Regional groundwater model of north-east Belgium

doi:10.1016/j.jhydrol.2006.11.006

Journal of Hydrology

Volume 335, Issues 1–2, 8 March 2007, Pages 133-139

https://doi.org/10.1016/j.jhydrol.2006.11.006 Get rights and content

Summary

Since 1975, the possibility to dispose of high-level radioactive waste in the Boom Clay formation is investigated in Belgium at the test site in Mol. This research involves detailed studies of the hydrogeological system at various scales. The regional groundwater model is the largest-scale model, providing boundary conditions for smaller-scale flow and transport models and for long-term predictions (e.g. climate change). The major challenges for its newest update were as follows: (1) to better understand regional groundwater flow, (2) to better calibrate the regional hydraulic parameters of the Boom Clay and the underlying aquifers, (3) to apply new data from the site investigation, and (4) to simulate the transient behaviour of the overexploited aquifers underlying the Boom Clay.

The new regional groundwater model using the MODFLOW 2000 code includes an updated geometry, new estimates of parameter values and a new conceptualisation of the hydrogeological system. Two steady states are simulated: the hypothetical original state (prior to pumping) and the equilibrium state of the aquifers assuming continuation of the present day pumping. The partly observed period of the transition between these states is simulated using a specific transient model.

The information included in the current set of observations is not sufficient to estimate the hydraulic properties of the major regional aquitard (the Boom Clay) using inverse modelling. However, low value of its hydraulic conductivity, based on large number of measurements, appears to be compatible with the model. Future data acquisition campaigns should also focus on increasing the confidence in the regional validity (transferability) of the measured values of the hydraulic conductivity.

Introduction

In 1975, SCK·CEN (the Belgian Nuclear Research Centre) started investigations on the possibility to dispose of high-level radioactive waste in the Boom Clay layer in north-east Belgium. Since 1982, ONDRAF/NIRAS (the Belgian Agency for Radioactive Waste and Enriched Fissile Materials) is responsible for the long-term management of all radioactive waste in Belgium. In this framework, ONDRAF/NIRAS and SCK·CEN are currently evaluating the feasibility and safety of a possible deep repository for high-level and medium-level radioactive waste in the Boom Clay (ONDRAF/NIRAS, 2001). Favourable properties of the Boom Clay are its low hydraulic conductivity (10⁻⁷ m/d), its thickness (about 100 m at the Mol site) and lateral homogeneity, its high sorption capacity, favourable geochemical (reducing) conditions and its high plasticity, which strongly contributes to its self-sealing capacity.

The regional groundwater model (Fig. 1) simulates the water flow in the groundwater system (Fig. 2) consisting of the host formation and the main surrounding aquifers. It covers almost 7000 km² and includes the most important regional natural boundary conditions (rivers, regional divides and the sea) that influence the regional groundwater system. The main applications of the regional groundwater model are as follows: understanding the hydrogeology of the delineated research area, providing boundary conditions for the local flow and transport models (Meyus et al., 1998) that are used in the performance assessments (Marivoet et al., 2002), and simulating the hydrogeological system behaviour for changing boundary conditions, e.g. due to climatic changes.

This paper presents the most recent update of the regional groundwater model incorporating the results of the data acquisition campaign started in 1996 (ONDRAF/NIRAS, 2001).

Section snippets

Regional groundwater system

The hydrogeological system of north-east Belgium (Fig. 1) consists of tertiary sedimentary layers that deepen towards the north-east and outcrop in the south. The layers form a series of outcrops that represents the main recharge zones of the aquifers both above and below the Boom Clay (Fig. 2).

The major aquitard in the studied area is the approximately 100 m thick (at the Mol site) Boom Clay separating the Neogene and Quaternary group of aquifers (upper aquifer) from the Lower Rupelian aquifer.

Problem description

The regional groundwater model runs in present-day conditions until a steady state is reached. The steady-state solution represents the system response to a specific set of boundary conditions and sources and sinks. Hence, changes in the system and especially the long term changes can be evaluated by adapting the boundary conditions of the regional model to specific evolution scenarios.

The preceding versions of the regional model were not able to simulate the observed groundwater levels below

Regional hydrogeological model update

The updated regional hydrogeological model, designated NEB-2002, and released in 2002, brought several improvements into the concept and parameters of the regional model. The main features are discussed below.

The boundary conditions for the upper aquifer are the net infiltration, the rivers and zero flow boundary conditions assigned to the groundwater catchment boundaries. The main boundary condition for the Lower Rupelian and Lede-Brussel aquifers is the infiltration in the outcrop area. At

Transient groundwater flow model of the deep aquifers

In order to fit the calculated levels in the aquifers located below the Boom Clay to the observed values we built a specific transient model. This model includes only the confined parts of the deep aquifers (deep groundwater circulation domain – Fig. 5). The unconfined areas of the deep aquifers are represented by fixed head boundary conditions. These boundaries represent the interfaces between the two groundwater circulation domains correctly because the regional groundwater model gives

Conclusions

The goal of the regional groundwater modelling for the performance assessment of the deep geological disposal of high-level radioactive waste was to improve the understanding of the regional groundwater system including the host formation. Ultimately, the regional groundwater model is also used to simulate effects of changing boundary conditions on the groundwater system and to provide the boundary conditions for smaller-scale models used for flow and transport calculations.

