First International Conference on Frontiers in Shallow Subsurface Technology

Nanoseismic monitoring techniques are able to detect extremely low-energy (ML > -4.0) signals generated by active subsurface instabilities within cavities, cliffs, landslides and sinkholes. The data is acquired by portable sparse seismic arrays, which are deployed, within minutes, in varying geometries as close as possible to a presumed zone of instability. Events detection is carried out by semi-automated pattern recognition-supported schemes4, which scan for broad-band energy spikes within continuous data records sampled at 200 to 500 Hz. The authenticity of source signals is verified either by true-scale simulation in the field -when possible- or through a multi-parameter validation process that uses a custom library of reference patterns. This comprehensive waveform characterization process includes full-spectral signal analysis, 3-D source location, waveform cross correlation and source magnitude evaluation. Study cases will be reviewed to show how failure generated within cavities5,6, cliffs7 and sinkholes1 can be located and characterized. Additional analyses include magnitude-time series, source time migration and event coincidence with external parameters as well as a custom calibration of source energy dissipation in unconsolidated material.

The properties of sand are important for its suitability for the use in construction works. For dredgers, this information needs to be available in advance of dredging activities. Shear wave velocities are linked to the strength of the sediment. In underwater environments, shear wave velocities can be extracted from seismic data using the dispersive properties of Scholte waves. We used forward modelling to investigate whether we can regard Scholte waves as their equivalent on land, Rayleigh waves. The difference between Rayleigh and Scholte waves is related to the relative water depth and the Poisson ratio distribution. For a halfspace of unconsolidated muddy sediments (high Poisson ratio) on deep water layer, the Scholte wave velocity is nearly 15% lower than the Rayleigh wave velocity. For sandstone (low Poisson ratio), there is no difference. When the water is shallow, the Scholte wave velocity deviates from Rayleigh wave velocity by less than 5%. Shear wave velocities were extracted from Scholte waves observed in data sets from the intertidal Wadden Sea (The Netherlands) and the river Danube near the village of Kulcs (Hungary). Based on the results from the forward modelling, we treated the Scholte waves as Rayleigh waves. The shear wave velocities in the top layer of the Wadden Sea sediments were approximately 150 m/s, which fall in the expected range for poorly consolidated material. The shear wave velocities of the Kulcs sediment were higher (400 to 600 m/s). The transition from clay to sand as apparent from a borehole log was successfully recovered in the subsurface shear wave velocity model.

During the last decades, the pressure for space in dense urbanized areas has led to an increasing use of the underground for a variety of functions such as infrastructure, utilities and buildings. In the last decade, an exponential growth in systems using the underground for thermal energy storage fueled by the need for renewable energy sources is added to this. These relative novel underground functions can compete with existing users of the underground. This includes both active (e.g. drinking water) and passive (groundwater dependent ecosystems) users of these so called ecosystems services. Dutch policy makers of various governmental levels (municipality, province, national) are currently exploring methods to deal with these conflicting interests. A few examples of issues to deal with are: 1) how do we deal with existing rights versus new claims for the underground? 2) how do we manage the exponential growth of subsurface functions of which the long-term environmental effects are unknown? KWR and University Utrecht have recently started a project funded by the Dutch Ministry of Housing, Spatial Planning and the Environment investigating these questions. In this project we will look at both the technical and legal aspects of subsurface planning and specifically focus on underground thermal energy storage in spatial planning. This extended abstract details the preliminary results of this project.

The climate change may pose threats to the way the Dutch live in deltas below sea level. Peat districts below sea level are at more risk, when peat gets oxidized to Green House Gas (GHG) emissions, followed by land subsidence due to low groundwater levels. However, new sustainable concepts can handle the climate change. Local materials like manure, sediment, water treatment sludge, debris from plants, and construction materials can reverse the soil subsidence. Inspiration comes from 500 BC, when farmers erected mounds for seasonal refuge. They mixed sods with manure and wooden debris to a fertile soil with high bearing capacity and resistance to failure. Today, the Cradle-to-Cradle (C2C) approach aims at a new life for waste materials in a transition of the rural farming to a sustainable one (the “Peat in balance” concept.

The infiltration rate in soils is important for determining the propagation of saturation, and hence for the stability of slopes subjected to overrunning water. The paper suggests a simple practical formula for interpretation of infiltration tests, validated by various laboratory and field test results.

