HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Related articles in JEQ Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via ISI Web of Science (1) Citing Articles via Google Scholar Google Scholar Articles by Rodvang, S. J. Articles by Ryan, M. C. Search for Related Content PubMed PubMed Citation Articles by Rodvang, S. J. Articles by Ryan, M. C. GeoRef GeoRef Citation Agricola Articles by Rodvang, S. J. Articles by Ryan, M. C. Related Collections Animal Waste Ground Water Quality Surface Water Quality Isotopes Nitrogen

TECHNICAL REPORT

Ground Water Quality

Changes in Ground Water Quality in an Irrigated Area of Southern Alberta

^a Irrigation Branch, Alberta Agriculture, Food and Rural Development, Agriculture Center, 100, 5401 First Avenue South, Lethbridge, AB, T1J 4V6 Canada
^b Department of Geology and Geophysics, University of Calgary, Calgary, AB, T2N 1N4 Canada

^* Corresponding author (landys{at}telusplanet.net).

Received for publication June 16, 2002.

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Few studies have documented spatial and temporal variations in ground water quality in areas with high densities of animal farming operations (AFOs), or the long-term effects on surface-water quality. Changes in ground water quality were characterized in an irrigated area with a high density of AFOs in southern Alberta, Canada to evaluate the effect on ground water quality of manure application to fields. Fifty-five piezometers in the oxidized zone were sampled once or twice annually from 1995 to 2001, and temporal changes were analyzed using mixed model analysis. Average NO₃^––N increased significantly from 12.5 to 17.4 mg L^–1 and average Cl^– increased significantly from 19.4 to 34.4 mg L^–1 in piezometers installed in an unconfined sand aquifer at locations receiving fertilizer and manure. Compared with these manured locations, nitrate and chloride concentrations were significantly lower in shallow aquifer water in areas of pasture or native range, and concentrations did not change significantly with time. Nitrate and chloride concentrations in shallow ground water in fine-textured manured locations did not change significantly. Ground water below about 6 m in till and fine lacustrine sediments contains ¹⁸O signatures indicative of recharge under pre-irrigation or glacially influenced conditions, suggesting this ground water has a low vulnerability to agricultural contamination. Evaluations suggest that shallow ground water discharge will cause NO₃^––N and Cl^– in the Oldman River to increase by factors of at least 4.3 and 1.3, respectively, with more significant effects in smaller streams and under low-flow conditions.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
A RECENT SURVEY of well water quality in Canada indicated that 20 to 40% of all rural wells have nitrate concentrations in excess of drinking water guidelines, with up to 60% of wells containing unsafe nitrate levels in regions with high-demand crops or animal farming operations (van der Kamp and Grove, 2001). Ground water nitrate and chloride contamination can result from the overapplication of manure as a fertilizer (Komor and Anderson, 1993; Wassenaar, 1995; Chang and Entz, 1996), from storage lagoons (Korom and Jeppson, 1994; Arnold and Meister, 1999), and directly from feedlot pens (Coote and Hore, 1979). Application of inorganic fertilizer at greater than agronomic rates can also contaminate ground water (Herbel and Spalding, 1993; Exner and Spalding, 1994).

Nitrate leaching occurs through both coarse- and fine-textured sediments (Rodvang and Simpkins, 2001). The extent to which ground water supplies and receiving surface water are compromised varies with hydrogeologic conditions, including climatic conditions and irrigation, and rates of leaching, ground water flow, and denitrification (Spalding and Exner, 1993).

Most previous investigations of the effect of AFOs on ground water quality have focused on manure storage areas or animal pens, or they have been general surveys of well-water quality. Little is known about the spatial and temporal variations in ground water nitrate or salt concentrations in areas with AFOs despite the fact that manure application to irrigated forage fields can have a greater effect on ground water quality than lagoons and animal pens (Harter et al., 2002). After 20 yr of manure application to barley (Hordeum vulgare L.), leaching losses in the glacial till landscape were minimal on non-irrigated land (except in years with unusually high precipitation), and significant on irrigated land (Chang and Entz, 1996). During the past 25 yr, the number of cattle in the province of Alberta has increased by more than 50%, to 5.8 million cattle in 1998. Over the same time period, hectares assessed for irrigation have increased by 66%, to more than 535000 ha (Jaipaul, 2001). Three percent of Alberta's land base is licensed for irrigation, with all irrigation concentrated in the southern one-fifth of the province. The Lethbridge Northern Irrigation District (LNID) in southern Alberta contains the highest concentration of livestock in Alberta, and 82% of the land base is irrigated.

Ground water quality in the eastern portion of the LNID was monitored between 1995 and 2001 to (i) identify aquifers that are vulnerable to agricultural contamination, (ii) determine the effects of intensive agriculture on ground water quality, and (iii) evaluate the potential effects of ground water contamination on surface water quality. To the authors' knowledge, this study is one of the first to document significant long-term changes in ground water quality below manured fields.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Site Description
The study area is located about 25 km north of Lethbridge in southern Alberta, Canada, bounded by the Oldman River on the south and the Little Bow River on the east. The topographic slope is east toward the Little Bow River (Fig. 1) . The Oldman River is a major tributary of the South Saskatchewan River and is an important water supply for agricultural and domestic use. Long-term average precipitation in the Lethbridge area is 400 mm yr^–1, compared with a mean growing-season potential evapotranspiration rate of 760 mm yr^–1. Annual irrigation amounts are generally about 300 to 450 mm yr^–1, significantly reducing the annual moisture deficit (Grace and Hobbs, 1986).

