Prepared in cooperation with the New Mexico Environment Department

Scientific Investigations Report 2004-5063

Questa Baseline and Pre-Mining Ground-Water Quality Investigation 4. Historical Surface-Water Quality for the Red River Valley, New Mexico, 1965 to 2001

By Ann S. Maest¹, D. Kirk Nordstrom², and Sara H. LoVetere²

The PDF for the report is 5,843 kb

Table of Contents

Abstract

Introduction

Methods for Compiling and Reviewing Historical Surface-Water Quality Data

Historical Surface-Water Quality

Summary

References

Figures

Figure 1. Study area and surface-water sampling locations.

Figure 2. Average daily discharge and timing of surface-water samples colle...

Figure 3. Average daily discharge and timing of surface-water samples colle...

Figure 4. Sulfate concentrations and average daily discharge in the Red Riv...

Figure 5. Sulfate concentrations and average daily discharge in the Red Riv...

Figure 6. Sulfate concentrations in the Red River at the Questa Ranger Stat...

Figure 7. Sulfate concentrations in the Red River upstream and downstream f...

Figure 8. Sulfate loads in the Red River at the Questa Ranger Station gage,...

Figure 9. Sulfate loads in the Red River upstream and downstream from the m...

Figure 10. Specific conductance in relationship to sulfate concentrations i...

Figure 11. Sulfate concentrations in relationship to total dissolved solids...

Figure 12. Sulfate concentrations in relationship to total dissolved solids...

Figure 13. Calcium concentrations in relationship to sulfate concentrations...

Figure 14. Calcium to sulfate molar ratio over time in the Red River at the...

Figure 15. Alkalinity values in the Red River upstream and downstream from ...

Figure 16. Zinc concentrations over time in the Red River upstream and down...

Figure 17. Specific conductance, pH, and average daily discharge in the Red...

Figure 18. Specific conductance and average daily discharge in the Red Rive...

Figure 19. Total and dissolved zinc concentrations and average daily discha...

Figure 20. Dissolved and total zinc concentrations and average daily discha...

Figure 21. pH, specific conductance, and average daily discharge in the Red...

Tables

Table 1. Red River monthly climate summary from January 1915 to December 20...

Table 2. Sampling events, dates, and analyses.

Table 3. Sample collection, preservation, and analysis methods, laboratory,...

Table 6. Definition of hydrologic conditions and summary of sampling events...

Table 4. Historical water-quality data at the Questa Ranger Station on the ...

Table 5. Water-quality data upstream from the Molycorp, Inc. molybdenum min...

¹Buka Environmental, Boulder, Colo.
²U.S. Geological Survey

Conversion Factors, Datum, and Abbreviations

Temperature in degrees Celsius (°C) may be converted to degrees Fahrenheit (°F) as follows:
°F=(1.8x°C)+32

Vertical coordinate information is referenced to the North American Vertical Datum of 1988 (NAVD 88). Elevation, as used in this report, refers to distance above the vertical datum. Specific conductance is given in microsiemens per centimeter at 25 degrees Celsius (µS/cm at 25°C).

Abbreviations used in this report:

Abbreviated water-quality units used in this retdort:

Abstract

Historical water-quality samples collected from the Red River over the past 35 years were compiled, reviewed for quality, and evaluated to determine influences on water quality over time. Hydrologic conditions in the Red River were found to have a major effect on water quality. The lowest sulfate concentrations were associated with the highest flow events, especially peak, rising limb, and falling limb conditions. The highest sulfate concentrations were associated with the early part of the rising limb of summer thunderstorm events and early snowmelt runoff, transient events that can be difficult to capture as part of planned sampling programs but were observed in some of the data. The first increase in flows in the spring, or during summer thunderstorm events, causes a flushing of sulfide oxidation products from scars and mine-disturbed areas to the Red River before being diluted by rising river waters.

A trend of increasing sulfate concentrations and loads over long time periods also was noted at the Questa Ranger Station gage on the Red River, possibly related to mining activities, because the same trend is not apparent for concentrations upstream. This trend was only apparent when the dynamic events of snowmelt and summer rainstorms were eliminated and only low-flow concentrations were considered. An increase in sulfate concentrations and loads over time was not seen at locations upstream from the Molycorp, Inc., molybdenum mine and downstream from scar areas. Sulfate concentrations and loads and zinc concentrations downstream from the mine were uniformly higher, and alkalinity values were consistently lower, than those upstream from the mine, suggesting that additional sources of sulfate, zinc, and acidity enter the river in the vicinity of the mine. During storm events, alkalinity values decreased both upstream and downstream of the mine, indicating that natural sources, most likely scar areas, can cause short-term changes in the buffering capacity of the Red River.