There is an increasing demand for detailed information on the spatial distribution of geological units and their physical and chemical properties. However, existing studies on sediment properties are scarce. The project TOPINTEGRAAL aims to build a geological model of the upper 30 to 50 meters of the subsurface and to characterize the lithological, hydraulic, geochemical and geotechnical properties of the sediments. A drilling campaign was launched in 2006 in order to obtain additional soil samples for the analyses and measurement of physical and geochemical properties. So far 157 cored wells were drilled. The results of the analysis and measurements of the samples are interpreted and published per GeoTOP-main area. In 2009 a first interpretation of the grain size and geochemical data of main area “Noord-Nederland” in the northern part of the Netherlands has been performed. The data and their interpretations will become a unique national database which will help researchers and planners to obtain better and more accurate results.

A recent development in the Netherlands is the implementation of a Key Register Subsurface (“Basisregistratie Ondergrond”). Part of this Key Register Subsurface is a new generation of 3D subsurface models that provide estimates of stratigraphy, lithology (clay, sand, peat) and sand-grain size class data at voxel-resolutions of 100*100*0.5m (referred to as GeoTOP models). These regional-scale models can be used as a starting point for detailed investigations of risks and opportunities that may evolve from intensive and multiple uses of the subsurface. An example of this is the planning of a new subway in the city of Rotterdam.

The Geological Survey of the Netherlands aims at building a 3D geological property model of the upper 30 meters of the Dutch subsurface. The model schematises the subsurface in millions of grid cells each measuring 100 by 100 meters in the horizontal directions and 0.5 meters in the vertical direction. Each grid cell of the model includes estimates of stratigraphy, lithofacies and lithology (clay, sand, peat) and if applicable, sand-grain size class data. Stochastic interpolation techniques are used to compute probabilities for these parameters, providing a quantification of model uncertainty. The model provides a sound framework for subsurface related questions on, amongst others, groundwater management, land subsidence, natural resources and infrastructural issues.

The main objective of this research is to enhance our understanding of the relation between pore-scale adsorption and Darcy-scale parameters using a 3D pore-network model. This helps to scale up from a simplified but reasonable representation of microscopic physics to the scale of interest in practical applications. This upscaling will be carried out in two stages: i) from pore scale to the scale of a tube, which is the main element of a pore-network model, and ii) from tube scale to the scale of a core represented by the pore-network model. The first stage of this upscaling from pore to tube scale has been reported in an earlier manuscript. There we found relationships between micro-scale parameters (such as equilibrium adsorption coefficient, kd, and Peclet number, Pe) and tube-scale parameters (such as attachment coefficient, katt, and detachment coefficient, kdet). Here, we perform upscaling by means of a 3D multi-directional network model, which is composed of a large number of interconnected pores bodies and pore throats (represented by tubes). We use the expressions which were developed in our earlier work to estimate tube-scale adsorption parameters for each and every pore in the network. Transport parameters are then upscaled over the whole network by fitting average concentration breakthrough curves at the outlet to the solution of classical advection-dispersion equation with adsorption. This procedure has resulted in relationships for upscaled adsorption parameters in terms of micro-scale coefficients.

Geomechanical models are indispensable for reliable design of engineering structures and processes and hazard and risk evaluation. Model predictions are however far from perfect. Errors are introduced by fluctuations in the input or by poorly known parameters in the model. To overcome these problems an inverse modelling technique to incorporate measurements into the deterministic model to improve the model results can be implemented. This allows for observations of on-going processes to be used for enhancing the quality of subsequent model predictions. In geomechanics several examples of inverse modelling exist where the improved model of the system is obtained by minimizing the discrepancy between the observed values in the system and the modelled state of the system within a time interval. This requires the implementation of the adjoint model. Even with the use of the adjoint compilers that have become available recently, this is a tremendous programming effort for the existing geomechanical model system. The Ensemble Kalman filter has been implemented to overcome this problem. The Ensemble Kalman filter analyses the state of the subsurface each time data becomes available. The Random Finite Element Method is used to simulate the heterogeneity of the subsurface. Very promising results of a conceptual example, based on the construction of a road embankment on soft clay, are presented. The Ensemble Kalman filter is not only used for a straight forward identification of the elastic Young’s modulus E of the foundation below the embankment, but also incorporates the determination of several critical parameters of the inverse modelling process.

