Polar organic chemical integrative samplers as an effective tool for chemical monitoring of surface waters - Results from one-year monitoring in France.

In an effort to support European Union Water Framework Directive goals, we have set up a national demonstrator project to identify the advantages and limitations of passive samplers for regulatory monitoring of polar contaminants in surface waters. Here we carried out successive 14 day-deployments of polar organic chemical integrative samplers (POCIS) for one year at three sites. In parallel, we used the passive sampler deployment/retrieval operations to collect spot water samples for comparative analysis. We observed that frequency of quantification was significantly higher in POCIS than spot samples for 29 contaminants, similar for 15, and lower for one, because POCIS lowered the limits of quantification for most contaminants (median value factor of 11). We built a database of sampling rates (Rs) according to quality indices to convert concentrations in POCIS to concentrations in water (23 contaminants with a high-quality median Rs value, 20 with an approximate Rs and two with no usable Rs). Several phenomena were observed over one-year monitoring period. For example, after a flood episode, dilution phenomenon in rivers is correctly observed by using POCIS sampling whereas significant concentration increased due to soil leaching is observed with both passive and spot sampling. Cases of episodic contamination that were missed by spot sampling were observed with POCIS as it was able to capture contamination of short duration but sufficient intensity. Contamination by pharmaceuticals was found to come from wastewater treatment plant discharges and showed relatively little variation over the course of the year in both POCIS and spot samples. POCIS enables more reliable annual monitoring of pesticide and pharmaceutical contamination than spot sampling. Furthermore, POCIS also improves the environmental quality standards based assessment of chemical status and on annual average concentrations compared to spot sampling. This study demonstrates the value and practicability of POCIS-based chemical monitoring for use in regulatory control networks.

To what extent can the biogeochemical cycling of mercury modulate the measurement of dissolved mercury in surface freshwaters by passive sampling?

Mercury (Hg) is a pollutant of global concern owing to its great toxicity even at very low concentrations. Its toxicity depends on its chemical forms evidencing the importance to study its speciation. Dissolved Hg (Hg(d)) and methylmercury (MeHg(d)) monitoring in surface freshwaters represents a great challenge because of their very low concentrations and substantial temporal variability at different timescales. The Hg(d) temporal variability depends on the environmental conditions such as the hydrology, water temperature, redox potential (Eh), and solar photo cycle. Passive samplers represent an alternative to improve the assessment of Hg(d) and MeHg(d) concentrations in surface freshwaters by integrating their temporal variability. An original sampling strategy was designed to assess the relevance of 3-mercaptopropyl DGT (Diffusive Gradient in Thin films) to integrate in situ the temporal variations of labile Hg (Hg(DGT)) and MeHg (MeHg(DGT)) concentrations. This strategy was implemented on two rivers to study the dynamics of Hg(d), Hg(DGT), MeHg(d) and MeHg(DGT) at diurnal and annual timescales. We evidenced that Hg(DGT) and MeHg(DGT) concentrations were generally consistent with discrete sampling measurements of Hg(d) and MeHg(d) in dynamic surface freshwaters. However, Hg(DGT) concentrations were overestimated (2-16 times higher) in case of low flow or low water depth, low suspended particulate matter (SPM) concentrations and elevated daily photoperiod. The most probable hypothesis is that such conditions promoted Hg0 production, and resulted in Hg0 uptake by DGT. Thus, attention should be paid when interpreting Hg(DGT) concentrations in surface freshwaters in environmental conditions that could promote Hg0 production.

The impact of dam flushing event on dissolved trace elements concentrations: Coupling integrative passive sampling and discrete monitoring.

Sediments accumulation in reservoirs induces water storage capacities reduction and flood risks increases rendering dam flushing or dredging events compulsory for security reasons. Short transient events like dam flushing monitoring is still a great challenge because the suspended sediments and contaminants concentrations increases could occur over only few hours/days and cover tens of kilometres. Since 1942, 21 dam flushing events have been performed on the Upper Rhône River (from Lake Geneva in Switzerland to Lyon in France) in order to evacuate accumulated sediments behind the Verbois dam (Switzerland). We designed an original sampling strategy to assess the 2016 dam flushing event consequences on the spatio-temporal dynamics of dissolved trace elements concentration and to reveal how passive sampling monitoring (Diffusive Gradient in Thin films, DGT) could improve this evaluation. Cr, Mn, Fe, Co, Ni, Cu, Zn, Cd, Pb, As and Hg dissolved concentrations were monitored by discrete and passive sampling at 3 stations over 160 km downstream the Verbois dam. Since dissolved Fe, Cu, Cd, Cr, Pb, Zn and Hg concentrations did not show great variations during the flushing event, the DGT efficiency was not fully demonstrated for these elements. In contrast, a sharp increase of Mn, Ni, Co and As dissolved concentrations (up to 22 times) was recorded, resulting mainly from a release from resuspended sediment. The dissolved As increase was mainly caused by reduced arsenic (AsIII) increase, even monitored 160 km downstream the Verbois dam. The DGT measurements were highly representative of trace elements concentrations and As speciation dynamics in comparison with discrete sampling. Although relatively high dispersion was highlighted for some elements DGT measurements during the flushing event, we showed that DGTs are robust and powerful time-integrative tools to monitor many trace elements more efficiently than discrete sampling during a short transient event on a large spatial scale.

