Human Impacts on the Water Quality of College Brook: A Fall 2011 reach study of College Brook in Durham, NH
Water quality is an important topic for terrestrial and aquatic ecosystems. Ultimately water quality impacts quantity. Humans impact water quality with land use activities such as agriculture, suburban/urbanization, and industry. A reach scale study was used to determine the human impacts on the water quality in two differently impacted portions of College Brook; a farm and athletic field impacted upper reach and a culvert and impervious surfaces impacted lower reach. The dissolved organic nitrogen was higher in the upper reach due to dairy farm runoff, and the nitrate levels were higher in the lower reach due to activity in the culvert. Dissolved organic carbon was higher in the upper reach due to decomposition in the culvert, ammonium behaved the same in both reaches but increased underground, phosphate levels were the same in both reaches due to a saturation from fertilizers, and chlorine concentrations and specific conductivity were higher in the lower reach due to runoff from impervious surfaces. The pH was similar in both reaches, but was lowest after it came out of the culvert. Temperature was higher in the lower reach due to warming in the culvert. The amount of total suspended solids was the same in each reach. The lower reach had lower dissolved oxygen due to warmer temperatures and conversion to CO2 during decomposition in the culvert.
Water is vital to survival. The quantity of freshwater available for human use is of great importance. Many people do not realize that water quantity is closely linked to water quality. Water contaminated by pollution and heavy nutrient loading reduces the quantity of available freshwater for drinking, agriculture, and industry (Peters & Meybeck, 2000). Peters and Meybeck (2000) argue that every land use decision is ultimately a water resources decision. Water quality is a combination of natural and anthropogenic factors (Mouri, Takizawa, & Oki, 2011). The anthropogenic impact on water quality is increasing, and human impacts are an important factor impacting freshwater quality (Allan & Castillo, 2007; Mouri et al., 2011; Peters & Meybeck, 2000).
Human activates such as agricultural practices, suburbanization and urbanization, placement of impervious surfaces, road salting, human waste disposal and industrial water use negatively affect water quality (Allan & Castillo, 2007; Benbow & Merrit, 2005; Berner & Berner, 1987; Malmqvist & Rundle, 2002; Mouri et al., 2011; Price & Leigh, 2006; USEPA, 2000; Whitehead, Wilby, Battarbee, Kernan, & Wade, 2009). The use of fertilizer on agricultural fields increases nitrogen and phosphorus in aquatic ecosystems (Allan & Castillo, 2007; Mouri et al., 2011; Peters & Meybeck, 2000; Price & Leigh, 2006). Nutrient loading leads to increased photosynthetic activity by algae and aquatic plants (Allan & Castillo, 2007; Mouri et al., 2011; Price & Leigh, 2006). When algae blooms die they decompose, thus decreasing dissolved oxygen (DO) (Malmqvist & Rundle, 2002).
Impervious surfaces increase nutrient loading by being quick conduits to streams (Whitehead et al., 2009). Road salt, sediments, and pet waste are rapidly delivered to aquatic systems due to impervious surfaces (Mouri et al., 2011; Price & Leigh, 2006). The application of road salt is increasing the conductivity of streams and rivers; 28% of sodium in rivers is from humans (Berner & Berner, 1987). Benbow and Merrit (2005) estimate that the US uses between 10-15 million tons of road salt a year. High conductivity is detrimental to aquatic life and can lead to un-potable drinking water (Allan & Castillo, 2007). Sediments wash off impervious surfaces into streams and rivers, which increases the total suspended sediment load (TSS) (Mouri et al., 2011; Price & Leigh, 2006). Increased TSS leads to decreased drinking water quality, increased turbidity, and less light availability for photosynthesis.
Water is often used for cooling factories and power plants. The heated water is put back into streams and rivers. Water temperature controls chemical and biological reaction rates, as well as having a control on pH (Peters & Meybeck, 2000; Whitehead et al., 2009). The rate of reactions and pH are essential to the health of aquatic ecosystem.
Humans impacts on all scales influence water quality (Peters & Meybeck, 2000). Understanding the human impacts on water quality will lead to an understanding of the overall health of an aquatic system. This study looks at a reach scale of the human impacts on water quality of College Brook in Durham, NH. The upper reach of College Brook has human impacts from dairy farming and athletic fields, whereas the lower reach has human impacts from a long culvert of the stream and runoff from impervious surface. The study is based on the hypothesis that both streams will show impacts from humans, but the human activities on the lower reach will lead to higher phosphate, dissolved organic nitrogen (DON), nitrate, ammonium, dissolved organic carbon (DOC), pH, water temperature, specific conductivity, and TSS. The lower reach is expected to have a lower DO than the upper reach.
