Abstract

Road salts are a dominant factor contributing to the increase in chloride (Cl-) concentrations in streams. Road salts dissolve in rainwater, and the resulting saline waters percolate into the ground. Saline solutions have been shown to increase heavy metal concentrations in soil porewater [1-3] which in turn enters streams, increasing the concentration of heavy metals in the streams. In this study, water sampling was conducted monthly; water quality parameters and the trace metals iron (Fe) and manganese (Mn) were measured for all samples for the year 2025. Water samples from Sites B, B’, and D had Cl– concentrations above the chronic aquatic limit for most of the year. Fe concentrations during 2025 were almost all below the drinking water limit except for November for Site B’ and December for both Site B, and B’, with the overall highest concentration being for Site B’ during the month of December. The highest Mn concentrations were measured at Sites C and D in January 2025 and Mn concentrations were above the drinking water limit of 50 μg/L for the entire year for Site B and B’. Correlations between Cl–, Fe, and Mn suggest that road salts play a role in releasing these metals from soil and sediments on the HCC campus.  Consistent, long-term monthly collection of such data will help form a more accurate understanding and illuminate trends regarding how road salts impact the health of streams through meta-analysis.

Introduction

Road salts have been shown to be effective in reducing the frequency and severity of motor vehicle accidents during icy winter conditions [4].  The most widely used deicing salt is sodium chloride (NaCl); other chloride salts such as calcium chloride (CaCl2) and magnesium chloride (MgCl2) are also commonly used for deicing purposes [5]. For example, during the winter of 2015-2016, the Maryland State Highway Administration spent about $56 million to maintain 17,842 miles of highway primarily by using rock salts [6].When snow melts or it rains, the salt runs off into freshwater ponds and streams, which increases the Cl- concentration and salinity of both surface water and ground water resources [7-9]. Although alternatives such as beet juice or acetate-based deicers are available, they are not as widely used as chloride salts due to cost. Different types of deicing materials affect the environment around them differently. Materials that are organic-based (beet juice and acetate deicers) are correlated with a decrease in dissolved oxygen [10]. Schuler and others hypothesized that this reduction in dissolved oxygen was due to the increase in phosphorous concentration, which would have caused an increase in algae population, increasing competition for oxygen between aquatic organisms [10].

Chloride (Cl-) ions have a negative effect on water quality with high quantities being toxic to aquatic organisms. The chronic and acute aquatic limits for Cl- in freshwater systems are 230 mg/L and 860 mg/L respectively [11, 12].  The chronic aquatic limit represents the level where continual Cl- concentrations at or above this level may result in negative impacts on freshwater organisms over a long period of time. In contrast, the acute aquatic limit represents a level where even short-term exposure to Cl- concentrations at or above this level can immediately lead to negative effects on the health of freshwater organisms [5, 11]. Such elevated concentrations of Cl- have also shown to be correlated with the mobilization of heavy metals in freshwater systems [1, 5, 13]. All these effects can harm aquatic ecosystems and biodiversity [14]. These increases in Cl- concentration have also been found to be long-lasting, so the negative effects would decrease slowly over time [14, 15].

Issues resulting from chloride pollution take a long time to be reversed, as Cl- is retained through two mechanisms: slow porewater movement and a cation exchange mechanism which helps retain salt [16]. the ion exchange reaction, Na+ exchanges with the other cations Ca2+, Mg2+, or K+ in the soil, which can take about 5 months to be released from the soil [10]. The most popular salt used to maintain roads is NaCl, which can be problematic, but is not as toxic to aquatic organisms as chlorides associated with K, Mg, and Ca [11]. NaCl has been shown to mobilize heavy metals like iron (Fe) and manganese which could then contaminate drinking water supplies [9, 16]. The chronic aquatic limit for Fe is set by the US Environmental Protection Agency (EPA) in the Aquatic Life Criteria table at 1 mg/L; however, this table does not list an acute aquatic limit for Fe [12, 17]. A recent study suggests that this increase of Fe into streams may promote the growth of iron metabolizing bacteria [18].

