Impacts on Water Quality in the Upper Elbow River Technology & Science Water

Impacts on water quality in the upper Elbow River Technology & Science Water

A. Sosiak* and J. Dixon** *Southern Region, Alberta Environment, 3rd Fl., 2938-11 St., NE, Calgary, AB, Canada T2E 7L7 (E-mail: [email protected]) **City of Calgary Waterworks and Wastewater, Laboratory Services, P.O. Box 2100, Stn M, Calgary, AB, Canada T2P 2M5 (E-mail: [email protected])

Abstract Recent work has found evidence of deterioration in water quality in the Elbow River upstream from Calgary, Alberta, Canada. We sampled this basin to describe spatial and temporal trends and factors 309–316 pp 10 No 53 Vol that could be contributing to this deterioration. Sources near Calgary generally contributed most of the total phosphorus (TP) entering the river, and most total suspended solids (TSS) during low to average flows. During high flows, a large influx of TSS occurred further upstream. Sources of TP could include runoff from residential developments and agriculture, and groundwater, while re-suspension of bed material, erosion, and storm sewers may contribute TSS. Increasing trends in dissolved phosphorus (TDP) and ammonia suggest that sources in this reach are also contributing dissolved nutrients. Appreciable loading of nitrate þ nitrite also occurred near Calgary, with a significant increasing trend in nitrate þ nitrite occurring at every Elbow River site and several tributaries. Fecal coliforms have increased significantly over time at a Q

downstream site. Runoff from residential developments, agriculture, or contributions from groundwater could 2006 Publishing IWA account for this trend. In the 2003 bacterial source tracking study, DNA markers from ruminant animals were found in samples from most locations sampled in this basin, even at some headwaters sites, but no human markers were found. Keywords Coliforms; deterioration; nutrients; suspended solids; water quality; watershed

Introduction The Elbow River is a relatively small river that provides drinking water to about one-half the population of Calgary, or approximately one-sixth the population of Alberta, Canada. In addition the Elbow River provides agricultural water supply, supports recreational activities, and provides habitat for fish and wildlife. Over the last 20 years, numerous studies reviewed in Sosiak and Dixon (2004) have examined water quality in Glenmore Reservoir and the upper Elbow River watershed. Beers and Sosiak (1993) found that water quality was generally good upstream of Calgary, but increases in particulate matter during peak flow conditions reduced water quality for short periods of time. They also reported few cases of any variables exceeding water quality guidelines except during peak flow. However, they found that near the mouth the Elbow River was influenced by urban stormwater and had elevated levels of various constituents. The Elbow River Watershed Study by Marshall Macklin and Monaghan (MMM, 1985) identified land use activities in the watershed that posed a potential risk to the water supply. The study found that water quality of the Elbow River and Glenmore Reservoir, although acceptable, was at risk because of continued urbanization and development in the watershed. Lewis and Seidner (1993) found an increasing trend in chlorine demand at the Glenmore Water Treatment Plant, which resulted in an increase in the production of disinfection by-products. Algal blooms of increasing intensity have been observed in recent years in the reservoir (Dixon et al., 1993; Watson et al., 1996, 2001), doi: 10.2166/wst.2006.326 309 and these blooms have led to consumer complaints about taste and odour in the treated water. Two major sources that could affect water quality in the Elbow River watershed are: (1) nonpoint source runoff from agriculture, recreation, and residential developments in the upper watershed; and (2) urban runoff from Calgary conveyed to the Elbow River and Glenmore Reservoir .Ssa n .Dixon J. and Sosiak A. by the storm sewer system. The main issues relative to water quality that were identified included: taste and odour, disinfection by-products, microbial pathogens, and turbidity. Recent work has found evidence of significant deterioration of water quality in the Elbow River upstream from Glenmore Reservoir. Sosiak (1999) found significant increases in total dissolved phosphorus, fecal coliforms, total coliforms and turbidity in the upper Elbow River at Highway 8 (Twin Bridges). Furthermore, water quality guidelines for the protection of aquatic life, irrigation and recreation were exceeded at the Twin Bridges site. No evidence of adverse impacts on the Glenmore Reservoir was found. There were then insufficient data to determine the cause of increasing trends in these variables. Accordingly, the City of Calgary and Alberta Environment (AENV) sampled the upper Elbow River basin intensively in 1999–2003, to describe spatial and temporal trends in concentration and factors that could be contributing to water quality deterioration.