The major-element water chemistry of the Red River is controlled by dissolution of calcite and gypsum and the oxidation of pyrite, and the river is generally not well buffered with respect to pH. During higher-flow periods, Red River water was diluted by calcium-carbonate waters, most likely from unmineralized Red River tributaries and areas upstream from scars. The effect of pyrite oxidation on Red River water chemistry was more pronounced after the early 1980's. Elevated zinc concentrations were most apparent during summer thunderstorm and rising limb times, which also were associated with a decrease in alkalinity and an increase in sulfate concentrations and conductivity. The water-quality results demonstrate that it is critical to consider hydrologic conditions when interpreting water chemistry in naturally mineralized or mined drainages.

Introduction

Molycorp, Inc. operates a molybdenum mine west of the town of Red River, New Mexico on approximately 18 km² of land surrounded by the Carson National Forest. The Red River forms the southern boundary of the mine site and flows to the west where it joins the Rio Grande River (fig. 1). State and federal authorities have raised concerns about potential injury to the Red River fishery and aquatic life since open-pit mining began in 1965. Water quality of receiving streams can change over time because of climatic events and because mining activities evolve and change in type and extent over time. Information is needed to discern the potential effects of climate and seasonal events from mining or other anthropogenic activities. This paper puts the available data on Red River water quality into historical perspective so that trends over time and their causes can be better understood.

Figure 1. Study area and surface-water sampling locations.

Historical surface-water quality evaluation is part of the Questa baseline and pre-mining ground-water quality investigation, hereinafter called the Questa pre-mining study, conducted by the U.S. Geological Survey (USGS) in cooperation with the New Mexico Environment Department (NMED). The overall objective of the investigation is to infer the pre-mining ground-water quality at the Questa Molycorp, Inc. mine site. As part of this investigation, the USGS, in cooperation with NMED, has compiled and evaluated historical surface-water quality data for the Red River.

Although a considerable number of surface-water samples in the Red River Valley were collected from 1965 to 2001, the data were not compiled in a single document, did not undergo data quality evaluation, and have not been utilized to interpret water-rock interactions. The purpose of this report is to compile and evaluate historical surface-water quality data in the Red River Valley between the Questa Ranger Station and the town of Red River. "Historical" data are defined as all data existing from the earliest available record (1965) until November 2001. The purpose of this report also is to delineate any spatial and temporal trends in the data and to obtain some preliminary information on water-rock interactions or any other factors affecting temporal trends in Red River water quality.

Physical Description of Study Area

The study area is located in Taos County in north-central New Mexico in the Red River drainage basin, a tributary to the Rio Grande within the Carson National Forest. The area is a rugged and altered terrain with steep slopes and V-shaped valleys. The study reach is in the Red River Valley between the Questa Ranger Station (elevation 7,452 ft or 2,271 m) at the west end and the town of Red River (elevation 8,680 feet or 2,646 m) at the east end (fig. 1). The Molycorp, Inc. Questa molybdenum mine, referred to as the mine site, is located on the north side of State Highway No. 38 and the Red River, approximately 20 km east of the Ranger Station. The mine site is approximately 15 km² and encompasses three tributary valleys to the Red River: Capulin Canyon, Goat Hill Gulch and Sulphur Gulch, from west to east respectively.

Mining activities produced extensive underground workings and an open-pit of approximately 914 m in diameter, covering approximately 0.66 km², near or in Sulphur Gulch (URS, 2001). Waste-rock piles cover steep slopes on the north side of the Red River between Capulin Canyon and Spring Gulch (a tributary valley of Sulphur Gulch). Hydrothermally altered bedrock is found in Capulin, Goat Hill, Sulphur, Hansen, Straight, and Hottentot drainages (fig. 1). Weathering of extensively altered rock has resulted in steep, highly erosive, sparsely vegetated "scars" that are clearly visible from the ground and in aerial photographs.

Climate and Vegetation

The Red River Valley is located within a semi-arid desert that receives precipitation throughout the year and sustains moderate biodiversity. Between 1915 and 2002, the annual average temperature was 4°C, the annual average precipitation and snowfall were 52 cm and 370 cm, respectively, and the daily temperatures generally fluctuated by 18°C throughout the year (table 1).

Climate and vegetation vary greatly within short distances because of differences in topography, weather, sun exposure, elevation, and sediment composition. The land-surface elevation in the study area ranges from 2,271 m at the Ranger Station to 2,704 m at the Zwergle gage upstream from the town of Red River. Orographic effects of mountainous topography lead to precipitation on the windward slopes and localized storms within tributary valleys. Major precipitation events include summer thunderstorms and winter-spring snowstorms. Thunderstorms are responsible for mass wasting in hydrothermally altered areas, producing debris flows that potentially impact vegetation, alluvial aquifers, and the Red River (K. Vincent, USGS, written commun., 2003). Winter snowpack contributes to ground water recharge through snowmelt infiltration and runoff.