View larger version (28K): [in this window] [in a new window]   Fig. 1. The study area in the eastern portion of the Lethbridge Northern Irrigation District.

Glacial till in the study area is generally 45 to 70 m thick. The till is a dense sandy-clay loam to clay texture, and contains very few coarse fragments (Rodvang et al., 1998). The till is overlain by up to 15 m of homogeneous glaciolacustrine plastic silty clay at lower elevations (Fig. 1 and 2a) . The fine-textured glaciolacustrine sediments are overlain by coarse- to fine-grained glaciolacustrine and fluvial sand in the eastern portion of the study area (Fig. 1 and 2a). The sand grades to fine- and medium-textured lacustrine deposits along the western and northern margins and reaches a thickness of at least 27 m near the junction of the Little Bow and Oldman Rivers (Fig. 2a). In this paper, "coarse-textured" refers to all coarse- and medium-textured glaciolacustrine and fluvial deposits, while "fine-textured" indicates all glacial till and fine-textured lacustrine sediments. Livestock density is highest in the area between Piezometer Nests 7 and 13, while the area between Nest 11 and the rivers is dominantly native range (Fig. 2a).

View larger version (76K): [in this window] [in a new window]   Fig. 2. A portion of the study area in cross-section (labeled A–A' in Fig. 1) showing (a) geology and hydrogeology and (b) 18O signatures and lower limit of tritium detection.

Ground Water Characterization, Sampling, and Analysis
Fifty piezometers were installed in 1994 or 1995 in 22 nests, at depths ranging from 2 to 75 m. These piezometers and an additional five domestic wells screened in the oxidized zone were regularly sampled on 10 occasions (June of 1995, 1996, 1997, 1999, 2000, and 2001; and November of 1994, 1996, 1997, and 1998). Slug tests were conducted on the majority of piezometers and water-table wells. Data were analyzed using the Hvorslev (1951) and the Bouwer and Rice (1976) methods.

Piezometers were bailed before sampling. Piezometers shallower than 8 m were sampled with a peristaltic pump, and piezometers deeper than 8 m were sampled with a bailer. Samples were filtered through 0.45-µm filter paper in the field or in the lab on the day of collection. Cation samples were preserved with 10% HCl. Samples were analyzed using standard methods for major cations, HCO₃^– (acid titration), and SO₄^2– (turbidimetric method; Greenberg et al., 1999). A TRAACS 800 was used to analyze Cl^– (Industrial Method 783-86T), NO₃^––N (782-86T), NO₂–N (784-86T), NH₄–N (780-86T), PO₄–P (781-86T), and total and dissolved P (787-86T).

Environmental isotopes, including ¹⁸O in water (from 54 piezometers), ¹⁵N in nitrate (from 19 piezometers), and enriched tritium (³H, from 22 piezometers) were analyzed on water from a representative number of piezometers in each geologic unit. The isotopic ratio of ¹⁸O to ¹⁶O (¹⁸O) increases as the temperature during precipitation increases (Clark and Fritz, 1997). A particularly light ¹⁸O signature (i.e., <–19 to –22) is, therefore, indicative of water that recharged under colder climatic conditions, such as glaciation (Remenda et al., 1996). The ¹⁸O ratios were determined using standard CO₂–water equilibration techniques (Hendry and Wassenaar, 1999). The isotopic ratio of ¹⁵N to ¹⁴N (¹⁵N) provides indications of nitrate source and denitrification. Animal or sewage waste is characterized by ¹⁵N of greater than +10, while ¹⁵N of soil organic nitrogen typically ranges from +4 to +9, and ¹⁵N of inorganic fertilizer is usually –4 to +4 (Wassenaar, 1995). The lighter ¹⁴N is used preferentially in denitrification, so the remaining nitrate is progressively enriched in ¹⁵N as denitrification proceeds (Wassenaar, 1995). Nitrate was concentrated using an ion exchange column, converted to AgNO₃, and combusted to N₂ gas for analysis of ¹⁵N on a mass spectrometer (Wassenaar, 1995). Isotopic values are reported as deviations of the isotopic abundance relative to international standards (VSMOW for ¹⁸O and AIR for ¹⁵N), with precisions of ±0.1 for ¹⁸O and ±0.3 for ¹⁵N. Enriched tritium, or the isotopic ratio of ³H to ²H, was analyzed using liquid scintillation counting with a detection limit of 0.8 TU. Tritium entered the atmosphere in large quantities during the testing of nuclear bombs in the 1950s and 1960s, so it is used to indicate the presence of ground water that recharged after 1953 (Clark and Fritz, 1997).

Geological Groups and Statistical Analysis
The 55 piezometers and domestic wells installed in oxidized sediments were divided into five groups based on geology, depth, and land use (Table 1). Piezometers in fine-textured sediments below 5 m were put into a separate group since isotopic and hydraulic data (see Results and Discussion section, below) indicated that this ground water is less vulnerable to contamination. Piezometers that were sampled irregularly were not included in the temporal statistical analysis (11 piezometers). The statistical groups contain some missing data (Table 1) due to lack of water or late installation date.

