Development of in-aquifer heat testing for high resolution subsurface thermal-storage capability characterisation

Highlights

• Distributed-temperature-sensing.

• Shallow subsurface thermal-storage.

• Shallow geothermal exploration.

• Direct-push in situ investigations.

Summary

The ongoing transition from fossil fuels to alternative energy source provision has resulted in increased geothermal uses as well as storage of the shallow subsurface. Existing approaches for exploration of shallow subsurface geothermal energy storage often lack the ability to provide information concerning the spatial variability of thermal storage parameters. However, parameter distributions have to be known to ensure that sustainable geothermal use of the shallow subsurface can take place – especially when it is subject to intensive usage. In this paper, we test a temperature decay time approach to obtain in situ, direct, qualitative, spatial high-resolution information about the distribution of thermal storage capabilities of the shallow subsurface. To achieve this, temperature data from a high-resolution Fibre-Optic-Distributed-Temperature-Sensing device, as well as data from conventional Pt₁₀₀-temperature-sensors were collected during a heat injection test. The latter test was used to measure the decay time of temperature signal dissipation of the subsurface. Signal generation was provided by in-aquifer heating with a temperature self-regulating electric heating cable. Heating was carried out for 4.5 days. After this, a cooling period of 1.5 weeks was observed. Temperature dissipation data was also compared to Direct-Push-derived high-resolution (hydro-)geological data. The results show that besides hydraulic properties also the bedding and compaction state of the sediment have an impact on the thermal storage capability of the saturated subsurface. The temperature decay time approach is therefore a reliable method for obtaining information regarding the qualitative heat storage capability of heterogeneous aquifers for the use with closed loop system geothermal storage systems. Furthermore, this approach is advantageous over other commonly used methods, e.g. soil-sampling and laboratory analysis, as even small changes in (hydro-)geological properties lead to strong variances in observed heat-storage capabilities at the investigated case study site. By using fibre-optic-thermometers, nearly every requested spatial resolution can be achieved and easily be adjusted to the needs of actual test sites for shallow geothermal storage exploration.

Introduction

The ongoing transition from fossil fuels to alternative energy source provision has resulted in an increased geothermal use of the shallow subsurface for functions such as heat generation, cooling, and energy storage. This technique is widely accepted as being capable of providing thermal energy with decreased CO₂ emissions, compared to fossil fuels and other conventional techniques (e.g. Lund and Boyd, 2015, Morrone et al., 2014, Blum et al., 2010, Lund, 2010). However, to ensure economically and environmentally sustainable usage, information about heat transport parameters is crucial for forecasting the thermal transport and storage behaviour of the subsurface (e.g. Vienken et al., 2015, Banks, 2014, Milenić et al., 2010, Omer, 2008, Ferguson, 2007).

A way to increase the lifetime and efficiency of shallow geothermal systems is to switch from unidirectional systems, for heating and cooling only, over to bidirectional systems allowing for regeneration and enhancement of the geothermal potential of the subsurface (e.g. Dickinson et al., 2009). Within bidirectional systems heat or cold is stored within the subsurface for later retrieval, so called underground thermal energy storage (e.g. Lee, 2010, Novo et al., 2010). This can be done by injecting tempered water into aquifers called aquifer thermal energy storage (ATES), commonly done by using open loop systems within aquifers (e.g. Sommer et al., 2014, Bloemendal et al., 2014) or using closed loop systems like borehole heat exchangers within aquifers as well as within the unsaturated subsurface (e.g. Lee, 2010).

To analyse the thermal storage potential of a given subsurface setting numeric modelling is often used (e.g. Bridger and Allen, 2014, Lee, 2010). Different input data has to be provided for modelling, e.g. thermal and hydraulic properties of the subsurface. Although a vast variety for hydraulic exploration methods exist, e.g. slug-test, pumping test, grains-size analysis, different tracer tests (e.g. Bakker et al., 2015, Vienken et al., 2012, Chia-Shyun and Ching-Rung, 2006, Butler et al., 2003), there is less variation for determining shallow subsurface thermal properties. A common way to gain knowledge of subsurface thermal properties is to rely on laboratory results originating from cores or borehole debris or using the information from geological surveys (sampling, maps) to derive the thermal properties from literature (e.g. VDI, 2010, Hamada et al., 2002).

