Helen French

Senior Research Scientist

(+47) 991 69 947
helen.french@nibio.no

Place
Ås O43

Visiting address
Oluf Thesens vei 43, 1433 Ås

To document

Abstract

Environmental assessments are required prior to remediation and redevelopment of contaminated sites. To date, regulatory guidelines are commonly based on total concentrations. Occasionally, simple leaching procedures are included in environmental assessment. Despite being essential for quantification of contaminant transport, analysis of hydraulic conductivity is rarely considered. Cost-effective methods that reflect both contaminant leaching and hydrogeological properties of contaminated soils are needed to ensure proper soil management. The aim of this study was to simultaneously evaluate contaminant leaching and hydraulic conductivity in soil using a combined column test (CCT) and compare this to the leaching results from batch tests (BT) and transport estimates derived from the empirical Hazen equation. Two soils of different origin were characterized using the CCT. By including physical and chemical factors affecting the release and retention of contaminants, the CCT provides an integrated assessment of leaching and transport of trace elements from soils. Additionally, the effect of soil compaction was investigated as a physical treatment to reduce leaching and transport in contaminated soils. Soil compaction did not demonstrate reduced leaching, but a less extensive contaminant transport was observed due to reduced hydraulic conductivity in the soil.

To document

Abstract

In cold climates, the use of de-icing chemicals in the winter can lead to groundwater contamination, especially when used in large quantities, such as at airports. Oslo Airport, Gardermoen, is situated on Norway’s largest rain-fed aquifer. Potassium formate is used to remove ice from runways and propylene glycol from airplanes; the organic parts are degradable. Most of the wells to monitor the spread of de-icing chemicals in the underlying aquifer have well screens near the groundwater level, while the runways and the source of de-icing chemicals are near the groundwater divides, where vertical flow is expected. The objective of this study is to demonstrate the importance of layers and time-varying recharge on the spreading of contaminant plumes in an aquifer near a groundwater divide. This is done with numerical modelling. The model results show increased vertical transport of the added tracer in the presence of horizontal layers, both continuous and discontinuous, in the aquifer. With certain distributions of hydraulic conductivity, Ks, we demonstrate that deeper monitoring wells are required. With the scenarios modelled here, time-varying recharge has a weaker effect on plume distribution. Measured concentrations of potassium and total organic carbon show the cyclic effect of seasonally varying recharge of contaminants, and an asymptotic accumulation of concentration over time, that is consistent with the model runs. In conclusion, groundwater monitoring systems near a groundwater divide should include multi-level samplers to ensure control of the vertical plume movement.

To document

Abstract

To improve risk assessment, control and treatment strategies of contaminated sites, we require accurate methods for monitoring solute transport and infiltration in the unsaturated zone. Highly spatio‐temporal heterogeneous infiltration during snowmelt increases the risk of contaminating the groundwater in areas where de‐icing chemicals are required for winter maintenance of roads and runways. The objective of this study is to quantify how the different processes occurring during snowmelt infiltration of contaminated meltwater affect bulk electrical resistivity. Field experiments conducted at Moreppen experimental lysimeter trench are combined with heterogeneous unsaturated soil modelling. The experimental site is located next to Oslo airport, Gardermoen, Norway, where large amounts of de‐icing chemicals are used to remove snow and ice every winter. Bromide, an inactive tracer, and the de‐icing chemical propylene glycol were applied to the snow cover prior to the onset of snowmelt, and their percolation through the unsaturated zone was monitored with water sampling from 37 suction cups. At the same time, cross‐borehole time‐lapse electrical resistivity measurements were recorded along with measurements of soil water tension and temperature. Images of two‐dimensional (2D) bulk resistivity profiles were determined and were temperature corrected, to compensate for the change in soil temperature throughout the melting period. By using fitted parameters of petrophysical relations for the Moreppen soil, the tensiometer data gave insight into the contribution of water saturation on the changes in bulk resistivity, while water samples provided the contribution to the bulk resistivity from salt concentrations. The experimental data were compared with numerical simulation of the same experimental conditions in a heterogeneous unsaturated soil and used to quantify the uncertainty caused by the non‐consistent resolutions of the different methods, and to increase our understanding of the resistivity signal measured with time‐lapse electrical resistivity tomography. The work clearly illustrates the importance of ground truthing in multiple locations to obtain an accurate description of the contaminant transport.