Of the 16 piezometers in the coarse-textured high-intensity agriculture group (Table 1), nine piezometers were installed on or adjacent to the coarse-textured manure-research plots, six were along the edges of irrigated fertilized fields, most of which receive periodic applications of manure, and one was adjacent to a feedlot pen. The five piezometers in the coarse-textured low-intensity agriculture group (Table 1) were installed on native range, pasture, or farmyard grass within 2 km of the Little Bow or Oldman Rivers. An additional six piezometers installed in medium-textured lacustrine in low-intensity agricultural areas were not monitored regularly; consequently they were not included in the statistical analysis.

Seven of the 13 piezometers installed in the fine-textured shallow group (Table 1) were located on the fine-textured manure-research plots, four were along the margins of irrigated fertilized fields, and two were adjacent to feedlot pens. Five of the 10 piezometers in the fine-textured deep group were installed on the fine-textured manure-research plots, and five were installed on pasture or farmyard grass more than 2 km from the rivers.

Changes in geochemical concentration with time were analyzed for four of the five piezometer groups (Table 1) using mixed model methodology (the PROC MIXED procedure) in SAS (Littell et al., 1996). Univariate analysis of variance (ANOVA) is the method most commonly applied to repeated measured data that make comparisons with time. However, ANOVA is only strictly valid if variance does not change with time, and if pairs of measurements are equally correlated regardless of the time lag between measurements (Huynh and Feldt, 1970). These requirements are generally not met for repeated measures in time (Littell et al., 1998). Therefore, the PROC MIXED procedure was used to model the variance and correlation structure of the repeated measures analysis, and then the estimated covariance structure was used to obtain generalized least squares estimates of treatment and time differences (Littell et al., 1998). PROC MIXED can account for missing data (Littell et al., 1996).

The goodness-of-fit criteria indicated that covariance was best described using the compound symmetry covariance structure, which specifies that measures at all times have the same variance, and that all pairs of measures on the same piezometer have the same correlation (Littell et al., 1998). Means were then compared using the Tukey–Kramer method, with probability adjusted for the number of measurements, to determine whether geochemical concentrations were significantly different among groups and with time (P < 0.05). Each sampling date was compared with every other sampling date for nitrate and all major ions. Data were skewed, so means were compared on log-transformed data, after adding a value of 1 to each concentration value to avoid working with negative log values.

Trends with time were also analyzed at each piezometer using linear regression between date and ion concentration (P < 0.05). Nitrate and Cl^– concentrations measured in 2001 in coarse-textured sediments were compared between the high- and low-intensity agriculture groups (Table 1) using a single-factor ANOVA (P < 0.05).

Evaluation of Potential Effect on Surface Water
The Oldman River is a major southern Alberta river that supplies water for domestic, municipal, recreational, and agricultural uses. The maintenance of water quality in this important river is the goal of a consortium of several governmental and non-governmental organizations (Oldman River Basin Water Quality Initiative, 2002). The ground water monitoring results were extrapolated to evaluate the potential for contaminated ground water discharge to decrease water quality in the river.

Ground water discharge to surface water in the Oldman River basin was estimated as the product of ground water recharge and surface area. The total surface area of coarse-, medium-, and fine-textured sediments in the Oldman River basin was calculated from a digitized map of surficial geology (Shetsen, 1987). Measured surface area (Table 2) was reduced by 90% because only 10% of land in the Oldman River basin is irrigated. It was assumed that the relative percentages of coarse-, medium-, and fine-textured sediments in the irrigated areas are similar to the percentages throughout the Oldman Basin, consistent with a visual inspection of the surficial geology.

Ground water recharge rates were estimated using three methods: (i) integral tritium method (Sukhija and Shah, 1976), (ii) water-table rise method (Lerner et al., 1990), and (iii) observed leaching rates for fertilizer-derived NO₃^––N and Cl^–. Details of the methods are described in Rodvang et al. (1998), and estimated recharge rates are summarized in Table 3. Recharge through medium-textured sediments was estimated as the mean of recharge through coarse- and fine-textured sediments.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

The geometric mean hydraulic conductivity of coarse sediments is about 500 times higher than that of shallow fine-textured lacustrine sediments and about 10⁴ times higher than that of shallow till (Table 2). Ground water flow is dominantly horizontal throughout the lacustrine sediments, becoming dominantly vertical in the till (Fig. 2a). Horizontal hydraulic gradients in coarse sediments ranged from 0.0075 to 0.009 m m^–1, corresponding to horizontal Darcy velocities of about 30 to 55 m yr^–1. Horizontal gradients in fine-textured lacustrine sediments ranged from 0.01 to 0.02 m m^–1, corresponding to Darcy velocities of about 0.05 to 0.24 m yr^–1. Ground water from the fluvial and coarse-, medium-, and fine-textured lacustrine sediments discharges to the Oldman and Little Bow Rivers either directly or via irrigation canals and surface drains. Gradients within the lacustrine basin are very flat. Local ground water flow directions vary with location and season due to minor changes in local topography, irrigation canals, and pumped wells. Irrigation canals tend to discharge to ground water during the irrigation season and receive ground water during the nonirrigation season (Zilkey, 2001).

