Water quality assessment of a small peri-urban river using low and high frequency monitoring

High-frequency data significantly enhances the prediction ability of point and interval estimation

Accurate prediction of dissolved oxygen (DO) dynamics is crucial for understanding the influence of environmental factors on the stability of aquatic ecosystem. However, limited research has been conducted to determine the optimal frequency of water quality monitoring that ensures continuous assessment of water health while minimizing costs. To address these challenges, the present study developed a hybrid stochastic hydrological model (i.e., ARIMA-GARCH hybrid model) and machine learning (ML) models. The objective of this study is to identify the best-performing model and establish the optimal monitoring frequency. Results revealed that high-frequency DO monitoring data exhibit greater variability compared to low-frequency data. Moreover, the ARIMA-GARCH model demonstrates promising potential in predicting DO concentrations for low-frequency monitoring data, surpassing ML models in performance. Furthermore, increasing the monitoring frequency significantly improves the prediction accuracy of models, regardless of whether point (with lower R2 values of 0.64 and 0.51 for daily detection than these of every 15 min (0.96 and 0.99) at CHQ and LHT, respectively) or interval predictions (with RIW higher values of 2.00 and 1.55 for daily detection higher than these of 0.02 and 0.16 in every 15 min at CHQ and LHT, respectively) are considered. Additionally, a 4 hourly monitoring frequency was found to be optimal for water quality assessment using each model. These findings identify the superior performing of the ARIMA-GARCH model and highlight the crucial role of monitoring frequency in enhancing DO prediction and improving model performance.

On inorganic tracers of wastewater treatment plant discharges along the Marque River (Northern France)

Increase of water quality in aquatic systems has become a hot button issue in recent decades. However, with the aim to implement an effective remediation strategy, the first step is to identify the sources of diffuse and point-source pollution using several tracers. In urban areas, B isotopes, Gd enrichment, Cl^- or carbamazepine concentrations can be used as wastewater treatment plant tracers. In this study, a focus was made on the quantification of a wide variety of inorganic compounds (elements, ions, isotopic ratios) all along the Marque River, a small stream located in Northern France receiving effluents coming from seven wastewater treatment plants (WWTPs). The objectives were (i) to determine the importance of the WWTPs discharge during low water events, (ii) to assess the efficiency of conventional tracers in quantifying the contribution of the WWTPs and (iii) to investigate new potential tracers less commonly used. The results have shown, through statistical analyses ANOVA (Analysis Of Variance) tests, PCA (Principal Component Analysis) and contribution calculations, that the WWTPs discharges strongly impact the water composition of all the watercourse and particularly during the first 6 km. However, due to high discharges of wastewaters not always well treated, some classical indicators (e.g. B, Rb/Sr) have shown limitations when used alone. The use of a set of relevant tracers including alkali metals could therefore be one solution for overcoming such a problem. Finally, other indicators like Rb/B or Gd/Pt ratios may also be a way to tackle this issue; they are indeed promising to discriminate the source of wastewaters.

Optimising punctual water sampling with an on-the-fly algorithm based on multiparameter high-frequency measurements

The way in which aquatic systems is sampled has a strong influence on our understanding of them, especially when they are highly dynamic. High frequency sampling has the advantage over spot sampling for representativeness but leads to a high amount of analysis. This study proposes a new methodology to choose when sampling accurately with an automated sampler coupled with a high frequency (HF) multiparameter probe. After each HF measurement, an optimised sampling algorithm (OSA) determines on-the-fly the relevance of taking a new sample in relation to previous waters already collected. Once the OSA was optimised, considering the number of HF parameters and their variabilities, it was demonstrated through a study case that the number of samples could be significantly reduced, while still covering periods of low and high variabilities. The comparison between the total HF dataset and the sampled subdataset shows that physicochemical parameter variability is preserved (Pearson correlations > 0.96) as well as the multiparameter variability (PCA axes remained similar with Tucker congruence > 0.99). This algorithm simplifies HF studies by making it easier to take samples during brief phenomena such as storms or accidental spills that are often poorly monitored. In addition, it optimises the number of samples to be taken to correctly describe a system and thus reduce the human and financial costs of these environmental studies.

