The innovative MIP-IN device combines (1) detection of pollutants by for instance a membrane interface probe (MIP) and (2) a simultaneous correlated injection (IN) during direct push of the device using a Geoprobe®. The MIP-IN device is the basis for a new detection-injection technology with the main advantage of the nearly simultaneous coupling of detection of pollutants at a certain depth and injection of a suitable amount of reactive agent at that precise spot. In this way, the injected reagent is more targeted towards the real location of the pollution with reduced remediation time and cost.
The MIP-IN concept was developed within the FP7 UPSOIL project (EU GA 226956). A first prototype version of the device was used for injection of guar gum stabilized zerovalent iron in a contaminated subsurface as part of the FP7 AQUAREHAB project (EU GA 226565). MIP-logs showed that chlorinated pollutants were present in the subsurface at distinct depths between 2 and 12 meter below ground surface, but the exact depths altered from spot to spot. The MIP-IN injection approach was used to inject guar gum stabilised micro-scale zerovalent iron slurry in a challenging sandy subsurface using high injection pressures, high flow and relatively low injection volumes. After the injection, undisturbed soil samples were taken to verify the distribution of the injected material. It was shown that the MIP-IN device was able to inject guar gum stabililized zerovalent iron at depths where CAHs were detected, and that the MIP-IN approach has potential. To deliver the reagent at the injection depth, the targeted radius of influence of 0.5 m was found suitable. Automatic logging of injection parameters (volumes, times, depths, ..), not yet available for the used prototype MIP-IN version, was identified as a crucial aspect for further improvement of the device and the injection.
Since end 2013, the MIP-IN device is being further developed and tested within the MIP-IN EUROSTAR-project (E!8246) where VITO (Belgium), Ejlskov (Denmark), Ecorem/ABO (Belgium) and Dekonta (Czech Republic) joined forces. The main goals of the MIP-IN EUROSTARS project are to (1) improve the MIP-IN device, (2) validate the MIP-IN device in relevant environments and define boundary conditions, and (3) develop an innovative MIP-IN based remediation strategy closely linked with site investigation, based on 3-D modelling of MIP data.
A set of different MIP-IN-probes has been developed and are being tested in the field in different geologies and for different reagent types. At a Danish site (clay) BOS200 has been injected at an oil contaminated site, while nanoscale ZVI was injected at a Czech site where chlorinated compounds were present. Early 2015, a field test in Belgium is scheduled, where EHC-L and guar gum stabilized micro-scale iron will be injected in a sandy soil using the MIP-IN device. An overview of the results of these field tests will be presented, with focus on the functioning of the MIP-IN device and its added value. Further, the potential of the MIP-IN based remediation strategy will be explained and illustrated with an example.
Since the 1980s, several subsurface investigations conducted to evaluate impacts at an industrial facility in California have identified volatile organic compounds (VOCs) in shallow soil and soil gas underlying the site. VOCs have also been identified in groundwater encountered at approximately 180 feet (ft) below ground surface (bgs). However, the assemblage of detected VOCs in the shallow horizons (soil/soil gas) is significantly different than that observed in the deeper groundwater. Shallow VOC impacts (i.e., <~55 ft) are predominantly tetrachloroethene (PCE) with minor amounts of benzene, toluene, ethylbenzene, and xylenes (BTEX), carbon tetrachloride, other substituted benzenes, and a notable near absence of trichloroethene (TCE). The deeper groundwater is characterized by a predominantly TCE signature, with very low PCE concentrations. Collectively, the site data suggest that PCE impacts from historic site uses were limited to shallow depths, and the observed TCE in groundwater beneath the site had likely migrated from an upgradient location off-site. However, additional lines of evidence were needed to confirm this conceptual model, support traditional site assessment data, and to direct the path of future investigation and/or remediation efforts.
ENVIRON conducted an investigation to evaluate the vertical distribution of VOCs in soil gas beneath areas of the site where elevated PCE concentrations had been identified in previous testing. Borings were advanced to approximately 100 ft bgs, and nested soil vapor probes were installed. Soil vapor samples were collected and analyzed for VOCs including PCE, TCE, cis-1,2-dichloroethene, trans-1,2-dichloroethene, and vinyl chloride. Based on the bulk compositional analyses, select soil vapor samples were collected for compound specific isotope analysis (CSIA) of carbon and chlorine. In addition, groundwater samples from monitoring wells at the site were submitted for CSIA (carbon, chlorine).
ENVIRON used the CSIA and bulk composition results in soil gas and groundwater, coupled with fate and transport modeling for the site to evaluate and ultimately confirm the original conceptual model – that groundwater impacts beneath the site originates from an upgradient source. Moreover, the data suggest that the shallow VOCs (predominantly PCE) in soil gas have not undergone reductive dechlorination during transport, further evidence that a putative source for the TCE in groundwater does not exist on-site. The isotopic data from soil vapor and groundwater, combined with traditional site characterization data helped to validate and refine the conceptual site model. New CSIA results from additional vapor probes installed in early 2015 will be included that provide further resolution of the isotopic conditions at the site.