Prevalent vegetation in the Red River Valley is controlled by elevation and includes four zones: pinyon-juniper woodland (from 1,829 to 2,286 m), mixed-conifer woodland (from 2,286 to 2,743 m), spruce-fir woodland (from 2,743 to 3,658 m), and alpine tundra (above 3,658 m) (Knight, 1990). Perennial grasses, shrubs, flowering vegetation, willows, and cottonwoods occur near the banks of the Red River. Extending from the river are widely spaced pinyon pines (Pinus edulis) and junipers (Juniperus monosperma). Gains in altitude give rise to an abundance of ponderosa pines (Pinus ponderosa) and limber pines (Pinus flexilis), while douglas-fir (Pseudotsuga menziesii) and white fir (Abies concolor) can be found at higher elevations (Briggs and others, 2003). This typical montane community, while diverse, is dominated by ponderosa pines.

Geology

Ground water passes through, and geochemically interacts with, the various types of earth materials discussed in this section, including fractured bedrock, soil and alluvium, and waste rock. This section summarizes the work of Schilling (1956), Rehrig (1969), Lipman (1981), and Meyer and Leonardson (1990, 1997), in addition to observations made by the USGS scientists working at the site.

The Taos Range of the Sangre de Cristo Mountains is composed of Precambrian metamorphic assemblages and granitic intrusives overlain by Tertiary volcanics. Late Oligocene to early Miocene granitic plutons and associated hydrothermal alteration were the source of molybdenite and other sulfide mineralization.

The Red River Valley is located along the southern edge of the Questa volcanic caldera and contains complex structural features and extensive hydrothermal alteration. In the Red River Valley, most of the visible rocks are Tertiary volcanics with smaller areas of Precambrian metamorphics and granitic stocks. The volcanics are primarily intermediate to felsic in composition (andesites to rhyolites); granites and porphyries have intruded the volcanics and are the apparent source of hydrothermal fluids and molybdenite mineralization. The hydrothermally altered volcanics often contain pyrite mineralization (generally 1-3 percent).

The mineral deposits in the Red River Valley are considered Climax-type deposits, which are associated with silica- and fluorine-rich rhyolite porphyry and granitic intrusives. Climax-type hydrothermal alteration produces zones of alteration assemblages, with a central zone of fluorine-rich potassic alteration, a quartz-sericite-pyrite zone (QSP), often with a carbonate-fluorite veinlet overprint, and a propylitic zone. Chlorite, epidote, albite, and calcite typically are found in the propylitic assemblages. In the potassic zone, rocks are altered to a mixture of biotite, potassium feldspar, quartz, fluorite, and molybdenite.

In the Red River Valley, ore deposits contain quartz, molybdenite, pyrite, fluorite, calcite, manganiferous calcite, dolomite, and rhodochrosite. Lesser amounts of galena, sphalerite, chalcopyrite, magnetite, and hematite also are present. The hydrothermal alteration related to mineralization overprints an older, regional propylitic alteration. In these areas, rocks can contain a mixture of quartz, pyrite, and illite clays replacing feldspars, chlorite, and epidote. Calcite is an important mineral in the Red River Valley because its dissolution can effectively neutralize the acid inflows so that pH values in the Red River tend to be neutral to alkaline (pH 7-8). Gypsum is commonly found throughout the Red River Valley and forms as a secondary product of acid-sulfate weathering when the products of pyrite oxidation react with calcite. Abundant minerals in waste rock produced by mining activities include chlorite, gypsum, illite, illite-smectite, jarosite, kaolinite, and muscovite (Gale and Thompson, 2001).

Scar-area bedrock outcrops are composed of andesite volcanic and volcaniclastic rocks, rhyolitic tuff, quartz latite, and rhyolite porphyry. The dominant alteration type in all scars is QSP; carbonates also are found in all scar areas. Most of the andesite and quartz latite has been propylitically altered and contains plagioclase feldspar and chlorite, with less QSP alteration. Rhyolite porphyry and tuff do not seem to have been substantially affected by propylitization. Depending on location within the weathering profile, altered rocks contain variable amounts of quartz, illite, chlorite, and plagioclase feldspar, with smaller amounts of pyrite, gypsum, rutile, jarosite, and goethite (Livo and Clark, 2002; Ludington and others, in press).

Ground Water

Ground water is influenced by the climate, geology, and anthropogenic activities in the Red River Valley. There are three major types of water-bearing units present: fractured bedrock, debris-fan deposits, and Red River alluvium. Bedrock constitutes the largest aquifer in the study area in terms of rock mass, but probably contains only small amounts of ground water because of low porosity and hydraulic conductivity that are controlled by fractures. Debris fans are composed of sediments shed from their watersheds, which are tributary to the Red River. Tributary watersheds with scars contain large, active debris fans with both coarse and fine-grained sediments and correspondingly variable porosity and infiltration rates. The chemistry of these sediments likely reflects the chemistry of their rapidly eroding and altered erosion scars. Sediments deposited by the Red River (alluvium), in contrast, generally consist of well washed sandy-gravel and are composed of a mix of the lithologies found in the entire Red River basin. The largest debris fans caused the Red River to aggrade behind the fans during the Quaternary. Thus water flowing in the shallow alluvial aquifers likely passes alternately through Red River alluvium and debris fan alluvium. Both the Red River alluvium and debris fan alluvium are less than several hundred meters wide and less than 60 m thick, but together they contain most of the ground water in the valley (K. Vincent, USGS, written commun., 2003).