All these factors are capable of negatively impacting the ecosystem; since microorganisms have been found to be particularly susceptible, this would have direct impacts on the food chain and biodiversity [8, 14, 19]. The effects of road salt on the ecosystem and the community within it depend on the environmental conditions, the organisms themselves, and the type of salt used [20]. Baumann and Marshner found that plant growth is reduced as the bacteria responsible for carbon mineralization are negatively affected by the increased salinity, due to osmotic stress [19]. Eaton and others found that lowbush blueberry yields were negatively correlated with how much deicing salt the plants were exposed to [21]. Such negative effects could have wide-ranging consequences within an ecosystem, as a reduction in plant growth or seed production could directly impact organisms which depend on plants to provide shelter and nutrition and would indirectly impact other organisms in the food chain [19].  Since road salt is a long-term contaminant in the environment that can affect the health of plants and microorganisms, it is important to find solutions to reduce road salt usage.

For these reasons, this study was conducted to help form a better understanding of how road salts impact water quality parameters and the concentration of trace metals in two stream systems on the Howard Community College (HCC) campus in Columbia, Maryland.  Howard County, in which the town of Columbia is located, has made a push towards using brine over rock salt, making it easier to plow snow while at the same time reducing rock salt usage. The county has also increased outreach to the community on how to reduce the use of rock salt [22]. Although Howard County is promoting Best Management Practices (BMPs) with respect to the application of deicing salts, are these practices being applied successfully?  What are the impacts of road salts and metals in the streams on the HCC campus? In this study, monthly measurements of water quality parameters and trace metal concentrations for the year 2025 were performed to address these questions by comparing the measured concentrations to water quality limits and to detect correlations between these parameters.  In addition, the current study is a continuation of work that has been underway since 2021; therefore, comparisons will be made with the data sets from previous years to detect longer-term trends in Cl- concentration and the concentrations of metals Fe and Mn.

Methods

Location

Water samples were collected monthly from five sites on two stream systems on the Howard Community College campus. Locations of sampling sites are shown in Figure 1. Site A and B are both located on the same stream (stream system 1) which runs parallel to Little Patuxent Parkway in Columbia, Maryland. The pool at Site B is downstream of Site A and receives water from a culvert (Site B’). Site C is a pond which receives water from another campus stream (stream system 2) that runs adjacent to Campus Drive. Both stream systems are part of the Patuxent watershed. Site D is supplied with water from 3 culverts with the right-most receiving water from storm drains on Campus Drive; the other two culverts transmit the water from a stream that emerges from another culvert at the edge of the HCC campus property and flows under Campus Drive. This stream, which we sampled at Site D, flows downstream to a pond where Site C samples are collected. The Sites A and D samples were collected roughly at the center of their respective streams, as these streams are shallow. Site B samples were collected from the edge of a pool, and the Site C samples were collected close to the edge of the pond; the depth of these water bodies made sampling from their centers impractical.

New on the HCC campus in 2025 is the construction site for the Workforce Development and Trades Center. The location of this construction site is shown in Figure 1; work began on May 1, 2025 and is ongoing. Runoff and sediment from this construction site could possibly impact both stream systems on the Howard Community College campus. Campus Drive slopes downhill from the construction site toward the stream and culverts where Site D is located, hence runoff from the construction site may flow this way. But the current construction site could also possibly impact the campus stream system where Sites A, B, and B’ are located. Sediment was observed flowing from the construction site on October 30, 2025 and in rain events since then. The sediment flowed east, toward the traffic circle at the main entrance to the college. If runoff continues to flow in this direction, it would enter the stream system that includes Sites A, B, and B’.

Figure 1: Map of the Howard Community College (HCC) campus showing the sampling sites and the location of the Workforce Development and Trades Center (under construction).

Additional information on the methods applied and selected results in this study can be found in this online Supplement: Supplement for JRIP paper by Taha et al. (2026)

Results

Weather/precipitation

Weather data used in this study is from the Washington Post daily weather page. Weather data recorded at BWI (Baltimore-Washington International) airport was used to approximate conditions in Columbia, Maryland. BWI airport is about 11 miles east of Columbia, Maryland. There was a total of 35.99 inches of rainfall at BWI in 2025; this was 9.01 inches below normal. In 2025, only two months had above average rainfall – May (3.00 inches above normal) and July (1.51 inches above normal); the other 10 months in 2025 had below-average rainfall. Similarly, rainfall was below normal leading up to nine of the twelve monthly sampling dates. In addition to May and July, rainfall was also above normal by 0.25” on February 20, 2025, although the last eight days of February 2025 were dry, so we wound up with below normal rainfall for the month.