In the Port of Rotterdam the long-term presence of various industrial activities has resulted in soil and groundwater contamination. This contamination is substantial, complex and is usually not limited to one particular site but affects (ground)water systems at a regional scale. This being the case, site-specific approaches are neither effective nor cost efficient. In these cases it is better to develop an integral approach at megasite level, in which risk management scenarios are combined and measures are prioritized. The natural resilience of the soil and groundwater system will help significantly in reducing the risk of contaminated groundwater, if managed properly. In the EU WELCOME project an integrated management strategy (IMS) has been developed and later on successfully applied at the Port of Rotterdam. The implementation has revealed that a significant cost-reduction of up to 40% can be achieved in comparison to a site by site remediation approach, if measures are considered following a risk based approach on megasite scale. Active remediation efforts will still be conducted within the framework of the Integrated Management Strategy in order to manage the risk of still present contaminations. Conventional remediation techniques need energy and produce CO2 which has raised doubts about the overall sustainability of these techniques. The introduction of green remediation approaches can significantly boost the sustainability of Area Management of Contamination. For megasites like the Port of Rotterdam the key to this ambition lies in the combination of several goals for energy and water. For the Port of Rotterdam it is possible to effectively combine industrial water use and the industrial heat surplus with the remediation of contaminated groundwater. By using the groundwater for cooling purposes and storing (part of) the heat surplus underground (Aquifer Thermal Energy Storage, or ATES), biological remediation can be stimulated under certain conditions. This will lead to both an improvement of the groundwater quality and a reduction of energy use and CO2 emissions.

At present the search for energy other than coal, oil or gas is popular and the exploitation of geothermal energy provides an interesting alternative. For the design of exploitation sites, numerical models are in use, which should be validated by field test and analytical solutions. In this paper different physical processes encountered in heat transport in porous media and validation by analytical tools are explored.

We conducted a series of laboratory measurements of the electrical properties of samples of water-saturated, unconsolidated sand as a function of frequency, water saturation, and salinity. We employ the parallel-plate capacitor technique to measure the complex impedance for frequencies between 200 Hz and 3 MHz. We performed main drainage and secondary imbibition cycles for unconsolidated sand-saline water systems at atmospheric pressure and temperatures between 21°C and 22°C. We found hysteresis in the electric permittivity and resistivity, caused by the redistribution of the water and air phases. The hysteretic effect becomes more pronounced at higher concentrations of salt. Also we found that the sand grain size does not affect the permittivity of dry sand. For the saturated sand, the situation changes considerably and the coarser grain size leads to a larger polarization effect. An explanation of this phenomenon is that for the coarse-grain size sand, the capillary pressure is lower than for the fine-grain size sand, allowing the water molecules to respond more easily to the external electric field. The electric properties at all saturation levels were found to depend on both frequency and salinity of the pore fluid, implying that a description by an effectively homogeneous medium should incorporate these parameters.

Time-Domain Reflectometry (TDR) provides a robust and widely used method for the monitoring and characterisation of soils using user-friendly cable testers. However, the simplicity of common TDR practice, often utilizing only measurements of reflection distances for electromagnetic signals in probes, causes a number of difficulties when applied to fine-grained soils, largely because they can be electromagnetically dispersive: that is their dielectric 'constants' vary with frequency. This paper therefore considers two ways in which TDR data can be made to greater use in geotechnical disciplines: supplementing TDR with frequency-domain data, and addressing the issues of standardisation and data sharing.

This paper investigates the potential of using forsterite olivine as an alternative for carbonate-containing limes in agriculture. The use of carbonate containing limes contributes to the anthropogenic CO2 emissions. Additional to replacement of agricultural lime, it may also increase the pH of soil and therefore the flux of bicarbonate from soil to the sea. Soil has been incubated using various amounts of olivine flour or lime in a relative acidic soil during eight months. The results show that during this period the effect of olivine on the soil pH is very small compared to lime. Compared to lime it is necessary to use approximately 35 times more olivine to get the same pH increase. Assuming similar costs for olivine and lime it is uneconomical to use olivine to replace lime. Currently also the Dutch Fertilisers Act does not allow olivine as a fertilizer due to the high amount of nickel compared to the neutralizing value.

Measurements of spectral induced polarization (SIP) data were performed during a biostimulation experiment consisting of electron donor amendment. The implementation of an appropriate error model for the phase data within a complex resistivity inversion algorithm provided images free of artifacts, exhibiting phase anomalies well correlated with spatiotemporal changes in geochemical conditions resulting from stimulated subsurface microbial activity.

Recently several in situ techniques are being developed to change the mechanical properties of soils by stimulating biochemical processes. A procedure named Biogrout is developed to increase strength and stiffness of unconsolidated sands, based on microbial-induced carbonate precipitation. Recently, this development resulted in a field scale experiment in which 40m3 of sand was cemented within 12 days, reaching average unconfined compressive strengths of 2 MPa with limited reduction in permeability. Pilot projects are in preparation. In order to optimize this process, the use of waste streams as substrate for biocementation has been investigated.

