Tryptophan-like and humic-like fluorophores are extracellular in groundwater: implications as real-time faecal indicators

Uncertainty in Environmental Micropollutant Modeling

Water pollution policies have been enacted across the globe to minimize the environmental risks posed by micropollutants (MPs). For regulative institutions to be able to ensure the realization of environmental objectives, they need information on the environmental fate of MPs. Furthermore, there is an urgent need to further improve environmental decision-making, which heavily relies on scientific data. Use of mathematical and computational modeling in environmental permit processes for water construction activities has increased. Uncertainty of input data considers several steps from sampling and analysis to physico-chemical characteristics of MP. Machine learning (ML) methods are an emerging technique in this field. ML techniques might become more crucial for MP modeling as the amount of data is constantly increasing and the emerging new ML approaches and applications are developed. It seems that both modeling strategies, traditional and ML, use quite similar methods to obtain uncertainties. Process based models cannot consider all known and relevant processes, making the comprehensive estimation of uncertainty challenging. Problems in a comprehensive uncertainty analysis within ML approach are even greater. For both approaches generic and common method seems to be more useful in a practice than those emerging from ab initio. The implementation of the modeling results, including uncertainty and the precautionary principle, should be researched more deeply to achieve a reliable estimation of the effect of an action on the chemical and ecological status of an environment without underestimating or overestimating the risk. The prevailing uncertainties need to be identified and acknowledged and if possible, reduced. This paper provides an overview of different aspects that concern the topic of uncertainty in MP modeling.

Seasonal variations in dissolved organic matter concentration and composition in an outdoor system for bank filtration simulation

Dissolved organic matter (DOM) in surface waters can vary markedly in character depending on seasonal variations such as rainfall intensity, UV radiations and temperature. Changes in DOM as well as temperature and rainfall intensity over the year can affect the biochemical processes occurring in bank filtration (BF). Identification and characterization of DOM in the surface water could help to optimize the water treatment and provide stable and safe drinking water. This study investigated year-long variations of DOM concentrations and compositions in a surface water of a circulated outdoor pond (research facility) connected to a BF passage. DOM was dominated by humic substances and a changing pattern of DOM in surface water was observed throughout the year. A significant increase of DOM (∼ 38%) in surface water was noted in August compared to November. The fluorescent DOM showed that DOM in summer was enriched with the degradable fraction whilst non-degradable fraction was dominated in winter. A constant (1.7 ± 0.1 mg/L) effluent DOM was recirculated in the system throughout the year. DOM removal through BF varied between 4% to 39% and was achieved within a few meters after infiltration and significantly correlated with influent DOM concentration (R2 = 0.82, p < 0.05). However, no significant (p > 0.05) change in the removal of DOM was observed in two subsurface layers (upper and lower). This study highlights the presence of a constant non-degradable DOM in the bank filtrate, which was not affected by temperature, redox conditions and UV radiations.

In Situ Water Quality Monitoring Using an Optical Multiparameter Sensor Probe

Optical methods such as ultraviolet/visible (UV/Vis) and fluorescence spectroscopy are well-established analytical techniques for in situ water quality monitoring. A broad range of bio-logical and chemical contaminants in different concentration ranges can be detected using these methods. The availability of results in real time allows a quick response to water quality changes. The measuring devices are configured as portable multi-parameter probes. However, their specification and data processing typically cannot be changed by users, or only with difficulties. Therefore, we developed a submersible sensor probe, which combines UV/Vis and fluorescence spectroscopy together with a flexible data processing platform. Due to its modular design in the hardware and software, the sensing system can be modified to the specific application. The dimension of the waterproof enclosure with a diameter of 100 mm permits also its application in groundwater monitoring wells. As a light source for fluorescence spectroscopy, we constructed an LED array that can be equipped with four different LEDs. A miniaturized deuterium-tungsten light source (200-1100 nm) was used for UV/Vis spectroscopy. A miniaturized spectrometer with a spectral range between 225 and 1000 nm permits the detection of complete spectra for both methods.

