Abstract. Introduction

Transcription

1 Abstract Nutrient Loading Patterns on an Agriculturally Impacted Stream System over Three Summers Elizabeth Diesel, Caitlan Zlatos, Ryan Mathur, and Evan Teeters Dept. of Geology, Juniata College, Huntingdon PA Spruce Creek and Warrior's Mark Run were studied from 2002 to 2004 to determine the effects of agricultural nutrient loading on water quality. These two streams are part of the Spruce Creek watershed, which contains both the state's premier trout fishery and largest Concentrated Animal Feeding Operation. The summers of 02, 03, and 04 represent different hydrologic conditions of the Spruce Creek watershed. The summer of 02 is typified by drought conditions and an overall decline in discharge, the summer of 03 is characterized by rainy conditions and variable discharge, and the summer of 04 is characterized as a combination of the hydrologic conditions of 02 and 03. Patterns in the nitrate concentrations behaved similarly during 03 and the start of 04, and similarly during 02 and the end of 04. During 03 and 04, increases in discharge created a decrease in nitrate concentration. This indicates that nitrate dilution occurs during discharge events. But during 02 and 04, decreases in discharge created small increases in nitrate concentration. The patterns in the phosphate concentrations are more complex than those of the nitrates. Phosphate concentrations are related to discharge, but each summer shows a different pattern in relation to concentration and discharge. Increases in discharge also caused large increases in turbidity during 03 and the start of 04. This increase in turbidity occurred along side a great increase of phosphate and nitrate concentrations. This correlation implies that increases in turbidity will increase nitrate and phosphate concentrations due to nutrients in sediments washed off of farming fields. The headwaters of Warrior s Mark Run were sampled during the summer of 04 and analyzed for nutrient concentrations. Phosphate concentrations for the headwater samples were comparable to concentrations found along the rest of the stream, indicating a source of phosphate that affects the water before agriculture is prevalent in the area. This source is likely the stream's underlying bedrock. concentrations at the headwaters are effectively zero, whereas the concentrations downstream are much higher. This increase in nitrate concentration points to an important source of nitrate downstream from the headwaters of Warrior s Mark Run, which could be an agricultural source. Introduction Spruce Creek is one of the many spring-fed, limestone streams flowing through central Pennsylvania. The stream is about seven miles long, and its waters provide an exceptional habitat to the large rainbow and brown trout for which it is known. Over the years, those who live along its banks and those who frequent its waters have noticed a

2 dramatic decline in the quality of Spruce Creek s fishery. The suspected reason for this decline in quality is nutrient loading from the agriculture that is found on the banks of Spruce Creek and its tributaries. In particular, there is a Concentrated Animal Feeding Operation (CAFO) found along the banks of Spruce Creek that is thought to influence water quality. In 2001, residents of the Spruce Creek watershed contacted Juniata College concerning monitoring the health of this stream, on which they rely for their income. This concern for the health of Spruce Creek prompted this baseline study. As can be seen in Figure 1, many sampling sites were chosen along Spruce Creek and Warrior s Mark Run in order to best capture an overall nutrient profile of the stream system. Site SGR was chosen as a generally unaffected site because it is upstream from the CAFO. Downstream from SGR are sites SFR, WSC, SGW, and SLJ. Site SFR is directly downstream from the CAFO. Sites WSC, SGW, and SLJ are downstream from the CAFO and the confluence of Spruce Creek and Warrior s Mark Run. Along Warrior s Mark Run, eight sampling sites were chosen to best represent the nutrient profile of the stream (see Figure 1). Sites HUS and HDS were chosen to represent the agriculturally unaffected headwaters of Warrior s Mark Run. Sites WCP and WCL were chosen to represent each of the agriculturally impacted tributaries to Warrior s Mark Run. Downstream from WCP and WCL are sites WBR, WTF, WMR, and WSC. These sites are both downstream from and flowing through agriculturally impacted areas. Materials and Methods Sampling Procedures Water samples were collected five days a week between the end of May and the beginning of August, and two days a week during the remaining months. In May and August, the time of sampling was kept constant, with sampling beginning at 9:00 am and ending at 11:00 am. From late August to early May, sampling usually began around 1:00 or 2:00 pm and ended around 3:00 or 4:00 pm. Sampling times were kept consistent to ensure that varied times of sampling did not affect our results. Parameters examined in the field include temperature, total dissolved solids, and discharge. Temperature and total dissolved solids measurements were collected using a HACH Sension5 portable meter, and discharge was calculated using a Global Water FP101water flow probe. The water samples were collected using 250 ml glass bottles and were kept in a cooler until they were ready for testing. Laboratory Procedures All water samples were tested in the laboratory for ph, turbidity, nitrates as N-NO3-, and phosphates and PO43-. ph was measured using a HACH Sension1 portable meter. Turbidity was measured using the HACH DR2000 direct reading spectrophotometer. s were measured using the HACH DR2000 with a cadmium reduction method, and