Values of ¹⁸O decrease with depth from relatively enriched values (less than –16) in unconfined coarse sediments to less than –19.2 below 5 m in fine sediments (Fig. 2b). The shift in the ¹⁸O signature to more enriched values corresponds to climatic warming at the end of the Pleistocene, about 10000 to 12000 years BP (Remenda et al., 1996). The depleted values thus correspond to older, pre-irrigation or glacially influenced ground water, consistent with the lack of detectable tritium in ground water with ¹⁸O values less than –19. Where fine-textured sediments occur at the surface, tritium was generally detected to maximum depths of 5 to 7 m (Fig. 2b), corresponding to typical fracture depths. In contrast, at locations with coarse-textured sediments at the surface, tritium was detected to at least a 14.5-m depth in underlying unoxidized fine and medium lacustrine sediments. The ¹⁸O values heavier than –19 also persisted to greater depths in locations with surficial coarse sediments (Fig. 2b). The presence of tritium and relatively enriched ¹⁸O values throughout the uppermost 5 m, and to greater depths below the unconfined aquifer, indicate that this ground water recharged predominantly within the last 40 or 50 yr (Freeze and Cherry, 1979). The depleted ¹⁸O values below 5 to 7 m in fine-textured sediments indicate much older ground water that recharged before irrigation or under a glacial influence.

Major-ion ground water chemistry in oxidized fine-textured sediments in the study area (data not shown) is similar to that described by Hendry et al. (1986) for another location in till in southern Alberta and by Keller et al. (1991) for an "unleached" till in Saskatchewan. The three dominant major ions, Na⁺, SO₄^2–, and Mg²⁺, occur at very high concentrations, particularly in the top 5 m of the unsaturated zone, where the median total dissolved solids concentration for 17 piezometers was 8172 mg L^–1. Chloride is also naturally high with a median value of 250 mg L^–1 for the 17 piezometers. Coarse sediments contain much lower inorganic-ion concentrations since they have been flushed with recent meteoric water, but they receive some high-salt ground water discharge from the underlying till and fine lacustrine sediments, and this process may partially account for the higher salt concentrations at some locations in the unconfined aquifer.

Ground water chemistry in piezometers installed in the brown oxidized sediments contained >2 mg L^–1 dissolved oxygen, lacked dissolved Fe²⁺, Mn²⁺, and arsenic, and often contained NO₃^––N. Ground water in piezometers installed in the gray reduced sediments did not contain detectable NO₃^––N, contained <2 mg L^–1 dissolved oxygen, and usually contained dissolved Fe²⁺, Mn²⁺, and arsenic, and smelled of H₂S. Bicarbonate and NH₄⁺ also generally increased across the redox boundary. Ground water chemistry data, therefore, suggest that any NO₃^––N below the redox boundary is denitrified, consistent with findings at other locations (Postma et al., 1991; Robertson et al., 1996). This paper, therefore, considers only ground water in oxidized sediments.

Nitrate and Chloride in Coarse-Textured Sediments
Nitrate in oxidized coarse and medium lacustrine and fluvial sediments in 2001 ranged from below detection to 52 mg L^–1 NO₃^––N. Several factors indicate that the NO₃^––N and some Cl^– are derived from agricultural sources. The highest NO₃^––N and Cl^– concentrations occur in areas with relatively high agricultural intensity (Fig. 3a and 3b) . Piezometers installed in high-intensity agricultural areas contain significantly higher NO₃^––N and Cl^– concentrations (P < 0.05) than areas with low agricultural intensity (Fig. 4a) . The drinking-water guideline for NO₃^––N (10 mg L^–1) was exceeded in 13 of 16 piezometers in intense agricultural areas in 2001, and in 2 of 11 piezometers in low-intensity areas. The ¹⁵N values in ground water nitrate in four samples from near the water table in coarse-textured sediments ranged from 7.9 to 15.4, suggesting a mainly manure source with a smaller contribution from inorganic fertilizer. There is also a slight tendency for NO₃^––N to increase with increasing Cl^– (Fig. 5) . The correlation between NO₃^––N and Cl^– is consistent with a manure source, due to the practice of adding salt to animal feed, and suggests that denitrification is not appreciable at most locations in the oxidized part of the aquifer. A component of Cl^– from natural sources in fine-textured sediments is suggested at some locations in the aquifer since Cl^– concentrations are relatively high in some low-intensity areas with low NO₃^––N (Fig. 3b and 5). Oxidized coarse-textured sediments are a maximum of about 6 m thick in most agriculturally intense areas (Fig. 2a), and NO₃^––N and Cl^– concentrations were relatively uniform with depth in these piezometers (Fig. 3a and 3b).

View larger version (23K): [in this window] [in a new window]   Fig. 3. Nitrate and chloride concentrations with depth in 2001. Different symbols represent different piezometer nests, except for squares, which indicate nests represented by a single piezometer in that group.

View larger version (21K): [in this window] [in a new window]   Fig. 4. Median nitrate and chloride in 2001 with agricultural intensity in (a) coarse-textured and (b) fine-textured sediments.

View larger version (21K): [in this window] [in a new window]   Fig. 5. Nitrate vs. chloride concentrations in coarse-textured sediments in 2001.

View larger version (43K): [in this window] [in a new window]   Fig. 6. Geometric mean nitrate and chloride concentrations with time. Bars indicate standard error of means. Number of samples is shown above the lines, and dates with the same letter below the lines are not significantly different.