College Brook is a small headwater stream that runs through the campus of the University of New Hampshire in Durham, NH. The UNH Fairchild Dairy farm drains into the headwaters of College Brook. The stream flows through culverts under two roads before flowing past fields and woods. The stream goes under campus for 200m and then flows out from under Spaulding Hall (Figure 1 and Figure 2). From here College Brook flows through a forested riparian zone on campus until it reaches the Durham Market Place parking lot before going underground again. College Brook is a series of pools, runs, and riffles.
The study looked at the 200m reach upstream of campus and a 200m reach below Spaulding Hall. The upper and lower reach were marked every 20m for data collection. The upper reach of college brook (UCB) is wooded, but drains a dairy farm, parking lots and athletic fields. The lower reach of college brook (LCB) is also wooded, but drains out of a 200m culvert and receives runoff from the impervious surfaces of campus.
Discharge in College Brook is variable over time. The overall trend from the beginning of September until early November is an increase in discharge (Figure 3). The dimensions of UCB and LCB are shown in Figure 4.
Continuous water depth was measured using a submerged Hobo pressure sensor. The Hobo logger was corrected for variations in atmospheric variation. A rating curve was created using depth and discharge measurements taken on 6 days for LCB (Sept. 7, 9, 16, 23, 30, and Oct. 7, 2011) and 8 days for UCB (Sept. 7, 9, 23, 30, and Oct. 7, 17, and twice on the 21st) (Figure 5). The rating curve was applied to the water depth measurements to get a flow record. A hydrograph was created from this data. Details can be found in Lab 0 (Wollheim, 2011a).
The stream dimensions were taken on each reach at 20m intervals. The cross sectional area was measured for width and depth and then averaged for each reach (Figure 4). Procedure details can be found in Lab 1 (Wollheim, 2011b).
Nutrient grab samples were collected at 0m UCB, 200m UCB, 0m LCB and 200m LCB on September 16, 2011. Two samples were collected at each site using the method described in section 3 of Lab 3 (Wollheim, 2011c). Due to lab constraints only one bottle from each site was analyzed in Bill McDowell’s Water Quality Lab.
Temperature, Specific Conductivity, pH, and Dissolved Oxygen
A Hydrolab data logger was used to measure temperature (oC), specific conductivity (uS/cm), pH, and DO (% and mg/L). The Hydrolab was placed in the stream every 20m along each reach. Procedure details can be found in section 2 of Lab 3 (Wollheim, 2011c).
Four particulate samples were taking in each reach, two samples at 0m UCB and LCB and two samples at 200m UCB and LCB. Each bottle was rinsed with stream water 3 times and filled as described in Lab 3 (Wollheim, 2011c). Each bottle was then processed for particulates. The particulate sampling process is described in detail in Lab 3 (Wollheim, 2011c).
Student’s T-tests were performed in order to compare differences between UCB and LCB. The function used was T.Test in Microsoft® Excel® 2008 for Mac.
Graphs of DON, nitrate, phosphate, DOC, chloride, and ammonium concentration are presented in Figure 6. DON was steady in UCB at ~0.31 mg N/L, but dropped between 200m and 400m to 0.097 mg N/L, and increased to 0.12 mg N/L from 400m to 600m. The nitrate concentration was similar at 0m to 200m, rising from 0.61 mg N/L to 0.65 mg N/L. Between 200m and 400m, the nitrate concentration increased to 1.07 mg N/L and then decreased to 0.97 mg N/L at 600m. The phosphate concentration from 0m to 600m was ~0.2 mg P/L, with a minimum of 0.01734 mg P/L and a maximum of 0.02105 mg P/L. DOC concentration decreased from 0m to 200m, and ranged from 5.16 mg C/L to 5.08 mg C/L. A drop occurred between 200m and 400m, with DOC going to 3.97 mg C/L. The level decreased at 600m to 3.88 mg C/L. The chloride concentration was 201.05 mg Cl/L at 0m, it increased at 200m to 208.27 Cl/l. It rose again at 400m to 266.59, and continued to increase to 284.75 mg Cl/L at 600m. The ammonium concentration started at 0.032 mg N/L at 0m and decreased to 0.015 mg N/L at 200m. It increased to 0.043 mg N/L at 400m, and decreased to 0.021 mg N/L at 600m.