Modeled predictions of human-associated and fecal-indicator bacteria concentrations and loadings in the Menomonee River, Wisconsin using in-situ optical sensors

Human sewage contamination of waterways is a major issue in the United States and throughout the world. Models were developed for estimation of two human-associated fecal-indicator and three general fecal-indicator bacteria (HIB and FIB) using in situ optical field-sensor data for estimating concentrations and loads of HIB and FIB and the extent of sewage contamination in the Menomonee River in Milwaukee, Wisconsin. Three commercially available optical sensor platforms were installed into an unfiltered custom-designed flow-through system along with a refrigerated automatic sampler at the Menomonee River sampling location. Ten-minute optical sensor measurements were made from November 2017 to December 2018 along with the collection of 153 flow-weighted discrete water samples (samples) for HIB, FIB, dissolved organic carbon (DOC), and optical properties of water. Of those 153 samples, 119 samples were from event-runoff periods, and 34 were collected during low-flow periods. Of the 119 event-runoff samples, 43 samples were from event-runoff combined sewer overflow (CSO) influenced periods (event-CSO periods). Models included optical sensor measurements as explanatory variables with a seasonal variable as an interaction term. In some cases, separate models for event-CSO periods and non CSO-periods generally improved model performance, as compared to using all the data combined for estimates of FIB and HIB. Therefore, the CSO and non-CSO models were used in final estimations for CSO and non-CSO time periods, respectively. Estimated continuous concentrations for all bacteria markers varied over six orders of magnitude during the study period. The greatest concentrations, loads, and proportion of sewage contamination occurred during event-runoff and event-CSO periods. Comparison to water quality standards and microbial risk assessment benchmarks indicated that estimated bacteria levels exceeded recreational water quality criteria between 34 and 96% of the entire monitoring period, highlighting the benefits of high-frequency monitoring compared to traditional grab sample collection. The application of optical sensors for estimation of HIB and FIB markers provided a thorough assessment of bacterial presence and human health risk in the Menomonee River.

Field evidences of fluorescent dissolved organic matter (FDOM) as potential fingerprints for agricultural and urban sources in river environment

Field evidences of the fluorescence differences between agricultural and urban river reaches are still lack. In this study, the middle reaches of Danhe River (DH) and Mihe River (MH) in Shouguang, China, were designed as agricultural and urban river reaches, respectively, to compare the the fluorescence differences in disparate river reaches using excitation-emission matrix coupled with parallel factor analysis (EEM-PARAFAC). Three fluorescence components were identified. C1 (Ex/Em=230,255,295 nm/420 nm) was categorized as humic-like fluorophores, C2 (Ex/Em=230,275 nm/330 nm) was recognized as tryptophan-like substances, and C3 (Ex/Em=215 nm/290 nm) was noted as tyrosine-like and phenylalanine-like compounds. The results showed that the FDOM posed significant differences between agricultural and urban river reaches (P < 0.001). The monitoring sites in DH were rich in C2 (1.90 ± 0.62 Raman Unit (RU), mean ± standard deviation), and the monitoring sites in MH were rich in C3 (1.32 ± 0.51 RU). Redundancy analysis revealed that C2 could be regarded as a fluorescence indicator of agricultural sewage in river environment, while C3 was recognized as a fluorescence indicator of domestic sewage in river environment. In conclusion, this study provided field evidences of FDOM as potential fingerprints for agricultural and urban sources in river environment.

A continuous, in-situ, near-time fluorescence sensor coupled with a machine learning model for detection of fecal contamination risk in drinking water: Design, characterization and field validation