Methods Mainstem and tributary sites (locations in Figure 1) were sampled from March to Septem- ber, 1999 to 2002, by AENV and The City of Calgary, usually on a bi-weekly or monthly basis. During each sampling visit, water temperature, dissolved oxygen, and conductivity were measured in the ﬁeld. Grab samples were also taken for pH, total organic carbon, the various forms of phosphorus and nitrogen, total suspended solids (as nonﬁlterable residue), and coliform bacteria. Most sampling was conducted during the open water

310 Figure 1 Elbow River and tributary sampling locations season because previous work indicated this was the period of greatest contaminant loading to the Elbow River and its tributaries. For AENV samples, Maxxam Analytics Inc. performed all physical and chemical analyses, and the Provincial Health Laboratory for Southern Alberta provided all coliform counts. For the City of Calgary, Laboratory Services, Waterworks and Wastewater, conducted all physical, chemical, and microbiolo- gical analyses. To better sample during periods of peak loading, additional daily composite samples Dixon J. and Sosiak A. were collected during the anticipated period of peak snowmelt and river flow at key Elbow River sites. Composite samples for TP and TSS analysis were collected on alternate days every two hours by AENV using automated samplers (ISCO 6700) during April 7 to July 8, 1999, May 1 to August 1, 2000, and April 15 to August 1, 2002. Automated samplers were installed in the Elbow River just upstream of the Hamlet of Bragg Creek, at Highway 22 Bridge, at the Glencoe Golf and Country Club (Glencoe GCC), at Twin Bridges, and at Weaselhead Bridge. Sample intakes were anchored mid-water column, several metres from the bank in the main river flow. The Laboratory for Foodborne Zoonoses, Population and Public Health Branch, Health Canada, Animal Disease Research Institute in Lethbridge, Alberta, tested water samples collected by AENV on four dates in 2003 (May 21, July 15, September 10, and Decem- ber 2) for Bacteroides DNA markers following methods in Bernhard and Field (2000), and the important pathogens E. coli O157:H7, Salmonella, and Campylobacter.In addition, samples were tested for E. coli, fecal coliforms, and Enterococcus (in July only) to provide other standard measures of fecal contamination. This bacterial source tracking work was designed to determine whether coliforms in the upper Elbow watershed are mainly from humans, for example through septic leachate, or from ruminant animals. A wide range of wild and domesticated animals that occur in this basin are ruminants, including cattle, bison, sheep, elk, and deer. To determine where deteriorating water quality has occurred in the Elbow basin, sites with at least four years of data for each variable were tested for monotonic trends, which indicate gradual increasing or decreasing concentration. Variables exhibiting significant seasonality were tested for monotonic trends using the seasonal Kendall test, with or without correction for significant serial correlation. Data that did not display significant seasonal variation were tested for monotonic trends using the Mann–Kendall test. The mass flux of key chemical variables was estimated for sites sampled intensively from Bragg Creek to Weaselhead Bridge, where daily flow estimates were available. For each site, the mass flux of each variable was estimated by six different methods using the computer program FLUX 4.5 (Walker, 1996).

Results and discussion The results of this study provide the most detailed analysis of surface water quality in the upper Elbow River to date. The key conclusions of this study are the following. (1) Sources near Calgary (Twin Bridges to Weaselhead Bridge) generally contributed most of the total phosphorus (TP) mass entering the Elbow River (Figure 2). Small increasing trends in total dissolved phosphorus (TDP) concentration and ammonia (Table 1) suggest that sources in this reach are also contributing dissolved nutrients. Sources could include runoff from residential developments, agriculture, and groundwater. No evidence of trends in tributary concentration that could account for these increasing trends in TDP was found (Table 1), but TP and TDP concentrations were much higher in tributaries than in the Elbow River (Figure 3). Concentrations from all sites in each reach have been plotted together in these graphs. 311 .Ssa n .Dixon J. and Sosiak A.

