• No results found

Assessing and modeling irrigation-induced selenium in the stream-aquifer system of the Lower Arkansas River Valley, Colorado

N/A
N/A
Protected

Academic year: 2021

Share "Assessing and modeling irrigation-induced selenium in the stream-aquifer system of the Lower Arkansas River Valley, Colorado"

Copied!
12
0
0

Loading.... (view fulltext now)

Full text

(1)

Assessing and Modeling Irrigation-Induced Selenium in the

Stream-Aquifer System of the Lower Arkansas River Valley,

Colorado

Alexander W. Herting 1

Graduate Research Assistant and MS candidate, Civil Engineering Department, Colorado State University, Fort Collins

Timothy K. Gates2

Professor, Civil Engineering Department, Colorado State University, Fort Collins

Abstract. Water quality data on dissolved selenium (Se) have been collected since April 2003 in a study area of the irrigated lower Arkansas River Valley, Colorado. Data have been obtained from 22 surface water locations and from 59 ground-water monitoring wells using low-flow sampling techniques. GIS mapping and statistical analysis were used to characterize the occurrence, severity, and spatial-temporal dis-tribution of Se in the study area. Results indicate dissolved Se concentrations ranging from about 0 to 3760 micrograms per liter (µg/l), with a median of 16.6 (µg/l) in the ground water, including three significant “hot spots”. River concentrations range from 4.2 to 23.0 µg/l, often exceeding stream standards for aquatic habitat. Relation-ships are explored between Se concentrations and more-easily- monitored indicators such as electrical conductivity, sulfate concentration, and nitrate concentration. Statis-tically significant non-linear relationships exist between Se, sulfate, and nitrate con-centration. Uranium (U) concentrations also are found to have a significant relation-ship with Se. Since high U concentrations have earlier been linked to marine shale and shale-derived soils in the Valley, the relationship between U and Se concentra-tions suggests a similar linkage between Se and the presence of shale formaconcentra-tions. This relationship will be explored further with additional sampling events to test its validity A model of the selenium transport process in the unconfined aquifer, as af-fected by irrigation practices, is currently being developed and will be calibrated using field data.

1. Introduction

The Lower Arkansas River Valley in Colorado is an intensively irrigated, alluvial basin comprised of underlying marine sedimentary rocks. With these characteristics, dissolution and mobilization of metals and minerals, such as se-lenium (Se), can lead to transport into canals and the Arkansas River and a consequent threat to the health and safety of animals and aquatic life. The study described herein aims to describe the extent to which these processes are

1

Civil Engineering Department

Colorado State University, Fort Collins, CO 80523-1372 Tel: (970) 491-5387

e-mail: aherting@engr.colostate.edu

2

Civil Engineering Department

Colorado State University, Fort Collins, CO 80523-1372 Tel: (970) 491-5043

(2)

ground for possible total maximum daily load (TMDL) development. Thirteen segments of the Arkansas River are designated as“water quality limited” with respect to Se and/or Fe and have been placed on the current Clean Water Act 303(d) list (CDPHE 2002) for TMDL development, along with six segments on the current Monitoring and Evaluation List (CDPHE 2002). In order to set appropriate water quality standards, the Colorado Water Quality Control Division of the Colorado Department of Public Health and Environment (CDPHE), is in the process of determining the nature of the ele-vated levels of Se in the stream-aquifer system of the Arkansas River.

The study area consists of a segment of irrigated alluvial valley extending about 60 km from Lamar, Colorado to the Colorado-Kansas border along the Arkansas River. The Lower Arkansas River Basin is underlain by seleniferous Upper Cretaceous and Tertiary marine sedimentary rocks, as well as shale out-croppings that serve as a source of Se in the overlying soils. The study area encompasses approximately 54,300 ha of land, including about 25,700 ha un-der irrigation. It is sufficiently large to capture the diversity of soils, hydro-geology, irrigation and drainage infrastructure, and crops that are characteristic of the portion of the Arkansas River downstream of John Martin Reservoir.

Approximately 59 monitoring wells and 22 surface water monitoring points have been sampled nineteen times over the past three years for dissolved Se, iron (Fe), pH, specific conductance (EC), temperature, dissolved oxygen (DO), oxidation-reduction potential (ORP), and major salt ions using low flow sam-pling techniques. Analytical analysis has been provided by Olson Biochemis-try Labs at South Dakota State University in Brookings, SD and Ward Labora-tories, Inc. in Kearney, NE. A map of the study region, showing GIS coverage of individual fields and locations of monitoring sites is shown in Figure 1.