Groundwater discharge can be an important contaminant source to surface water bodies and must be addressed when responding to legislation like the EU Water Framework Directive, which aims to protect and restore surface water bodies. Since contaminated sites are a major source of contaminants in groundwater resources it is important to evaluate the risks posed by them to streams. Such risk assessments are the basis the selection of appropriate and cost-effective remediation actions. This is a challenge because little is known about how contaminant discharge to stream varies because of stream meandering and changes in the water levels in streams and in aquifers.
This study aimed to develop a model of groundwater discharge to streams that incorporates the stream morphology and the time varying water levels in streams and in aquifers. The model was applied in order to determine the likely location of groundwater discharge in streams, and determine the origin of that groundwater. The models also showed that time-varying stream water level and groundwater head affect the discharge. The model was developed for a field site at Grindsted stream, a study site located in Denmark. The study aimed to use the model to analyze groundwater discharge measurements obtained at the field site. The work provided new insight on groundwater/surface water interaction and the interpretation of field data.
The project successfully developed a three-dimensional COMSOL Multiphysics model for groundwater discharging into Grindsted stream. The model accounts for the geometry of the stream and the geological heterogeneity of the aquifer. In addition, it includes the time variability in stream and aquifer water levels.
The study site was characterized by an extensive field campaign. The model was used to design the field monitoring and then, once data was collected, was compared with Point Velocity Probe and stream temperature measurements which provided data on the location, magnitude and direction of groundwater discharge to the stream. The results were also compared with time series of head data at monitoring wells located next to the stream. The model was shown to reproduce field data very well, leading to confidence in results.
It was observed that the discharge into the stream is highly dependent on the gradient between the stream and the aquifer. Thus, temporal variability in the discharge is correlated to changes over time of the gradient. In addition, the geological heterogeneity of the aquifer underneath the stream was shown to affect the groundwater flow, the discharge into the stream and the provenience of the discharged water.
This study showed that the presence of meanders had high impacts on the discharge and on the groundwater flow in proximity of the stream. The results indicated that the upper part of the aquifer mostly discharges in the outward pointing meanders, while groundwater from the lower part enters the stream from the opposite side. Furthermore, the groundwater discharge velocity is higher in the outer part of the meander bends. The observation was supported by contaminant concentration measurements in the groundwater collected nearby the stream. This suggested that groundwater discharge through a stream bank does not necessarily originate from the same side of the stream. Stream monitoring data should therefore be carefully interpreted in order to avoid misunderstandings based on the assumption that groundwater originates from an area on the same side as a given stream bank.
The results of this study showed that groundwater discharge to streams varied greatly with location, depth and time. Furthermore, the stream morphology and the geological heterogeneity has a great impact on stream-aquifer interaction. This study also indicated that mathematical models are very useful for interpreting field data and for designing monitoring campaigns. The joint application of field investigations and modelling tools can be used for better understanding the contaminant mass discharge into streams. This may improve the risk assessment and reduce the costs for remediation and monitoring, in order to fulfill the requirements in the EU Water Framework Directive.
Introduction: Water quality in rivers typically depends on the degree of urbanization or the population density in a catchment. Transport of many pollutants in rivers is coupled to transport of suspended particles, potentially dominated by storm water overflows and mobilization of legacy contamination of sediments. Concentration of these pollutants strongly sorbed to suspended particles cannot be diluted by water directly, but depends on the mixture of “polluted” urban and “clean” background particles (Schwientek et al. 2013). In the current study, the total concentration of polycyclic aromatic hydrocarbons (PAHs), the amount of total suspended solids (TSS) and turbidity were measured on a monthly basis in water samples from 5 neighbouring catchments with contrasting land use in Southwest Germany and in 3 sub-catchments of the Bode River in Eastern Germany over up to 1.5 years. In addition, single flood events with large changes in turbidity were sampled at high temporal resolution. Suspended particles representing different time periods of pronounced events were also characterized towards type and geochemistry (total organic carbon and carbonate content, grain size distributions). Using on-line monitoring of turbidity (by optical backscattering sensors) mass flow rates of PAH over time were calculated.
Results and discussion: Linear correlations of turbidity and TSS where obtained over all catchments investigated and over an extended turbidity range (up to 2000 NTU for the flood samples). Linear correlations were also obtained for the total amount of PAH and suspended sediment concentrations even for very high turbidity or TSS values (> 2000 NTU or mg l-1, respectively). From the linear regressions concentrations of PAHs on suspended particles were obtained – which varied by catchment. The values comprise a robust measure of the average sediment quality in a river network or catchment and may be correlated to the degree of urbanization represented by the number of inhabitants per total flux of suspended particles. PAH concentrations on suspended particles were stable over a large turbidity range (up to 2000 NTU) confirmed by samples taken during flood events. No pronounced effects due to changing particle size or origin have been observed for the catchments investigated (< 150 squared km; Rügner et al., 2014). Results of on-line turbidity monitoring showed that high turbidity/discharge events account for major proportion of pollutant fluxes (this study: 90% PAH flux > 90 NTU; events representing only 2.5 % of the observed time period).