Figure 2 Mass ﬂux of total phosphorus from the Elbow River and tributaries

(2) Appreciable loading of nitrate þ nitrite occurred between Twin Bridges and Weaselhead in 1999 and 2002 (Figure 4). Sources could include loadings from storm sewers, atmospheric deposition, agriculture, groundwater, and Lott Creek. A signiﬁcant increasing trend in nitrate þ nitrite concentration occurred at every Elbow River site from Bragg Creek to the Weaselhead Bridge and several tributaries (Table 1), but the reason for these trends is not understood. Tributaries between Highway 22 and the Glencoe GCC were major sources of nitrate þ nitrite (Figure 4), in particular Pirmez Creek, which contributed more nitrate þ nitrite than any other tributary (Sosiak and Dixon 2004). However, nitrate þ nitrite was well below Canadian water quality guidelines (Figure 5).

Table 1 Sen slopes (units/yr) for signiﬁcant monotonic trends in variables sampled in the Elbow River and tributaries

Site TP TDP NO3 1 NO2 NH3 TKN TSS Turb TC FC mg/L mg/L mg/L mg/L mg/L mg/L NTU no/100 mL no/100 mL

Mainstem Cobble Flats ns ns ns – – – – þ1.1 – Above Bragg Creek – – þ1.7 2,0.01 ns 149.6 ns ns ns Hwy 22 Bridge 21.0 ns þ1.6 ns ns ns ns ns ns Glencoe GCC ns ns þ2.3 ns ns þ412.2 – – ns Twin Bridges ns þ , 0.01 þ1.4 þ , 0.01 ns ns ns þ6.1 þ0.5 Weaselhead Bridge 23.4 þ , 0.01 þ1.9 ns ns ns þ0.21 þ39.6 ns Tributaries Little Elbow ns ns þ2.3 ns – ns ns þ0.3 – River McLean Creek ns – ns 2,0.01 – ns ns þ30.2 Bragg Creek ns ns þ3.6 ns 26.7 ns ns ns ns Pirmez Creek ns ns – – – ns – – – Millburn Creek ns ns – – – ns – – – Springbank ns ns – – – ns – – – Creek Lott Creek ns 20.3 ns ns ns ns ns þ71.7 ns

Abbreviations: TP (total phosphorus), TDP (total dissolved phosphorus), NO3 þ NO2 (nitrate þ nitrite nitrogen), NH3 (ammonia nitrogen), TKN (total Kjeldahl nitrogen), TN (total nitrogen), TSS (total suspended solids), Turb (turbidity in nephelometric turbidity units), TC (total coliform bacteria), and FC 312 (fecal coliform bacteria). ns ¼ no signiﬁcant trend .Ssa n .Dixon J. and Sosiak A.

Figure 3 Phosphorus concentrations in the Elbow River and tributaries. Bars in boxes are medians, tops and bottoms of boxes represent the 25th and 75th percentiles, and whiskers represent the minimum and maximum values

Figure 4 Mass ﬂux of nitrate þ nitrite (kg per sampling season) from tributaries and non-point sources for Elbow River reaches in 1999, 2000 and 2002

(3) Sources between Twin Bridges and Weaselhead Bridge contributed most total suspended solids (TSS) entering the Elbow River during low to average flows (Figure 6). Possible sources include the re-suspension of bed material, bank erosion, and storm sewers. TSS loadings from Elbow River tributaries were too low to plot on Figure 6. During the high flows in 2002, a large influx of TSS mass occurred between Bragg Creek and Highway 22 (Figure 6). Bank erosion at various sites and other potential sources in that reach were identified during an aerial survey in November 2002. The spatial pattern of TSS loading and deposition changed considerably from year to year depending on flows and other factors (Figure 6). (4) The greatest change in E. coli counts in the Elbow River occurred each year in the lower reach of the study area, ending at Weaselhead Bridge (Figure 7); fecal coliforms have increased significantly over time (1979–2002) at Twin Bridges alone (Table 1). Runoff from residential developments and agriculture, or contributions from groundwater could account for this increasing trend. Elbow River

Figure 5 Nitrate þ nitrite concentrations in the Elbow River and tributaries 313 .Ssa n .Dixon J. and Sosiak A.

