By Anna Willig and Lauren McGrath | Willistown Conservation Trust Watershed Protection Program
Cover Photo by Jennifer Mathes
Since 2018, the Watershed Protection Program has monitored water quality at ten sample sites in the headwaters of Darby, Crum, and Ridley Creeks (Map 1). Every four weeks, the team visited each of the ten sites to take in-stream measurements and collect samples for analysis in the lab. We are proud to present our findings on water quality based on analysis of our data collected from 2018 through 2021, which includes 41 monitoring visits and over 7500 different measurements.
August is National Water Quality month, and each week we will publish excerpts from one chapter from the report. Last week, Chapter 2 focused on discharge, turbidity, and total suspended solids. Chapter 1 gave an introduction to water chemistry. This week, we are focusing on specific conductivity, chlorides, and nutrients. The full report, which includes more information than is provided in the blog posts, can be found here.
To better understand potential sources of pollution to the headwaters of Darby, Crum, and Ridley Creeks, we examined specific conductivity, salts, and nutrients. Specific conductivity is a broad water quality measurement that reflects the presence of ions in the water. These ions include compounds that are formed when salts and nutrients dissolve in water. Higher specific conductivity measurements indicate a higher concentration of ions. However, specific conductivity does not provide insight into the type or concentration of ions in the streams. To explain changes in specific conductivity over time or across sites, we monitored salt and nutrient concentrations.
Chloride is an ion that forms when salts and, to a lesser extent, fertilizers dissolve in water. Elevated chloride concentrations can be toxic for stream organisms. Nutrients, mainly nitrogen and phosphorus, enter streams from fertilizer runoff and leaky septic and sewer systems, all of which increase specific conductivity. When nitrogen and phosphorus are too high, rapid algal growth can occur. This eventually leads to a depletion of dissolved oxygen. Chloride, nitrogen, and phosphorus are all naturally present in low concentrations in streams. However, changes in local land use — specifically increases in development and impervious surface cover — can increase the concentration of these compounds, threatening water quality. An impervious surface is any surface that water cannot pass through, such as buildings, roads, parking lots, and sidewalks.
All stream samples have elevated specific conductivity (Figure 1). Elevated specific conductivity is driven by salts and nutrients, as indicated by high chloride, total nitrogen, and total phosphorus concentrations, though sites in Crum Creek are less impacted than Ridley and Darby Creek sites. While chloride and total nitrogen do not exceed levels deemed unsafe for human consumption (there are no such regulations for total phosphorus), they are still present in excess, potentially posing a threat to stream organisms. Warm water temperatures (see Chapter 1) may exacerbate the hazards posed by elevated chloride to stream life.
Monitoring specific conductivity, chloride concentration, and nutrients reveals the importance of land protection for maintaining and improving water quality in our area. Catchments, or drainage areas, with low impervious surface cover have better water quality, as demonstrated by lower specific conductivity, chloride concentration, and nutrient concentration, than catchments with high impervious surface cover (Figure 2). The most impaired sites are Darby Creek at Waterloo Mills (DCWM1) and West Branch Ridley Creek (WBRC1), both of which have 20% impervious cover in their surrounding catchments and a high percentage of developed land. The least impaired site is West Branch Crum Creek (WBCC1), which has 9% impervious cover in its catchment, the lowest of all sites. Improving water quality, especially in Darby and Ridley Creeks, will require a reduction in excessive road salt and fertilizer use. Limiting development upstream of WBCC1 is crucial to protecting and maintaining the health of Crum Creek.
For a primer on statistical tests and how to read boxplots and scatterplots, click here.
Figure 1. Specific conductivity from January 2018 through December 2021 (a) across ten sample sites in the headwaters of the Darby, Crum, and Ridley Creeks and (b) over time.
Specific conductivity measures how well electricity travels through water. Pure water is a poor conductor and has a low specific conductivity. Ions – often from salts (such as sodium chloride), nitrogen and phosphorus-based nutrients, and metal cations – all increase specific conductivity. As many of these ions come from anthropogenic sources, specific conductivity provides insight into the general impact of human activities on waterways. Higher specific conductivity indicates a more heavily impacted stream, but it does not indicate which compounds are entering waterways. Though there are no federal or state standards for specific conductivity, natural background levels of specific conductivity in the sampled stretches of stream are estimated to be between 75 and 95 µS/cm, which is far lower than any values measured.1 Elevated specific conductivity is expected due to the long history of human activity and development within the study area.
There are significant differences in specific conductivity between sites. Specific conductivity is significantly lower at all Crum Creek sites than at DCWM1 and most Ridley Creek sites (Figure 1a). Interestingly, mean specific conductivity is significantly higher at WBRC1 than Main Stem Ridley Creek (RC1), despite their physical proximity (Figure 1a, Map 1). Check out this blog post to learn more about these two sites.
Figure 2. The relationship between percent impervious surface cover and mean specific conductivity at ten sample sites in the headwaters of Darby, Crum, and Ridley Creeks from January 2018 through December 2021. Error bars represent standard error. The blue line represents a linear trendline and the shaded region shows the 95% confidence interval.