The Impacts of Precipitation on Fluorescent Dissolved Organic Matter (FDOM) in an Urban River System

Precipitation is considered a key factor influencing the fluorescent dissolved organic matter (FDOM) of urban rivers. However, the multiple effects of precipitation on FDOM in urban rivers and the long-term impacts of precipitation on the spatial patterns of FDOM are seldom known. Spatiotemporal variations of FDOM at 36 sites from the urban rivers of Jinan City during dry and wet seasons were investigated in this study. Four components were identified using an excitation–emission matrix and parallel factor analysis. Overall, the total fluorescence intensities in dry and wet seasons ranged from 6.59 to 35.7 quinine sulfate units (QSU) and 3.42 to 69.3 QSU, respectively. Significant variations were found for different components that C2 and C3 declined but C4 increased in the wet season (p < 0.05). The temporal variations for different components could be explained by the different combined effects of precipitation dilution and flushing. Three different reference FDOM sources, including background water, spring water, and wastewater treatment plant (WWTP) outlets, were illustrated using principal coordinate analysis (PCoA). The places of FDOM in most sites were more closed to the PCoA location of WWTP outlets in the dry season while central shifted in the wet season. The changes of FDOM sources in the wet season could be explained by the mixed effect of precipitation. In conclusion, this study provided new insights into the multiple impacts of precipitation on FDOM in urban river systems, and also data support for precise pollution discharge and water resource management.

Spatiotemporal variations of surface water quality in a medium-sized river catchment (Northwestern Germany) with agricultural and urban land use over a five-year period with extremely dry summers

Medium-sized rivers, which are used for intensive agriculture and urban infrastructure, are subject to manifold hydrochemical stressors. Identifying and monitoring these stressors is important for river basin management and a functioning ecosystem. To understand the spatiotemporal variation of surface water quality in a highly modified lowland river, the Münstersche Aa River (Northwestern Germany) with 62% of land used for agriculture and 26% urban/residential area, was exemplarily studied. A total of 519 samples were collected using two automated high-frequency samplers and five catchment-wide sampling campaigns. They covered the five-year period 2015-2020 and included two extremely dry summers. The Münstersche Aa catchment is dominated by low permeable strata resulting in surface water runoff (Baseflow Index: 0.41) which leads to a high amplitude of discharge variation (mean discharge: 0.7 m³/s) with high flow conditions in winter/spring, and low discharge during summer/fall. In wintertime, maximum nitrate concentrations (up to 73 mg NO₃/L) and loads (up to 1300 t NO₃/a; up to 98% in winter) correlate with high-flow conditions. δ¹⁸O and δ¹⁵N isotopic analysis indicated manure from farmland as the major source of nitrate whereas the impact of municipal wastewater treatment plants was neglectable. Increased nitrate concentrations are linked to the higher proportion of farmland in the upper catchment (77%) compared with the lower catchment (47%). In summertime, at extremely low flow conditions, surface water consisted of up to 100% of treated wastewater, resulting in the highest measured chloride, sodium and potassium concentrations. The river is impacted by strongly seasonal and different stressors, which can be expected to intensify with ongoing climate change. Results from this study may help to adapt monitoring schemes for the Münstersche Aa but also for other lowland streams with comparable land-use targeting the goals of the Water Framework Directive.

Development and validation of DGT passive samplers for the quantification of Ir, Pd, Pt, Rh and Ru: A challenging application in waters impacted by urban activities

Platinum group elements (PGEs) are among the least abundant in the continental crust. They have become excellent tracers of anthropogenic activities, particularly due to their use in catalytic converters or in the medical industry. However, their quantification in environmental matrices is still problematic because of their low concentrations combined with the presence of interfering elements. Preconcentration methods are therefore necessary to measure accurate concentrations. In this study, the quantification of Ir, Rh, Ru, Pd and Pt was studied in depth by focusing on two resins: AG MP-1 (anion exchange) and Purolite® S-920 (chelating) with the aim of developing passive Diffusive Gradients in Thin films (DGT) samplers as in-situ pre-concentration tools. The characteristics of both resins (e.g. adsorption, elution, selectivity, etc.) were studied and the diffusion coefficients of PGEs in different matrices were determined. For the first time, carcinostatic platinum-based drugs were also studied. Better rates and percentages of adsorption were observed for S-920 while AG MP-1 was more selective with regard to spectral interferents and easier to elute. The diffusion coefficients of PGEs were resin-dependent, particularly for carcinostatic platinum-based drugs. For the first time, the applicability of these DGT samplers dedicated to PGEs was demonstrated in the field after their deployment in two wastewater treatment plants in Northern France for which concentrations were found to range from few pg L^-1 (Ir, Ru) to few ng L^-1 (Pt).