Based on mixed model analysis, changes in inorganic ion concentrations were not significant for any ion except Cl^– in the high-intensity agriculture group. However, each piezometer in coarse-textured sediment with increasing NO₃^––N or Cl^– based on linear regression also exhibited significant increases in one to three other inorganic ions (Table 4, Fig. 7) . For piezometers where NO₃^––N and Cl^– did not change with time, inorganic ion concentrations did not usually change with time either. Although some Cl^– and other inorganic ions can occur naturally in coarse-textured sediments, increasing trends suggest agricultural effects. Changes in inorganic ion concentrations may be related to direct leaching from manure and to cation exchange reactions and/or precipitation–dissolution reactions involving carbonate minerals and gypsum. Major ions in manure include N, K, Ca, Cl, P, Mg, and Na (American Society of Agricultural Engineers, 2000). Dantzman et al. (1983) found that Mg²⁺, Ca²⁺, Mn²⁺, soluble salts, and Fe²⁺ all increased in the top 1.5 m of the soil after feedlot pens had been in place for 15 yr.

View larger version (24K): [in this window] [in a new window]   Fig. 7. Ion concentrations with time at a 5-m piezometer at the coarse-textured manure plots. Significantly increasing concentrations based on linear regression are indicated.

Nitrate N and Cl^– concentrations in the deep fine-textured group were greater than 50 mg L^–1 in all piezometers (Fig. 3c and 3d). Elevated NO₃^––N (100 to 450 mg L^–1) and Cl^– are common in oxidized till and fine lacustrine sediments below the level of tritium detection in southern Alberta (Hendry et al., 1984; Rodvang et al., 1998). Nitrate tends to increase with depth throughout oxidized fine-textured sediments. The highest NO₃^––N concentrations occur between 6 and 12 m below ground (Fig. 3c). Chloride and inorganic ions peak at shallower depths than NO₃^––N, resulting in similar Cl^– concentrations in shallow and deep fine-textured sediments (Fig. 3d and 4b) and in nonlinear relationships between nitrate and inorganic ions (data not shown).

Fertilizer was rarely applied in southern Alberta before 1953, and manure was present in relatively low quantities, indicating that NO₃^––N and Cl^– in this ground water were probably not derived from agriculture. Ground water with elevated NO₃^––N and Cl^– contains high Mg²⁺, Na⁺, Cl^–, and SO₄^2– concentrations, suggesting relatively low rates of flushing with recent precipitation and irrigation. Hendry et al. (1984) concluded that the nitrate is derived from exchangeable NH₄⁺. The ¹⁵N in geologic nitrate from five samples ranged from 10.7 to 24.5 with median and average values of 16.7. Mixed model analysis indicated that nitrate and major-ion concentrations did not change significantly with time in oxidized deep till and fine lacustrine sediments (Fig. 6e and 6f), as expected for old ground water. Similarly, NO₃^––N and Cl^– did not increase in any piezometers based on linear regression (Table 4).

Nitrate and Chloride in Oxidized Shallow Fine-Textured Sediments
Nitrate N was <1 mg L^–1 in all four piezometers in fine-textured sediments above 5 m in areas with low agricultural intensity (Fig. 3c). Geochemical changes with time were not analyzed statistically for this group (Tables 1 and 4) due to irregular sample collection. Nitrate N ranged from 0.1 to 74 mg L^–1 in the 13 piezometers installed in areas with relatively high agricultural intensity, with a median concentration of 14 mg L^–1 (Fig. 3c and 4b). Chloride was not always higher in high-intensity agricultural areas (Fig. 3d and 4b), consistent with the occurrence of geologic Cl^– above 5 m.

Although NO₃^––N increased significantly in 4 of the 13 piezometers in the oxidized shallow fine-textured high-intensity group based on linear regression (Table 4), the changes in NO₃^––N and inorganic ions with time were not significant based on mixed modeling (Fig. 6g and 6h).

Nitrate at six piezometers in this group is clearly derived from agriculture since it occurs in tritiated ground water and decreases to nil below the 5-m depth. Nitrate at the remaining seven piezometers increased with depth below the tritiated zone, indicating that geologic nitrate may occur in the tritiated zone. However, NO₃^––N increased with time in some of these shallow piezometers (Table 4), particularly during recharge events, suggesting an agricultural contribution.

Potential Effect on Surface Water
Most ground water that enters the unconfined aquifer flows laterally and discharges to surface water, or is evapotranspired. Only a very small percentage recharges to deeper ground water. There was an approximate steady state between leaching and ground water flow at several piezometers that were already adversely affected by agriculture in 1995. This suggests that shallow ground water chemistry in the study area will eventually approach large-scale long-term steady state, where contaminants leaching from the surface will be balanced by contaminants discharging to surface water. The increase in NO₃^––N and Cl^– with time at many locations in the aquifer, particularly piezometers on native range near the rivers, suggests that surface water may be increasingly affected by ground water contaminants as ground water continues to transport contaminants to irrigation canals and the rivers.