There was a total of 12.7” of snow at BWI for the winter from December 2024-March 2025. There was a significant snow event on January 6, 2025 (6.6” of snow) and rain, sleet, and snow fell on January 19 (0.19” rain equivalent). Temperatures were very cold in January 2025 and the snow and ice lingered. As a result, roads were treated and retreated with deicing salts in January. In February 2025, there was a freezing rain/rain event on February 6 and 3.7” of snow fell on February 12, 2025. There was no additional snow recorded in March 2025.

Water Quality Summary

Table 1 shows yearly ranges and averages of dissolved oxygen (%), pH, conductivity (𝜇S/cm), and turbidity (NTU) for all sites in 2025.  The full 2025 water quality data set for this study can be viewed here: Water Quality Data 2025.  Here we summarized the water quality results for 2025 and compare them to US EPA guidelines for healthy streams as well as to the 2023 results of Shaner [25] from these same sites.

Dissolved oxygen is expressed as a percentage of the saturated value at the water temperature, atmospheric pressure, and conductivity values applicable at the time the sample was collected. Dissolved oxygen was measured in the field in mg/L; the percentage of the saturated value was calculated using the dissolved oxygen calculator on the USGS website [26].  Nearly all dissolved oxygen values measured for this study were in the “supportive” range (greater than 6.5 mg/L) indicated by the US EPA [27]. Only one value, measured at Site D on 8/29/2025, during a period of drought, was in the “stressful” range (4.69 mg/L).  In comparison to the 2023 results of Shaner [25], the average percent dissolved oxygen values at Sites A and B were slightly higher in 2025, while average percent dissolved oxygen values were somewhat lower at Sites C and D in 2025.

Values of pH between 6.5 and 8 are considered “optimal” by the US EPA [28].  In this study, pH values measured for all sites fell well within this range throughout 2025. While Shaner [25] observed some pH values of 6.5, at the low end of the optimal range, in 2023, the lowest pH value observed in 2025 was 6.77 at Site D on 12/16/2025.

Turbidity values measured at all sites in 2025 were generally at levels considered safe for aquatic life [29], and there are explanations for the occasional higher turbidity values that we observed. For example, turbidity at Site C on 1/27/2025 and 2/20/2025 was high due to the pond being frozen and the sediment being stirred up in the process of breaking through the ice to collect water samples. Turbidity at Site B on 2/20/2025 was high because the pool at Site B was still largely filled with sediment at that time – a fire hydrant blow-out on Little Patuxent Parkway on October 27, 2024 (R. Marietta, pers. comm. 1-23-2025) had filled the pool at Site B with sediment and it took several months for this sediment to work its way downstream. We also have in our field notes that debris was present in the samples from Sites A and B collected on 8/29/2025 that affected the measured turbidity.

Conductivity and Chloride

Conductivity and Cl- concentrations are closely related quantities. Conductivity is a measure of total ions in solution, measured in units of 𝜇S/cm (microsiemens per centimeter). Chloride (measured in mg/L) is one of the major anions present in natural waters as well as in water affected by anthropogenic sources of chloride such as road salts.

The US EPA [30] has determined that most rivers in the United States have conductivities between 50 and 1500 𝜇mhos/cm (this is equivalent to 𝜇siemens/cm) and ideal conductivities to support a healthy fish population are between 150 and 500 𝜇mhos/cm. In this study, conductivity values for all five sampling sites range between 406 and 12450 𝜇S/cm.

When examining the data collected from all sites for the year (Figure 2), Cl- and conductivity have a strong significant positive correlation across all sites (rho = 0.93, p<0.001, see Figure 2).  Site D, in particular, exhibited a wide range in measured conductivity and Cl- concentrations, yet these wide-ranging values fit this same trend. The Cl- and conductivity relationship from this study can be compared with Figure 8 in Haq et al. [3], which also demonstrates the strong linkage between these parameters.

Figure 3 shows chloride concentrations versus time for all sampling sites in 2025. Site-specific trends include: at Site A, four of the monthly Cl- concentrations were above the chronic aquatic limit of 230 mg/L (May to August 2025); the remainder were below this level. At Sites B, B’, and D, the majority of the measured Cl- concentrations were above the chronic aquatic limit (11/12, 12/12, and 10/12 months respectively). Chloride concentrations at Site B were above the chronic aquatic limit almost all year but, like Site A, were highest in the summer months. Site B’ (the culvert) had Cl- concentrations above the chronic aquatic limit throughout the year. At Site C (the pond), 6/12 (half of the months in the year) measured Cl- concentrations were above the chronic aquatic limit. The highest Cl- concentrations at Site C occurred from February through May 2025. After May, Cl- concentrations remained at or below the chronic aquatic limit. Site D had spikes in Cl- concentration in January, February, and December 2025 that exceeded the acute aquatic limit (860 mg/L). In addition, Cl- concentrations were at or above the chronic aquatic limit at Site D throughout the summer and most of the Fall.

