Evaluating the impact of anisotropy and low-permeability layers on high temperature aquifer thermal energy storage performance

The seasonal storage of heat has the potential to reduce greenhouse gas emissions, since it accounts for the seasonal disbalance between heat production and demand in renewable-based energy-systems. High temperature aquifer thermal energy storage (HT-ATES) is a heat storage technology utilising the subsurface and therefore provides high storage capacity with limited above-ground usage, which is especially required in urban areas. Since the temperature difference between the ambient groundwater and the injected water (> 50 °C) results in density differences, convective buoyancy flow can be induced by HT-ATES. This process leads to an uneven heat distribution over the aquifer thickness, reduced storage efficiency and increased thermal impacts. The occurrence and intensity of buoyancy flow is site-specific, since it depends on operational parameters, such as injection temperature, as well as geological parameters, such as aquifer thickness and especially vertical and horizontal permeability.

The geological site considered for HT-ATES storage is located in Hamburg, Germany, using the Miocene Lower Braunkohlensande (brown coal sands) as storage aquifer. This sedimentological formation was deposited in a coastal transition regime between terrestrial and shallow-marine settings and consists mainly of sands. Peat swamps and lagoons formed brown coal, silt and clay layers, which have the potential to hinder convection due to their low permeability, depending on their lateral extent in relation to the size of the induced heat plume by HT-ATES. Lithological classifications of 25 wells provide the data basis for the geological analysis.

The aim of this study is to evaluate the influence of thin low permeability layers on induced buoyancy flow and thus HT-ATES performance, as measured by heat recovery. To this end, a site-specific numerical HT-ATES model is created, which simulates the coupled thermo-hydraulic processes. Different scenarios with varied vertical permeability as well as the number and lateral extent of low-permeability layers show the effect on density-driven buoyancy flow and HT-ATES efficiency. Increasing the vertical permeability by a factor of 10 results in an efficiency decrease from 78 % to 57 % in the 10^th storage cycle. The findings serve as a process understanding basis for complex heterogeneous facies models of the HT-ATES site, which will be based on the geostatistical evaluation of site data.