The recent update of

Acknowledgements

This update of the regional hydrogeological model forms part of the programme on geological disposal of high-level and long-lived radioactive waste that is carried out by ONDRAF/NIRAS, the Belgian Agency for Radioactive Waste and Fissile Materials. The views expressed in this paper do not necessarily correspond to those of ONDRAF/NIRAS.

References (12)

M. Aertsens et al.
Spatial variability of transport parameters in the Boom Clay
Applied Clay Science
(2004)
Anderman, E.R., Hill, M.C., 2003. MODFLOW-2000, The US Geological Survey Modular Ground-Water Model – Three Additions...
Bernier, F., Bastiaens, W., 2004. Fracturation and self-healing processes in clays - the SELFRAC project. In: Davies,...
De Cannière, P., Put, M., Neerdael, B., 1995. Hydraulic characterisation of the boom clay formation from the HADES...
R. Demyttenaere
The post-paleozoic geological history of north-eastern Belgium
Mededeling van de Koninglijke Academie voor Wetenschappen, Letteren en Schone Kunsten van Belgie
(1989)
Marivoet, J., Sillen, X., Mallants, D., De Preter, P., 2002. Performance assessment of geological disposal of...

There are more references available in the full text version of this article.

Cited by (32)

Palaeohydrological reconstruction (1500-2000AD) of a drift sand landscape using pedogeomorphological and historical data (Campine area, NE Belgium)
2015, Catena
In order to build confidence in describing the hydrological cycle under future global change, hydrological models not only need to be calibrated and validated with recent data, but should also be tested against different boundary conditions from the past. This requires building a dataset of hydrological state variables against which the modelling output can be tested. We use pedogeomorphological and historical data to monitor hydraulic heads at specific time slices during the last 500 years for a sandy interfluve in the Nete catchment, NE Belgium. The nature of the archive is temporally and spatially highly discontinuous but allows to observe a systematic drop in phreatic groundwater levels during the second half of the 19th century. It is concluded that the approach yields promising results and that it might be of use elsewhere in drift sand landscapes across the European sand belt.
Regional groundwater flow in hard rocks
2015, Science of the Total Environment
Citation Excerpt :
With the rapid increase of computer power and the advent of Geographic Information System (GIS), the partial approaches to regional flow modeling, including the assessment of leakage structures and preferential flow paths, could be integrated into complex algorithms, being MODFLOW (McDonald and Harbaugh, 1988) and the associated graphic user interfaces (e.g., Visual MODFLOW; Waterloo Hydrogeological, 2001) the most widely used packages, because of their flexible modular structure, complete coverage of hydrogeological processes and public domain free availability. Among the studies employing MODFLOW in the analysis of regional groundwater flow, one can refer the ones by Dufresne and Drake (1999), Gedeon et al. (2007) and Scanlon et al. (2003). The data required to operate these models are immense, and for that reason their applicability may be hampered when these data are lacking or fragmented.
The territory of continental Portugal has a geologic history marked by the Hercynian orogeny, and to the north of this country the Hercynian large-scale tectonic structures are typically represented by long and deep NW–SE trending ductile shear zones and NNE–SSW trending fragile faults. These structures are elements of mineral and thermal water circuits that discharge as springs in more than one hundred locations. The purpose of this study is to investigate if these structures are also used by shallower non-mineral groundwater, integrated in a large-scale regional flow system. Using an original combination of water balance and recession flow models, it was possible to calculate catchment turnover times based solely on groundwater discharge rates and recession flow parameters. These times were then used to classify a group of 46 watersheds as closed or open basins, and among the later class to identify source and sink basins, based on innovative interpretations of relationships between turnover time and catchment area. By definition, source basins transfer groundwater to sink basins and altogether form a regional flow system. Using a Geographic Information System, it could be demonstrated the spatial association of open basins to the Hercynian ductile and fragile tectonic structures and hence to classify the basins as discharge cells of a regional flow system. Most of the studied watersheds are sub-basins of the Douro River basin, one of the largest regional catchments in the Iberian Peninsula, being located in its mouth area. Because the largest part of open basins is sink, which by definition tends to dominate in the mouth area of regional catchments, it is proposed as an extension of the studied area conceptual boundaries towards the Douro River basin headwaters, where the corresponding sources could be searched for.
Groundwater resources management through the applications of simulation modeling: A review
2014, Science of the Total Environment
The global population is increasing rapidly and expected to touch the 9.5 billion mark by 2050 from the current 7.2 billion. The management of the groundwater resources is a challenging task worldwide against the backdrop of the growing water demand for industrial, agricultural, and domestic uses and shrinking resources. Moreover, this task has been hampered significantly due to declining/rising groundwater levels and associated contamination. A broad range of solutions could be considered to address the aforementioned problems of groundwater management, but the effectiveness of all the solutions and their combinations cannot be verified with field experiments. Given their predictive capability, simulation models are often the only viable means of providing input to management decisions, as they can forecast the likely impacts of a particular water management strategy. This paper presents a comprehensive review on the simulation modeling applications for the management of groundwater resources. The past papers on the overview of groundwater simulation models, use of remote sensing and GIS in groundwater modeling, and application of simulation models in arid and semiarid regions are described in detail. Conclusions are drawn where gaps exist and more research needs to be focused.