Chloride comparison 2022 through 2025

Table 2 shows the yearly average and standard deviation, maximum, and minimum Cl- concentrations from all five sampling sites for 2022 through 2025. This allows for a comparison of trends in Cl- over a four-year period. In 2022, average Cl- concentrations at all sites were relatively high.  A season total of 14” of snow fell in the winter of 2022 (December 2021 to March 2022). During the following winter, December 2022 to March 2023, there was almost no frozen precipitation and consequently, there was minimal salt applied to the roads. In 2023, Shaner [25] observed that yearly average Cl- concentrations decreased at all five sampling sites. During the winter of 2024 (December 2023 to March 2024) there was a moderate amount of snow (11.2” total) and temperatures were generally at or above the normal range in January and February 2024. Wilson and Carmody [32] found that the yearly average Cl- concentrations in 2024 decreased further at all sampling sites, except Site D. Site D showed an increase from 2023 to 2024, which could be due to a large construction project nearby. Runoff from this construction site flowed onto Campus Drive and drained into the pool at Site D. This study, however, has documented a reversal of this decreasing trend in 2025. There was 12.7” of snow during the winter of 2025 (December 2024 to March 2025), 1.5” more than was recorded in the winter of 2024. It stands to reason, then, that road salt application, in general, was at least as high as, and possibly higher than in the winter of 2024. Yearly average Cl- concentrations increased at all five sites in 2025, ranging from very small increases at Sites A and C to greater increases at Sites B, B’, and D (in increasing order).

Manganese

Figure 4 shows the Mn concentrations for all sites for January to December 2025. The Mn concentrations were all above the drinking water limit of 50 µg/L for Sites B and B’. Site C had the highest concentration of 648 µg/L, recorded in January, when the campus was covered in snow; Site C had an average Mn concentration of 87.4 µg/L. After January 2025, the Mn concentration for Site C was below the drinking water limit except during March, May, and December. After May, Mn concentrations remained below the drinking water limit until December 2025.

Site B and B’ were all above the drinking water limit of 50 µg/L throughout 2025. The Mn concentration was highest for Site B during April, remained elevated until September, and had a sharp decline during October. The Mn concentration increased markedly for Sites B and B’ during November and December, peaking at B’ during November with a concentration of 246.9 µg/L. Site B also had the highest average Mn concentration in 2025 (198.8 ± 84.3 µg/L). Notice that some of the trends seen in the chloride vs. time plot (Figure 3) are also seen in Figure 4.

Site D had high Mn concentrations in January, February (decreasing but above average), August, and December 2025; all of these were well in excess of the drinking water limit for Mn. For the year 2025, Site D had seven samples with Mn concentrations above the drinking water limit (50 µg/L) and five samples with Mn concentrations below this limit. On the other hand, the trend for Mn at Site A does not follow the trend seen for Cl- in which Cl- was elevated from May through August. Instead, samples from Site A generally had Mn concentrations below the drinking water limit except for the samples from May and September both of which were collected during rainy periods.

Iron

Figure 5 shows Fe concentrations across the five sampling sites for January to December 2025. The Fe concentrations typically were lower during rainy periods than compared to the previous month for Sites A and B. Site D, however, exhibited modest increases in Fe concentration during rainy periods in March, May, and September 2025. Samples from all sites were below the drinking water limit for Fe from January to October 2025. In November 2025, the Fe concentration at Site B’ exceeded the drinking water limit and in December 2025, Fe concentrations at both Sites B and B’ exceeded the drinking water limit, with the highest Fe concentration for 2025 (778.22 μg/L) being recorded at Site B’ on December 16, 2025.