Temperature, Specific Conductivity, pH, and Dissolved Oxygen
Comparison graphs between UCB and LCB for temperature, specific conductivity, pH, and DO are displayed in Figure 7. The temperature in UCB ranged from 13.24oC to 13.67oC. The mean (μ) was 13.413oC with a standard deviation (σ) of 0.147oC. The temperature in LCB ranged from 15.13oC to 16.93oC. The μ was 16.37oC with a σ of 0.581oC. A t-test between UCB and LCB gave a p-value <0.0001, indicating the two means were significantly different for temperature. The specific conductivity of UCB ranged from 904 uS/cm to 937 uS/cm. The μ was 922.75 uS/cm with a σ of 14.69 uS/cm. The specific conductivity of LCB ranged from 1263 uS/cm to 1370 uS/cm. The μ was 1296.825 uS/cm with a σ of 30.95 uS/cm. A t-test between UCB and LCB gave a p-value <0.0001, indicating the two means were significantly different for conductivity. The pH of UCB ranged from 6.675 to 7.6. The μ was 7.239 with a σ of 0.276. The pH of LCB ranged from 5.34 to 7.04. The μ was 6.667 with a σ of 0.529. A t-test between UCB and LCB gave a p-value of 0.025, indicating the two means were significantly different for pH. The DO in UCB ranged from 93.95% to 97.35%. The μ was 95.328% with a σ of 0.907%. The DO in LCB ranged from 31.3% to 52.5%. The μ was 45.676% with a σ of 6.94%. A t-test between UCB and LCB gave a p-value <0.0001, indicating the two means were significantly different for DO.
The ash free dry mass in UCB was 0.0004 g/L for both samples at 0m, and 0.00026 g/L for both samples at 200m. The μ was 0.00035 g/L with a σ of 9.998 x 10-5 g/L. The ash free dry mass in LCB was 0.00026 g/L for the only weighed sample at 0m and 0.00026 g/L for both samples at 200m. The μ was 0.00026 with a σ of 4.34 x 10-6 g/L. There were too few samples to perform a statistical analysis. A bar chart (Figure 8) displays the lower error bar for UCB overlaps with the TSS value for LCB.
The human activities surrounding College Brook are impacting water quality. Several parameters were more heavily impacted in LCB as expected, however other parameters were more impacted in UCB. The nutrient chemistry in each reach cannot be statistically compared due to limited samples, however a look at nutrient trends between the two reaches is a valuable starting point in understanding human impacts on nutrients in College Brook. Inferences can be made about what might be occurring in each reach of College Brook.
DON was expected to be higher in LCB, however, the level of dissolved organic nitrogen was higher in UCB. The land use at the headwaters of College Brook is dairy farming. Cow urine drains into UCB. Urine is high in urea, a form of DON. Another explanation for higher DON in UCB is the leaching from allochthonous inputs, which are missing from College Brook for 200m where the stream is underground. Underground DON is lost. It is possibly being mineralized and nitrified. It is clear that the conversion of DON does not stop at mineralization because there would be a corresponding increase in ammonium. Nitrite was not measured, so it is not possible to confirm the DON was nitrified. It is also possible the DON was further converted into nitrate. This could explain in the increase in nitrate between 200m and 400m.
Nitrate levels followed the hypothesized trend. The concentration is steady between 0m and 200m UCB. The increase of nitrate from 200m to 400m could be the result of a lack of aquatic life (particularly autotrophic) to take up and use the nitrate added by field fertilization. In UCB nitrate was used by aquatic life, keeping the concentration stable. The nitrate levels decline again once the stream emerges from underground, where photosynthesis occurs again. It is possible there is an additional source of nitrate from under the buildings (e.g. a sewage leak). If nitrate levels at 0m LCB was only due to nitrate from fertilizer than phosphate would also have increased here. Phosphate did not behave as hypothesized. Since phosphate can be taken up and stored for late use, it is possible the loading of phosphate from fertilizers has saturated the system. This would explain why there is not much change along College Brook, because there is a lack of demand.
DOC was not higher in LCB as hypothesized. DOC was higher in UCB. There are no litter inputs between 200m and 400m underground, and so it follows that DOC does not increase. For DOC to decrease as it does respiration must be occurring. The culvert must have heterotrophic activity within it. The underground culvert created a place for CO2 outgassing. It would be interesting to determine if culverts under roads, and in suburbanized or urbanized areas are a large source of CO2 to the atmosphere. A decrease in DOC downstream will also have an impact on aquatic life, as DOC is an important source of energy for many organisms (Allan & Castillo, 2007).
Ammonium did not behave as hypothesized. In both reaches there is a decrease in ammonium, however underground there is an increase. Ammonium comes from animal waste and so it follows that UCB at 0m would have higher concentrations than at 200m where farm animal waste would be diluted. The increase underground further supports the idea that there is a sewage leak underground (as mentioned for nitrate). Sewage leaks and overflow are important human causes of degrading water quality (Whitehead et al., 2009).