To document

Abstract

Assessing redox conditions in soil and groundwater is challenging because redox reactions are oxygen sensitive, hence, destructive sampling methods may provide contact with air and influence the redox state. Furthermore, commonly used redox potential sensors provide only point measurements and are prone to error. This paper assesses whether combining electrical resistivity (ER) and self-potential (SP) measurements can allow the mapping of zones affected by anaerobic degradation. We use ER imaging because anaerobic degradation can release iron and manganese ions, which decreases pore water resistivity, and produces gas, which increases resistivity. Also, electrochemical differences between anaerobic and aerobic zones may create an electron flow, forming a self-potential anomaly. In this laboratory study, with four sand tanks with constant water table heights, time-lapse ER and SP mapped changes in electrical/electron flow properties due to organic contaminant (propylene glycol) degradation. Sampled pore water mapped degradation and water chemistry. When iron and manganese oxides were available, degradation reduced resistivity, because of cation release in pore water. When iron and manganese oxides were unavailable, resistivity increased, plausibly from methane production, which reduced water saturation. To bypass the reactions producing methane and release of metallic cations, a metal pipe was installed in the sand tanks between anaerobic and aerobic zones. The degradation creates an electron surplus at the anaerobic degradation site. The metal pipe allowed electron flow from the anaerobic degradation site to the oxygen-rich near surface. The electrical current sent through the metal pipe formed an SP anomaly observable on the surface of the sand tank. Time-lapse ER demonstrates potential for mapping degradation zones under anaerobic conditions. When an electrical conductor bridges the anaerobic zone with the near surface, the electron flow causes an SP anomaly on the surface. However, electrochemical differences between anaerobic and aerobic zones alone produced no SP signal. Despite their limitations, ER and SP are promising tools for monitoring redox sensitive conditions in unsaturated sandy soils but should not be used in isolation.

To document

Abstract

Degradation of organic chemicals in natural soils depends on oxidation-reduction conditions. To protect our groundwater resources we need to understand the degradation processes under anaerobic conditions. Available iron and manganese oxides are used as electron acceptors for anaerobic degradation and are reduced to the dissolved form of metallic cations in pore water. To monitor this process is a challenge, because anaerobic conditions are difficult to sample directly without introducing oxygen. A few studies have shown an impact of iron reduction on spectral induced polarisation (SIP) signature, often associated with bacterial growth. Our objective is to study the impact of iron and manganese oxide dissolution, caused by degradation of an organic compound, with spectral induced polarisation signatures. Twenty-six vertical columns (30 cm high, inner diameter 4.6 cm) were filled with a sand rich in oxides (manganese and iron) with a static water table in the middle. In half of the columns, a 2 cm high contaminated layer was installed just above the water table. As the contaminant degrades, the initial oxygen is consumed and anaerobic conditions form Every three days over a period of one month, spectral induced polarisation (twenty frequencies between 5mHz and 10 kHz) data were collected on six columns: three contaminated replicates and three control replicates. Chemical analysis was done on twenty columns assigned for destructive water sampling, ten contaminated columns and ten control. The results show an increase of the real conductivity associated with the degradation processes, independent of frequency. Compared with the pore water electrical conductivity in the saturated zone, the real conductivity measurement revealed the formation of surface conductivity before iron was released in the pore water. In parallel, we also observed an evolution of the imaginary conductivity in both saturated and unsaturated zones at frequencies below 1 Hz. Overall, the anaerobic reduction of iron and manganese oxide during the organic degradation increased both the conductive and polarisation component of the complex conductivity.

To document

Abstract

A physically-based, distributed hydrological model (MIKE SHE) was used to quantify overland runoff in response to four extreme rain events and four types of simulated land use measure in a catchment in Norway. The current land use in the catchment comprises arable lands, forest, urban areas and a stream that passes under a motorway at the catchment outlet. This model simulation study demonstrates how the composition and configuration of land use measures affect discharge at the catchment outlet differently in response to storms of different sizes. For example, clear-cutting on 30% of the catchment area produced a 60% increase in peak discharge and a 10% increase in total runoff resulting from a 50-year storm event in summer, but the effects on peak discharge were less pronounced during smaller storms. Reforestation of 60% of the catchment area was the most effective measure in reducing peak flows for smaller (2-, 5- and 10-year) storms. Introducing grassed waterways reduced water velocity in the stream and resulted in a 28% reduction in peak flow at the catchment outlet for the 50-year storm event. Overall, the results indicate that the specific effect of land use measures on catchment discharge depends on their spatial distribution and on the size and timing of storm events.

To document

Abstract

The recent flooding episode in Norway from May this year shows the necessity of understanding the processes of water discharge from small tributaries feeding the larger river systems. The major objective of the recently started ExFlood project is to define and analyze measures to combat negative impact of extreme weather events on infrastructure in small watershed areas in Norway and to incorporate this in a land use planning tool. Urban, agriculture, nature, and forest areas and infrastructure elements demands different approaches concerning impacts of and opportunities for extreme weather events. The approach of the ExFlood project is to reduce the peak flow and delay the peak time to avoid damages on infrastructure. Three municipalities from different climate regions in Norway contribute to the project where the planning tool will be tested, and an experimental catchment site is selected to conduct in depth process studies.