Conclusions: Turbidity may be used as proxy for the total concentration of suspended solids (TSS) and particle-bound pollutants in river water. From regressions of total PAH vs. TSS (or turbidity) concentrations on suspended sediments (Csus) may be calculated. For calculation in principle a single event is sufficient. Contamination of suspended particles depends on urban pressure per total suspended particle flux. The findings are promising for other particle-bound contaminant fluxes (PCBs, phosphorus, and several heavy metals, etc.).
References:
Schwientek M, Rügner H, Beckingham B, Kuch B, Grathwohl P. 2013. Integrated monitoring of transport of persistent organic pollutants in contrasting catchments. Environ Poll 172:155-162.
Rügner H, Schwientek M, Egner M, Grathwohl P. 2014. Monitoring of event-based mobilization of hydrophobic pollutants in rivers: Calibration of turbidity as a proxy for particle facilitated transport in field and laboratory. Sci Tot Environ 490: 191-198.
Acknowledgement:
The authors thank the Ministry of Science, Research and Arts of Baden-Württemberg (AZ Zu 33-721.3-2) and the Helmholtz Centre for Environmental Research - UFZ, Leipzig. The study was also supported by the European Communities 7th Framework Programme under Grant Agreement no 603629-ENV-2013-6.2.1-GLOBAQUA.
Plumes of dissolved contaminants may be widely distributed in the saturated zone, i.e. to great depths and over large areas. Also the concentration levels of contaminants in the plume generally are much lower than the concentrations found in the hotspot area. Characterization and delineation of contaminant plumes therefore typically require many investigation points at great depths. Often the delineation of a contaminant plume is conducted using traditional drilling techniques and installation of screened wells. However, this method is both time-consuming and resource-intensive and only a limited number of discrete depths can be screened in each well. Furhermore, screen depths are often chosen based on expected flow and contaminant distribution patterns in the saturated zone, rather than on detailed hydrogeological data and vertical contaminant distribution.
On behalf of The Capital Region of Denmark, Department of Regional Development, and in cooperation with experts from Geoprobe Systems (US), NIRAS A/S (DK) has tested a novel tool, Low-Level MIHPT (LL-MIHPT), for delineation and characterization of contaminant plumes in groundwater. LL-MIHPT is based on the existing high resolution direct push investigation tools MIP and HPT developed by Geoprobe Systems. The HPT probe (Hydraulic Profiling Tool) is used to continuously map the hydrogeological conditions (permeability), while the MIP probe is used to continuously map VOC contamination in soil and groundwater. The MIHPT system is a combination of the MIP and HPT systems and has proved to be very efficient for field investigations in hotspots and areas with high contaminant levels. However, the detection limits of the standard MIHPT system are too high for delineation of contaminant plumes where the concentration levels are significantly lower than in the source zones. The Low-Level MIHPT system is developed with the objective to detect contamination at low concentrations and thus provides a means for conducting more time efficient and cost-effective delineation of contaminant plumes in unconsolidated saturated formations.
The LL-MIHPT systems has been tested at two sites located in the towns of Farum and Slangerup in Denmark as part of ongoing field investigations at the two sites.
The purpose of this project was to test the LL-MIHPT technique for delineation of contaminant plumes in groundwater at two sites with different geological formations; sandy and clayey, respectively. The objectives have thus been to determine at which concentration level the LL-MIHPT system could detect the site specific contaminants and to investigate the correlation between observed LL-MIHPT responses and results from analysed water samples from targeted depths.
9 LL-MIHPT logs to 20-25 meters below surface have been carried out. At each log water samples were collected at specific depths with the GeoProbe for verification of the observed responses from the LL-MIHPT and for correlation of contamination levels. For further correlation of the LL-MIHPT data core samples were collected at three locations.
The results from the field tests show that it is possible with the LL-MIHPT to track relatively low concentrations of chlorinated solvents and BTEX’s in the saturated zone. Hence, for chlorinated solvents a detection limit in the order of 10 ug/L can be expected. For comparison the detection limit for chlorinated solvents with the standard MIP system is in the order of 1-10 mg/L.
Based on the results and experiences obtained from the field tests the new LL-MIHPT system shows good promise for delineation of contaminant plumes in the saturated zone with simultaneous retrieval of hydrostratigraphic data from the saturated zone. Thus, LL-MIHPT logs followed by depth specific groundwater sampling with the GeoProbe system is considered to be an optimal set-up for delineation and characterization of contaminant plumes in saturated zones in unconsolidated geological formations.
The field tests were conducted and evaluated in the fall of 2013 and spring of 2014. Since then NIRAS A/S has used the LL-MIHPT system at field investigations at several other sites. Thus, the presentation will include results from the field tests in Farum and Slangerup complemented with the most recent data.