The grab samples showed higher chloride concentrations in LCB. The hypothesis of higher conductivity in LCB was also supported by specific conductivity data from the Hobo data logger. LCB receives runoff from the roads, sidewalks, stairways, and paths throughout campus. The University of New Hampshire uses road salt to melt snow. The salt accumulates over time and is continually transported to LCB throughout the year. High conductivity is detrimental to aquatic life, farming, and drinking water (Allan & Castillo, 2007).
As hypothesized, LCB had a higher temperature than UCB. Warming occurred during the underground travel of College Brook. Temperature change alters the DO saturation point. Colder water can hold more DO. The prediction of LCB having lower DO is confirmed, however it is not entirely due to the temperature change. If the temperature was the only factor influencing DO, than the %DO would be close in UCB and LCB. Instead, LCB had about half the %DO. The remaining DO loss must be due to a conversion to CO2 during the decomposition seen between 200m and 400m. The culvert has again shown to have a negative impact on water quality. Since the end of the study it has been revealed there is a sump pump in Spaulding Hall that drains into the underground culvert.
This study is limited in its assessment on human impacts on water quality due to the small number of samples taken, a lack of repetition, and a narrow timescale. Although this study cannot prove that human impacts degrade water quality, it does suggest that human activities play a role in water quality degradation. Climate change models predict that with changing conditions, water is going to be very susceptible to decreasing quality (Chung, Park, & Lee, 2011; Hamlet, 2011; Peters & Meybeck, 2000; Whitehead et al., 2009). It is therefore critical to mitigate the negative impacts humans have on water quality now. Fortunately, rivers are capable of self restoration once human pollutants are controlled (Allan & Castillo, 2007). In order to lessen the human impacts on College Brook, vegetative buffers could be placed to slow runoff from impervious surfaces, better management practices could be applied to fertilizing the athletic fields (Bernhardt et al., 2005), UNH could reduce the amount of road salt used or switch to an alternative such as sanding, and the sump pump in Spaulding Hall could be rerouted to a sewer pipe. In order to fully understand the management needs of UCB and LCB a more comprehensive, longer time scale study is needed.
Allan, J. D., & Castillo, M. M. (2007). Stream Ecology: Structure and function of running waters (Second Edition ed.). Dordrecht, The Netherlands: Springer.
Benbow, M., & Merrit, R. (2005). Road Salt -WQ-831. In J. Lehr & J. Keeley (Eds.), Water Encyclopedia. Wiley, New York.
Berner, E., & Berner, R. (1987). The Global Water Cycle: Prentice-Hall, Englewood Cliffs, NJ.
Bernhardt, E. S., Palmer, M. A., Allan, J. D., Alexander, G., Barnas, K., Brooks, S., et al. (2005). Ecology – Synthesizing US river restoration efforts. Science, 308(5722), 636-637.
Chung, E. S., Park, K., & Lee, K. S. (2011). The relative impacts of climate change and urbanization on the hydrological response of a Korean urban watershed. Hydrological Processes, 25(4), 544-560.
Hamlet, A. F. (2011). Assessing water resources adaptive capacity to climate change impacts in the Pacific Northwest Region of North America. Hydrology and Earth System Sciences, 15(5), 1427-1443.
Malmqvist, B., & Rundle, S. (2002). Threats to the running water ecosystems of the world. Environmental Conservation, 29(2), 134-153.
Mouri, G., Takizawa, S., & Oki, T. (2011). Spatial and temporal variation in nutrient parameters in stream water in a rural-urban catchment, Shikoku, Japan: Effects of land cover and human impact. Journal of Environmental Management, 92(7), 1837-1848.
Peters, N. E., & Meybeck, M. (2000). Water quality degradation effects on freshwater availability: Impacts to human activities. Water International, 25(2), 185-193.
Price, K., & Leigh, D. S. (2006). Comparative water quality of lightly- and moderately-impacted streams in the southern Blue Ridge Mountains, USA. Environmental Monitoring and Assessment, 120(1-3), 269-300.
USEPA. (2000). National water quality inventory, EPA Publication 841-R-02-001. Retrieved. from.
Whitehead, P. G., Wilby, R. L., Battarbee, R. W., Kernan, M., & Wade, A. J. (2009). A review of the potential impacts of climate change on surface water quality. Hydrological Sciences Journal-Journal Des Sciences Hydrologiques, 54(1), 101-123.
Wollheim, W. (2011a). Lab 0: An Introduction to College Brook.
Wollheim, W. (2011b). Lab 1: Basic physical characteristics of College Brook study reach: Stream gauging, geomorphology, substrate size.
Wollheim, W. (2011c). Lab 3: Water Quality Characterization of College Brook study reaches: Canopy Cover, Light, Nutrients, Water Quality.