Correlations: Chloride and Metals

Correlations between chloride and the metals iron and manganese were plotted to determine whether chloride might play a role in releasing metals from soil and sediment into the streams on the HCC campus as has been seen in previous studies [1, 2, 13]. Because Mn2+ has a similar ionic radius and the same charge as Fe2+, it typically occupies Fe lattice sites in iron-bearing minerals [31]. Therefore, correlations between Mn and Fe were investigated to determine whether the release of these two elements as minerals break down within soils is related as well and whether other factors may be involved.

For Spearman rank correlation analysis results with a p<0.05, the strength of the statistically significant positive correlation was characterized by the subsequent rules. If rho fell into the range of 0.4 – 0.59, the relationship was termed as a weak significant positive correlation. If rho fell into this range of 0.6 – 0.79, it was considered a moderate positive correlation. If rho was 0.8 – 0.99, it was considered a very strong positive correlation [32].

Manganese and chloride have a weak significant positive correlation across all sites (rho = 0.56, p<0.01). In Figure 6, Mn is plotted versus Cl- for Site D only. Mn and Cl- at Site D have a very strong significant positive correlation at Site D (rho = 0.92, p<0.001).

Including the data across all five sites, there is a weak significant positive correlation between Fe and Cl- (rho = 0.41, p<0.01). When just stream system 1’s (Sites A, B, and B’) data are compared, there is a strong significant positive correlation between Fe and Cl- (rho = 0.64, p<0.001). However, when Sites C and D are compared alone or combined, there is no significant correlation between Fe and Cl-.

For data from all five sites, there is a strong significant positive correlation between Mn and Fe (rho = 0.69, p<0.01). When just stream system 1’s (Sites A, B, and B’) data is compared, there is a strong significant positive correlation between Mn and Fe (rho = 0.77, p<0.001). However, when Sites C and D are compared alone or combined, there is no significant correlation between Mn and Fe.

Discussion

In this study, water samples were collected monthly from five sites on two stream systems on the HCC campus and water quality parameters and dissolved Fe and Mn concentrations were measured to monitor the health of the streams and to attempt to better understand the interaction between road salts, dissolved chloride, and these dissolved metals. Here we summarize the most noteworthy results and trends discovered in 2025. The lowest conductivity values measured in this study are at the high end of the optimal range (150 to 500 𝜇S/cm) and the highest values measured (up to 12450 𝜇S/cm) were well above the range typical for rivers in the United States (50 to 1500 𝜇S/cm). Similarly, Cl- concentrations at the five sampling sites in this study are above the chronic aquatic limit of 230 mg/L for much of the year and three Cl- concentrations at Sites D exceeded the acute aquatic limit. Both findings indicate that conductivity and Cl- levels in the streams on the HCC campus are detrimental to fish and other aquatic life for much of the year.

Table 2 shows that there was a decrease in the yearly average Cl- concentrations at all sites from 2022 to 2023 and again from 2023 to 2024, except for Site D.  This trend, however, was reversed in 2025, with an increase in average Cl- concentrations seen at all sites. The largest increases occurred at Sites B’ (66.1 mg/L Cl- increase) and D (194.6 mg/L increase). In fact, the average Cl- concentration for 2025 at Site D was above the acute aquatic limit of 860 mg/L Cl- (although Cl- concentrations for only three months exceeded the acute aquatic limit). The increase in average Cl- concentrations observed in this study in 2025 could have resulted from the cumulative effect of two winters with moderate snowfall and the inevitable application of deicing salts to the roads, the lag time involved as the Cl- from the winters of 2024 and 2025 worked its way through the groundwater into the streams, in addition to the cumulative effect of two years with below-average rainfall [33] and possible contamination from the construction site of the Workforce Development and Trades Center.

Manganese concentrations were above the drinking water limit for most or all of 2025 at Sites B, B’, and D, with a yearly-high Mn concentration of 648 µg/L documented at Site C in January 2025.  Fe concentrations, although mostly below the drinking water limit in 2025, nevertheless merit continued monitoring because Fe concentrations above the drinking water limit were recorded in November and December 2025 at Sites B and B’ as well as in previous years at these sites [18, 25].