(3)

2. Data Analysis

2.1 General Statistics

Histograms of the data collected and analyzed to date (through the January 10-14, 2006 sampling event) were plotted as shown in Figures 2 to 4 for ground water, all surface water, and Arkansas River samples, respectively. In order to visually examine the relevance of measured concentrations of CSe along the study area, current standards are indicated on each histogram. For ground water comparison, the 20 (µg/l) Colorado groundwater standard for ag-ricultural use (CDPHE 2001a) is shown. The Arkansas River, irrigation ca-nals, and drains are compared against the Colorado chronic aquatic life stan-dard (also referred to as the table value stanstan-dard, TVS) of 4.6 µg/l (CDPHE 2001b) and the temporary lower Arkansas River segment 1c standard of 14 µg/l (CDPHE 2003). Data indicate levels of Se in the study area that exceed use protective standards.

0 50 100 150 201 251 CSe (µg/L) 0 100 200 300 400 500 600 700 800 Nu m b e r o f Ob se rv a ti o n s

Colorado Ground Water Agricultural Standard CSe= 20 mg/L

Figure 2. Total dissolved selenium in ground water samples collected April 25 – May 3 2003 to January 10-14, 2006.

(4)

2 6 11 15 19 24 28 33 37 42 46 CSe (µg/L) 0 20 40 60 80 100 N u m b er of O b se rva ti ons

Colorado Temporary Standard

CSe (ch) = 14 µg/L

Colorado TVS

CSe (ch) = 4.6 µg/l

Figure 3. Total dissolved selenium in surface water samples collected from April 25 – May 3 2003 to January 10-14, 2006. 2 4 6 8 10 12 14 16 18 20 22 24 26 CSe (µg/L) 0 5 10 15 20 25 30 35 40 Nu m b e r o f Ob se rv a ti o n s Colorado T VS CSe (ch) = 4.6 µg/l

Colorado T emporary Standard

CSe (ch) = 14 µg/L

Figure 4. Total dissolved selenium in Arkansas River samples collected from April 25 – May 3, 2003 to January 10-14, 2006.

Summary statistics of CSe for all ground water, surface water, and river data are presented in Table 1. The Se concentrations in ground water samples range over five orders of magnitude. This skews the sample mean right of the me-dian, as is apparent from the large variance and skewness. The maximum value of 3760 µg/l was determined from a sample taken at a monitoring well north of Lamar on June 30, 2004. The samples collected throughout the rest of the sampling period at this well also indicated consistently high CSe. The

(5)

monitoring well was destroyed by field harvesting operations in late summer 2004. Two new wells were installed nearby in November 2005. One well was dry and the other contained a selenium concentration of only 15.7 µg/l when sampled during January 2006.

.

Table 1. Summary statistics of dissolved Se (µg/l).

Ground

Water

Surface Water Arkansas River

Size of Population 954.0 210.0 135.0 Mean 70.6 13.9 10.7 Median 16.6 11.5 10.2 Maximum 3760.0 46.2 23.0 Minimum 0.4 0.5 4.2 Standard Deviation 373.5 10.6 4.0 Skewness 8.9 1.2 0.6

2.2 Spatial and Temporal Statistical Analysis

From the histograms, it is clear that the probability distributions of the data are not normally distributed. Therefore Se concentrations were analyzed in time series using nonparametric statistics to avoid biased results. The box and whisker plots shown in Figures 5 and 6 illustrate a summary of the statistical analysis. The central squares on the plot represent the median Se concentration at each point in time. The top and bottom of the rectangles represent the 75th and 25th percentiles, while the whiskers extend to the non-outlier range by the data set. The highest and lowest values were omitted from the data sets to de-crease biasing by extreme values. Outliers were systematically determined as values that are higher than the 75th percentile value by more than two times the difference between the 25th and 75th percentile values. The outliers are shown on the plot.

Figure 5 indicates the median of each ground water data set is not substan-tially changing over time. However, it appears that the variability in CSe in ground water changes throughout the year. The degree to which this occurs depends on recharge rates, drainage rates, and river flow rates, which will be explored.

Figure 6 indicates seasonal patterns of CSe in the Arkansas River. There is higher CSe during low flow months and lower CSe during high flow months. The low flows occur during fall and winter months while high flows occur dur-ing the sprdur-ing and summer months.