Sampling of suspended particulate matter using particle traps in the Rhône River: Relevance and representativeness for the monitoring of contaminants.

Monitoring hydrophobic contaminants in surface freshwaters requires measuring contaminant concentrations in the particulate fraction (sediment or suspended particulate matter, SPM) of the water column. Particle traps (PTs) have been recently developed to sample SPM as cost-efficient, easy to operate and time-integrative tools. But the representativeness of SPM collected with PTs is not fully understood, notably in terms of grain size distribution and particulate organic carbon (POC) content, which could both skew particulate contaminant concentrations. The aim of this study was to evaluate the representativeness of SPM characteristics (i.e. grain size distribution and POC content) and associated contaminants (i.e. polychlorinated biphenyls, PCBs; mercury, Hg) in samples collected in a large river using PTs for differing hydrological conditions. Samples collected using PTs (n = 74) were compared with samples collected during the same time period by continuous flow centrifugation (CFC). The grain size distribution of PT samples shifted with increasing water discharge: the proportion of very fine silts (2-6 µm) decreased while that of coarse silts (27-74 µm) increased. Regardless of water discharge, POC contents were different likely due to integration by PT of high POC-content phytoplankton blooms or low POC-content flood events. Differences in PCBs and Hg concentrations were usually within the range of analytical uncertainties and could not be related to grain size or POC content shifts. Occasional Hg-enriched inputs may have led to higher Hg concentrations in a few PT samples (n = 4) which highlights the time-integrative capacity of the PTs. The differences of annual Hg and PCB fluxes calculated either from PT samples or CFC samples were generally below 20%. Despite some inherent limitations (e.g. grain size distribution bias), our findings suggest that PT sampling is a valuable technique to assess reliable spatial and temporal trends of particulate contaminants such as PCBs and Hg within a river monitoring network.

Metal measurement in aquatic environments by passive sampling methods: Lessons learning from an in situ intercomparison exercise.

Passive sampling devices (PS) are widely used for pollutant monitoring in water, but estimation of measurement uncertainties by PS has seldom been undertaken. The aim of this work was to identify key parameters governing PS measurements of metals and their dispersion. We report the results of an in situ intercomparison exercise on diffusive gradient in thin films (DGT) in surface waters. Interlaboratory uncertainties of time-weighted average (TWA) concentrations were satisfactory (from 28% to 112%) given the number of participating laboratories (10) and ultra-trace metal concentrations involved. Data dispersion of TWA concentrations was mainly explained by uncertainties generated during DGT handling and analytical procedure steps. We highlight that DGT handling is critical for metals such as Cd, Cr and Zn, implying that DGT assembly/dismantling should be performed in very clean conditions. Using a unique dataset, we demonstrated that DGT markedly lowered the LOQ in comparison to spot sampling and stressed the need for accurate data calculation.

Long-term survey of heavy-metal pollution, biofilm contamination and diatom community structure in the Riou Mort watershed, South-West France.

In a metal-polluted stream in the Riou Mort watershed in SW France, periphytic biofilm was analyzed for diatom cell densities and taxonomic composition, dry weight and metal bio-accumulation (cadmium and zinc). Periphytic diatom communities were affected by the metal but displayed induced tolerance, seen through structural impact (dominance of small, adnate species) as well as morphological abnormalities particularly in the genera Ulnaria and Fragilaria. Species assemblages were characterized by taxa known to occur in metal-polluted environments, and shifts in the community structure expressed seasonal patterns: high numbers of Eolimna minima, Nitzschia palea and Pinnularia parvulissima were recorded in Summer and Autumn, whereas the species Surirella brebissonii, Achnanthidium minutissimum, Navicula lanceolata and Surirella angusta were dominant in Winter and Spring. Commonly used indices such as the Shannon diversity index and Specific Pollution Sensitivity Index reflected the level of pollution and suggest seasonal periodicity, the lowest diversities being observed in Summer.