Petrophysical analysis of regional-scale thermal properties for improved simulations of geothermal installations and basin-scale heat and fluid flow

Development of geothermal energy and basin-scale simulations of fluid and heat flow both suffer from uncertain physical rock properties at depth. Therefore, building better prognostic models are required. We analysed hydraulic and thermal properties of the major rock types in the Molasse Basin in Southern Germany. On about 400 samples thermal conductivity, density, porosity, and sonic velocity were measured. Here, we propose a three-step procedure with increasing complexity for analysis of the data set: First, univariate descriptive statistics provides a general understanding of the data structure, possibly still with large uncertainty. Examples show that the remaining uncertainty can be as high as 0.8 W/(m K) or as low as 0.1 W/(m K). This depends on the possibility to subdivide the geologic units into data sets that are also petrophysically similar. Then, based on all measurements, cross-plot and quick-look methods are used to gain more insight into petrophysical relationships and to refine the analysis. Because these measures usually imply an exactly determined system they do not provide strict error bounds. The final, most complex step comprises a full inversion of select subsets of the data comprising both laboratory and borehole measurements. The example presented shows the possibility to refine the used mixing laws for Petrophysical properties and the estimation of mineral properties. These can be estimated to an accuracy of 0.3 W/(m K). The predictive errors for the measurements are 0.07 W/(m K), 70 m/s, and 8 kg/m^3 for thermal conductivity, sonic velocity, and bulk density, respectively.

💡 Research Summary

The paper addresses the critical problem of uncertain subsurface rock properties that hampers both geothermal energy development and basin‑scale heat‑fluid flow modelling. Focusing on the Molasse Basin in southern Germany, the authors measured thermal conductivity, bulk density, porosity, and compressional wave velocity on roughly 400 core samples representing the main lithologies (Upper Jurassic limestones and Lower Triassic sandstones). The study proposes a three‑step analytical workflow of increasing sophistication.

Step 1 employs univariate descriptive statistics to obtain mean values, quartiles and confidence intervals for each lithostratigraphic unit. When the entire formation is treated as a single dataset, the thermal conductivity uncertainty can be as large as 0.8 W (m·K)⁻¹; however, subdividing the formation into petrophysically homogeneous sub‑units reduces the 50 % confidence interval to about 0.1 W (m·K)⁻¹.

Step 2 introduces quick‑look cross‑plot methods based on classic mixing laws: Wyllie’s equation for acoustic slowness, a linear density mixing rule, and the geometric Woodside‑Messmer law for thermal conductivity. By plotting measured thermal conductivity against slowness, the data fall within a triangular domain defined by end‑members (calcite, shale, water). The authors show that end‑member values must be calibrated to the dataset; for example, the shale slowness and limestone conductivity required adjustment from literature values to 220 µs m⁻¹ and 3.1 W (m·K)⁻¹ respectively. This step provides rapid insight into mineral fractions and validates the applicability of the mixing models.

Step 3 performs a full inversion on selected subsets that combine laboratory measurements with borehole log data (density, gamma‑ray, sonic). The inversion simultaneously refines mixing‑law parameters and estimates mineral volume fractions (e.g., shale content) with an accuracy of about 0.3 W (m·K)⁻¹ for the matrix conductivity. Predictive errors achieved are 0.07 W (m·K)⁻¹ for thermal conductivity, 70 m s⁻¹ for sonic velocity, and 8 kg m⁻³ for bulk density, which are sufficiently low for direct use in geothermal design calculations and regional numerical models.

The authors discuss the trade‑offs among the three approaches: simple statistics are fast but may retain large uncertainties; cross‑plots are intuitive but rely on well‑calibrated end‑members; full inversion yields the highest accuracy at the cost of computational effort and data completeness. By applying this tiered methodology, the study demonstrates a practical pathway to reduce the risk of geothermal drilling and to provide robust, region‑specific petrophysical parameters for basin‑scale simulations.