The potential effect of ground water discharge on water quality in the Oldman River was investigated for six potential combinations of ground water contamination and recharge (Table 5). Scenarios with low agricultural intensity (Scenarios 1, 2, and 3) are based on median NO₃^––N and Cl^– concentrations measured in 2001 in the five piezometers installed in coarse-textured sediments on native range near the rivers. The high-intensity scenarios (4, 5, and 6) are based on median concentrations in all 21 piezometers in coarse-textured sediments. Agricultural land use in the detailed study area is currently more intense than land use in most of the irrigated region, but agricultural intensity is increasing throughout the basin. Scenarios 4, 5, and 6, therefore, represent a situation of increased agricultural intensity in the basin.

Investigations suggest that nitrate removal in riparian zones in the study area varies from partial to complete, depending on riparian zone lithology and hydrogeologic conditions (McCallum, 2001; Rodvang and Riemersma, 2002). The wide variation in nitrate removal with site-specific conditions due to one or more of the processes of plant uptake, denitrification, and dilution is consistent with the findings of other researchers (McMahon and Bohlke, 1996; Cey et al., 1999; Devito et al., 2000). Current research data does not permit the prediction of riparian nitrate removal in a range of landscapes (Devito et al., 2000), so potential losses of 10, 50, and 90% were investigated (Table 6).

View this table: [in this window] [in a new window]   Table 6. Potential effect of ground water discharge on water quality in the Oldman River for different scenarios of recharge and ground water concentration.

Estimates in Table 6 include only discharge from irrigated land, which accounts for 10% of land in the Oldman basin. Smaller water bodies, including canals and surface drains, will be more significantly affected. In addition, more significant increases in the Oldman River would occur during periods of low flow (Table 6).

	CONCLUSIONS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Hydraulic, isotopic, and geochemical data indicate that shallow ground water in an irrigated area of southern Alberta is vulnerable to contamination in both coarse- and fine-textured sediments. Coarse-textured fluvial and lacustrine sediments constitute an unconfined aquifer that is up to 27 m thick and covers an area of about 9000 ha. For 16 piezometers installed in the aquifer on or adjacent to irrigated cropped fields receiving inorganic fertilizer, and in most cases manure, NO₃^––N was detected at concentrations of up to 52 mg L^–1, and Cl^– at up to 85 mg L^–1. For these 16 piezometers, average NO₃^––N increased from 12.5 mg L^–1 in the 1994 to 1996 period to 17.4 mg L^–1 in the 1999–2001 period, and average Cl^– increased from 19.4 to 34.4 mg L^–1 during the same time period. The increases in both parameters were statistically significant.

Compared with these piezometers, NO₃^––N and Cl^– were substantially lower in five piezometers installed in the aquifer on native range or pasture at the down-gradient end of the study area. Although linear regression indicated that nitrate increased significantly in three and Cl^– in two of the five low-agricultural-intensity piezometers (suggesting lateral ground water transport from upslope locations), these temporal changes were not significant for the group as a whole. For all piezometers in the unconfined aquifer, increases in NO₃^––N and Cl^– with time were often accompanied by increases in other inorganic ions.

Ground water in oxidized till and fine lacustrine sediments at depths greater than 6 m contained ¹⁸O values ranging from –19.2 to –21.1 and no detectable enriched tritium, indicating that it recharged before agriculture was introduced on the prairies. Elevated NO₃^––N and Cl^– (>50 mg L^–1) and other inorganic ions in this ground water are derived from geologic sources, and concentrations did not change significantly with time.

Nitrate and Cl^– from geologic sources were present in till and fine lacustrine sediments above 6 m at some locations. However, NO₃^––N was clearly derived from manure and/or fertilizer at some locations, as indicated by depth profiles, relationship to land use, and seasonal fluctuations. Nitrate and Cl^– did not increase significantly between 1995 and 2001 for this group, but NO₃^––N increased significantly in 4 of the 13 piezometers based on linear regression.

Ground water in the unconfined aquifer flows predominantly from high-intensity areas to low-intensity areas near the rivers, transporting NO₃^––N and Cl^– as it flows. If agricultural intensity remains constant at current levels, NO₃^––N and inorganic ion concentrations leached from the surface will eventually be balanced by ground water flow and discharge to rivers. Concentrations will continue to fluctuate seasonally, but annually averaged concentrations will eventually reach steady state concentrations throughout the aquifer, as is currently the case at some piezometers in the high-intensity portion of the aquifer. Nitrate and Cl^– concentrations throughout the aquifer will depend on agricultural intensity. Contaminant concentrations will remain spatially variable for much longer periods of time in shallow till and fine lacustrine sediments. Where till and fine lacustrine sediments occur at ground surface, ¹⁸O signatures indicate that ground water below 5 m recharged predominantly under pre-irrigation or glacially influenced conditions. This suggests that ground water below depths of about 6 m in fine-textured sediments will not be significantly affected by agriculture.