Calibration of spatial aquitard distribution using hydraulic head changes and regularisation
2011, Journal of Hydrology
Irregularly expressed aquitards have a significant impact on hydraulics and redox conditions. The method expands the scope and applicability of characterising spatial structures close to well fields. The test site is an aquifer system with two aquifers divided by an aquitard (glacial till), located at a well field for drinking water abstraction. During operation, the wells are frequently switched and hydraulic head data are recorded at 10 wells in both aquifers. The data contain information about the impact of each abstraction well on each observation well. We develop an inverse modelling procedure to calibrate spatial aquifer parameters from this data. Instead of heads we calibrate to selected head differences to keep the model concise to short term fluctuations and to reduce data. The calibrated model parameters are storativity of the upper aquifer, hydraulic conductivity of both aquifers and leakage of the aquitard. The calibration is carried out spatially by pilot points with the software PEST using regularisation. Different cross validations show that the leakage can be calibrated in a physically meaningful way, but not the hydraulic conductivities and not the storativity. Geostatistic assumptions are not required. The calibrated spatial aquitard distribution coincides with bore profiles and explains anomalies of redox conditions.
Combined use of groundwater modeling and potential zone analysis for management of groundwater
2011, International Journal of Applied Earth Observation and Geoinformation
Citation Excerpt :
MODFLOW developed by USGS (McDonald and Harbaugh, 1988) is the widely used software in the community of groundwater modelers. Several case studies were carried out through MODFLOW to understand and predict the groundwater behavior (Palma and Bentel, 2007; Gedeon et al., 2006; Idrysy et al., 2006; Kumar, 2001). As per the literature review, several studies have been performed to find out the groundwater potential zones using different methods, but the real impact/influence of these potential zones on aquifer has not been examined with the help of available technique i.e. groundwater modeling.
A methodology for groundwater evaluation has been developed by the combined use of numerical model and spatial modeling using GIS. The developed methodology has been applied on the sub-basin of the Banganga River, India. Initially, the groundwater potential zones have been delineated by spatial modeling. Different thematic maps of the basin like geology, geomorphology, soil, drainage, slope factor and landuse/landcover have been used to identify the groundwater potential zones. Further, the groundwater flow model for the study area has been developed in the MODFLOW. The groundwater flow vector map has been developed and superimposed on the potential zone map to validate the results of spatial modeling. Finally, the different scenarios have been conceptualized by varying the discharge of the wells and purposing the location for new rainwater harvesting structures. Results reveal that increasing the discharge of the wells in the potential zones put less stress on the aquifer. The suggested locations of rainwater harvesting structures also help to reduce the overall decline of groundwater in the area. The hydrological and spatial modeling presented in this study is highly useful for the evaluation of groundwater resources and for deciding the location of rainwater harvesting structures in semi-arid regions.
Comparative reflections on the use of modelling tools in conflictive water management settings: The Mancha Occidental aquifer, Spain
2010, Environmental Modelling and Software
Citation Excerpt :
Resolution varies as a function of the available data. Model development follows a standard implementation methodology (Anderson and Woessner, 1992; Refsgaard et al., 2007) and is based on a finite-difference numerical code (Harbaugh and McDonald, 1996; Chiang and Kinzelbach, 2001; Chiang, 2005) whose reliability and robustness have been tested extensively (Prommer et al., 2002; Bradley, 2002; Lin and Medina, 2003; Martinez-Santos et al., 2004; Martinez-Cortina and Cruces, 2005; Scibek and Allen, 2006; Gedeon et al., 2007). Aquifer geometry is defined throughout by the interpretation of over 400 borehole logs.
Participatory methods provide an increasingly accepted path to integrated assessment. This paper reflects on the role of two participatory modelling initiatives implemented in a highly conflictive setting: the Mancha Occidental aquifer, Spain. The methodologies are described within the context of the case study, examining their potential relevance to integrated assessment from a conceptual standpoint. The strengths and weaknesses of each approach are analysed in absolute and relative terms, attending to the different stages of the modelling process. The focus then shifts to explore the implications of this work within the context of participatory integrated assessment and scenario analysis. This serves the purpose of establishing the reasons why the tools have been useful in the eyes of stakeholders, and how the case-specific findings of this project may be relevant to other settings.

View all citing articles on Scopus

View full text