Environ. Sci. Technol. 2001, 35, 1299-1302
Application of Stable Isotopes To Identify Problems in Large-Scale Water Transfer in Grand Canyon National Park NEIL L. INGRAHAM* Earth Science Division, Directorate for Geosciences, National Science Foundation, 4201 Wilson Boulevard, Arlington, Virginia 22230 KIM ZUKOSKY Las Vegas Springs Preserve Project, Las Vegas Valley Water District, 1001 South Valley View Boulevard, Las Vegas, Nevada 89153 DAVID K. KREAMER Water Resources Management Program and Department of Geoscience, University of NevadasLas Vegas, 4505 Maryland Parkway, Las Vegas, Nevada 89154-4029
Waters on, and below, the South Rim of the Grand Canyon were sampled for stable isotopic analysis to determine the hydrologic effects of the transcanyon pipeline. The transcanyon pipeline transports North Rim water discharging at Roaring Spring across the Grand Canyon to South Rim. Ultimately this water is discharged through the sewage treatment plant at the Clearwell Overflow wash on the surface expression of the Bright Angel Fault. The North Rim water is some 8 per mil more depleted in δD than most of the water issuing from springs on the South Rim except for that from Indian Garden Spring which lies below the Clearwell Overflow wash. Such a composition of Indian Garden Spring must come from discharged wastewater on the rim, percolating downward approximately 1000 m vertically through the Bright Angel Fault. The difference in stable isotopic composition of the North Rim water renders it not only traceable in Indian Garden Spring water, but the proportions may be determined as well which result in projecting an admixture of up to half the total discharge. Curiously however, Indian Garden Spring contains no appreciable amounts of the anions associated with wastewater. More recently, a leak in the transcanyon pipeline was discovered above Indian Garden Spring, suggesting that a portion of that spring’s discharge may have its origin in water directly from the pipeline. Nevertheless, these data provide information relevant to the National Park Service policy of precluding anthropomorphic forces impacting national parks. In addition, the stable isotopic ratios of park water provide a mechanism to assess the potential for future degradation, as well as the origin of any future degradation, of the water quality of Indian Garden Spring.
Introduction Stable isotopes of oxygen and hydrogen found in water can be used as conservative tracers to identify the original source 10.1021/es0015186 CCC: $20.00 Published on Web 02/20/2001
2001 American Chemical Society
of groundwater as well as define groundwater flow characteristics. This ability lies in the fact that the natural ratios of the stable isotopes observed in water (HDO/H2O, H218O/ H216O) are generally only altered during physical changes in the phase of the water (e.g., evaporation or condensation) that primarily occur in the atmosphere. The natural process of redistribution of the stable isotopes of water during a change of state results in the isotopic separation between the phases such that the heavier molecules (HDO, H218O) are enriched in the phase of lower energy (solid rather than liquid and liquid rather than vapor), while the lighter molecule (H216O) is concentrated in the remaining phase. In general, the systematics of stable hydrogen and oxygen isotopic fractionation are similar, thus their behavior in the hydrologic cycle is also similar. This similarity gives rise to a covariance between the stable hydrogen and oxygen isotopic concentrations found in most meteoric water which was observed first by Friedman (1) and later identified by the relationship of δD ) 8δ18O + 10 (2) and termed the meteoric water line (MWL). Once recharged to the groundwater system, alteration of the isotopic ratios can only occur under very unique circumstances. Thus the atmospheric history of the groundwater is locked in the isotopic ratios allowing determination of atmospheric and surface origins of the water. There are two major factors that control the degree of separation of the stable isotope during condensation and precipitation at any location: temperature of condensation and the composition of the condensing vapor. Thus, geographic variations are observed in meteoric water (3, 4) including the observation that precipitation is more depleted (containing less of the heavier molecules) at higher elevations. The depletion is the result of both controlling factors; precipitation at the higher elevations is the result of increased rain out due to continuous cooling of the air mass exacerbated by colder temperatures. Variations in the stable isotopic compositions of precipitation due to elevational differences have been reported to be approximately 40 per mil in δD per 1000 m (5) on the west slope of the Sierra Nevada. Others have reported variations of 3.2 per mil depletion in δ18O per 1000 m increase in elevation for Switzerland (6).