Evaluations suggest that shallow ground water discharge will cause NO₃^––N and Cl^– in the Oldman River to increase by factors of at least 4.3 and 1.3, respectively, with more significant effects in smaller streams and under winter low-flow conditions. If agricultural intensity increases throughout the Oldman River basin to approach current conditions in the study area, NO₃^––N could rise by a factor of at least 13.5 to exceed the aquatic life guideline for total nitrogen, and Cl^– could double in concentration. More information on recharge rates and nitrate removal in riparian zones will improve the accuracy of the prediction of surface water effects.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 

• Alberta Environment. 1999. Surface water quality guidelines for use in Alberta: 1999. Environ. Sci. Div. and Water Management Div., Alberta Environ., Edmonton, AB, Canada.
• American Society of Agricultural Engineers. 2000. Manure production and characteristics. ASAE Standard D384.1. ASAE, St. Joseph, MI.
• Arnold, S.D., and E.A. Meister. 1999. Dairy feedlot contributions to groundwater contamination: A preliminary study in New Mexico. J. Environ. Health 62:16–19.
• Bouwer, H., and R.C. Rice. 1976. A slug test for determining hydraulic conductivity of unconfined aquifers with completely or partially penetrating wells. Water Resour. Res. 12:423–428.
• Cey, E.E., D.L. Rudolph, R. Aravena, and G. Parkin. 1999. Role of the riparian zone in controlling the distribution and fate of agricultural nitrogen near a small stream in southern Ontario. J. Contam. Hydrol. 37:45–67.
• Chang, C., and T. Entz. 1996. Nitrate leaching losses under repeated cattle feedlot manure applications in southern Alberta. J. Environ. Qual. 25:145–153.
• Clark, I., and P. Fritz. 1997. Environmental isotopes in hydrogeology. Lewis Publ., Boca Raton, FL.
• Coote, D.R., and F.R. Hore. 1979. Contamination of shallow groundwater by an unpaved feedlot. Can. J. Soil Sci. 59:401–412.
• Dantzman, C.L., M.F. Richter, and F.G. Martin. 1983. Chemical elements in soils under cattle pens. J. Environ. Qual. 12:164–168.[Abstract/Free Full Text]
• Devito, K.J., D. Fitzgerald, A.R. Hill, and R. Aravena. 2000. Nitrate dynamics in relation to lithology and hydrologic flow path in a river riparian zone. J. Environ. Qual. 29:1075–1084.[Abstract/Free Full Text]
• Exner, M.E., and R.F. Spalding. 1994. N-15 identification of nonpoint sources of nitrate contamination beneath cropland in the Nebraska Panhandle: Two case studies. Appl. Geochem. 9:73–81.
• Freeze, R.A., and J.A. Cherry. 1979. Groundwater. Prentice Hall, Englewood Cliffs, NJ.
• Grace, B., and E.H. Hobbs. 1986. The climate of the Lethbridge agricultural area: 1902–1985. Lethbridge Res. Stn. Mimeo Rep. 3. Res. Branch, Agric. Canada, Lethbridge, AB.
• Greenberg, A.E., A.D. Eaton, and L.S. Cleseri (ed.) 1999. Standard methods for the examination of water and wastewater, 20th ed. Am. Public Health Assoc., Washington, DC.
• Harter, T., H. Davis, M.C. Mathews, and R.D. Meyer. 2002. Shallow groundwater quality on dairy farms with irrigated forage crops. J. Contam. Hydrol. 55:287–315.[Medline]
• Hendry, M.J., J.A. Cherry, and E.I. Wallick. 1986. Origin and distribution of sulfate in a fractured till in southern Alberta, Canada. Water Resour. Res. 22:45–61.
• Hendry, M.J., R.G.L. McCready, and W.D. Gould. 1984. Distribution, source and evolution of nitrate in a glacial till of southern Alberta, Canada. J. Hydrol. (Amsterdam) 70:177–198.
• Hendry, M.J., and L.I. Wassenaar. 1999. Implications of the distribution of D in pore waters for groundwater flow and the timing of geologic events in a thick aquitard system. Water Resour. Res. 35:1751–1760.
• Herbel, M.J., and R.F. Spalding. 1993. Vadose zone fertilizer-derived nitrate and ¹⁵N extracts. Ground Water 31:376–382.
• Huynh, H., and L.S. Feldt. 1970. Conditions under which mean square ratios in repeated measures designs have exact F-distributions. J. Am. Stat. Assoc. 65:1582–1589.
• Hvorslev, M.J. 1951. Time lag and soil permeability in groundwater observations. U.S. Army Corps of Eng. Waterway Exp. Stn. Bull. 36. U.S. Gov. Print. Office, Vicksburg, MS.
• Jaipaul, R. 2001. Cattle and calves on farms in Alberta [Online]. Available at http://www.agric.gov.ab.ca/economic/stats/cattle.html (verified 4 Nov. 2003). Alberta Agric., Food and Rural Development, Edmonton, AB, Canada.
• Keller, C.K., G. van der Kamp, and J.A. Cherry. 1991. Hydrogeochemistry of a clayey till. 1. Spatial variability. Water Resour. Res. 27:2543–2554.
• Komor, S.C., and H.W. Anderson. 1993. Nitrogen isotopes as indicators of nitrate sources in Minnesota sand-plain aquifers. Ground Water 31:260–270.
• Korom, S.F., and R.W. Jeppson. 1994. Nitrate contamination from dairy lagoons constructed in coarse alluvial deposits. J. Environ. Qual. 23:973–976.[Abstract/Free Full Text]