We designed and validated a sensitive, continuous, in-situ, remotely reporting tryptophan-like fluorescence sensor and coupled it with a machine learning model to predict high-risk fecal contamination in water (>10 colony forming units (CFU)/100mL E. coli). We characterized the sensor's response to multiple fluorescence interferents with benchtop analysis. The sensor's minimum detection limit (MDL) of tryptophan dissolved in deionized water was 0.05 ppb (p <0.01) and its MDL of the correlation to E. coli present in wastewater effluent was 10 CFU/100 mL (p <0.01). Fluorescence response declined exponentially with increased water temperature and a correction factor was calculated. Inner filter effects, which cause signal attenuation at high concentrations, were shown to have negligible impact in an operational context. Biofouling was demonstrated to increase the fluorescence signal by approximately 82% in a certain context, while mineral scaling reduced the sensitivity of the sensor by approximately 5% after 24 hours with a scaling solution containing 8 times the mineral concentration of the Colorado River. A machine learning model was developed, with TLF measurements as the primary feature, to output fecal contamination risk levels established by the World Health Organization. A training and validation data set for the model was built by installing four sensors on Boulder Creek, Colorado for 88 days and enumerating 298 grab samples for E. coli with membrane filtration. The machine learning model incorporated a proxy feature for fouling (time since last cleaning) which improved model performance. A binary classification model was able to predict high risk fecal contamination with 83% accuracy (95% CI: 78% - 87%), sensitivity of 80%, and specificity of 86%. A model distinguishing between all World Health Organization established risk categories performed with an overall accuracy of 64%. Integrating TLF measurements into an ML model allows for anomaly detection and noise reduction, permitting contamination prediction despite biofilm or mineral scaling formation on the sensor's lenses. Real-time detection of high risk fecal contamination could contribute to a major step forward in terms of microbial water quality monitoring for human health.

Identification of groundwater pollution from livestock farming using fluorescence spectroscopy coupled with multivariate statistical methods

Extensive livestock farming has highly threatened groundwater quality, thereby necessitating a rapid and effective method to identify groundwater quality in such areas. Fluorescence spectroscopy has been recognized as an interpretable method for tracking anthropogenic influences on water quality, but its applicability in identifying the groundwater pollution from livestock farming remains unknown. In this study, the fluorescence characteristics of dissolved organic matter (DOM) in groundwater from a typical livestock farming area were investigated by using fluorescence excitation emission matrix (EEM)-parallel factor analysis (PARAFAC) coupled with multivariate statistical methods. The results showed that livestock farming significantly altered the content and composition of DOM in groundwater, and these effects were mainly observed in shallow groundwater in the study area. Hierarchical cluster analysis based on fluorescence parameters divided the groundwater samples into three clusters with significantly different pollution degrees: Cluster A, unpolluted; Cluster B, highly polluted; Cluster C, moderately polluted. In particular, the intensity of tryptophan-like fluorescence was high in the polluted groundwater but was almost undetectable in the unpolluted groundwater, suggesting that it is a potential indicator of groundwater quality. Principal component analysis based on the fluorescence parameters explained 91.5% of the variance with the first two principal components, and revealed that the degree of pollution dominated the fluorescence characteristics of groundwater in the study area. In addition, NO₃^- was abundant in Clusters B and C, while it was low in Cluster A, validating the analysis results of fluorescence spectroscopy. These findings indicated that DOM fluorescence was sensitive to livestock farming pollution and could be applied to identify, monitor, and assess groundwater pollution from livestock farming.

In-situ fluorescence spectroscopy is a more rapid and resilient indicator of faecal contamination risk in drinking water than faecal indicator organisms

Faecal indicator organisms (FIOs) are limited in their ability to protect public health from the microbial contamination of drinking water because of their transience and time required to deliver a result. We evaluated alternative rapid, and potentially more resilient, approaches against a benchmark FIO of thermotolerant coliforms (TTCs) to characterise faecal contamination over 14 months at 40 groundwater sources in a Ugandan town. Rapid approaches included: in-situ tryptophan-like fluorescence (TLF), humic-like fluorescence (HLF), turbidity; sanitary inspections; and total bacterial cells by flow cytometry. TTCs varied widely in six sampling visits: a third of sources tested both positive and negative, 50% of sources had a range of at least 720 cfu/100 mL, and a two-day heavy rainfall event increased median TTCs five-fold. Using source medians, TLF was the best predictor in logistic regression models of TTCs ≥10 cfu/100 mL (AUC 0.88) and best correlated to TTC enumeration (ρ_s 0.81), with HLF performing similarly. Relationships between TLF or HLF and TTCs were stronger in the wet season than the dry season, when TLF and HLF were instead more associated with total bacterial cells. Source rank-order between sampling rounds was considerably more consistent, according to cross-correlations, using TLF or HLF (min ρ_s 0.81) than TTCs (min ρ_s 0.34). Furthermore, dry season TLF and HLF cross-correlated more strongly (ρ_s 0.68) than dry season TTCs (ρ_s 0.50) with wet season TTCs, when TTCs were elevated. In-situ TLF or HLF are more rapid and resilient indicators of faecal contamination risk than TTCs.