Some spatial differences in specific conductivity between sites can be explained by the percent impervious cover in the surrounding watershed. There is a significant positive relationship between mean specific conductivity and percent impervious surface cover (Figure 2). As impervious surface cover reflects the density of human development and specific conductivity reflects human influence on streams, this relationship is unsurprising.
Specific conductivity generally remains constant throughout the year, though it can spike in winter (Figure 1b). Road salts, which are applied in the winter, form chloride when dissolved in water, increasing specific conductivity. The maximum specific conductivity at each site was recorded on days when there was notable snowmelt, indicating that runoff containing road salts is responsible for these spikes.
Figure 3. Monthly analysis of chloride concentration via Quantab strips from January 2019 through December 2021 (a) across ten sample sites in the headwaters of the Darby, Crum, and Ridley Creeks and (b) over time.
Chloride is naturally present in streams at low concentrations due to the weathering of rocks and soils. Road salt is the main anthropogenic source of chloride in streams, though fertilizers can also contribute chloride. Long-term increases in chloride have been reported nation-wide.2 Chloride impacts ecosystems by altering the microbial communities that form the base of the food chain and by disrupting ion transport in aquatic plants and animals. Though much is still unknown about the effects of chronic exposure to elevated chloride, warmer temperatures increase the toxicity of chlorides to stream insects, with consequences for the rest of the stream ecosystem.3 Click here to learn more about the impact of elevated chloride levels on streams.
Chloride concentration in the sample area has not exceeded the Pennsylvania Department of Environmental Protection potable water supply standard of 250.3 ppm, though WBRC1 and DCWM1 have reached 247 ppm (Figure 3a).4 There are significant differences in monthly chloride concentration between sites. There are no significant differences in monthly chloride concentration between Crum Creek sites (Figure 3a). However, in Ridley Creek, chloride concentration is significantly lower at Ridley Creek State Park (RCSP1) than at WBRC1, indicating either a reduction in chloride entering the stream or dilution as discharge increases (Figure 3a). Chloride concentration in Darby Creek is comparable to most Ridley Creek sites, with RCSP1 and Crum Creek sites having lower concentrations (Figure 3a). Similar to specific conductivity, chloride is significantly higher at WBRC1 than at RC1, despite their proximity (Figure 3a, Map 1). Chloride concentration is generally higher in winter months than in other seasons due to road salt applications (Figure 3b).
Nutrients are a group of chemical compounds that are essential to the growth and survival of living organisms. The two most common nutrients are nitrogen and phosphorus, which enter the water through animal waste, fertilizer runoff, and leaky septic and sewer systems, and cycle through the environment in a complex system. An excess of nutrients in streams can trigger a process called eutrophication, which is the rapid growth of vegetation and algae that ultimately reduces dissolved oxygen as vegetation dies and decomposes.5
i. Total Nitrogen
Figure 4. Total nitrogen from January 2018 through March 2020 (a) across ten sample sites in the headwaters of the Darby, Crum, and Ridley Creeks and (b) over time. The dashed line represents the recommended maximum total nitrogen threshold for streams in this ecoregion.
Total nitrogen measures the concentration of nitrates, nitrites, and ammonia in the water. Total nitrogen dynamics are mostly driven by nitrates, which are present in higher concentrations than nitrites and ammonia. The Pennsylvania Department of Environmental Protection standard for the maximum concentration of nitrates and nitrites in potable water supply is 10 mg/L.4 Total nitrogen does not approach 10 mg/L at any sample sites.
Though total nitrogen concentration is not high enough to be dangerous for human consumption, concentrations are high enough to impair streams. Two studies of nutrient concentrations in streams in this area have found that the total nitrogen threshold for streams that are not impaired by nutrients is 2.225 – 2.3 mg/L.6-7 In the sample area, the 2.3 mg/L threshold is exceeded in 141 out of 289 samples: at least 20 times by each Ridley Creek site and at most 8 times by each Darby and Crum Creek site (Figure 4b). Furthermore, natural background concentrations of total nitrogen are estimated to be 0.10 – 0.30 mg/L for this region, which is far lower than measured concentrations at all sites (Figure 4b).8 Total nitrogen concentrations are much higher than natural background levels and regularly exceed recommended thresholds, indicating that excess nitrogen is an impairment.
There are significant differences in total nitrogen between sites (Figure 4a). In Ridley Creek, most downstream sites have significantly lower total nitrogen than most upstream sites, which could be due to dilution with greater volumes of water (Figure 4a). Despite close physical proximity, total nitrogen is significantly higher at WBRC1 than RC1 (Figure 4a, Map 1). There are no significant differences in total nitrogen between sample sites in Crum Creek, and Crum Creek sites and DCWM1 tend to have significantly lower total nitrogen than Ridley Creek sites (Figure 4a). Total nitrogen does not vary seasonally (Figure 4b).
ii. Total Phosphorus
Figure 5. Total phosphorus concentration from January 2018 through March 2020 (a) across ten sample sites in the headwaters of the Darby, Crum, and Ridley Creeks and (b) over time. The shaded region represents estimated natural background concentrations of total phosphorus and the gray hashed line represents the recommended maximum total phosphorus threshold for streams in this ecoregion.