Site Description and Geologic Setting The Grand Canyon National Park is in the northwest corner of Arizona. The Grand Canyon is on average between 13 and 16 km wide and has a maximum depth of about 1.8 km. The canyon divides the Colorado Plateau into the Kaibab, Kanab, Unikaret, and Shivwits Plateaus on the north, and the Coconino and Hualapai Plateaus on the south (7). The North Rim is over 300 m higher in elevation than the South Rim. As a result, the North Rim receives more precipitation at about 50 cm/yr compared to the 40 cm/yr observed for the South Rim, albeit the rainfall is highly variable in both time and space. Soils and alluvial materials are thin on the plateaus, and thus runoff and infiltration is generally rapid (8). This study was performed in the central region of the canyon near the Grand Canyon Village which is bounded by the Kaibab Plateau (North Rim) and the Coconino Plateau (South Rim) as shown in Figure 1. The regional dip of the Coconino Plateau is toward the southwest, away from the canyon’s South Rim. Therefore, the general groundwater flow direction is generally away from the canyon rim, and only secondary porosities in the form of faults and fractures, or local folding perturbing the regional dip, could direct water toward the canyon from the south. The Bright Angel Fault * Corresponding author e-mail:
[email protected]. VOL. 35, NO. 7, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1299
FIGURE 1. Location map of the sample sites on the South Rim of the Grand Canyon and the structural control of the springs. is one of the largest regional faults. This fault trends generally northeast to southwest from the Coconino Plateau, across the canyon and past the North Rim. The eastern block is down thrown with a displacement of some 60 m near the south rim (9). The Bright Angel Fault is clearly a controlling conduit for groundwater discharge as several springs are observed along its trend including Roaring Spring on the North Rim and Indian Garden Spring on the South Rim (10, 9). Smaller faults nearby such as Hermit Fault that trends nearly subparallel to the Bright Angel Fault to the west also behave as conduits for groundwater flow. Hermit Fault appears to control discharge at Hermit, Hawaii, Dripping, and Santa Maria Springs. The water resources of the South Rim are meager. The area supports only intermittent streams and has a very deep groundwater system. The groundwater table varies, but is approximately 2040 m above sea level at the South Rim, or approximately 650 m below land surface. Available potable water has been a problem since the creation of the park. Originally the source of the water was shallow local wells or it was imported in on rail cars. In the early 1920s a pipeline from Indian Garden Spring was constructed; however, it became inadequate by the 1950s. In 1970 a pipeline from Roaring Springs on the North Rim of the canyon was finished which supplies North Rim water to the Grand Canyon Village on the South Rim. The transcanyon pipeline delivers from 1.44 to 1.59 m3/min of Roaring Spring water to the South Rim. The ultimate fate of this water is the Grand Canyon sewer treatment plant that discharges 458 000 m3/yr into 1300
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 7, 2001
the Clearwell Overflow approximately 1.5 km due south of the rim. Clearwell Overflow in turn discharges into a wash, which is the surface expression of the Bright Angel Fault. The National Park Service has been entrusted with the responsibility for protecting the spring flow within the canyon and for the ecosystems those spring support. Any potentially adverse effects of spring water quantity and quality changes are therefore of great concern to park managers and those in the surrounding Kaibab National Forest (11).
Procedures Samples were collected at Monument, Hermit, Hawaii, Santa Maria, Dripping, Horn, and Indian Garden Springs. Imported North Rim water was collected from the transcanyon pipeline at the Indian Garden pump house, and discharge from the Clearwell Overflow was collected where it receives discharge from the Grand Canyon sewer treatment plant. The sample collection sites, their discharge, structural control, and elevation, are listed in Table 1. All water samples for stable isotope analysis were collected in 125 mL glass bottles with polyethylene sealed caps; anions were collected in chemically resistant 250 mL polyethylene screw-cap storage bottles and refrigerated as soon as possible after collection.