(6)

Median 25%-75% Non-Outlier Range Outliers Extremes A p ri l 25 -M ay 5, 200 3 May 29-Ju n e 11 20 0 3 June 30-Ju ly 8 2 0 0 3 Ju ly 27 -3 1 2003 O c tober 25-N o v em b er 2 2003 Ja n u ar y 12-16 2 0 0 4 Mar ch 15-1 8 20 0 4 A p ri l 30 -M ay3 200 4 Ju ne 1-4 20 0 4 June 28-Ju ly 1 2 0 0 4 A ugus t 2-5 2 0 0 4 N o ve mbe r 4 -8 200 4 Ja n u ar y 10-14 2 0 0 5 Mar ch 14-2 0 20 0 5 Ju n e 2 7 -J uly2 2005 Ju ly 18 -2 3 2005 A ugus t 1 5 -22 2005 N o v em b er 19-2 2 20 0 5 Ja n u ar y 10-14 2 0 0 6 0 20 40 60 80 100 120 140 160 180 200 220 CSe ( µg/L) Extreme Points

Figure 5. Box and Whiskers plot of dissolved Se in the ground water at continuously-sampled sites in the study area.

Median 25%-75% Non-Outlier Range Outliers Extremes A p ri l 2 5 -M ay 5, 20 0 3 May 29-Ju ne 1 1 2003 June 30-Ju ly 8 2 003 Ju ly 2 7 -31 2003 O ctober 25-No vem b er 2 2003 Ja nuar y 12-16 2 004 Mar ch 1 5 -18 2004 A p ri l 3 0 -M ay3 20 0 4 Ju n e 1 -4 2 004 June 28-Ju ly 1 2 004 A ugus t 2 -5 2 004 N o vember 4 -8 2 0 0 4 Ja nuar y 10-14 2 005 Mar ch 1 4 -20 2005 Ju n e 2 7 -J u ly2 2005 Ju ly 1 8 -23 2005 A u gu st 1 5 -22 2005 N ovem b er 19 -2 2 2005 Ja nuar y 10-14 2 006 2 4 6 8 10 12 14 16 18 20 22 24 CSe ( µg /L )

Figure 6. Box and Whiskers plot of dissolved Se at continuously-sampled sites in the Arkansas River.

(7)

Figures 7 and 8 provide examples of GIS contour maps of CSe developed for summer and winter sampling events in 2005, respectively. The contours were estimated in GMSTM (BYU 2005) utilizing the “natural-neighbor” inter-polation method. The study area has relatively low Se concentrations uni-formly throughout the middle portion. However, the eastern portion of the study area has two distinct hot-spots, which change in magnitude depending on the season.

Ground water samples collected during the period of July 18 – 23, 2005 were analyzed for dissolved uranium concentration, CU. A GIS color-gradient contour map shown in Figure 9 reveals significant spatial variation in CU within the study area. High CU values are generally found in areas of high CSe, as was earlier noted in a study by Zielinski et al. (1995). Additional samples for uranium from 15-20 ground water monitoring wells are planned for the March 2006 sampling period. The wells chosen for uranium sampling will be located where the highest values of CSe have been detected. All of the ground water monitoring wells will be sampled for uranium during the May sampling event.

2.3 Water Quality Relationships

Relationships were explored between CSe and several in-situ monitored variables including EC, DO, and ORP. Dissolved Se is thermodynamically in equilibrium as selenate, Se(VI), in highly oxidized environments such as the lower Arkansas River Valley, although it can be stable as selenite [Se(IV)] in fairly oxidized environments. In order to empirically explore the possibility of determining Se concentrations from ORP measurements, a laboratory analysis was conducted for selenite in the July 27 - 31, 2003 data set. Using total dis-solved selenium and selenite analysis, the amount of selenate was calculated. Approximately half of the samples analyzed for selenite were below the detec-tion limit and the data indicate that over 90 percent of Se in the study area is oxidized to selenate. This suggests that waters in the lower Arkansas River valley may be sufficiently oxidized and that the ORP may not be a limiting factor for Se dissolution.

(8)

Figure 8. Contour plot of CSe from samples collected during January 10-14, 2006. Although there does not appear to be a statistically significant relationship between Se concentration and ORP, the combination of low dissolved Fe and selenite concentrations indicates that the majority of the ground water and all of the surface waters in the Arkansas River Valley may be sufficiently oxi-dized to allow for the dissolution and mobilization of CSe. Therefore, ORP monitoring may not be useful in quantifying CSe in the Valley. However, there are a limited number of cases where ground water maintains high levels of dis-solved salts but does not contain high levels of CSe. In some of these cases the ORP values are very low.