Figure 6 Mass ﬂux of total suspended solids (kg per sampling season) from tributaries and non-point sources for Elbow River reaches in 2000 and 2002

Figure 7 Escherichia coli counts in the Elbow River and tributaries

tributaries generally had higher E. coli and fecal coliform counts than the mainstem sites in low to average ﬂow years (Figure 7). Fecal coliform counts in Pirmez Creek and Millburn Creek often exceeded water quality guidelines (Figure 7). DNA markers from ruminant animals were found in samples at all but four locations in the Elbow River basin on at least one sampling date in 2003, even in the headwaters at Cobble Flats (Figure 8). Ruminant markers were most often detected in Elbow River samples collected in June, while markers were often found in tributary samples in May and September. No human markers were found at any sampling site. Of the human pathogens for which samples were tested, there was only one positive result for Salmonella. For further information on the results of this study, refer to Sosiak and Dixon (2004). Results of this study did not provide consistent evidence of impacts on surface water quality in the Elbow River from the Hamlet of Bragg Creek. However, recent work by

Figure 8 Incidence of DNA markers for ruminant and humans each month in May, July, September and 314 December 2003 in the Elbow River and its tributaries researchers at the University of Calgary (Manwell, 2005) has determined that sources from the Hamlet of Bragg Creek contributed about 10% of the chloride in the downstream Elbow River, and stressed the need to consider land uses on the river-connected alluvial aquifer when selecting river sampling sites. The Elbow’s alluvial aquifer is comprised mainly of permeable sands and gravels, and is hydraulically well connected to the Elbow River. The location of any new land uses that have the potential to affect the shallow groundwater (and by corollary the river water) quality should be carefully Dixon J. and Sosiak A. considered, in particular upgradient of the Glenmore Reservoir.

Recommendations Bank erosion. The upper Elbow River is naturally prone to bank erosion. Its banks are readily eroded in various locations during high flows, which occur during years with high runoff from mountain snowmelt. Analysis of water quality trends has detected increased turbidity and TSS levels in recent decades. During the same time period, necessary channel modifications have been made at various locations, to improve bank stability and reduce erosion. However, erosion continues at some of these sites and downstream, with turbidity levels in the Elbow River increasing. This suggests that the current approach to bank protection has not proved effective. Other approaches to erosion protection should be considered. For example, restrictions on development of the flood plain and the flood fringe have been used elsewhere to pro- tect naturally unstable channels like the upper Elbow River. Development in vulnerable sections of the Elbow River flood plain and flood fringe should be restricted.

Agricultural impacts. Except for nitrogen and other nonpoint impacts along the mainstem Elbow, this study found little evidence that agricultural activities on tributaries have had a signiﬁcant impact on water quality in the Elbow River. Indirect agricultural impacts on the Elbow River via groundwater were not assessed in this study, but could be an important pathway. However, impaired water quality for irrigation and contact recreation was found in some tributaries and some of this impairment probably reﬂects current agricultural practices. The Farmers of the Elbow Watershed (FEW) have implemented a program to improve agricultural practices and reduce water quality impacts in the Elbow basin. Their efforts to improve agricultural practices on tributary streams should be supported and enhanced.

Urban runoff. Sampling during low to average ﬂows provided evidence of appreciable TSS, TP, and coliform loadings from sources near Calgary. The bulk of this loading is probably from urban sources, including various storm sewers that discharge to this reach. The individual contribution of current and future storm sewer systems needs to be carefully assessed, and improved where feasible and cost-effective. The adoption of best management practices by stakeholders within the watershed can also help address this issue and should be encouraged. This could include low impact development practices such as retaining runoff on site, or improved methods of sewage disposal.

