Evaluating Solute Sources and source distribution in the upper Gila River Basin

Abstract:

The upper Gila River is located in the southwestern New Mexico, and categorized as one of the last free flowing streams in the United States. It exhibits substantial spatial and temporal variability of solute concentrations across the watershed. Significant trends of salinization are evident by increase of concentrations of major ion, as well as many trace elements, in downstream progression. Understanding the dynamics of chemical variability of this system is vital for evaluation of the Gila River as a sustainable source of water for the downstream communities.

Deriving its headwaters in the Mogollon-Datil volcanic field, the upper Gila River has a significant number of perennial springs. Volumetrically small such endogenic sources of recharge could have a substantial impact on chemical composition of the stream and lead to salinization of the system. Impairment of the water quality in the Gila River is most evident during prolonged drought conditions when the system discharge is reduced to base flow conditions.

Commonly accepted climatologic projections suggest reduced amounts of annual precipitation and increase in temperature, which signifies prolonged base flow conditions. During low flow regimes springs have been determined to be the primary sources of recharge in the upper Gila River.

Primary objectives of this project are to evaluate spatial distribution of endogenic sources of recharge and their impact on chemical composition of the stream. Concurrently it is important to take in to consideration recent wildfires that have impacted large areas of multiple sub-watersheds in the headwaters of major tributaries. Ash flows as well as resulting erosional processes of impacted regions could revile to be significant contributors of solutes and sources of impairment for the stream.

Acknowledgment: I thank Dr. Laura Crossey and Dr. Abdul-Mehdi Ali for contributing to data analysis and providing necessary instrumentation to perform this study.

Outline:

I. Purpose:

1. Create Gila River Profile using GIS

· Plot elevation and length of streams within the system

2. Compare sources of solutes and confluence mixing

· Evaluate impact of springs on chemical composition of streams

· Major confluences

· Downstream progression

· Variability between sampling dates

3. Quantifying fire impacted areas

· Estimate percent of burnt area of the impacted sub-watersheds

II. Regional setting:

1. Location and spatial parameters

2. Land use

3. Hydrology

4. Climate

III. Methods:

1. Data collection:

2. Laboratory analysis

3. Geographic Information Systems Analysis

4. Displaying acquired data

IV. Results

Introduction

For well over a century, agriculture, ranching, mining, flood control, and municipal development have placed an increasing demand on the scarce water resources of the southwestern United States. Today, both surface and groundwater remain major water sources for numerous urban, ranching, and agricultural communities along the Gila drainage basin. Water quality is of great concern due to the general scarcity of renewable water supplies in the region (Phillips et al., 2003; Mills, 2003; Plummer et al., 2004; Newell et al., 2005; Anning et al., 2007).

In order to understand the complex interactions and impact of the water-quality-impairing sources on the Upper Gila subbasin, it is essential to approach the problem with consideration of surface and groundwater as an integrated system (Winter et al., 1998). Although the Gila River is the dominant surface water feature in the region, numerous springs, many of endogenic origin, may have significant impact on the hydrochemical composition of the stream. Many of these springs emerge along faults and fractures of the underlying volcanic bedrock predominant to the regional geology of the Mogolon-Datil volcanic field. Although volumetrically small in comparison to the dominant surface stream, recent research suggests such springs are geochemically important deep fluid transports of many components, including elements derived from interaction with the Earth’s upper mantle (Newell et al., 2005; Crossey et al., 2006; Crossey et al., 2009). Previous studies suggest that deep-seated faults in the bedrock create a complex of conduits which allows deeply derived fluids to ascend to the surface. However, understanding the hydraulic properties of fault fluid transport remains vague, largely due to the technological constrains which limit the acquisition of detailed property distribution from within active faults (Newell et al., 2005; Crossey et al., 2006; Crossey et al., 2009, Fairley, 2009). Fault fluid transport modeling largely demonstrates conductive capacity orders of magnitude higher along the fault plane than surrounding matrix (Summers, 1982, Fetter, 1994; Lallahem et al., 2007). In the deeper flows, through fractured bedrock, longer term geochemical interactions determine the chemistry of waters that eventually discharge to the stream. Conversely, during snowmelt and storm events, faults may transmit significant amounts of flow through the subsurface resulting in the hillside springs with chemical alteration interrelated with landscape cover features (geology, soil, and vegetation types) (Winter et al., 1998; Acuña and Dahm, 2006). During periods of low rainfall, streams in mountainous terrains are reduced to base flow and exist due to drainage from saturated alluvium in the valley bottoms and perennial springs from bedrock fractures. Mixing of these chemically different waters results in geochemical reactions that affect the chemistry of the water in the stream (Winter et al., 1998). Quantifying discharge of a given spring presents a challenge due to the inability to measure volumes contributed by numerous discharge points along a given fault system and conducted through surface flows as well as shallow alluvium of the inner valley fill.