While chemical analyses of ground water were not obtained prior to mining in the Red River Valley, a substantial amount of historical data are currently available (LoVetere and others, 2003). Alluvial, debris-fan, and shallow bedrock ground water is dominantly calcium-sulfate type water. Most wells developed in the Red River Valley were installed to monitor water quality down gradient from mine waste deposits (waste rock dumps and tailings piles) and/or scar areas.

Surface Water

The Red River originates at an elevation of approximately 3,658 m near Wheeler Peak and flows north and then westward for roughly 55 km to its confluence with the Rio Grande River at an altitude of 2,012 m. The drainage area upstream from the Questa Ranger Station gage is 293 km². Peak streamflow usually occurs from late May to mid-June, with snowmelt-related flows beginning in late March and increasing through mid-April. Summer thunderstorms are prevalent from July through September. The mean annual discharge of the Red River at the Questa Ranger Station gage has ranged from 12.8 to 103 ft³/s between 1930 and 2001, while the average daily discharge ranged from 2.5 to 557 ft³/s, with an average of 42.9 ft³/s, between 1965 and 2001 (U.S. Geological Survey, 2003a).

The main drainages in the vicinity of the mine site are Capulin Canyon, Goat Hill Gulch, and Sulphur Gulch on the north side of the Red River (fig. 1). Upstream from the mine site, Little Hansen, Hansen, Straight, and Hottentot Creeks drain scar areas, while Mallette and Bitter Creeks drain non-scar areas on the north side of the Red River. Bear Canyon and Columbine, Pioneer, and Placer Creeks drain largely unmineralized land on the south side of the river. Downstream from the mine site the Red River joins with Cabresto Creek, entering from the north side of the Red River, before it discharges to the Rio Grande.

Springs and shallow alluvial ground water discharge to the Red River rendering it a gaining stream over much of its length (Smolka and Tague, 1989). Between the town of Red River and the Questa Ranger Station gage, there are roughly 25 ephemeral seeps and springs along the banks of the Red River and approximately 20 intermittent seeps and springs in tributary drainages on the north side of the river (South Pass Resource, Inc., 1995; Steffen, Robertson, and Kirsten, 1995; Robertson GeoConsultants, Inc., 2001). The majority of the seeps and springs are acidic (pH 2-4) with high specific conductance, total dissolved solids (TDS), and metal concentrations. Springs downgradient from scar and mined areas on the north side of the Red River often have a milky aluminum hydroxide precipitate that affects the color and turbidity of the Red River (Vail Engineering, Inc. 1989) (see cover photo).

Mining History

The history of the Molycorp, Inc. mine site and other related information is available from Molycorp, Inc. (www.molycorp.com), NMED, and the U.S. Environmental Protection Agency (USEPA; www.epa.gov/superfund/sites/npl/nar1599.htm). The following is a summary of that information and other data that pertain to the study of surface water in the Red River Valley.

A pair of prospectors first discovered molybdenite in Sulphur Gulch in 1914. Small-scale underground mining took place between 1919 and 1958. In 1954 there were over 56 km of underground mine workings. Molycorp, Inc. began removing the rock overburden at Sulphur Gulch in 1964, and the first molybdenite ore was extracted from the open pit in 1965. Tailings were transported by pipeline from the mine to the tailings facility, located near the town of Questa, New Mexico, just upstream of the confluence of the Red River and the Rio Grande. More than 200 spills from the tailings pipeline occurred between 1964 and 1982 (URS, 2002). Water used in the mill operation was produced from the Red River and the Red River alluvial aquifer.

In 1983, Molycorp, Inc. ceased open-pit mining and initiated a new phase of underground mining in Goat Hill Gulch. The switch effectively stopped the dumping of waste rock in Capulin Canyon, along the north slope of the Red River, and in Goat Hill, Sulphur, and Spring Gulches, and increased the volume of tailings transported by pipeline to the impoundment in Questa. It is estimated that roughly 328 million tons of waste rock (295 million metric tons) were deposited near the open pit between 1964 and 1983 (Robertson GeoConsultants, Inc., 2000c; Slifer, 1996).

Low market values for molybdenum caused the mine to shut down between 1986 and 1989 and again in 1992. From 1992 to 1995, while the underground mine was shut down, pumping of ground water from the underground mine stopped, and the workings were allowed to flood. Point-source discharges from the tailings facility at Questa were monitored through a discharge permit issued by USEPA, and two discharge points for stormwater runoff from the mine site were added when the permit was renewed in 1993 (Slifer, 1996). A discharge permit for the mine site was issued in November 2000. The permit required quarterly sampling, which was initiated in June 2001. In 2002, Molycorp, Inc. and USEPA entered into an Administrative Order on Consent (AOC) to define the nature and extent of contamination from the mine site. Currently, Molycorp, Inc. is required to collect soil, rock, water, animal, and plant samples for the Remedial Investigation as defined in the AOC.