3 phosphates were measured using the HACH DR2000 with a sulfuric acid reduction method. Sampling Sites Streams Roadways Agricultural Areas Huntingdon County Pennsylvania WCP Warrior s Mark Run WCL WBR WTF SF WSC WMR Spruce Creek SGW LJR SLJ SGR Figure 1: Map of Sampling Sites Results and Discussion Hydrologic differences between summers As can be seen in Figure 2, the summers of 2002, 2003, and 2004 were significantly different in terms of hydrologic conditions. The summer of 2002 was a dry summer, with discharges well below 150 cfs. The average discharge for this summer was 34 cfs. The summer of 2003 was a wet summer, with discharges as high as 650 cfs, and an average discharge of 201 cfs. The summer of 2004 was a median between the wet and dry summers with discharges between 125 and 225 cfs; the average discharge for this summer was 155 cfs.

4 Discharge vs. Time (SGR) Discharge (cfs) Summer 2002 Summer 2003 Summer Time (days) Nutrient Transport Figure 2: Discharge at SGR over three summers The concentration of nitrate present within the stream system is dependent upon amount of water input into that system (Viney, et al, 2000). When comparing discharge trends to nitrate concentration trends, it can be seen in Figure 3 that during the summer of 2002, low discharges, with an average of 34 cfs, coincided with high nitrate concentrations, with an average of 10 mg/l. During the summer of 2003, highly variable discharge rates, ranging from 100 to 650 cfs, caused a large amount of fluctuation within the nitrate concentrations between 3 and 12 mg/l with an average of 8.39 mg/l. The summer of 2004 shows a moderate variability in the discharge rate, ranging from 125 to 225 cfs, and an almost daily fluctuation in the nitrate concentration between 2.75 and 9.75 mg/l with an average of 6.45 mg/l. Phosphate concentrations are also dependent on the amount of water input into the stream system, as can be seen in Figure 4. By comparing discharge rates to phosphate concentrations for the summer of 2002, it can be seen that the low discharge rates correlate with low to absent phosphate concentrations, the average of which is essentially 0 mg/l. The summer of 2003 showed a high degree of variability for both discharge rates and phosphate concentrations, which varied between 0.05 and 0.8 mg/l. During the summer of 2004, there were moderate fluctuations in discharge and phosphate concentrations, with phosphates ranging from 0.05 to 0.25 mg/l with an average of 0.12 mg/l.

5 s over Three Summers (SGR) s (mg/l) Summer 2002 Summer 2003 Summer Time (days) Figure 3: Concentrations at SGR over three summers Phosphates over Three Summers (SGR) Phosphates (mg/l) Summer 2002 Summer 2003 Summer Time (days) Figure 4: Phosphate Concentrations at SGR over three summers

6 It is possible that the differences in nitrates and phosphates are due to different sources for the nutrients. The source of phosphates could be fertilizers whereas the source of nitrates could be animal wastes. Another possible explanation is that nitrates and phosphates enter the stream system differently. Our data indicates that nitrates are present in the system regardless of discharge but phosphates appear to enter most noticeably during increases in discharge. It is possible then that nitrates enter the system through groundwater inputs, whereas phosphates enter the system through overland flow. As illustrated in Table 1, increases in turbidity typically coincide with increases in discharge, which usually signify a rain event. Increases in turbidity typically coincide with decreases in total dissolved solids, which should also create a decrease in nitrate and phosphate concentrations, being that nitrates and phosphates are dissolved species. On the contrary, nitrate and phosphate concentrations increase as turbidity increases. This would suggest that the increased sediment load is likely responsible for the increases in nutrient concentrations. This sediment load likely contains many particles that have been recently washed off of agricultural fields, like particles of fertilizer and animal waste. These particles, while not dissolved, will still add nitrates and phosphates to the stream system. Turbidity Average Conc in mg/l Average Phosphate Conc in mg/l FTU FTU FTU FTU FTU FTU Table 1: Ranges of Turbidity and the average and Phosphate Concentrations that correspond CAFO vs. Traditional Agriculture As can be seen in Table 2(below), average nitrate concentrations along Spruce Creek and Warrior s Mark Run are comparable between all sampling sites, although the discharges vary greatly from site to site. This table also illustrates that the total load of nitrates moving through the stream system varies between sampling sites.