Optical Properties of Water for Prediction of Wastewater Contamination, Human-Associated Bacteria, and Fecal Indicator Bacteria in Surface Water at Three Watershed Scales

Relations between spectral absorbance and fluorescence properties of water and human-associated and fecal indicator bacteria were developed for facilitating field sensor applications to estimate wastewater contamination in waterways. Leaking wastewater conveyance infrastructure commonly contaminates receiving waters. Methods to quantify such contamination can be time consuming, expensive, and often nonspecific. Human-associated bacteria are wastewater specific but require discrete sampling and laboratory analyses, introducing latency. Human sewage has fluorescence and absorbance properties different than those of natural waters. To assist real-time field sensor development, this study investigated optical properties for use as surrogates for human-associated bacteria to estimate wastewater prevalence in environmental waters. Three spatial scales were studied: Eight watershed-scale sites, five subwatershed-scale sites, and 213 storm sewers and open channels within three small watersheds (small-scale sites) were sampled (996 total samples) for optical properties, human-associated bacteria, fecal indicator bacteria, and, for selected samples, human viruses. Regression analysis indicated that bacteria concentrations could be estimated by optical properties used in existing field sensors for watershed and subwatershed scales. Human virus occurrence increased with modeled human-associated bacteria concentration, providing confidence in these regressions as surrogates for wastewater contamination. Adequate regressions were not found for small-scale sites to reliably estimate bacteria concentrations likely due to inconsistent local sanitary sewer inputs.

Laboratory In-Situ Production of Autochthonous and Allochthonous Fluorescent Organic Matter by Freshwater Bacteria

This work investigates the origin and range of fluorescent organic matter (FOM) produced in-situ by environmentally sourced freshwater bacteria. Aquatic FOM is an essential component in global carbon cycling and is generally classified as either autochthonous, produced in-situ via microbial processes, or allochthonous, transported into aquatic systems from external sources. We have demonstrated that, within laboratory model systems, environmentally sourced mixed microbial communities and bacterial isolates can produce and/or export FOM associated with both autochthonous and allochthonous material. This study focuses on fluorescence peak B, T, M, C and C+, exploring (1) the cellular nature of FOM produced, (2) FOM exported as extracellular material into the water column and (3) the impact of physical cell lysis on FOM signature. For the laboratory model systems studied, Peak T fluorescence is retained within bacterial cells (>68%), while Peak C fluorescence is mainly observed as extracellular material (>80%). Peak M is identified as both cellular and extracellular FOM, produced by all isolated freshwater microorganisms investigated. The origin of Peak C+ is postulated to originate from functional metabolites associated with specific microorganisms, seen specifically within the Pseudomonas sp. monoculture here. This work challenges the binary classification of FOM as either allochthonous or autochthonous, suggesting that FOM processing and production occurs along a dynamic continuum. Within this study, fluorescence intensity data for the environmental bacteria isolate monocultures are presented as enumeration corrected data, for the first time providing quantitative fluorescence data per bacterial colony forming unit (cfu). From this, we are able to assess the relative contribution of different bacteria to the autochthonous FOM pool and if this material is cellular or extracellular.

A year-long cyclic pattern of dissolved organic matter in the tap water of a metropolitan city revealed by fluorescence spectroscopy

Delivering drinking water with stable quality in metropolitan cities is a big challenge. This study investigated the year-long dynamics of dissolved organic matter (DOM) in the tap water and source water of a metropolitan city in southern China using fluorescence spectroscopy. The DOM detected in the tap water, and source water of Shenzhen city was season and location-dependent. A year-long cyclic trend of DOM was found with predominate protein-like fluorescence in the dry season compared to the humic-like enriched DOM in the wet season. A general DOM pattern was estimated by measuring the shift in dominant fluorescence regions on the excitation-emission matrix (EEM). The difference in fluorescent DOM (FDOM) composition (in terms of the ratio of protein-like to humic-like fluorescence) was above 200% between wet and dry seasons. The taps associated with reservoirs receiving water from the eastern tributary of Dongjiang River showed significant changes in protein-like contents than the taps with source water originating from the western part of the river. This study highlights the importance of optimizing drinking water treatment plants' operational conditions after considering seasonal changes and source water characteristics.