Purpose

In order to better understand longitudinal distribution of the sources of solutes distribution in the Gila River and its many tributaries a detailed river profile had to be constructed.

Projecting chemical analysis data of major ions for better understanding and interpretation of complex mixing processes of contributing surface and endogenic sources of solutes to the upper Gila River system.

Quantifying areas of sub-watersheds impacted by wildfires.

Regional Setting

Location and spatial parameters

The Gila River watershed is located on the New Mexico-Arizona border in the Datil-Mogollon section of the Transition-Zone province (Hawley 1969, 1975, 1986: Morrison 1991). Drainage basin portion located in New Mexico has an area of approximately 9,300km² (3,590mi²). The upper Gila River subbasin derives its headwaters in the Mogollon – Datil volcanic field and has an area of 7,780 km² (2,645 mi²). The Continental Divide defines northern, eastern, and southern watershed boundaries, while the northwestern boundary is confined by the San Francisco watershed. High plateaus of the Mogollon-Datil volcanic field overlook deeply carved canyons which are predominant geomorphic features of the entire Upper-Gila watershed. Elevation of the Mogollon-Datil section ranges from 3,320m (10,892ft) at Whitewater Baldy in the Mogollon Mountains to 1,463m (4,800 ft.) at Redrock in the Gila River Valley.

Land Uses

Landcover of the Upper Gila subbasin is comprised primarily of extensive forested areas of high elevations and rangeland in most of the rest of the basin. An urban and agricultural development is only found in one location along the river valley, located near the Gila-Cliff subbasin boundary. No major consumptive water use has been recorded in the Upper Gila subbasin (Hawley et al., 2000).

Hydrology

The surficial layer of Upper Quaternary valley fill associated with the Upper Gila River, comprises a shallow aquifer unit, with saturated thicknesses of up to 30 m (100 ft). Even though this hydrogeologic unit is areally limited by the narrow canyon walls, it is recognized to be of key importance to the recharge of perennial and intermittent tributaries.

Climate

The Gila River system falls in to a category of arid to subhumid climatic regions with an annual precipitation ranging from 23.1cm (9.1 in) at the Virden Station to 51 cm (20 in) at the Pinos Altos station located on the continental Divide (elev. 2,134 m; 7,000 ft). The average spatial variability in temperature ranges from 15.1oC at Redrock (elev. 1,483 m; 4,900 ft) to 11.6^oC at Gila Hot Springs (elev. 1,707m; 5,600ft). The monsoonal regimes, which occur from July through September and contribute nearly half of the annual precipitation in the region (NCDC, 1999, Hawley et al., 2000). However, historical records from 1929 and 1973 report 35% (51 cm (20 in)) of the annual precipitation had been contributed by winter precipitation (Gabin and Lesperance 1977). Such significant amounts of winter precipitations are attributed to quasiperiodic climate patterns, and are driven by the southern oscillation cycles. Controversially, observations from winter 2010-2011 report substantial dry conditions throughout Gila watershed exhibiting a sharp contrast from winters in the previous several years. A number of climatologic models project considerable increase of average annual temperatures and significant decrease in annual precipitation. Dryer and hotter conditions suggest an increase in frequency and persistency of base flow conditions. In response to low flow conditions, stream water quality will degrade due to salt loading contributed by perennial springs.

Data Collection

Campaign sampling

Samples were collected precisely following the EPA protocol. Each of the sampling locations were evaluated in upstream to downstream progression. To evaluate an impact of a given tributary on the chemical composition of the stream, field parameters and samples were collected approximately 100 meters above and below each confluence with perennial tributary as well as endogenic contributing sources. Sampling locations were field-referenced using Garmin 60c portable GPS device. Readings of temperature, pH, conductivity, specific conductance, and dissolved oxygen (percent saturation and mg/L) field parameters were taken and recorded in triplicates, to minimize potential errors, using an Oakton pH/CON 300 Series and Oakton DO6 series field instruments. Field measurements from each stream location were taken midstream in approximately two inches below the water surface. Springs were sampled at the closest possible proximity to the orifice.

Sample collection and preservation

Concurrently to collection of field parameters, two volumetric aliquots were collected into 125ml bottles, pre-conditioned by triple-rinse with sampling water. Preservation method varies by the objective of future analysis. Aliquot intended for alkalinity and anion analysis is sealed in a sterile storage bottle, eliminating atmospheric void, referred to as headspace, from the container to minimize degassing. A second volume of the sample was passed through a 0.45μm filter, fixed with small amount of concentrated nitric acid and stored in a sterile, triple-rinsed with filtered sample, container for future cation analysis. All collected samples must be refrigerated to 4 °C as soon as possible after collection and remain cooled until the analysis.