• Lerner, D.N., A.S. Issar, and I. Simmers. 1990. Groundwater recharge: A guide to understanding and estimating natural recharge. Int. Contributions to Hydrogeol. Vol. 8. Int. Assoc. Hydrogeol., Heise Publ., Hannover, Germany.
• Littell, R.C., P.R. Henry, and C.B. Ammerman. 1998. Statistical analysis of repeated measures data using SAS procedures. J. Anim. Sci. 76:1216–1231.[Abstract/Free Full Text]
• Littell, R.C., G.A. Milliken, W.W. Stroup, and R.D. Wolfinger. 1996. SAS system for mixed models. SAS Inst., Cary, NC.
• McCallum, J. 2001. The extent of denitrification in groundwater under a manured field in southern Alberta. M.S. thesis. Univ. of Calgary, AB, Canada.
• McMahon, P.B., and J.K. Bohlke. 1996. Denitrification and mixing in a stream-aquifer system: Effects on nitrate loading to surface water. J. Hydrol. (Amsterdam) 186:105–128.
• Natural Resources Conservation Board. 2001. Standards and administration regulations. Agricultural Operation Practices Act [Online]. Available at http://www.agric.gov.ab.ca/ministry/acts/regs/ar267-2001.html (verified 4 Nov. 2003). Queen's Printer Bookstore, Edmonton, AB, Canada.
• O'Brien, R.O., C.K. Keller, and J.L. Smith. 1996. Multiple tracers of shallow ground-water flow and recharge in hilly loess. Ground Water 34:675–682.
• Oldman River Basin Water Quality Initiative. 2002. Water hauler's bulletin [Online]. Available at http://www1.agric.gov.ab.ca/$department/newslett.nsf/all/whb1826?opendocument (verified 4 Nov. 2003). Alberta Agric., Food and Rural Development, Edmonton, AB, Canada.
• Olson, B.M., D.R. Bennett, R.H. McKenzie, T. Ormann, and R.P. Atkins. 1999. Manure and nutrient management to sustain soil and groundwater quality under irrigated silage barley. Alberta Agric., Food and Rural Development, Lethbridge, AB, Canada.
• Postma, D., C. Boesen, H. Kristiansen, and F. Larsen. 1991. Nitrate reduction in an unconfined sandy aquifer: Water chemistry, reduction processes, and geochemical modeling. Water Resour. Res. 27:2027–2045.
• Rehm, B., S.R. Moran, and G.H. Groenewold. 1982. Natural groundwater recharge in an upland area of central North Dakota, U.S.A. J. Hydrol. 59:293–314.
• Remenda, V.H., G. van der Kamp, and J.A. Cherry. 1996. Use of vertical profiles of ¹⁸O to constrain estimates of hydraulic conductivity in a thick, unfractured aquitard. Water Resour. Res. 32:2979–2987.
• Robertson, W.E., B.M. Russell, and J.A. Cherry. 1996. Attenuation of nitrate in aquitard sediments of southern Ontario. J. Hydrol. (Amsterdam) 180:267–281.
• Rodvang, J., and S. Riemersma. 2002. Groundwater at a buffer strip site in southern Alberta. Interim report. Irrigation Branch, Alberta Agric., Food and Rural Development, Lethbridge, AB, Canada.
• Rodvang, S.J., R. Schmidt-Bellach, and L.I. Wassenaar. 1998. Nitrate in groundwater below irrigated fields in southern Alberta. CAESA Project RES-041-93. Alberta Agric., Food and Rural Development, Lethbridge, AB, Canada.
• Rodvang, S.J., and W.W. Simpkins. 2001. Agricultural contaminants in Quaternary aquitards: A review of occurrence and fate in North America. Hydrogeol. J. 9:44–59.
• Shetsen, I. 1987. Quaternary geology, southern Alberta. 1:500,000 scale. Alberta Res. Council, Edmonton, AB, Canada.
• Spalding, R.F., and M.E. Exner. 1993. Occurrence of nitrate in groundwater—A review. J. Environ. Qual. 22:392–402.[Abstract/Free Full Text]
• Sukhija, B.S., and C.R. Shah. 1976. Conformity of groundwater recharge rate by tritium method and mathematical modeling. J. Hydrol. (Amsterdam) 30:167–178.
• Van der Kamp, G., and G. Grove. 2001. Well water quality in Canada: An overview. p. 39–42. In Proc. 2nd Joint Conf. of the Int. Assoc. of Hydrogeol. and the Canadian Geol. Soc., Calgary, AB. 16–19 Sept. 2001. Canadian Geotech. Soc., Calgary.
• Wassenaar, L. 1995. Evaluation of the origin and fate of nitrate in the Abbotsford Aquifer using the isotopes of ¹⁵N and ¹⁸O in NO₃^–. Appl. Geochem. 10:391–405.
• Zilkey, M.M. 2001. Fate of nutrients under an irrigated manured field, southern Alberta. M.S. thesis. University of Calgary, AB, Canada.

Related articles in JEQ:

This Issue in Journal of Environmental Quality

JEQ 2004 33: 413-418. [Full Text]  

This Article Abstract Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Related articles in JEQ Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via ISI Web of Science (1) Citing Articles via Google Scholar Google Scholar Articles by Rodvang, S. J. Articles by Ryan, M. C. Search for Related Content PubMed PubMed Citation Articles by Rodvang, S. J. Articles by Ryan, M. C. GeoRef GeoRef Citation Agricola Articles by Rodvang, S. J. Articles by Ryan, M. C. Related Collections Animal Waste Ground Water Quality Surface Water Quality Isotopes Nitrogen