The correlation between Mn and Cl- indicates that the behavior of these two parameters may be related.  This is particularly apparent at Site D (Figure 6). This correlation is consistent with the idea that Mn and other metals are released from adsorption sites in sediment and soil by an ion exchange mechanism in response to interaction with saline groundwater [e.g., 1, 13]. The significant positive correlation between Mn and Fe at all sites further suggests that both metals may be released from soil and sediment by the same mechanism. However, the slightly weaker correlation between Fe and Cl- and lack of correlation at Site D suggests that another factor is at work on Fe and that this factor has less effect on the Mn, hence the correlation between Mn and Cl- is stronger. The presence of iron-related bacteria at Site B was demonstrated by Wilson and Carmody [33] and metagenomic analysis described by these authors [18] found that the following genera of iron-related bacteria are present: Magnetospirillum and Shewanella.  Shewanella are ferric iron-reducing bacteria. Magnetospirillum takes up both FeO and Fe2O3 to form magnetite crystals inside of specialized organelles-like structures within their cells. Magnetospirillum could potentially take up a significant amount of both ferrous and ferric iron from the environment. Shewanella, on the other hand, produces ferrous iron which is more soluble than ferric iron. The action of Shewanella could potentially lead to more iron in solutions in the stream water on campus. The action of both types of bacteria could account for the lack of correlation between Cl- and Fe at Site D.

Future Work

In 2026, the continuation of this research project will involve a more in-depth exploration of microbial communities. This aims to determine the presence of specific genera of bacteria that metabolize Fe and Mn, their effects on the dissolved Fe and Mn concentrations in the streams, and the effects of high concentrations of Cl- and metals on the activity of microorganisms and the overall health of the streams.  In addition, researchers in 2026 will continue to monitor possible impacts from the Workforce Development and Trades Center construction on the local streams.  The current 2025 study did not document definitive impacts from this construction, but it is possible that, due to the construction site’s relatively large distance from the streams, there may be a lag time before potential impacts become apparent. Therefore, monitoring potential impacts of the continuing construction project will be a focus of the research in 2026.

Authors and mentors involved with this ongoing research will also have the goal to spread the word on the HCC campus and within the local community about the importance of Best Management Practices (BMPs) with regard to the application of deicing salts.  Although this information is available to the local community [22], it is a message that bears frequent repetition because it is so easy to over-apply deicing salts and many studies have documented the severity of the consequences of this practice [e.g., 5, 8, 15, 20]. In the 2024-2025 Salt Watch Report [34] produced by the Isaak Walton League of America, the figure on page 9 shows that of 1,437 chloride measurements made in Maryland by Salt Watch volunteers between July 1, 2024 and June 30, 2025, 551 measurements (38%) were above the chronic aquatic limit of 230 mg/L. There is evidently a lot of work to do in terms of getting the word out on avoiding over-application of deicing salts in Maryland.

In 2026, this ongoing research project will be in its sixth year of monitoring water quality, chloride, and trace metal concentrations in the stream systems on the HCC campus. It is only by continuing this work for several years that longer-term trends can be discovered, as shown in Table 2 above. Long-term monitoring by Professor S. S. Kaushal and his co-workers at the University of Maryland have yielded insights into the health and safety of the rivers that serve as drinking water sources for the city of Washington, DC and the surrounding area [35]. Our ongoing study provides data on Cl- concentrations and other important parameters that can inform the community at HCC and in Columbia, Maryland on how the local streams are faring in the face of road salt application, construction projects, and other potentially detrimental anthropogenic influences.

Acknowledgements

We would like to acknowledge and thank Dr. Marlena Wilson for her involvement with the field work, sample preparation, and water quality measurements for this study.  We also thank Thomas Carmody, Eva Rose, Somaera Choudhary, Yanyu Arias, and Jasmine Kapya, the Physical Sciences lab aides at Howard Community College, for their support of this project.

Contacts: hasanmuhammad.taha@howardcc.edu, abdul.wahab@howardcc.edu, rcarmody@howardcc.edu, and hpie@howardcc.edu

References

[1]    Y. Wen, “Impacts of Road De-icing Salts on Manganese Transport to Groundwater in Roadside Soils,” DiVA Portal, Aug. 2012. Retrieved from https://www.diva-portal.org/smash/get/diva2:580615/FULLTEXT01.pdf

[2]    S. Kim and C. Koretsky, “Effects of road salt deicers on sediment biogeochemistry,” Springer, vol. 112, pp. 343–358, Mar. 2012, doi: https://doi.org/10.1007/sl%200533-012-9728-x.

[3]    S. Haq, S. S. Kaushal, and S. Duan, “Episodic salinization and freshwater salinization syndrome mobilize base cations, carbon, and nutrients to streams across urban regions,” Biogeochemistry, vol. 141, no. 3, pp. 463–486, Oct. 2018, doi: https://doi.org/10.1007/s10533-018-0514-2.