Acknowledgements

The authors wish to acknowledge Cheryl Naus for her assistance in preparing the map of the study area. Katie Walton-Day and Laura Bexfield of the USGS provided helpful technical reviews on a draft of the manuscript. Molycorp, Inc. provided documents and data on historical water quality. This report also would not have been possible without the support of the National Research Program.

Table 1. Red River monthly climate summary from January 1915 to December 2002.

Data source: Western Regional Climate Center, 2003.

Methods for Compiling and Reviewing Historical Surface-Water Quality Data

For the purposes of this report, historical water-quality data are defined as those from samples collected on or before the November 12, 2001 sampling of Molycorp, Inc. The location with the longest period of record for water-quality data is the Questa Ranger Station gage on the Red River, where there has been continuous streamflow-gaging by the USGS since October 1, 1924. The USGS streamflow-gaging station (site number 08265000) is located downstream from the Molycorp, Inc. Mine in Questa, New Mexico (fig. 1). Because of the long-term water-quality and flow record, the historical surface-water data compilation has focused on the Questa Ranger Station gage. In addition, data from locations upstream from the mine site and downstream from natural scar areas are also included in the compilation.

The historical surface-water quality data for the Red River are contained in Federal and State agency reports, consultant reports, and on computer files or databases that are not associated with any report. One of the main purposes of compiling and evaluating the historical surface-water quality information is to gather all the available data in one document and to describe its quality to the extent possible. Vail Engineering, Inc., URS, and others, under contract to Molycorp, Inc., have compiled historical surface-water quality data in the past. However, not all parameters were reported, a comprehensive evaluation of the methods used to collect and analyze the samples was not conducted, geochemical interpretations were not included, and/or the data were not evaluated with respect to water flow dynamics. The more complete review and evaluation presented herein will aid in data comparison and analysis. For this report, original sources were reviewed for information on collection and analysis methods, and the data were gathered from these original sources. Once compiled and evaluated for quality, the historical surface-water quality data are useful for evaluating chemical, temporal, and spatial trends and for establishing an historical framework for comparing any recent surface-water quality data collected.

Water-quality sampling events, dates, parameters, and locations are presented in table 2. The U.S. Department of Health, Education, and Welfare (USHEW), the USEPA, and Molycorp, Inc. conducted water-quality sampling on the Red River, its tributaries, and the Rio Grande in the mid-1960's to the mid-1970's. NMED and Vail Engineering, Inc., under contract to Molycorp, Inc., collected the majority of the water-quality samples on the Red River and its tributaries in the 1980's and 1990's. The source of the data, and, if available, the reference for the report containing the data are listed in table 2. Table 2 shows that when metals were determined on Red River and tributary samples, the most common analyte suite was aluminum, iron, manganese, and zinc. For a number of sampling events, sulfate was the only anion determined. Flow was usually only available for samples collected at the Questa Ranger Station gage, although the USGS streamflow-gaging station located upstream from the town of Red River (Zwergle gage, site number 08264500, see fig. 1) operated from May 1, 1963 to December 31, 1973. Water-quality samples were collected from the Zwergle gage location on the Red River in 1997 and 1998 (Allen and others, 1999).

Sample collection, preservation and analysis methods, laboratories, and quality assurance/quality control (QA/QC) plans used for surface-water quality sampling are presented in table 2. When a document containing surface-water quality information was associated with the analytical data, it was reviewed for sample collection, preservation, analytical methods, and other quality-control issues. The references for the documents are listed in both tables 2 and 3, and full citations can be found in the reference section at the end of this report. For several documents and sampling events, not all QA/QC information was available; for these instances, the entries are indicated by two short dashes (--) in table 3.

In table 3, the collection methods column includes frequency of sampling for a given sampling event, whether samples were composited, whether replicate field samples were collected, where in the stream the samples were collected (for example, mid-depth), whether samples were collected by pump or grab, and whether the sampling was synoptic. Most of the samples collected before the mid-1980's were unfiltered (except for Faith, 1974). If samples collected before the mid-1980's were filtered, the filtration was for suspended sediment and metals, most notably suspended aluminum and iron. When filtering was conducted, 0.45-µm filters were used; no historical water-quality samples were filtered through 0.1-µm filters. Few samples were analyzed for both total and dissolved constituents, although limited samples collected after 1984 were analyzed for both. Similarly, preservation of samples was not common before the mid-1980's (again, Faith, 1974 is an exception to this). Unfortunately, Faith did not collect any samples at the Questa Ranger Station gage.

Table 3 also lists analytical methods used to determine metals and other parameters. Samples collected in the 1960's and early 1970's used colorimetric or flame atomic-absorption spectrometry (AAS) for determination of metals, and consequently, high detection limits were common. However, Faith, (1974) used graphite-furnace AAS for the determination of trace metals, iron and manganese. Standard Methods (American Public Health Association, 1981) and USEPA methods (U.S. Environmental Protection Agency, 1983) were used by New Mexico Department of Game and Fish (NMDGF) (Jacobi and Smolka, 1984; Smolka and Tague, 1987 and 1989; NMDGF, 1992 and 1993) since the mid-1980's. Allen and others (1999) of the University of New Mexico, was the only report in which inductively-coupled argon-plasma—mass spectrometry (ICP-MS) was used for determination of trace metals; other sampling efforts generally used inductively-coupled argon-plasma—atomic emission spectrometry (ICP-AES) for determination of trace and minor metals.