Figure 9. Contour plot of U concentration from samples collected during July 18 – 23 2005.

Using the least-squares criteria, a two parameter power function was fit to the groundwater data using the Levenberg-Marquardt estimation procedure

(9)

(Moré 1977). Extreme data points stand out in the ground water relationship and it is unclear from the depiction what type of relationship (if any) exists. In order to examine the relationships within the bulk of the data set, extreme val-ues were removed where the EC was greater than 12 decisiemens per meter (dS/m) and where CSe was greater than 240 (µg/l) as shown in Figure 10. There appears to be considerable uncertainty in the relationship, as indicated by the scatter about the fitted curve.

0 2 4 6 8 10 EC (dS/m) 12 0 50 100 150 200 250 300 350 400 CSe ( µg/ L ) M odel: CSe=1.07*(EC) 2.08 r2=0.34

Figure 10. Se concentration versus EC in the ground water samples (excluding data when EC>12 dS/cm or Se>240 µg/L).

Further relationships were explored between CSe and the major salt ions. The most promising bivariate relationship was between CSe and sulfate concen-tration,

4

SO

C , in the surface water as shown in Figure 11. A multivariate

nonlinear regression with and nitrate concentration, , explained even more of the variance in C

4

SO

C CNO3

Se, as may be seen in the observed values of CSe ver-sus the values predicted by the fitted multivariate equation as shown in Figure 12.

(10)

0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 2800 CSO4 (mg/L) 0 10 20 30 40 CSe ( µg/ L ) M odel: CSe=0.241*10-3*CSO4 1.47 r2=0.33

Figure 11. Se concentration versus SO4 concentration for all surface water

sam-ples. 0 5 10 15 20 25 30 35 40 Predicted CSe (µg/L) 0 5 10 15 20 25 30 35 40 45 50 Ob se rv e d C Se ( g/ L )

Model: CSe=(0.48*10-3)CSO4+(0.756)CNO3

r2=0.48

Figure 12. Observed Se concentrations in surface water compared to concen-trations predicted by a multivariate relationship with SO4 and NO3

concentra-tions.

3. Conclusions and Future Research Plans

Overall, selenium concentrations in the study area are higher than the 20 (µg/l) Colorado groundwater standard for agricultural use (CDPHE 2001a). The selenium concentrations vary across the Arkansas River Basin, but three

(11)

main “hot spots” have been detected in the region. There does not appear to be significant seasonal fluctuation in the central tendency of dissolved Se concen-trations in the continuously-monitored ground water data. However, it appears that the variability of Se in the ground water slightly increases during the irri-gation season (from approximately June and July) and decreases during the off season (approximately November through March ). Further analysis exploring relationships between Se concentrations and average recharge rates, drainage rates, and river flow rates are needed to verify this phenomenon.

Concentrations of dissolved Se in the Arkansas River have routinely ex-ceeded the current standard for aquatic habitat. The mean sampled river con-centration is a little over double the standard.

The ground water and surface waters in the Arkansas River Valley may be sufficiently oxidized to allow for the dissolution and mobilization of Se. There-fore, ORP monitoring may not be useful in quantifying Se concentration in the Valley.

Significant relationships exist between Se concentration and EC and SO4 concentration. Nonlinear forecasting relationships were developed between Se concentration and EC, SO4, and NO3 concentration data. Uranium sampling was performed on July 18-23, 2005, and it appears that there is a significant re-lationship between uranium and selenium concentrations. Since high U con-centrations have earlier been linked to marine shale and shale-derived soils in the Valley (Zeilinski et al 1995), the significant relationship between CSe and CU suggests a similar linkage between CSe and the presence of shale forma-tions. Additional sampling will be performed in March and May of 2006 to help enhance this relationship. Also, monitoring wells in a region upstream of the current study area will be sampled during summer 2006 to further explore selenium and uranium concentrations.

A flow and balance model of the unconfined aquifer in the study area using Groundwater Modeling System (GMSTM) software version 6.0 (BYU 2005) is currently being developed and calibrated against field data with salinity and se-lenium concentrations. A steady state model is initially being created, and transient flow and transport models will be produced once the steady state model is calibrated and running properly. This model will be used to predict selenium concentrations under a variety of improved water management alter-natives.

Acknowledgements

The cooperation of more than 80 Arkansas Valley farmers and the financial support of the Colorado Department of Public Health and the Environment (Colorado Nonpoint Source Program) and the Colorado Agricultural Experi-ment Station are gratefully acknowledged. The assistance of a number of stu-dents at Colorado State University in field data collection is also much appre-ciated.