[4]    D. Kuemmel and R. Hanbali, “Accident Analysis of Ice Control Operations,” e-Publications@Marquette, Jun. 01, 1992. Retrieved from https://epublications.marquette.edu/cgi/viewcontent.cgi?article=1001&context=transportation_trc-ice

[5]    W. D. Hintz, L. Fay, and R.A. Relyea, “Road salts, human safety, and the rising salinity of our fresh waters,” Frontiers in Ecology and the Environment, vol. 20, no. 1, pp. 22-30, Feb 2022, doi: https://www.jstor.org/stable/pdf/27115929.pdf

[6]    Maryland Department of Transportation, 2025, “Operations Facts and Figures for the 2024/2025 Winter Season”, Maryland.gov [Online]. Retrieved from https://roads.maryland.gov/OC/Baltimore_Vicinity_Winter_Fact_Sheet.pdf

[7]    E. Foley and A. D. Steinman, “Urban lake water quality responses to elevated road salt concentrations,” Science of The Total Environment, vol. 905, p. 167139, Dec. 2023, doi: https://doi.org/10.1016/j.scitotenv.2023.167139.

[8]    T. Usman, L. Fu, and L. F. Miranda-Moreno, “Quantifying safety benefit of winter road maintenance: Accident frequency modeling,” Accident Analysis & Prevention, vol. 42, no. 0001-4575, pp. 1878–1887, May 2010, doi: https://doi.org/10.1016/j.aap.2010.05.008.

[9]    J. Thorslund and M. T. H. van Vliet, “A global dataset of surface water and groundwater salinity measurements from 1980–2019,” Scientific Data, vol. 7, no. 1, p. 231, Jul. 2020, doi: https://doi.org/10.1038/s41597-020-0562-z.

[10]  M. S. Schuler et al., “How common road salts and organic additives alter freshwater food webs: in search of safer alternatives,” Journal of Applied Ecology, vol. 54, no. 5, pp. 1353–1361, Feb. 2017, doi: https://doi.org/10.1111/1365-2664.12877.

[11]  United States Environmental Protection Agency, “Ambient water quality criteria for chloride – 1988,” Feb. 1988. Retrieved from: https://www.epa.gov/sites/default/files/2018-08/documents/chloride-aquatic-life-criteria-1988.pdf

[12]  United States Environmental Protection Agency, “National recommended water quality criteria – aquatic life criteria table,” last update February 13, 2026. Retrieved from https://www.epa.gov/wqc/national-recommended-water-quality-criteria-aquatic-life-criteria-table#a

[13]  M. S. Schuler and R. A. Relyea, “A Review of the Combined Threats of Road Salts and Heavy Metals to Freshwater Systems,” BioScience, vol. 68, no. 5, pp. 327–335, Mar. 2018, doi: https://doi.org/10.1093/biosci/biy018.

[14]  K. R. James, B. Cant, and T. Ryan, “Responses of freshwater biota to rising salinity levels and implications for saline water management: a review,” Australian Journal of Botany, vol. 51, no. 6, p. 703, 2003, doi: https://doi.org/10.1071/bt02110.

[15]  S. S. Kaushal et al., “Increased salinization of fresh water in the northeastern United States,” Proceedings of the National Academy of Sciences, vol. 102, no. 38, pp. 13517–13520, Sep. 2005, doi: https://doi.org/10.1073/pnas.0506414102.

[16]  T. Baraza and E. A. Hasenmueller, “Road salt retention and transport through vadose zone soils to shallow groundwater,” Science of The Total Environment, vol. 755, p. 142240, Feb. 2021, doi: https://doi.org/10.1016/j.scitotenv.2020.142240

[17]  Iowa Department of Natural Resources, “Iron criteria and implementation for Iowa surface waters,” 2020. Retrieved from  https://www.iowadnr.gov/media/7117/download?inline#:~:text=A.-,The%20Recommended%20Iron%20Criteria,acute%20or%20a%20chronic%20value.