Incipient instabilities, which are presently developing within the chalk of ancient (ca. 1100 years old) bell-shaped caverns at the Bet Guvrin National Park (Israel), have been the aim of multiple investigations for more than a decade. An innovative and independent validation approach is presented here, whereby numerical predictions of material instabilities are compared with the spatial distribution of extremely weak (ML > -4.0) failure signals detected by nanoseismic monitoring. Our data show that it is indeed possible to detect a series of extremely weak spiky broad-band signals. Their characterization and location indicate that they are unequivocally generated by incipient brittle failure within the chalk material of the bell-shaped caves, thus verifying numerical predictions. Furthermore, our investigations suggest that numerical modelling is best used to guide nanoseismic monitoring, which, in turn, can quantify the spatio-temporal distribution of actual incipient failure.

We apply seismic interferometry to shallow-subsurface shear-wave data recorded using active sources. We cross-correlate the active data to obtain virtual shot gathers and virtual common-offset sections and compare the results to the active data.

One of the most challenging issues in the experimental investigation of interfacial area and its properties for two-phase flow in a micro-model is the visualization method. A micro-model is a micro-structure that represents a porous medium. The choice of the appropriate method depends on the dimensions of a micro-model and whether the experiments concern steady-state or dynamic effects. In the case of relatively small micro-models for steady-state or dynamic experiments, the use of a microscope in combination with a camera is the most effective method. When it comes to experiments where the micro-model is bigger than the optical frame of a microscope, the direct use of cameras is preferable. In this work, we present the visualization method for an elongated micro-model, with dimensions of the flow network of 1x10 mm2. The purpose of the experiment is the investigation of the role of fluid-fluid interfacial area in two-phase flow under dynamic conditions. Because of the dimensions of the micro-model, the use of a microscope is impossible without moving the micro-model. In our setup, four high resolution cameras with frame rate of 15 fps will be used. Each camera will cover an area of 1x1.7 mm2, thus a total area of 1x6.8 mm2 will be observed at any time. These cameras will record the distribution of phases. Then, with the aid of a computer and proper software, specific interfacial area and phase saturation will be determined as a function of time.

We use a 2D Finite-Difference-Time-Domain electromagnetic (EM) wave simulator to model a borehole radar system located in a shallow oilfield production environment. Main aim of this study is to investigate if radar technology is suitable for detecting and monitoring fluid saturation changes in the proximity of production wells. An analysis of the main constraints for an implementation of a radar system as permanent down-hole sensor has been performed. The main constraints are given by the conductivity of the formation and by the effect of the borehole casing on the emitted EM signal. An antenna shield is necessary to reduce the destructive interference between the emitted signal and the reflections by the metal parts of the wellbore. Moreover, different reservoir scenarios are considered. The heterogeneity of the background formation strongly affects the retrieval of the target reflections and gradual fluid saturation changes reduce the amount of the reflections.

This study investigates whether the low efficiency of electricity generation from low quality geothermal energy sources can be improved by using night sky radiation. Useful energy or work can be generated from a heat source when there is a temperature difference ΔT, e.g., between the heat source (T > T0) and a heat sink T0. It is also possible to have a low temperature heat "source" (T < T0) as long as the Carnot efficiency factor (T0-T)/T0 is sufficiently large. By way of example we use a geothermal energy plant from which a maximum of useful energy (electricity) is to be extracted. We only give a schematic exergy analysis of geothermal energy extraction, which includes the exergy necessary for circulating water and material costs for a 20-year project. The useful energy produced is calculated using a device that contains series of Peltier elements that are exposed on one side to the produced hot water and on the other side to a vessel loosing heat to the night sky. The result is compared to a similar device where the cold side assumes an ambient temperature of 10o C. As expected, the process using night cooling is the more efficient process, but the present efficiency of Peltier elements is insufficient for an exergetically viable electricity production. It turns out that the use of night sky radiation leads to about the same efficiency as groundwater cooling.

The beach is no longer the frontier of human activity. The shallow subsurface in the North Sea, for example, is used by numerous industries (Figure 1). The North Sea seabed is the site of communication and electricity transfer cables; pipelines that transport hydrocarbons, water and waste; military exercises and munitions dumps; sand extraction (currently over 25 million m3/year); and platforms, windmills and other structures for extracting and harnessing energy sources. Some areas have been set aside for ecological and historical resource preservation, but most of the sea will become ‘busier’ and more crowded as existing offshore industrial activities expand and new uses are introduced. New uses of the North Sea may be geothermal energy production, artificial islands for airports and energy, a network of electricity interconnectors, CO2 storage, and harnessing tides to produce electricity. All these activities, even those associated with ships and floating structures, require structures that are based on and interact with the shallow subsurface: they all require ‘foundations.’