Chemical analyses of surface-water samples were conducted at commercial, government, and university laboratories. Very few of the sampling efforts included sampling plans or QA/QC plans, such as quality assurance project plans (QAPPs). However, sampling performed by New Mexico State agencies usually did include a sampling or QA/QC plan.

Table 2. Sampling events, dates, and analyses.

[Alk, alkalinity; App, appendix; assoc, associated; BLM, Bureau of Land Management; CG, campground; up, upstream; CN, cyanide; DIM, dissolved inorganic matter; diss, dissolved; DO, dissolved oxygen; DOC, dissolved organic carbon; DOM, dissolved organic matter; hard, hardness; down, downstream; ENSR, ENSR Consulting and Engineering; Inc.; FWPCA, Federal Water Pollution Control Authority; NMEID, New Mexico Environmental Improvement Division; NMED, New Mexico Environment Department; NM BMMR, New Mexico Bureau of Mines and Mineral Resources; NMDGF, New Mexico Department of Game and Fish; params, parameters; POTW, publicly-owned treatment works; RGC, Robertson GeoConsultants; SC, specific conductance; SPRI, South Pass Resources, Inc.; SRK, Steffen, Robertson, and Kirsten; T, temperature; TDS, total dissolved solids; TSS, total suspended sediment; trib, tributary; turb, turbidity; UNM, University of New Mexico; USEPA, U.S. Environmental Protection Agency; USGS, U.S. Geological Survey; WC, Woodward-Clyde Consultants; WWTF, waste water treatment facility; --, no data]

Table 3. Sample collection, preservation, and analysis methods, laboratory, and quality control plans.

[AAS, atomic absorption spectrometry; alk, alkalinity; APDC, ammonium pyrrolidine dithiocarbamate; APHA, American Public Health Association; CLP, contract laboratory procedure; EDTA, ethylenediaminetetraacetic acid; EMSL-LV, Environmental Monitoring Systems Laboratory, Las Vegas; ENSR, ENSR Consulting and Engineering; EP, extraction procedure; FWPCA, Federal Water Pollution Control Authority; GFAAS, graphite furnace atomic absorption spectrometry; HDPE, high-density polyethylene; hard, hardness; ICP-AES, inductively-coupled argon-plasma—atomic emission spectrometry; ICP-MS, inductively-coupled argon-plasma—mass spectrometry; Lab., laboratory; NMDGF, New Mexico Department of Game and Fish; NMEID, New Mexico Environmental Improvement Division; µg/L, micrograms per liter; NMED, New Mexico Environment Department; µm, micrometer; QAPP, quality assurance project plan; QA/QC, quality assurance/quality control; refrig, refrigerated; RGC, Robertson Geoconsultants; TAL, target analyte list; TDS, total dissolved solids; TSS, total suspended solids; UCLA, University of California at Los Angeles; UNM, University of New Mexico; USGS, U.S. Geological Survey; USEPA, U.S. Environmental Protection Agency; --, no information]

Historical Surface-Water Quality

Table 4 contains the water-quality data for the Red River at the Questa Ranger Station gage from 1965 to 2001. Although all of the constituents for which determinations were attempted are included in table 2, constituents that were rarely determined or that had concentrations consistently near or below detection were excluded from table 4. A smaller set of water-quality data from stations upstream from the Questa Ranger Station gage but downstream from hydrothermal scar areas also is included in table 5. The source of the water-quality data is indicated in table 4 and corresponds to the 'Reference' columns in tables 2 and 3. The source for all water-quality data in table 5 is Vail Engineering, Inc. (written commun., 2002).

Hydrologic Conditions for Historical Water-Quality Sampling Events

In mineralized or mined areas, it is common to see relationships between concentrations and flow, or between concentrations and changes in hydrologic conditions determined from the hydrograph record. For example, concentrations of metals in streams draining mined areas can increase during the rising limb of the hydrograph before dilution occurs (Jambor and others, 2000). Stored acid and metals from the oxidation of pyrite and other sulfides (unsaturated-zone waters and efflorescent salts) are flushed from soils, mine wastes, and unmined but mineralized areas by melting snow or rain and washed into receiving streams. To determine if similar mechanisms were operating in the study area, hydrographs for all years containing sampling events were examined to find the hydrologic conditions during sampling. Average daily discharge values were taken from the U.S. Geological Survey web site (USGS, 2003a).