Sample Analysis The oxygen isotope ratios of the collected water were determined by heating 10 µL of water with 100 mg of guanidine hydrochloride for 8-10 h at 260 °C to produce
TABLE 1. Discharge of the Springs Sampled for This Study spring
discharge, L/m
Roaring
2650
Indian Garden Horn Dripping Monument Hermit Hawaii Santa Maria Clearwell Overflow
1150 4 4 4 4 4 2 850
transcanyon pipe
1500
structural control (fault) Bright Angel (North Rim) Bright Angel Bright Angel Hermit Hermit Hermit Hermit Bright Angel
elevation, m
TABLE 2. Stable Isotopic Composition of the Samples Collected for This Study and the Date of Collection source Roaring Spring (North Rim)
1150 1100 1700 1000 1310 1100 1560 1980
ammonia and carbon dioxide (12). When cooled, the gases combine to produce solid ammonium carbamate. CO2 is then produced by heating the ammonium carbamate with 0.5 mL of 100% H3PO4 at 80 °C for 1 h. The CO2 is then purified and directly introduced into a Finnigan Mat Delta E mass spectrometer. The hydrogen isotope ratios were determined by the quantitative conversion of a 10 µL aliquot of the water sample to hydrogen gas using zinc as a reducing agent (13). The water sample is introduced into a glass capsule containing approximately 50 g of zinc shot and then is baked at 435 °C for about 1 h. The glass capsule is then cracked, and the resultant H2 gas is released directly into a Nuclide 3-60 double collector mass spectrometer. The reproducibility of the δD values is reported as 1 per mil while that of the δ18O values is 0.2 per mil. All data are reported in the standard δ-notation as a per mil variation from the SMOW standard where
Indian Garden Spring
Horn Spring Dripping Spring
Monument Spring
Hermit Spring
Hawaii Spring Santa Maria Spring
18
δ(D, O) per mil ) {R(sample) - R(SMOW)/R(SMOW)} × 1000 and where R ) D/H or
Clearwell Overflow 18O/16O.
Results The stable isotopic composition of all samples collected for this study are listed in Table 2 and are shown along with the meteoric water line (MWL) in Figure 2. The stable isotopic composition of imported North Rim water as collected at the Indian Garden pump house was measured at -13.5 per mil in δ18O and -96 per mil in δD. Most sampled springs (Monument, Hermit, Hawaii, Dripping, and Santa Maria Springs) on the South Rim have a more enriched isotopic value than that for the North Rim water at about -11.8 per mil in δ18O and -89 per mil in δD. The compositions of these springs are remarkably similar to each other and consistent through time. Horn and Dripping Springs are a little more depleted at about -90 per mil in δD, while Indian Garden Spring has the most depleted stable isotopic values of the sampled South Rim Springs, yet still temporally consistent at -12.5 per mil in δ18O and -92.5 in δD. The stable isotopic composition of the Clearwell Overflow tends to fluctuate along the MWL from a δD of about -92 to -96 per mil.
Discussion As predicted based on elevational differences, the stable isotopic composition of the water issuing from Roaring Spring on the North Rim, as collected at the Indian Garden pump house, is almost 8 per mil more depleted in δD than most of the water issuing from springs on the South Rim. The isotopic composition of this water may have only slightly been altered by its use at the South Rim Village, thus the water would retain its depleted nature indicating its North Rim origin as shown in Figure 2, when discharged along the
date collected
δD
δ18O
4-93 9-93 9-92 12-92 1-93 3-93 4-93 5-93 6-93 7-93 8-93 10-93 4-93 6-93 9-93 10-92 2-93 4-93 9-93 9-92 12-92 4-93 7-93 9-93 2-93 4-93 7-93 9-93 9-92 4-93 9-93 12-92 2-93 4-93 9-93 9-92 1-93 2-93 3-93 4-93 8-93 9-93
-95 -97 -93 -93 -92 -92 -92 -93 -93 -93 -93 -92 -90 -91 -90 -90 -90 -89 -90 -89 -88 -87 -89 -89 -87 -89 -90 -89 -88 -88 -90 -90 -88 -86 -90 -92 -96 -95 -94 -93 -92 -92
-13.5 -13.5 -12.5 -12.2 -12.4 -12.4 -12.3 -12.5 -12.6 -12.6 -12.4 -12.5 -11.6 -12.2 -11.8 -12.4 -12.5 -12.0 -12.0 -12.0 -11.7 -11.5 -12.0 -11.9 -11.9 -12.0 -11.8 -11.7 -12.0 -11.8 -12.1 -11.8 -11.9 -11.4 -12.2 -12.3 -13.2 -13.2 -13.0 -13.0 -12.8 -12.8
Clearwell Overflow. In addition, the mixing of this water with other waters of different compositions will result in a stable isotopic composition that can be calculated by mass balance of the different waters. It is this property which makes the stable isotopic analysis of waters appealing as a tool in determining possible different sources of waters. It is precisely this endeavor that supported an initial observation that water in the Clearwell Overflow might be a mixture of sewage treatment plant water (ultimately North Rim Water) and local precipitation (Figure 2). Mass balance calculations could also support a conclusion that the water issuing from Indian Garden Spring is a mixture of local South Rim water and water discharged along the Clearwell Overflow. The remarkable homogeneity and stability in the isotopic composition of the spring waters, as shown in Figure 2, renders contamination easily identifiable. However, paradoxically, Indian Garden Spring does not contain high concentrations of conservative (mobile) anions such as chloride and nitrate, as shown in Table 3, as would be expected if sewage treatment water were percolating from the Clearwell Overflow located on the Rim only 1000 m vertically higher. The stable isotopic composition of the Clearwell Overflow fluctuates in stable isotopic composition between being similar to North Rim Water and a more enriched composition. This fluctuation is likely the result of the contribution of VOL. 35, NO. 7, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1301
Indian Garden Springs, a distance of approximately 1000 vertical meters, of only 5 y or less. Shortly after this study concluded its measurements, a leak was discovered in a gasket in a buried section of the transcanyon pipeline directly upgradient from Indian Garden Spring. Although the size of this puncture was small, the outflowing water jetted a resultant hole 3 m deep due to high water pressure. This leak was never quantified in magnitude or duration, yet it provides a plausible explanation as to why Indian Garden Spring has a partial North Rim isotopic signature but no evidence of the anionic composition of Clearwell Overflow.