(12)

References

Brigham Young University (BYU). 2005. Environmental Modeling Research Labora-tory, GMS Version-6.0.

Colorado Department of Public Health and Environment (CDPHE). 2001a. “Regula-tion Number 41 – The Basic Standards for Ground water, 65.” Water Quality Control Commission, Denver, Colo.

Colorado Department of Public Health and Environment (CDPHE). 2001b. “Regula-tion Number 31 – The Basic Standards and Methodologies for Surface Water, 169.” Water Quality Control Commission, Denver, Colo.

Colorado Department of Public Heath and Environment (CDPHE). 2002. “Water Quality Limited Segments Still Requiring TMDLs – Colorado’s 2002 303(d) List and Monitoring and Evaluation List, 46.” Water Quality Control Division, Den-ver, Colo.

Colorado Department of Public Health and Environment (CDPHE). 2003. “Regula-tion Number 32 - Classifica“Regula-tion and Numerical Standards for Arkansas River Ba-sin, 81.” Water Quality Control Division, Denver, Colo.

Donnelly, J. P. 2005. “Assessing irrigation-induced selenium and iron in the Lower Arkansas River Valley in Colorado”. MS Thesis, Colorado State Univ., Fort Collins, Colo.

Donnelly, J. P., and Gates, T. K. 2005. “Assessing irrigation-induced selenium and iron in the lower Arkansas River Valley, Colorado.” Proc. of the World Water & Environmental Resources Congress 2005, ASCE-EWRI, Anchorage, Alaska. Fujii, R., and Deverel, S. J. 1989. “Mobility and distribution of selenium and salinity

in ground-water and soil of drained agricultural fields, Western San Joaquin Val-ley of California.” Selenium in Agriculture and the Environment, Soil Science

So-ciety of America, Inc., New Orleans, LA.

McNeal, J. M., and Balistrieri, L., S.. 1989. “Geochemistry and occurrence of sele-nium: An overview.” Selenium in Agriculture and the Environment, Soil Science

Society of America, Inc., New Orleans, LA.

Moré, J., J. 1977. “The Levenberg-Marquardt algorithm: Implementation and theory.”

Lecture Notes in Mathematics. (Berlin). 630, 105-116.

White, A. F., Benson, S.M., Yee, A.W., Wollenberg, H. A., and Flexser, S. 1991. “Groundwater contamination at Kesterson Reservoir, California .2. Geochemical parameters influencing selenium mobility.” Water Resources Research, 27,

1085-1098.

Zielinski, R. A., Asherbolinder, S., and Meier, A. L. 1995. “Uraniferous waters of the Arkansas River Valley, Colorado, USA – A function of geology and land-use.” Applied Geochemistry, 10(2), 133 – 144.

Figure

Figure 1.  The study area and sampling locations.
Figure 2.  Total dissolved selenium in ground water samples collected April 25 – May 3  2003 to January 10-14, 2006
Figure 3. Total dissolved selenium in surface water samples collected from April  25 – May 3 2003 to January 10-14, 2006
Table 1.  Summary statistics of dissolved Se (µg/l).
+6

References

Related documents

46 Konkreta exempel skulle kunna vara främjandeinsatser för affärsänglar/affärsängelnätverk, skapa arenor där aktörer från utbuds- och efterfrågesidan kan mötas eller

Both Brazil and Sweden have made bilateral cooperation in areas of technology and innovation a top priority. It has been formalized in a series of agreements and made explicit

The increasing availability of data and attention to services has increased the understanding of the contribution of services to innovation and productivity in

Närmare 90 procent av de statliga medlen (intäkter och utgifter) för näringslivets klimatomställning går till generella styrmedel, det vill säga styrmedel som påverkar

Den förbättrade tillgängligheten berör framför allt boende i områden med en mycket hög eller hög tillgänglighet till tätorter, men även antalet personer med längre än

Det har inte varit möjligt att skapa en tydlig överblick över hur FoI-verksamheten på Energimyndigheten bidrar till målet, det vill säga hur målen påverkar resursprioriteringar

Detta projekt utvecklar policymixen för strategin Smart industri (Näringsdepartementet, 2016a). En av anledningarna till en stark avgränsning är att analysen bygger på djupa

Industrial Emissions Directive, supplemented by horizontal legislation (e.g., Framework Directives on Waste and Water, Emissions Trading System, etc) and guidance on operating