Improved Representation of Flow and Water Quality in a North-Eastern German Lowland Catchment by Combining Low-Frequency Monitored Data with Hydrological Modelling

Achievements of good chemical and ecological status of groundwater (GW) and surface water (SW) bodies are currently challenged mainly due to poor identification and quantification of pollution sources. A high spatio-temporal hydrological and water quality monitoring of SW and GW bodies is the basis for a reliable assessment of water quality in a catchment. However, high spatio-temporal hydrological and water quality monitoring is expensive, laborious, and hard to accomplish. This study uses spatio-temporally low resolved monitored water quality and river discharge data in combination with integrated hydrological modelling to estimate the governing pollution pathways and identify potential transformation processes. A key task at the regarded lowland river Augraben is (i) to understand the SW and GW interactions by estimating representative GW zones (GWZ) based on simulated GW flow directions and GW quality monitoring stations, (ii) to quantify GW flows to the Augraben River and its tributaries, and (iii) to simulate SW discharges at ungauged locations. Based on simulated GW flows and SW discharges, NO3-N, NO2-N, NH4-N, and P loads are calculated from each defined SW tributary outlet (SWTO) and respective GWZ by using low-frequency monitored SW and GW quality data. The magnitudes of NO3-N transformations and plant uptake rates are accessed by estimating a NO3-N balance at the catchment outlet. Based on sensitivity analysis results, Manning’s roughness, saturated hydraulic conductivity, and boundary conditions are mainly used for calibration. The water balance results show that 60–65% of total precipitation is lost via evapotranspiration (ET). A total of 85–95% of SW discharge in Augraben River and its tributaries is fed by GW via base flow. SW NO3-N loads are mainly dependent on GW flows and GW quality. Estimated SW NO3-N loads at SWTO_Ivenack and SWTO_Lindenberg show that these tributaries are heavily polluted and contribute mainly to the total SW NO3-N loads at Augraben River catchment outlet (SWO_Gehmkow). SWTO_Hasseldorf contributes least to the total SW NO3-N loads. SW quality of Augraben River catchment lies, on average, in the category of heavily polluted river with a maximum NO3-N load of 650 kg/d in 2017. Estimated GW loads in GWZ_Ivenack have contributed approximately 96% of the total GW loads and require maximum water quality improvement efforts to reduce high NO3-N levels. By focusing on the impacts of NO3-N reduction measures and best agricultural practices, further studies can enhance the better agricultural and water quality management in the study area.

Comparison of POCIS passive samplers vs. composite water sampling: A case study

The relevance of Polar Organic Chemical Integrative Samplers (POCIS) was evaluated for the assessment of concentrations of 46 pesticides and 19 pharmaceuticals in a small, peri-urban river with multi-origin inputs. Throughout the period of POCIS deployment, 24h-average water samples were collected automatically, and showed the rapid temporal evolution of concentrations of several micropollutants, as well as permitting the calculation of average concentrations in the water phase for comparison with those estimated from POCIS passive samplers. In the daily water samples, cyproconazol, epoxyconazol and imidacloprid showed high temporal variations with concentrations ranging from under the limit of detection up to several hundreds of ngL^-1. Erythromycin, cyprofloxacin and iopromide also increased rapidly up to tens of ngL^-1 within a few days. Conversely, atrazine, caffeine, diclofenac, and to a lesser extent carbamazepine and sucralose, were systematically present in the water samples and showed limited variation in concentrations. For most of the substances studied here, the passive samplers gave reliable average concentrations between the minimal and maximal daily concentrations during the time of deployment. For pesticides, a relatively good correlation was clearly established (R²=0.89) between the concentrations obtained by POCIS and those gained from average water samples. A slight underestimation of the concentration by POCIS can be attributed to inappropriate sampling rates extracted from the literature and for our system, and new values are proposed. Considering the all data set, 75% of the results indicate a relatively good agreement between the POCIS and the average water samples concentration (values of the ratio ranging between 0,33 and 3). Note further that this agreement between these concentrations remains valid considering different sampling rates extracted from the literature.

Spatial distribution and temporal variation of chemical properties of drainage watercourses in rural and peri-urban areas of Novi Sad (Serbia)-a case study

Waters are among to the most vulnerable environmental resources exposed to the impact of various point and non-point pollutants from rural/urban activities. Systematic and long-term monitoring of hydro-resources is therefore of crucial importance for sustainable water management, although such practice is lacking across many (agro-)hydro-ecosystems. In the presented study, for the first time, the spatial distribution (covering almost 9000 ha) and temporal variation (2006-2013) in certain quality parameters was characterized in drainage watercourses Tatarnica and Subic, whose catchment is rural and suburban areas close to the city of Novi Sad, Republic of Serbia. Based on majority of observed parameters, both watercourses belonged to I and II water quality classes, with occasional presence of certain parameters (e.g., suspended solids, total phosphorus; ammonium) at extreme values exacerbating both watercourses to classes IV and V. The value of the synthetic pollution index (i.e., a combined effect of all considered parameters) showed a higher degree of water pollution in watercourse Subic (on average 2.00) than Tatarnica (on average 0.72). Also, cluster analysis for watercourse Tatarnica detected two groups of parameters (mostly related to nutrients and organic matter), indicating more complex impacts on water quality during the observed period, in which elucidation thus established water quality monitoring program would be of great importance.