel, R. C. Thomas, J. A. LaBar, K. A. Strevett, D. Fuller, W. Strosnider, W. J. Andrews, J. Bays, and R. C. Knox. 2009. “Challenges in Design and Construction of a Large Multi-Cell Passive Treatment System for Ferruginous Lead-Zinc Mine Waters,” pp. 871–92 in Proceedings, Joint Conference of American Society of Mining and Reclamation 26th Annual Conference and 11th Billings Land Reclamation Symposium, Billings, Mont.

U. S. Environmental Protection Agency (USEPA). 2010. Fourth Five-Year Review Report for the Tar Creek Superfund Site, Ottawa County, OK. 246 pp.

USEPA. 2011. “Full-Scale PTS Uses Six-Step Process to Treat Polluted Mine Discharge”, Technology News and Trends, Issue 53, May 2011.

 

 

 

 

 

Publication Date: November 2013

Permission is granted to refer to or quote from this publication with the customary acknowledgment of the source (see suggested citation and disclaimer).

 

This web site is owned by ITRC.

1250 H Street, NW • Suite 850 • Washington, DC 20005

(202) 266-4933 • Email: [email protected]

Terms of Service, Privacy Policy, and Usage Policy

 

ITRC is sponsored by the Environmental Council of the States.