Total Phosphorus is the total concentration of all phosphorus-containing compounds in streams. Natural background concentrations of total phosphorus are estimated to be 0.025 – 0.060 mg/L for this region.8 In the study area, 55 out of 289 measurements exceed the upper end of this range and 189 exceed the lower end, suggesting that phosphorus may be present in excess (Figure 5b). However, an analysis of nutrient concentrations from 2000 to 2019 in streams in Southeastern Pennsylvania found that the maximum total phosphorus concentration for a stream that is not impaired by nutrients is 0.035 mg/L.7 In the sample area, this threshold is exceeded in almost half of the samples collected (140 out of 289 samples: 24 times each by WBRC1, RCAB1, RCOK1, and RCSP1 and only up to 11 times each by all other sites), indicating that phosphorus is regularly present at high enough concentrations to impair Ridley Creek and is occasionally an impairment in Crum and Darby Creeks.
There are significant differences in total phosphorus between sites. Total phosphorus does not vary significantly between sites in Darby and Crum Creeks (Figure 5a). In Ridley Creek, WBRC1 has significantly higher total phosphorus than RCSP1 and RC1 (Figure 5a). Though total phosphorus does not show strong seasonal variation, total phosphorus trended higher and had a broader spread from mid-2019 through 2020 than in 2018 (Figure 5b).
- Specific conductivity is elevated in all streams, indicating that they are impacted by human activities.
- Specific conductivity is related to impervious surface cover in the watershed, highlighting the importance of protecting open space and limiting development.
- Chloride, nitrogen, and phosphorus concentrations are all elevated, contributing to specific conductivity and threatening stream health.
- Limiting runoff of road salt by sweeping up excess after storms and reporting large piles on roads to municipalities is critical to improving stream health.
- Reducing chemical fertilizer use or switching to using compost or other soil amendments can limit the amount of nutrients entering streams. Head over to the Farm Program to learn more about soil health.
- Native plants act as filters for water, pulling out nutrients and other pollutants before they enter streams. Adding native plants to lawns and gardens is a great way to improve water quality while also creating habitat for wildlife.
To read the full “State of our Streams Report,” click here.
Map 1. Willistown Conservation Trust’s sampling sites. Five sample locations are within the Ridley Creek watershed, four are within the Crum Creek Watershed, and one is within the Darby Creek Watershed. Sampling was conducted at each site every four weeks from January 2018 through December 2021.
This report was made possible through a grant from the William Penn Foundation. The WIlliam Penn Foundation, founded in 1945 by Otto and Phoebe Haas, is dedicated to improving the quality of life in the Greater Philadelphia region through efforts that increase educational opportunities for children from low-income families, ensure a sustainable environment, foster creativity that enhances civic life, and advance philanthropy in the Philadelphia region. In 2021, the Foundation will grant more than $117 million to support vital efforts in the region.
The opinions expressed in this report are those of the author(s) and do not necessarily reflect the views of the William Penn Foundation.
1. Olson, J. R. & Cormier, S. M. Modeling spatial and temporal variation in natural background specific conductivity. Environ. Sci. Technol. 53, 4316–4325 (2019).
2. Kaushal, S. S. et al. Freshwater salinization syndrome: from emerging global problem to managing risks. Biogeochemistry 154, 255–292 (2021).
3. Jackson, J. K. & Funk, D. H. Temperature affects acute mayfly responses to elevated salinity: implications for toxicity of road de-icing salts. Philos. Trans. R. Soc. B Biol. Sci. 374, 20180081 (2019).
4. Pennsylvania Department of Environmental Protection. 25 Pa. Code Chapter 93. Water Quality Standards § 93.7. Specific Water Quality Criteria. https://www.pacodeandbulletin.gov/Display/pacode?file=/secure/pacode/data/025/chapter93/chap93toc.html&d=reduce (2020).
5. United States Geological Survey. Phosphorus and Water. Water Science School https://www.usgs.gov/special-topic/water-science-school/science/phosphorus-and-water?qt-science_center_objects=0#qt-science_center_objects (2018).
6. USEPA. Ambient Water Quality Criteria Recommendations: Rivers and Streams in Ecoregion IX. 108 (2000).
7. Clune, J. W., Crawford, J. K. & Boyer, E. W. Nitrogen and Phosphorus Concentration Thresholds toward Establishing Water Quality Criteria for Pennsylvania, USA. Water 12, 3550 (2020).
8. Smith, R. A., Alexander, R. B. & Schwarz, G. E. Natural Background Concentrations of Nutrients in Streams and Rivers of the Conterminous United States. Environ. Sci. Technol. 37, 3039–3047 (2003).
By Anna Willig and Lauren McGrath | Willistown Conservation Trust Watershed Protection Program