Geochemical data analysis

Once chemical compositions of the waters were obtained, complex analytical software

(PHREEQC) was utilized to accurately model mixing conditions and end members of these processes. PHREEQC was developed by the USGS (Parkhurst and Appelo, 1999) to perform low-temperature aqueous geochemical calculations. The code was used to evaluate the state of saturation with respect to mineral phases as well as to compute equilibrium pCO2 based on pH, alkalinity, and ionic strength. It was also used to speciate all water sampled and calculate saturation indices of calcite, gypsum, dolomite, halite, and carbonate species, as well as perform binary mixing modeling and inverse modeling to identify potential end member chemistries.

Geographic information systems analysis

In order to better understand longitudinal distribution of endogenic sources of recharge and their spatial relationship to fault conduits, a river profile was produced. The following steps describe step by step process used to create streams, delineate the Upper Gila watershed, and analyze features’ attributes and extract of necessary data.

Defining streams and watersheds

a. Add DEM of the desired area. Using symbology assign appropriate colors and check hillshade box (Right-click on DEM layer > properties > symbology > color ramp).

b. Projection. Make sure DEM and data frame in the same projection (Data management > projections and transformations > raster > project raster).

c. Fill (special analyst > hydrology > fill) and save output as raster.

d. Flow direction. Using spatial analyst and based and fill raster calculate flow direction (spatial analyst > hydrology > flow direction).

e. Flow accumulation. Using spatial analyst calculate flow accumulation (spatial analyst > hydrology > flow accumulation).

f. Stream link. Using spatial analyst define streams by minimum cell confluence in to single cell.

g. Stream to feature. River shapefile was created using stream to feature tool found in spatial analyst toolbox.

h. Delineating watershed. Delineation of the watershed was performed using spatial analyst watershed tool.

River Profile:

a. Flow length. Using flow direction calculate flow length (spatial analyst > hydrology > flow length).

b. Assigning headwater points. Add new “point” shapefile (Arc catalog > right click (desired folder) > new > shapefile (feature type = point, select correct projection). Using editor tool assign headwaters of each stream (right-click on newly created shapefile > edit feature > start editing). Create point for the beginning of each desired stream. In editor tool select ‘stop editing’ and save edits.

c. Cost path. Using spatial analyst tool determine the distance (spatial analyst > distance > cost path). Input raster = headwaters shapefile, Cost distance = flow accumulation raster, backlink raster = flow direction raster, output raster = user defined. (Calculates the least-cost path from a source to a destination).

d. Extracting data. Using the sample tool extract x, y, z, length and accumulation data in to a .dbf file (spatial analyst > extraction > sample). Select layers: filled DEM, flow length, flow accumulation. Input location raster or point feature = costpath raster. Resampling technique = ‘BILINEAR’.

e. Defining main stream. Set the identify tool to read the flow length raster and identify the headwater with greatest value. Repeat the steps described above for the identified point to extract the values.

f. Final data visualization was created as a graph using Excel based on the data extracted from GIS. Springs and faults were extracted and plotted using same methods as described above.

Displaying Major Ion Concentrations

a. Data management. Using Excel, data acquired from chemical analysis was formatted for further GIS analysis.

b. Importing data. Acquired chemistry data was imported in to GIS as a data table and projected using corresponding longitude and latitude coordinates.

c. Displaying data In order to adequately compare and interpret the results, concentrations of major ion was displayed as a pie chart and rendered proportionate to the sum of the major ions. All the major confluences and significant sources of solutes were exported as separate maps for enhanced visual comparison.

Fire Impact

a. Add fire impact layer.

b. Clip. Clip the layer to the Upper Gila Watershed layer.

c. Add subbasin layer. Using select by polygon create separate layers for each impacted sub-basin.

d. Clip fire impact layer to created individual sub-basins.

e. Using spatial statistics tool calculate area of each impacted region.

Conclusions

Elevation profile illustrated that majority of the springs are located in the upper reaches of the upper Gila watershed. Contrary to previous assumptions endogenic sources of solutes do not correlate with near proximity to the major fault lines.

Projections of the chemical analysis indicate that spring have a significant impact on major ion concentrations. Furthermore the springs located several kilometers apart have a nearly identical chemical composition.

Changes in hydrochemical compositions below evaluated confluences could possibly be used to quantify volumetric relationships of individual strams.

The fire impacted regions cover significant area. Subbasin 1 = 316km², Subbasin 2 = 173km², Subbasin 3 = 238km².

Future work

Further analysis of correlations between fire impacted areas will be performed once continuous monitoring data from deployed dataloggers will be uploaded.

Data obtained by hydrochemical modeling will be projected using GIS and compared to actual values produced by chemical analysis of the samples.