Since the mid 1990s, subsurface thermal transport parameters have commonly been determined in situ using the Thermal Response Test (TRT) (e.g. Beier et al., 2013, Sanner et al., 2005, Gehlin, 2002). The aim of this test is to determine effective thermal conductivity of the subsurface. To achieve this, a borehole-heat-exchanger (BHE) is heated or cooled, over 72 h, and the resulting inflow/outflow temperature of the heat-exchanging fluid is measured and analysed e.g. Gehlin (2002). A disadvantage of this approach is the fact that in the original design, only an integral value of thermal conductivity over the entire length of the BHE can be estimated. Nevertheless, different thermal usage of the subsurface requires different kinds of information regarding the thermal properties of the subsurface. For example, thermal extraction or injection systems rely on advective heat-transport for regeneration but within thermal storage systems such advection decreases the storage recovery efficiency. To overcome this disadvantage, an enhanced TRT was developed using distributed temperature sensing along the installation length, to obtain denser temperature information and horizontally layered information. (e.g. Acuña and Palm, 2013, Raymond et al., 2011, Fujii et al., 2009). The development of depth dependant TRTs proves the need for better spatial resolution of subsurface heterogeneity. This is especially important within systems of shallow depth (30–40 m) where changes in thermal properties of geological units of limited thickness may already impact geothermal systems performance (e.g. Sommer et al., 2013, Ferguson, 2007).

With the advent of fibre-optic distributed-temperature-sensing (DTS) technology becoming available for hydrology and groundwater research (e.g. Liu et al., 2013, van de Giesen et al., 2012, Hausner et al., 2011, Tyler et al., 2009, Moffet et al., 2008, Selker et al., 2006), scientists now have at their disposal a tool for obtaining spatial and temporal high resolution temperature measurements. This tool is already used, e.g. in hydrogeological characterisation for different geological settings (e.g. Bakker et al., 2015, Read et al., 2013, Liu et al., 2013, Leaf et al., 2012). Raman-DTS systems measure the Raman scattering along the fibre, using the ratio of anti-Stokes to Stokes peaks for temperature measurement, as it is temperature dependant. With the help of travel time inversion, it is possible to identify the location of the fibre where the signal originates from. This enables high spatial resolution measurements along the fibre. A more detailed overview of DTS principals and methods is available from van de Giesen et al., 2012, Hausner et al., 2011, Selker et al., 2006.

The initial aim of this study was to test ‘in-aquifer heating’ to ascertain and characterise the qualitative heat-storage capability of heterogeneous aquifers using temperature signal decay times and observing temperature evolution within a heterogeneous Quarternary aquifer.

In-aquifer heating describes the process of creating thermal pulses directly within the water column of the borehole, instead of injecting heated water into the well. In this study, a self-regulating electrical heating cable was used in combination with high-resolution Fibre-Optic-Distributed-Temperature-Sensing (FO-DTS) to gather high resolution spatial and temporal temperature data. To improve the spatial resolution from the metre scale (inherent resolution of common DTS-devices, e.g. Agilent Technologies, 2007) to the centimetre scale for vertical borehole measurements, the fibre was wrapped around tubes, which has been a successfully used in other studies (Liu et al., 2013, Vogt et al., 2010). Direct-Push technology was used for in situ vertically high-resolution (hydro-)geological profiling, to enable evaluation and better comparability of the approach used.

Nevertheless, the approach using temperature decay times as proxy for thermal storage capability has several main benefits, including how it enables us to directly describe thermal-storage behaviour of the shallow subsurface and its high measurement resolution compared to conventional TRT. Additionally, by using in-aquifer heating instead of adding volumes of heated water into the aquifer, as is commonly done during thermal tracer tests, we can also avoid disturbing the natural flow of the aquifer without using complex pumping systems to maintain the balance of groundwater levels. This is especially interesting for installing underground thermal energy storage systems relying on closed loop systems (e.g. BHEs) as those systems are not changing the hydraulic gradients like open loop systems and therefore rely on the natural groundwater flow.