7 No Agriculture Average Concentration (mg/l) Average Discharge (cfs) flow through system (kg of dissolved NO3-/day) HUS Traditional Agriculture WCP WCL WBR WTF WSC CAFO Agriculture SGR SFR Table 2: Comparison of Average Concentration and Discharge on the Headwaters of Warrior s Mark Run (No Agriculture), Warrior s Mark Run (Traditional Agriculture), and Spruce Creek (CAFO Agriculture) for summer 2003 Even though nitrate concentrations are similar on Spruce Creek and Warrior s Mark Run, the total amount of nitrate in kilograms per day varies greatly between the two streams. The reason for this variation is that Spruce Creek, with its larger discharge, is able to dilute its total nitrate load to concentrations that are comparable to those on Warrior s Mark Run. This indicates that the agriculture along Spruce Creek is contributing a larger total nitrate load into the stream system than Warrior s Mark Run. Warrior s Mark Run Headwaters and Stream waters The Warrior s Mark Run headwater samples consistently contain a lower concentration of nitrate than the Warrior s Mark Run stream water samples, as seen in Table 3. 6/8/2004 6/22/2004 7/29/2004 HUS 0.89 mg/l 0.44 mg/l 0.44 mg/l HDS 0.89 mg/l 0.44 mg/l 0.44 mg/l WCP 5.54 mg/l 9.52 mg/l 7.53 mg/l WCL 4.43 mg/l 8.20 mg/l 7.53 mg/l WBR 5.54 mg/l mg/l 9.97 mg/l WTF 2.22 mg/l 6.42 mg/l 8.20 mg/l WMR 3.52 mg/l 6.82 mg/l 7.92 mg/l Table 3: Comparison of nitrate concentrations of Warrior s Mark Run headwater spring samples (WHU & WHD) with Warrior s Mark Run stream samples (WCP, WCL, WBR, WTF, WMR)

8 The headwaters of Warrior s Mark Run consistently contain a lower concentration of nitrate than its stream waters. This indicates that there is a significant source of nitrate directly downstream from the headwaters. Water interaction with bedrock in stream bottoms is very unlikely; the rock units that underlie the stream are typically limestone and dolomite, with an occasional shale or sandstone unit. These rocks do not have the chemical makeup necessary to add such significant concentrations of nitrate to the stream system. The likely source of this nitrate is from the surrounding agricultural land use. Phosphate concentrations of the Warrior s Mark Run headwaters typically have similar concentrations as the stream waters, as seen in Table 4. Therefore there is a significant source of phosphates near the headwaters. This source could be the underlying bedrock, the Bald Eagle sandstone and conglomerate, which contains phosphatic minerals. 6/8/2004 6/22/2004 7/29/2004 HUS 0.06 mg/l 0.11 mg/l 0.06 mg/l HDS 0.08 mg/l 0.16 mg/l 0.06 mg/l WCP 0.13 mg/l 0.25 mg/l 0.15 mg/l WCL 0.11 mg/l 0.24 mg/l 0.19 mg/l WBR 0.09 mg/l 0.20 mg/l 0.16 mg/l WTF 0.10 mg/l 0.13 mg/l 0.13 mg/l WMR 0.09 mg/l 0.12 mg/l 0.14 mg/l Table 4: Comparison of phosphate concentrations of Warrior s Mark Run headwater spring samples (WHU & WHD) with Warrior s Mark Run stream samples (WCP, WCL, WBR, WTF, WMR) Effects of Nutrient Input on Water Quality The highest and lowest nitrate concentration for each sampling spot varied from summer to summer, as seen in Table 5. All nitrate concentrations are higher than the EPA EcoRegion XI standard for stream water (EPA, 2000). levels for ambient water should be 0.31 mg/l or less to ensure a healthy stream system (EPA, 2000). During the summers of 2002, 2003, and 2004, there was not a single nitrate concentration that was at or below this level. Indeed there is a water quality problem on the Spruce Creek Watershed Highest 2002 Lowest 2003 Highest 2003 Lowest 2004 Highest Concentration Concentration Concentration Concentration Concentration 2004 Lowest Concentration EPA Standard in mg/l in mg/l in mg/l in mg/l in mg/l in mg/l in mg/l SGR SFR SGW SLJ LJR WSC

9 WMR WTF WBR WCL WCP Table 5: Highest/Lowest Concentration for Sampling Sites over Three Summers Summary and Conclusions There is a definite problem with the water quality of the Spruce Creek stream system; nutrient concentrations in the water are far too high to sustainably support a trout population. Implementation of Best Management Practices (BMPs) along Spruce Creek and Warrior s Mark Run could help to reduce the effects of agricultural pollution on these coldwater stream systems (Wang, et al, 2002). An easier and less expensive alternative to a BMP would be to use woody draws, or small drainage areas covered by trees and shrubs to lessen runoff and other agricultural inputs (Qiu, et al, 2002). Acknowledgements Juniata College, Geological Society of America, Dennis Thompson, Alan Bright, Lori O Connell References [EPA] Environmental Protection Agency Dec. Ambient water quality criteria recommendations: Information supporting the development of state and tribal nutrient criteria: Rivers and streams in nutrient ecoregion XI. < Accessed 2004 Mar 16. Qiu Z, Prato T, Godsey L, Benson V Integrated assessment of uses of woody draws in agricultural landscapes. JAWRA 38: Viney N, Sivapalan M, Deeley D A conceptual model of nutrient mobilisation and transport applicable at large catchment scales. J of Hydrology 240: Wang L, Lyons J, Kanehl P Effects of watershed best management practices on habitat and fish in Wisconsin streams. JAWRA 38: