Continuous Water Quality Monitoring Research Articles

WATER QUALITY ANALYSIS FROM A MINING AREA. CASE STUDY: TARGU OCNA SALT MINE, ROMANIA

To resolve the water source quality problems for different water demand it is very important to establish in each zone the water sources (surface or/and groundwater) and the main source of pollution for these. By a continuous or periodical water quality monitoring, in industrial zone, can be avoided the major risks of environmental pollution and human health negative implications. The salt load and the specific salt composition, differing from natural rivers from a common area compared with surface water from a salt mining zone. The present paper aims to establish some surface and groundwater quality analysis parameters from an industrial area. The present study is focused on the Targu Ocna mining area, with salt extraction activities and the quality analyzed rivers were Trotus, Slanic, Valcica, two drilling and a fountain. In this sense, were measured 12 water quality parameters, some values were obtained in situ with portable equipment s and other parameters were determined in the research laboratory.

Filtration for improving surface water quality of a eutrophic lake

Algal blooms and the presence of cyanotoxins in surface water restrict the public from accessing lakes and beaches for drinking and recreational activities. An effort was taken in this on-site study to improve the surface water quality of a eutrophic lake, which has been under a swimming advisory for many years. A floating filtration unit with non-woven geotextiles as a sole filter media was tested for removing algae, nutrients, and suspended solids from overlying water under different lake conditions. Three non-woven geotextiles of different pore sizes were examined in different combinations and lake water quality was monitored for different physico-chemical, biological parameters. A YSI-EXO2 multiparameter probe was used for continuous online water quality monitoring during filtration. Depending on the initial water quality, excellent removal efficiency was observed as follows: 85–98% turbidity, 98–100% total suspended solids (TSS), 57–88% total phosphorus (TP), 33–66% chemical oxygen demand (COD) and 80–96% chlorophyll a (Chl. a.). The filtered lake water quality satisfied the norm set for oligotrophic lakes for TP and Chl. a. Results from this on-site study are very promising, showing the potential applicability of geotextile filtration as an ecologically attractive technique to improve the surface water quality of small aquatic bodies.

Metal Fractionation in Surface Sediments of the Brahmaputra River and Implications for Their Mobilization.

The Brahmaputra River is the largest tropical river in India that flows along the Himalayan regions and it is the lifeline of millions of people. Metal fractionation in the Brahmaputra River’s surface sediments and its correlation with turbidity are assessed in this study. The interaction between metal fractions and the overlying water is studied using multivariate statistical analyses. The strong positive correlation between NH4 of the overlying water and the exchangeable fractions in sediments signifies that the metals in the exchangeable fractions can be substituted by NH4. Subsequently, these metals can be released into the overlying water. The fluctuation in turbidity from 73 to 875 NTU indicates a large variation in the suspended matter concentration, and a higher concentration of suspended matter could provide attachment sites for pollutants such as metals. Significant variation in turbidity manifests a potentially high risk of pollution. In addition, the observation of local people along the Brahmaputra River turning its color to muddy indicates the need for continuous monitoring of water quality and an assessment of pollution is crucial. Although the Brahmaputra River’s risk assessment code is at low risk, the exchangeable fractions of Ni and Zn are present at all sites. Thus, the Brahmaputra River requires early preventive measures and management strategies to control metal pollution. This study contributes to an understanding of the fluctuation of turbidity of a tropical river. We provide baseline data for policymakers, and the importance of further intensive studies on metal pollution in the Himalayan Rivers is highlighted.

Emerging Tools for Design and Implementation of Water Quality Monitoring Based on IoT

The article provides a conceptual framework for developing real-time water monitoring system based on IoT technology. The process, strategy and knowledge base for multidisciplinary research on IoT systems and prerequisites for real-world application of IoT technology into continuous water quality monitoring are discussed. The study expands current efforts aimed at leveraging customized IoT solutions for better instrumentation and the continued integration of sensor data into networks. The process of system design from scratch and base components of IoT-based water quality monitoring system for surface water are described. While the focus of this article is on system design, opportunities to improve the system components for the management of water resources with continuous water quality monitoring are much broader. In this view, perspectives and development issues of IoT-based water quality monitoring are also discussed.

Remote-sensing-based algorithms for water quality monitoring in Olushandja Dam, north-central Namibia

AbstractFrequent and continuous water quality monitoring of Olushandja Dam in Namibia is needed to inform timely decision making. This study was carried out from November 2014 to June 2015 with Landsat 8 reflectance values and field measured water quality data that were used to develop regression-analysis-based retrieval algorithms. Water quality parameters considered included turbidity, total suspended solids (TSS), nitrates, ammonia, total nitrogen (TN), total phosphorus (TP) and total algae counts. Results show that turbidity levels exceeded the recommended limits for raw water for potable water treatment while TN and TP values are within acceptable values. Turbidity, TN, and TP and total algae count showed a medium to strong positive linear relationship between Landsat predicted and measured water quality data while TSS showed a weak linear relationship. The regression coefficients between predicted and measured values were: turbidity (R2 = 0.767); TN (R2 = 0.798,); TP (R2 = 0.907); TSS (R2 = 0.284,) and total algae count (R2 = 0.851). Prediction algorithms are generally the best fit to derive water quality parameters. Remote sensing is recommended for frequent and continuous monitoring of Olushandja Dam as it has the ability to provide rapid information on the spatio-temporal variability of surface water quality.

Evaluation of the impact of suspended particles on the UV absorbance at 254 nm (UV254) measurements using a submersible UV-Vis spectrophotometer.

There is an increasing need to use online instrumentation for continuous monitoring of water quality. However, industrial applications using online instruments, such as submersible UV-Vis spectrophotometers, may require the use of alternative techniques to remove particle effect rather than performing a physical filtration step. Some submersible UV-Vis spectrophotometers have built-in generic particle compensation algorithms to remove the filtration step. This work studied the influence of suspended particles on the measurements of a submersible UV-Vis spectrophotometer as well as the performance of the built-in particle compensation technique under laboratory-controlled conditions. Simulated water samples were used in the combinations of standard particles from laboratory chemical and natural particles extracted from water systems with ultrapure water and treated water from a drinking water treatment plant. Particle contributions to the UV absorbance at 254 nm (UV254) measurements of water samples varied differently when particle types or concentrations changed. The compensated UV254, measured by the submersible instrument using the built-in generic particle compensation algorithms, was compared with laboratory UV254, analysed by the bench-top instrument with the physical filtration method. The results indicated that the built-in generic compensation algorithms of the submersible UV-Vis spectrophotometer may generate undercompensated UV254 or overcompensated UV254 for various surface waters. These findings provide in-depth knowledge about the impact of suspended particles on the measurements of submersible UV-Vis spectrophotometers; source water dependence; and why site-specific calibration is often needed to get accurate measurements.

Using machine learning to understand the implications of meteorological conditions for fish kills

Fish kills, often caused by low levels of dissolved oxygen (DO), involve with complex interactions and dynamics in the environment. In many places the precise cause of massive fish kills remains uncertain due to a lack of continuous water quality monitoring. In this study, we tested if meteorological conditions could act as a proxy for low levels of DO by relating readily available meteorological data to fish kills of grey mullet (Mugil cephalus) using a machine learning technique, the self-organizing map (SOM). Driven by different meteorological patterns, fish kills were classified into summer and non-summer types by the SOM. Summer fish kills were associated with extended periods of lower air pressure and higher temperature, and concentrated storm events 2–3 days before the fish kills. In contrast, non-summer fish kills followed a combination of relatively low air pressure, continuous lower wind speed, and successive storm events 5 days before the fish kills. Our findings suggest that abnormal meteorological conditions can serve as warning signals for managers to avoid fish kills by taking preventative actions. While not replacing water monitoring programs, meteorological data can support fishery management to safeguard the health of the riverine ecosystems.

Spatial Variability and Associated Risk Assessment of Nitrate and Ammonium Concentration in Hail Haor Wetland, Northeastern Bangladesh

Nitrate-nitrogen and ammonium-nitrogen concentration were evaluated to assess the current nutrient condition of an ecologically as well as economically important wetland of Bangladesh. The contamination condition of surface water nitrate and ammonium of Hail Haor wetland was assessed to understand its probable risk to human health using the water samples collected from total-fifty monitoring stations, 25 each for two different seasons within 2018 to 2019. Nitrate concentration was measured using the spectrophotometer by colorimetric method, whereas ammonium was quantified using the micro Kjeldahl's distillation method. Statistical and geo-spatial analysis revealed an extensive understanding of the temporal and spatial variability as well as possible source identification of the nutrients in the studied area. Medium to a low level of nitrate ranging from 0.95 up to 9.25 mg/L and high ammonium with values from 0.32 up to 1.92 mg/L was a sporadic trend observed in wetland water, with low water season having more concentration than that of wet high water season. Hazard quotients (HQ) obtained from wetland water for four individual age groups recognized wetland water to be safe for consumption (HQ<1) based on its nitrate level. Wetlands can provide valuable ecosystem services, conserve the native and exotic wetland inhabitants and mitigate flood along with their purifying properties of nutrient removal to lessen the public health concerns about nitrate-nitrogen in drinking water. Therefore, appropriate rein in measures, incentivizing water use efficiency, and continuous monitoring of water quality will ensure safety for wetland habitats, also may aware of the risks or opportunities associated with using wetland water for both drinking and agricultural use.

Hydrogeochemical evaluation, suitability, and health risk assessment of groundwater in the watershed of Godavari basin, Maharashtra, Central India

In this investigation, the geochemical progression of a total of 31 groundwater samples of pre-monsoon season was assessed with categorization based on entropy weight water quality index and risk assessment on public health in the semi-arid area of Godavari basin, Maharashtra, Central India. Graphically, the major groundwater types identified were Ca-HCO3, mixed Ca-Mg-Cl, and mixed Ca-Na-HCO3. Based on [Mg2+/Na+] with [Ca2+/Na+] and [HCO3-/Na+] with [Ca2+/Na+] plots, carbonate and silicate weathering were identified as a major geochemical process governing groundwater chemistry. The presence of reverse ion exchange process was authenticated by (Ca2+ + Mg2+) vs. (HCO3- + SO42-) and Na+ + K+-Cl- vs. (Ca2+ + Mg2+)-(HCO3- + SO42-) plots. The saturation index values for calcite and dolomite showed that these minerals were in dissolution state. The dissolution of gypsum, dolomite, and anhydrite increased Ca2+ load in groundwater which accelerated the precipitation of calcite. The high toxic level of NO3- (> 45mg/L) was identified in 64.5% of the 31 groundwater samples, whereas F- concentration exceeded the threshold value in 12.9% of samples. Based on the entropy weight water quality index values, 70% of the samples were found to have moderate quality for drinking. In addition, health risk evaluation showed that the total hazard, due to fluoride and nitrate through oral pathways, was much higher than that through the dermal pathway. Children were found to be at high risk due to the consumption of NO3- and F- contaminated water. The calculated irrigation water quality index (IWQI) diverge from 7.4-89.2, expressing excellent to good quality for irrigation. Based on the irrigation water quality index, 90.3% of samples were found excellent for irrigation and 6.4% of good quality for irrigation. Authors recommend that continuous water quality monitoring programs along with effective management practices should be developed to avoid excessive extraction of groundwater.

Estimating Chlorophyll-a of Inland Water Bodies in Greece Based on Landsat Data

Assessing chlorophyll-a (Chl-a) pigments in complex inland water systems is of key importance as this parameter constitutes a major ecosystem integrity indicator. In this study, a methodological framework is proposed for quantifying Chl-a pigments using Earth observation (EO) data from Landsat 7 Enhanced Thematic Mapper Plus (ETM+) and 8 Operational Land Imager (OLI) sensors. The first step of the methodology involves the implementation of stepwise multiple regression (MLR) analysis of the available Chl-a dataset. Then, principal component analysis (PCA) is performed to explore Greek lakes’ potential interrelationships based on their Chl-a values in conjunction with certain criteria: their characteristics (artificial/natural), typology, and climatic type. Additionally, parameters such as seasonal water sampling and the date difference between sampling and satellite overpass are taken into consideration. Next, is implemented a stepwise multiple regression analysis among different groups of cases, formed by the criteria indicated from the PCA itself. This effort aimed at exploring different remote sensing-derived Chl-a algorithms for various types of lakes. The practical use of the proposed approach was evaluated in a total of 50 lake water bodies (natural and artificial) from 2013–2018, constituting the National Lake Network Monitoring of Greece in the context of the Water Framework Directive (WFD). All in all, the results evidenced the suitability of Landsat data when used with the proposed technique to estimate log-transformed Chl-a. The proposed scheme resulted in the development of models separately for natural (R = 0.78; RMSE = 1.3 μg/L) and artificial lakes (R = 0.76; RMSE = 1.29 μg/L), while the model developed without criteria proved weaker (R = 0.65; RMSE = 1.85 μg/L) in comparison to the other ones examined. The methodological framework proposed herein can be used as a useful resource toward a continuous monitoring and assessment of lake water quality, supporting sustainable water resources management.

Direct Mass Spectrometry with Online Headspace Sample Pretreatment for Continuous Water Quality Monitoring

This study investigates the use of selected ion flow tube mass spectrometry with an automated headspace pretreatment system for the continuous surveillance of water quality at wastewater treatment plants (WWTPs) and rivers. The reaction rates of the target compounds introduced using the headspace method were similar to those of the mass scan library, with a margin of error of <10%. Novel quantitative formulae were derived for the water samples of the target compounds, and the linearity of the calibration curves for both the purified and effluent matrix (0.1–2.0 mg/L) showed a coefficient of determination of 0.98–0.99 for most compounds. The detection limit for 74% of the target substances was 0.02–0.10 mg/L, and the average recoveries were 111.6% and 104.7% for the low- and high-concentration spiked samples, respectively, which are comparable to those of the headspace gas chromatography-mass spectrometry system. However, the variability in individual concentrations was still large, due to the unstable control of sample injection flow and pressure. Herein, 79% of the 28 compounds met one-tenth of the proposed method detection limit criteria for emergency operations in WWTP. Field experiments showed that the system was easy to maintain and could be used to monitor chemical accidents.

Severe western Canadian wildfire affects water quality even at large basin scales

Wildfires can have severe and lasting impacts on the water quality of aquatic ecosystems. However, our understanding of these impacts is founded primarily from studies of small watersheds with well-connected runoff regimes. Despite the predominance of large, low-relief rivers across the fire-prone Boreal forest, it is unclear to what extent and duration wildfire-related material (e.g., ash) can be observed within these systems that typically buffer upstream disturbance signals. Following the devastating 2016 Fort McMurray wildfire in western Canada, we initiated a multi-faceted water quality monitoring program that suggested brief (hours to days) wildfire signatures could be detected in several large river systems, particularly following rainfall events greater than 10 mm. Continuous monitoring of flow and water quality showed distinct, precipitation-associated signatures of ash transport in rivers draining expansive (800–100,000 km2) and partially-burned (<1–22 percent burned) watersheds, which were not evident in nearby unburned regions. Yields of suspended sediment, nutrients (nitrogen, phosphorus) and metals (lead, others) from impacted rivers were 1.2–10 times greater than from those draining unburned regions. Post-fire suspended sediment concentrations in impacted rivers were often larger than pre-fire 95% prediction intervals based on several years of water sampling. These multiple lines of evidence indicate that low-relief landscapes can mobilize wildfire-related material to rivers similarly, though less-intensively and over shorter durations, than headwater regions. We propose that uneven mixing of heavily-impacted tributaries with high-order rivers may partially explain detection of wildfire signals in these large systems that may impact downstream water users.

Impacts of gold mine effluent on water quality in a pristine sub-Arctic river

Impacts of mining on water quality are a great concern in the Arctic region. This study evaluated the impact of pre-treated mine effluent on river water quality. The study was conducted along the Seurujoki River in sub-Arctic Finland, which is impacted by Kittilä gold mine. The study analyzed water quality and hydrological data upstream and downstream of the mining area over an eight-year period, including a tailing dam leakage event in 2015. The analysis focused on water quality determinants such as electrical conductivity (EC), sulfate, antimony, manganese, and total nitrogen (Ntotal). Descriptive statistics on river water at four stations along the river corridor showed negative impacts of mining activities on the recipient water body. In order to find an indicator for water quality, correlation analysis between the water quality determinants was carried out. It identified EC as a good indicator for continuous water quality monitoring, especially to detect mining accidents such as partial failure of a tailings dam. The results showed increasing contaminant concentrations due to mining as more mine effluent was generated over time. A linear mixed model was developed to predict the coefficient of different elements affecting EC at river water monitoring stations impacted by mining effluents. The results provide new information on how to assess mining water impacts and plan future water quality monitoring.

Monitoring for the abundance and distribution of phytoplankton along the hypersaline Bardawil Lagoon, in Northern Sinai, Egypt (Multivariate Analysis Approach)

Bardawil Lagoon is a source of wildlife and high-quality fish, so, it is imperative that it should be subject to continuous monitoring of both water quality and the phytoplankton composition. Samples were collected from ten sites representing different habitats of the lagoon, from summer 2017 to spring 2018. The water properties were determined; noticeable low concentrations of dissolved inorganic nutrients were recorded. A total of 106 species belonging to six groups were recorded, among them Bacillariophyta (69 species) and Dinophyta (26 species). The obtained results clarify that Bacillariophytes prevailed in the eastern sites of Bardawil; on the contrary, Dinophytes occupied the sovereignty in the western sites of the lagoon. The most abundant Bacillariophytes were Licmophora gracilis, Gyrosigma acuminatum, Fragilaria construens, and Nitzschia sigmoidea; which formed more than 50% of total abounded diatoms. While, Protoperidinium leonis, Prorocentrum gracile and Protoperidinium granii were the dominant Dinophytes, giving it maximum at El-Rewak. From reviewing previous studies on the lagoon, it becomes clear that the phytoplankton composition varied widely from previously recorded, including the dominant species and their rates of sovereignty, which reflected economically on its fish productivity. Besides, the water quality in Bardawil fluctuated from slightly-polluted to moderately-polluted according to the diversity index, as a result of the recent, noticeable increase in human activities, especially fishing, which will be in turn reflected in the environment. Therefore, continuous follow-up through ecological assessment and monitoring studies of Bardawil became an urgent necessity.

Investigation and Analysis on Rural Domestic Sewage Discharge in Key Watersheds

In order to grasp the latest changes in the quality and quantity of domestic sewage in the Taihu Lake Basin and Chaohu Lake Basin, and to scientifically guide the construction of rural sewage treatment demonstration projects, two typical villages were selected from Taihu Lake Basin and Chaohu Lake Basin respectively for fieldwork continuous monitoring of sewage water quality and quantity. The results show that the typical rural per capita daily drainage in the plain area of the Taihu Lake Basin in summer is 89.4 L·cap−1·d−1, which is significantly higher than that in the typical rural area of the Chaohu Lake Basin (56.5 L·cap−1·d−1). The amount of sewage hourly discharge fluctuated distinctly, and the coefficient of variation of sewage discharge in the two villages ranged from 2.4 to 6.1, showing obvious characteristics of “morning and evening peaks” or “morning, noon and evening peaks”. In addition, the rural sewage water quality in the two places shows the same pattern, and the pollutant concentration of sewage from different sources is significantly different. It was found that the toilet-flush sewage in Wangjiatang village provided 84.42% of TN and 92.94% of NH4 +-N in the total domestic sewage, while toilet-flush sewage in Qianying village provided 74.31% of TN and 84.56% of NH4 +-N in the total domestic. Therefore, when constructing a sewage treatment demonstration project, the ideological principles of sewage classification and treatment from the source should be considered.

Development of the delta-normal stress combining CE-QUAL-W2 as a novel method for spatio-temporal monitoring of water quality in Karkheh Dam Reservoir.

Continuous monitoring of water quality in dam reservoirs is a typically difficult and costly operation. In this study, the results of computer modeling with the CE-QUAL-W2 model were combined with data mining techniques to develop a new method called "delta-normal stress" for identifying the critical temporal and spatial monitoring ranges. For this purpose, long-term variations of three quality parameters including nitrite-nitrate level, dissolved oxygen (DO) level, and water temperature near the outlet of the dam, which is the point of interest for reservoir exploitation, were analyzed. Based on this analysis, the time intervals and depth ranges with the highest frequency of significant variations in terms of each parameter were identified. The results showed that given the difference between the delta-normal stress trend of temperature and that of other parameters in Karkheh Dam Reservoir, temperature can be monitored at much lower sampling resolutions and using cheaper methods and equipment without sacrificing accuracy. Based on the frequency of occurrence of delta-normal stress of more than 20% above the total average, the key sampling times and locations for nitrite-nitrate and DO levels were determined to be the periods of January-February, February-March, and March-April, and depths of 60, 55, 50, and 5m, respectively.

Multivariate statistical analysis applied to the evaluation of groundwater quality in the central-southern portion of the state of Bahia - Brazil

The objective of this study was to identify and evaluate the variables responsible for contributing to possible natural and/or human contamination in groundwater of the semiarid region of the state of Bahia, seeking to subsidize water quality monitoring and management actions in the area. To do so, multivariate analysis techniques regarding factorial analysis in principal components and cluster analysis were used. The factorial analysis allowed the grouping of variables into two principal factors that explained 93% of total accumulated variance. Variables were strongly related to concentrations of metals and salinity in the water. The cluster analysis was used to classify water sources according to the quality of waters into three clusters in each factor. The natural background of the rocks of the municipality of Boquira was shown to influence water resources. A continuous (during dry and rainy seasons) monitoring of water quality from wells and springs located upstream and downstream from contamination sources is recommended, even if these waters are not used for public supply, to determine possible contamination plumes from contaminated material.

Estimation of high frequency nutrient concentrations from water quality surrogates using machine learning methods

Continuous high frequency water quality monitoring is becoming a critical task to support water management. Despite the advancements in sensor technologies, certain variables cannot be easily and/or economically monitored in-situ and in real time. In these cases, surrogate measures can be used to make estimations by means of data-driven models. In this work, variables that are commonly measured in-situ are used as surrogates to estimate the concentrations of nutrients in a rural catchment and in an urban one, making use of machine learning models, specifically Random Forests. The results are compared with those of linear modelling using the same number of surrogates, obtaining a reduction in the Root Mean Squared Error (RMSE) of up to 60.1%. The profit from including up to seven surrogate sensors was computed, concluding that adding more than 4 and 5 sensors in each of the catchments respectively was not worthy in terms of error improvement.

Hydrogeochemical baseline in a human-altered landscape of the central Pacific coast of Costa Rica.

Groundwater pollution in tropical and human-altered coastal landscapes is receiving novel attention due to decreasing in annual recharge as a consequence of recurrent droughts and overexploitation, whereby saline intrusion, point and diffuse source contamination, and water conflicts are common denominators. This study presents a detailed groundwater evaluation in a coastal aquifer within the central Pacific coast of Costa Rica. Three sampling campaigns including major ions, heavy metals, and fecal coliform analyses were conducted between July 2013 and March 2014 across 17 wells within the alluvial and fissured units of the Jacó aquifer. The groundwater system is classified as mixed HCO3--Ca2+-Mg2+ type. Coliforms presence was found in two wells, nearby Mona Creek headwaters and near the coastal line. Heavy metal concentrations were below quantification limits in most of the wells; however, chromium concentrations up to 6.56μg/L were quantified within the coastal line and central portion of the alluvial aquifer in 20 out of the 48 samples. The spatial distribution of major ions (K+, Na+, Ca2+, Mg2+, Cl-, SO42-, and HCO3-) exhibited an increasing trend towards the central portion of the alluvial aquifer, which may be potentially associated with the large unregulated urban expansion, invoking a need of a continuous water quality monitoring program in this touristic hot spot. This study provides useful information for other similar coastal aquifers in Central America, whereby increasing population growth and unregulated touristic, industrial, and agricultural activities are posing a truly challenge to ensure water security and sustainability parallel to the economic development in a changing climate.