[18]  M. Wilson and R. Carmody, “Possible Microbial Impact on Iron and Manganese Concentration in a Stream Ecosystem at Howard Community College,” Maryland Collegiate STEM Conference, Baltimore, MD, April 12, 2025. Retrieved from https://static1.squarespace.com/static/60148d36cfc2c95d80a06f4c/t/67e2dc0abcad930dc8d845c0/1742920716951/2025+MCSC+Conference+Program+-+Draft+03+24.pdf (Abstract on p.26)

[19]  K. Baumann and P. Marschner, “Effects of salinity on microbial tolerance to drying and rewetting,” Biogeochemistry, vol. 112, no. 1–3, pp. 71–80, Nov. 2011, doi: https://doi.org/10.1007/s10533-011-9672-1.

[20]  W. D. Hintz and R. A. Relyea, “A review of the species, community, and ecosystem impacts of road salt salinisation in fresh waters,” Freshwater Biology, vol. 64, no. 6, pp. 1081–1097, Mar. 2019, doi: https://doi.org/10.1111/fwb.13286.

[21]  L. J. Eaton, J. Hoyle, and A. King, “Effects of deicing salt on lowbush blueberry flowering and yield,” Canadian Journal of Plant Science, vol. 79, no. 1, pp. 125–128, Jan. 1999, doi: https://doi.org/10.4141/p97-115.

[22]  Howard County Office of Community Sustainability, “2025 Annual Report: HoCo Climate Forward,” Howard County, MD, September 2025. Retrieved from https://livegreenhoward.com/wp-content/uploads/2025/09/CLIMATEFORWARDAnnualReport.pdf

[23]  United States Environmental Protection Agency, “Secondary drinking water standards: guidance for nuisance chemicals,” www.epa.gov. Last update June 2, 2025. Retrieved from http://www.epa.gov/sdwa/secondary-drinking-water-standards-guidance-nuisance-chemicals

[24]  S. S. Kaushal et al., “Freshwater salinization syndrome: from emerging global problem to managing risks,” Biogeochemistry, vol. 154, no. 2, pp. 255–292, Apr. 2021, doi: https://doi.org/10.1007/s10533-021-00784-w.

[25]  S. Shaner, “Chloride and heavy metals in HCC’s stream systems,” Journal of Research in Progress, vol. 7, pp. 52 – 75, May 2024. Retrieved from https://pressbooks.howardcc.edu/jrip7/chapter/shaner-chloride-and-heavy-metals-in-hccs-stream-systems/

[26]  U.S. Geological Survey, “DOTABLES: Dissolved oxygen solubility tables,” Version 3.6, 8/27/2018. Retrieved from https://water.usgs.gov/water-resources/software/DOTABLES/

[27]  United States Environmental Protection Agency, “Fact sheet on water quality parameters: dissolved oxygen,” July 2021. Retrieved from  https://www.epa.gov/system/files/documents/2021-07/parameter-factsheet_do.pdf

[28]  United States Environmental Protection Agency, “Causal Analysis/Diagnosis Decision Information System: pH,” last update January 22, 2026. Retrieved from  https://www.epa.gov/caddis/ph

[29]  In-Situ Support, “What are typical turbidity values in natural environments?” last update November 19, 2024. Retrieved from https://in-situ.com/us/faq/water-quality-information/turbidity-faqs/what-are-typical-turbidity-values-in-natural-environments?

[30]  United States Environmental Protection Agency, “5.9 Conductivity: what is conductivity and why is it important?” last update March 6, 2012. Retrieved from  https://archive.epa.gov/water/archive/web/html/vms59.html

[31]  C. S. Hurlbut, Jr. and C. Klein, Manual of Mineralogy (after James D. Dana) 19th ed., New York: John Wiley & Sons, 1977.

[32]  LibreTexts Statistics, “12.5: Interpretation of r-Values,” last update Oct. 21, 2024.  Retrieved from https://stats.libretexts.org/Bookshelves/Introductory_Statistics/Statistics%3A_Open_for_Everyone_(Peter)/12%3A_Bivariate_Correlation/12.05%3A_Interpretation_of_r-Values

[33]  M. Wilson and R. Carmody, “Possible microbial impact on iron and manganese concentration in a stream ecosystem at Howard Community College,” February 17, 2025, unpublished.

[34]  Salt Watch, “Salt Watch Annual Report 2024-2025,” Izaak Walton League of America, July 2025. Retrieved from Salt-Watch-Annual-Report-2024-2025.pdf

[35]  K. Patel, “A salt crisis is looming for U.S. rivers,” The Washington Post, June 4, 2025. Retrieved from https://www.washingtonpost.com/climate-environment/2025/06/04/freshwater-rivers-saltier/