Impact
FIGURE 2. Stable isotopic composition of the samples collected for this study. Also shown is the Meteoric Water Line (δD ) 8δ18O + 10; 2) for reference.
TABLE 3. Concentrations of Conservative (Mobile) Anions in the Clearwell Overflow and Indian Garden Spring in mg/L chloride source
max
min
Clearwell Overflow 180.3 77.9 Indian Garden 11.0 15.0 Spring
Literature Cited
nitrate
average 125.6 12.8
max
min
167.8 50.3 3.9 2.8
average 109.4 3.3
discharged sewage treatment plant water into the wash in varying amounts. On the basis of only the depleted nature of the stable isotopic composition of the Indian Garden Spring water, without the benefit of the anionic composition of the water, the same scenario could be used to describe that spring water. That is, the structural control afforded by the Bright Angel Fault allows water from the Clearwell Overflow wash to infiltrate and percolate to eventual discharge at Indian Garden Spring. Following this assumption, isotopic mass balance calculations would project that up to half of the water discharging at Indian Garden Spring originates at the Clearwell Overflow. Because the sewage treatment outfall began operation only 5 y before the sampling of stable isotopes as part of this study, the evidence, if viewed alone, would indicate a groundwater travel time from the Rim to
1302
9
Whether by percolation from the sewage discharge at the Clearwell Overflow or by a leak in the transcanyon pipeline, the observation that perhaps up to half of the water discharging at Indian Garden Spring ultimately originates as North Rim water indicates that anthropomorphic forces have impacted this spring. The National Park Service seeks to reduce human intrusion and disruption of natural resources in our parks. In addition, if a rather direct hydrologic connection exists between the Clearwell Overflow and Indian Garden Spring as afforded by the Bright Angel Fault, there is a potential for future negative consequences on the spring in terms of altering or degrading water quality. Stable isotopic methods can provide a basis for quantitatively assessing these effects.
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 7, 2001
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13)
Friedman, I. Geochim. Cosmochim. Acta 1953, 4, 89-103. Craig, H. Science 1961, 133, 1702-1703. Dansgaard, W. Tellus 1964, 16, 436-438. Friedman, I., Redfield, A. C. Shoem, B.; Harris, J. Rev. Geophys. 1964, 2, 177-224. Friedman, I.; Smith, G. I. Science 1970, 169, 467-470. Siegenthaler, U.; Oeschger, H. Nature 1980, 285, 314-317. Beus, S.; Morales, M. Grand Canyon Geology; Oxford University Press: Oxford, U.K., 1990; 518 pages. Metzger, D. U.S. Geological Survey Water Supply Paper, 1961, 1475-C, pp 100-130. Huntoon, P. W. Water Resour. Res. 1974, 10, 579-590. Goings, D. B. Unpublished master’s thesis, University of Nevada, Las Vegas, 1985. Environmental Impact Statement, USDA Forest Service, 1998. Dugan, J. P., Jr.; Borthwick, J.; Harmon, R. S.; Gagnier, M. A.; Gahn, J. E.; Kinsel, E. P.; Macleod, S.; Viglino, J. A.; Hess, J. W. Anal. Chem. 1985, 57, 1734-1736. Kendall, C.; Coplen, T. Anal. Chem. 1985, 57, 1437-1440.
Received for review July 24, 2000. Revised manuscript received January 11, 2001. Accepted January 11, 2001. ES0015186