Microbial source tracking (MST) in Chattahoochee River National Recreation Area: Seasonal and precipitation trends in MST marker concentrations, and associations with E. coli levels, pathogenic marker presence, and land use

Escherichia coli levels in recreational waters are often used to predict when fecal-associated pathogen levels are a human health risk. The reach of the Chattahoochee River that flows through the Chattahoochee River National Recreation Area (CRNRA), located in the Atlanta-metropolitan area, is a popular recreation area that frequently exceeds the U.S. Environmental Protection Agency beach action value (BAV) for E. coli. A BacteriALERT program has been implemented to provide real-time E. coli estimates in the reach and notify the public of potentially harmful levels of fecal-associated pathogens as indicated by surrogate models based on real-time turbidity measurements from continuous water quality monitoring stations. However, E. coli does not provide information about the sources of fecal contamination and its accuracy as a human health indicator is questionable when sources of contamination are non-human. The objectives of our study were to investigate, within the Park and surrounding watersheds, seasonal and precipitation-related patterns in microbial source tracking marker concentrations of possible sources (human, dog, and ruminant), assess correlations between source contamination levels and culturable E. coli levels, determine which sources best explained model-based E. coli estimates above the BAV and detection of esp2 (a marker for the esp gene associated with pathogenic strains of Enterococcus faecium and Enterococcus faecalis), and investigate associations between source contamination levels and land use features. Three BacteriALERT sites on the Chattahoochee River were sampled six times per season in the winter and summer from December 2015 through September 2017, and 11 additional stream sites (synoptic sites) from the CRNRA watershed were sampled once per season. Samples were screened with microbial source tracking (MST) quantitative PCR (qPCR) markers for humans (HF183 Taqman), dogs (DogBact), and ruminants (Rum2Bac), the esp2 qPCR marker, and culturable E. coli. At the BacteriALERT sites, HF183 Taqman concentrations were higher under wet conditions DogBact concentrations were greater in the winter and under wet conditions, and Rum2Bac concentrations were comparatively low throughout the study with no difference across seasons or precipitation conditions. Concentrations of HF183 Taqman, DogBact, and Rum2Bac were positively correlated with culturable E. coli concentrations; however, DogBact had the largest R2 value among the three markers, and the forward stepwise regression indicated it was the best predictor of culturable E. coli concentrations at the BacteriALERT sites. Recursive partitioning indicated that BAV exceedances of model-based E. coli estimates were best explained by DogBact concentrations ≥3 gene copies per mL (CN/mL). Detections of esp2 at BacteriALERT sites were best explained by DogBact concentrations ≥11 CN/mL, while detections of esp2 at synoptic sites were best explained by HF183 Taqman ≥29 CN/mL. At the synoptic sites, HF183 Taqman levels were associated with wastewater treatment plant density. However, this relationship was driven primarily by a single site, suggesting possible conveyance issues in that catchment. esp2 detections at synoptic sites were positively associated with development within a 2-km radius and negatively associated with development within the catchment, suggesting multiple sources of esp2 in the watershed. DogBact and Rum2Bac were not associated with the land use features included in our analyses. Implications for Park management include: 1) fecal contamination levels were highest during wet conditions and in the off season when fewer visitors are expected to be participating in water-based recreation, 2) dogs are likely contributors to fecal contamination in the CRNRA and may be sources of pathogenic bacteria indicating further investigation of the origins of this contamination may be warranted as would be research to understand the human health risks from exposure to dog fecal contamination, and 3) high levels of the human marker at one site in the CRNRA watershed suggests more extensive monitoring in that catchment may locate the origin of human fecal contamination detected during this study.