Criteria used to discriminate sampling events based on hydrologic conditions are defined in table 6 and used in tables 4 and 5. As shown in table 6, five hydrologic conditions were defined: rising limb, peak, falling limb, storm, and low flow. Rising limb was defined as starting when flows first increased as a result of snowmelt (usually in April) and ending when flows reached 50 percent of the peak flow for that water year. Peak flow was defined as starting and ending at 50 percent of peak flow on either side of the peak (usually late April/early May). Falling limb started after the peak when flows fell to 50 percent of peak values and ended when flows fell to 25 percent of peak values. Low flows started at 25 percent of peak flow after the peak and continued to the start of the rising limb the following year. Storm events are defined as increases in flow superimposed on the low flow portions of the hydrograph (summer thunderstorms). To determine percent of peak flow, average daily flows were compared to maximum daily discharge (peak flow). The relevant peak flow was defined as the previous peak. For example, if the sampling date was between October 1, 1999, and the start of the rising limb in 2000, the relevant peak flow for these dates would be the peak flow for 1999, not 2000.

Sampling dates at the Questa Ranger Station gage from water years 1966 to 1982 and 1983 to 2001 are shown on hydrographs in figures 2 and 3, respectively. In total, there were 190 separate sampling dates when water quality samples were collected on the Red River at the Questa Ranger Station gage. On a small number of additional dates, only flow was measured, but those events are not included in figures 2 and 3 or in table 4. Figure 2 shows the wide variability in peak discharge values, ranging from a low of 29 ft³/s in 1977 to a high of 557 ft³/s in 1979. The earliest sampling events were in November 1965 and November 1971, both during low flow. A number of samples collected from 1977 through 1979 were average monthly composites (Molycorp, Inc., 1979; marked with an 'A' in table 4), and some of the most consistent sampling coverage for different flow conditions was during this time. Most samples from water years 1983 to 2001 (fig. 3) were collected during low flow conditions, although there were five summer thunderstorm samples collected in 1988 and a number of non-low-flow samples collected in 1992, 1994, 1997, and 1998. Table 6 shows the years in which samples from different hydrologic conditions were collected. Relationships between hydrologic conditions (for example, low flow) and concentrations are discussed in the following sections.

Table 6. Definition of hydrologic conditions and summary of sampling events.

[NA, not applicable]

¹If date falls between October 1 and start of rising limb, peak flow used is from previous year.

Figure 2. Average daily discharge and timing of surface-water samples collected in the Red River at the Questa Range Station gage in water years 1966 to 1982.

Figure 3. Average daily discharge and timing of surface-water samples collected in the Red River at the Questa Range Station gage in water years 1983 to 2001.

Evaluation of Selected Water-Quality Data in Previous Reports

The sample collection, preservation, and analysis methods are presented in table 3. Analytical issues that may limit the usefulness of previous Red River water-quality studies are discussed in this section. In addition, the results of a limited number of studies are discussed to provide a context for the presentation of historical water-quality data.

Baseline surveys of Red River water quality were conducted by the Federal Water Pollution Control Administration (1966) and the USEPA (1971). The reports concluded that the water quality was good in 1966 and that the chemical quality and biological conditions remained very good in 1971, with the exception of the influences of tailings pipeline breaks (USEPA, 1971). In the early 1970's, the NMDGF, however, noted that fish were absent in the middle reach of the Red River where there had been 572 fish per mile in 1960 (Slifer, 1996).

In 1980, the USEPA performed a preliminary assessment of the Red River-Questa site (Melancon et al., 1982). The study found biologic toxicity, high concentrations of arsenic, and concentrations of cadmium and silver that exceeded water quality standards for the protection of aquatic life. Four surveys were performed between 1984 and 1988 by different bureaus within NMED that indicated progressive degradation in the Red River (Slifer, 1996; URS, 2001). The Bureau of Land Management (BLM) studied Red River water quality between 1978 and 1986 (Garn, 1985; USGS, 2003b), and documented a downstream increase in concentrations of metals and other constituents, at times exceeding water quality standards, and related the impacts to mining. Smolka and Tague (1987, 1989) conducted water studies that showed increased loading from storm events, but concentrations in the Red River did not exceed aquatic toxicity thresholds. However, biological indices were substantially reduced downstream from the Molycorp, Inc. Mine. Summaries of these and other water-quality investigations are discussed and reviewed in the Historical Surface-Water Quality section.

Arsenic, iron, lead, and zinc were determined using colorimetric methods in FWPCA (1966). The detection limits for lead (20 micrograms per liter (µg/L) and zinc (1,000 µg/L) in the FWPCA study are elevated compared to those for graphite furnace AAS, ICP-MS, or even flame AAS and preclude comparisons to criteria for the protection of aquatic life in the Red River. Similarly, detection limits for many of the elements determined using colorimetric methods in the Pennak (1972a) report prohibit comparisons to water-quality standards or criteria. For example, colorimetric detection limits were 1 milligram per liter (mg/L) for silver, 0.5 mg/L for arsenic, and 2 mg/L for beryllium. However, for analytes determined by flame AAS, detection limits were in the parts per billion range (for example, 20 µg/L for copper and 40 µg/L for iron).