References

Acuña V.¸ Dahm C. N., 2007. Impact of monsoonal rains on spatial scaling patterns in water

chemistry of a semiarid river network.Journal of Geophysical Research, Vol. 112.The

American Geophysical Union.

Anning, D.W., Bauch, N.J., Garner, S.J., Flynn, M.E., Hamlin, S.N., Moore, S.J.,Schaefer, D.H., Anderholm, S.K., Spangler, L.E., 2007. “Dissolved solids in basin-fillaquifers and streams in the southwestern United States.” United States GeologicalSurvey Scientific Investigations Report 2006-515.

Bartolino, J.R., and J.C. Cole, 2002. "Ground-water resources of the middle Rio Grande

basin, New Mexico." United States Geological Survey Circular 1222.

Bauer, P.W., P.S. Johnson, and S. Timmons, 2007. “Springs of the Rio Grande gorge,

Taos County, New Mexico: Inventory, data report, and preliminary geochemistry.”

New Mexico Bureau of Geology and Mineral Resources Final Technical Report.

Open-File Report 506.

Crossey, L.J., Fischer, T.P., Patchett, P.J., Karlstrom, K.E., Hilton, D.R., Newell, D.L.,

Huntoon, P., 2006. “Dissected hydrologic system at the Grand Canyon: Interaction

between deeply derived fluids and plateau aquifer waters in modern springs and

travertine.” Geology 34.1: 25-28.

Crossey, L.J., Karlstrom, K.E., Springer, A., Newell, D., Hilton, D.R., Fischer, T., 2009.

“Degassing of mantle-derived CO2 and 3He from springs in the southern Colorado

Plateau region - flux rates, neotectonics connections, and implications for

understanding the groundwater system.” Geological Society of America Bulletin in

press.

Chapin C. E., Dunbar N. W., 1994.A Regional Perspective on Arsenic in Water of the Middle Rio Grande Basin, New Mexico.The Water Future of Albuquerque and Middle Rio Grande Basin. New Mexico Water Resources Research Institute, New Mexico Bureau of Mines and Mineral

Resources Campus Station Socorro, NM.

Fairley J., 2009. “Modeling fluid flow in a heterogeneous, fault-controlled hydrothermal system” Geofluids9, 153–166University of Idaho, Moscow, Idaho.

Hawley, J.W., Hibbs. J.B, Kennedy J.F., Creel J. B., Remmenega M.D., Johnson M., Lee M.M.,

Dinterman p., 2000.Trans-International Boundary Aquifers in Southwestern New

Mexico.

Mills, S.K., 2003. “Quantifying Salinization of the Rio Grande Using Environmental

Tracers.”M.S. Thesis New Mexico Institute of Mining and Technology.

Newell, D.L., Crossey, L.J., Karlstrom, K.E., Fischer, T.P., 2005. “Continental-scale links

between the mantle and groundwater systems of the western United States: Evidence from travertine springs and regional He isotope data.” GSA Today 15.12: 4-10.

Newell, D.L., 2007. “Hydrogeochemistry of CO2-rich mineral springs: Implications for

tectonics and microbiology”. Ph.D. Dissertation, University of New Mexico. 156p.

Phillips F. M., Mills S., Hendrickx M.H., Hogan J., 2003. Environmental Tracers Applied to

Quantify Causes of Salinity in Arid-Region Rivers: Results from the Rio Grande Basin,

Southwestern USA. Water Resources Perspectives: Evaluation, Management and Policy, 327-

334.

Plummer, L.N., Bexfield, L.M., Anderholm, S.K., Sanford, W.E., Busenberg, E., 2004.

“Geochemical characterization of ground-water flow in the Santa Fe Group aquifer

system, Middle Rio Grande Basin, New Mexico.” United States Geological Survey

Water-Resources Investigations Report 2003-4131.

Rawling, G., Timmons, S., Newton, T. Walsh, P., Land, L., Kludt, T., Timmons, M.,

Johnson, P., Felix, B., 2008. “Sacramento Mountains Hydrogeology Study” New

Mexico Bureau of Geology and Mineral Resources Open File Report 512.

Schwab G. E., Summers W.K., Colpitts R.M., Teuten C.E., Young W. K., 1982. “Pumping Tests of Well Campbell et al. No. 2 Gila Hot Springs, Grant County, New Mexico.” U.S. Department of Energy.

Winter, T.C., Harvey, J.W., Franke, O.L., Alley, W.M., 1998. “Ground water and surface

water: A single resource.” United States Geological Survey Circular 1139.

Witcher J. C., Lund J. W., 2002. Gila Hot Springs. GHC Bulletin. Southwest Technology DevelopmentInstitute NMSU, Las Cruces, NM, Geo-Heat Center.