Monitoring. The bacterial source tracking study in 2003 was completed during a relatively dry summer. The sampling should be repeated during a wetter year, when there could be more movement of septic leachate into streams and rivers. The protozoan parasites Giardia and Cryptosporidium are currently sampled by the City of Calgary at several sites in the Elbow watershed. More intensive sampling throughout the watershed would help to determine the relative contributions of agriculture, wildlife, and urban 315 sources. Current monitoring of the upper Elbow basin should continue, as this will allow an ongoing evaluation of improvements under these recommendations.

Conclusions Sources near Calgary contributed much of the TP and nitrate þ nitrite entering the Elbow River, and most TSS during low to average flows. Fecal coliforms have also .Ssa n .Dixon J. and Sosiak A. increased significantly over time at a downstream site. During high flows, a large influx of TSS occurred further upstream. Individual sources of these constituents cannot be determined from this study, but sources of TP, nitrate þ nitrite, and fecal coliforms could include runoff from residential developments, agriculture, and groundwater, while re-suspension of bed material, erosion, and storm sewers may contribute TSS. Increasing trends in TDP and ammonia suggest that sources in this reach are also contributing dissolved nutrients. Although some tributaries contributed appreciable nitrate þ nitrite, they were minor sources of TSS and this study found little evidence that agricultural activities on tributaries have affected water quality in the Elbow River. In the 2003 bacterial source tracking study, DNA markers from ruminant animals were found in samples from most locations sampled in this basin, even at some headwaters sites, but no human markers were found. Accordingly, these results provide no evidence that coliforms in the upper Elbow River were from human sources such as septic leachate.

References Beers, C. and Sosiak, A. (1993). Water quality of the Elbow River. Environmental Assessment Division, Alberta Environmental Protection, Calgary, Alberta, Canada. Bernhard, A.E. and Field, K.G. (2000). A PCR assay to discriminate human and ruminant feces on the basis of host differences in Bacteroides-Prevotella genes encoding 16S rRna. Appl. Environ. Microbiol., 66(10), 4571–4574. Dixon, R.W.J., Hardisty, H.M., McCauley, E. and Hargesheimer, E.E. (1993). Three-tiered approach to tracking taste-and-odour events. Proceedings of the American Water Works Association Annual Conference. San Antonio, TX, USA, pp. 440–444 Lewis, C.M. and Seidner, R.T. (1993). Impact of disinfection strategy on distribution system water quality. Calgary’s experience. Proceedings of the 5th National Conference on Drinking Water. Winnipeg, Manitoba, Canada. Manwell, B. (2005). Groundwater – surface water interaction and water quality in the Lower Elbow River, Alberta MSc. thesis, Department of Civil Engineering, University of Calgary, Alberta, Canada. Marshall Macklin and Monaghan Ltd (MMM) (1985). Elbow River Watershed Study. Phase 2 Report. Volumes 1 to 3. Prepared for the City of Calgary, Alberta, Canada. Sosiak, A. (1999). Evaluation of recent trends in water quality in the Elbow River upstream from Glenmore Reservoir. Alberta Environment, Water Management Division, Calgary, Alberta, Canada. Sosiak, A. and Dixon, J. (2004). Impacts on water quality in the upper Elbow River. Prepared by Southern Region, Alberta Environment and the City of Calgary (available at www3.gov.ab.ca/env/info/infocentre/ publisting.cfm). Walker, W.W. (1996). Simpliﬁed procedures for eutrophication assessment and prediction: user manual. Prepared for the U.S. Army Corps of Engineers. USAE Waterways Experiment Station. Vicksburg, MI, USA. Instruction Report W-96–2. Watson, S.B., McCauley, E., Hardisty, H. and Dixon, J. (1996). Chrysophyte blooms in oligotrophic Glenmore Reservoir (Calgary, Canada). Nova Hedwig., 114, 193–217. Watson, S.B., Satchwill, T. and McCauley, E. (2001). Under-ice blooms and source-water odour in: a nutrient-poor reservoir: biological, ecological, and applied perspectives. Freshwat. Biol., 46, 1–15.

316