Although the later Pennak (1976 to 1984) reports discussed water quality, no metals were determined, and the focus was predominantly on benthic aquatic life. The cause of a "yellowish-tan" precipitate was attributed to Molycorp, Inc. operations in May 1978 and to naturally-occurring streamside deposits upstream from the Eagle Rock campground in October 1978.

The USEPA report by Melancon and others (1982) concluded that concentrations of total arsenic, cadmium, and silver exceeded acute aquatic life criteria in the Red River. However, cadmium and silver detection limits were 7.5 and 12.0 µg/L, respectively, and the arsenic detection limit was 110 µg/L. Arsenic concentrations are higher than those reported in any other water-quality study of the Red River. For example, the highest arsenic concentration reported at the Questa Ranger Station gage (table 4) from 1966 to 2001 was 22 µg/L; arsenic concentrations reported by Melancon and others (1982) ranged from 633 µg/L upstream from the town of Red River to 1,400 µg/L approximately 200 m upstream from the State fish hatchery, which is downstream from the Questa Ranger Station. Arsenic was determined by ICP-AES, which is one of the least sensitive methods available for arsenic. The arsenic values reported in Melancon and others (1982) were not considered reliable because all other arsenic values measured in the Red River were lower. Furthermore, arsenic mineralization is rare in the Red River Valley (Schilling, 1956).

Only three of eight locations sampled by Melancon and others (1982) had detectable cadmium concentrations, and these ranged from 7.8 µg/L at one of the upstream control locations to 11.5 µg/L at the other upstream control location. These values are less than twice the detection limit of 7.5 µg/L and may not be reliable. Similarly, silver was detected at only two of eight locations; the values were 16.0 and 19.2 µg/L, which are only slightly higher than the detection limit of 12.0 µg/L. Silver also is highly insoluble (<1 µg/L) in most freshwaters. Consequently, conclusions by Melancon and others (1982) about exceedences of arsenic, cadmium, and silver aquatic criterion values are questionable.

The detection limits for lead (50 µg/L), copper (10 µg/L), and cadmium (5 µg/L) in ENSR Consulting and Engineering, Inc. (1988) were higher than many detected concentrations in other studies (especially for dissolved metal concentrations). Detection limits for zinc (10 µg/L) in the ENSR Consulting and Engineering, Inc. (1988) study were acceptable, because measured zinc concentrations in the Red River at the Questa Ranger Station gage and other locations were usually substantially higher than 10 µg/L.

The Allen and others (1999) report, prepared for the New Mexico Office of the Natural Resources Trustee, concluded that water quality, sediment chemistry, and aquatic life impacts in the Red River were related more to mining than natural sources at the Molycorp, Inc. mine. Allen and others (1999) compared metal concentrations in sediment from two lakes, Fawn Lakes and Eagle Rock Lake, dug as borrow pits for road construction. Fawn Lakes are located upstream from the mine site near the Elephant Rock Campground, and Eagle Rock Lake is located downstream from the mine site, approximately 700 m downstream from the Questa Ranger Station (see fig. 1). Metal enrichment appeared in the Eagle Rock Lake sediments but not in the Fawn Lakes sediments. Water-quality data were collected during 1997 and 1998 by Allen and others (1999) at five sites along the Red River from near the Questa gage to above the mine site. Their data showed increases in concentrations of sulfate, calcium, magnesium, zinc, nickel, and aluminum at the gage relative to upstream concentrations, and the concentrations increased through the reach adjacent to the mine. Alkalinity decreased from upstream to downstream locations. They concluded that the source of these elevated concentrations and decreased alkalinity values was mining activities, although they also noted that trace metal concentrations in waters upstream from the mine and adjacent to scar areas were elevated relative to those in waters upstream from all scar areas. They found that concentrations at any one site tended to increase during the dry season and decrease during the wet season.

Historical Water-Quality Trends over Time

Table 4. Historical water-quality data at the Questa Ranger Station on the Red River from 1965 to 2001.

[A, average (monthly composite); Alk, alkalinity; avg, average (monthly composite); Avg, average; BLM, Bureau of Land Management; _d, dissolved; DOC, dissolved organic carbon; FWPCA, Federal Water Pollution Control Authority; final, dissolved if available, total if dissolved not available, unknown if neither available; J, estimated value; JTU, Jackson turbidity units; Max, maximum; mg/L, milligrams per liter; µS/cm, microsiemens per centimeter; mm/dd/yy, month/day/year; NMDGF, New Mexico Department of Game and Fish; NMED, New Mexico Environment Department; NM ONRT, New Mexico Office of the Natural Resource Trustee; NTU, nephelometric turbidity units; RGC, Robertson GeoConsultants; _s, suspended; SC, specific conductance; SPRI, South Pass Resources, Inc.; SRK, Steffen, Robertson & Kirsten, Inc.; SU, standard units; Temp, temperature; _t, total; TDS, total dissolved solids; TSS, total suspended sediment; USEPA, U.S. Environmental Protection Agency; USGS, U.S. Geological Survey; <, less than; _?, not known if total or dissolved; --, no data]