Petrophysical Data and Core-Calibrated Multimineral Interpretation
Becker, J. & Marnat, S. (2024): Petrophysical Data and Core-Calibrated Multimineral Interpretation. Nagra Arbeitsbericht NAB 24-17.
pdf NAB 24-17 Petrophysical Data and Core-Calibrated Multimineral Interpretation(20.48 MB)
Basic rock properties, such as mineralogy and porosity, fundamentally influence barrier integrity as they control transport properties and self-sealing, sorption and diffusion parameters. Hence, the characterisation of these parameters, next to, e.g. hydrogeological, geomechanical and geochemical parameters, is of key importance. However, standard laboratory measurements cannot easily produce continuous profiles of mineralogy or porosity since they usually have a sample spacing of several metres. Other measurements techniques, e.g. core scanning, have the potential to produce quasi-continuous profiles (as the spacing between their measurements may be limited to a few millimetres). However, as the spacing decreases, measurement time increases. In addition, these measurement techniques necessitate (almost) perfectly preserved core material. Hence, a combined approach, the so-called multimineral interpretation, has been applied. This method is capable of combining all available data (i.e. geophysical logs, laboratory measurements and core-scanning measurements) to develop quasi-continuous profiles of borehole mineralogy and porosity.
To calculate the continuous borehole profiles at any given depth using the multimineral interpretation approach, a mineralogical and fluid content as well as porosity are assumed (based on expert judgement and prior knowledge), and the theoretical response of the available measurements is calculated. Through local and overall error minimisation, the assumed volumes are adjusted until the errors between the calculated theoretical response and the actual measurements of the available data have been minimised and result in continuous profiles of mineralogy and porosity. This workflow obviously carries an intrinsic uncertainty with respect to interpretation. Other factors that can influence interpretation include, e.g., borehole conditions impacting the data quality of the measurements.
As previously mentioned, these interpretations result in quasi-continuous curves of mineralogyand porosity along the borehole walls. Mineralogical data include clay as well as quartz and feldspar, carbonate, halite, anhydrite, pyrite, goethite and siderite contents and their respective uncertainties. The interpretations are homogenised between the different boreholes to allow comparison of the results, both within a siting region and also across the three siting regions. This workflow has been applied to the measurements conducted in 12 boreholes (Bözberg-1-1 and Bözberg-2-1, Bachs-1-1, Stadel-2-1 and Stadel-3-1, Bülach-1-1, Marthalen-1-1, Rheinau-1-1 and Trüllikon-1-1, Benken, Riniken and Weiach-1). These are summarised here and have been detailed in a number of separate reports. However, the Benken, Riniken and Weiach-1 boreholes were drilled several decades ago and, the acquired data quality for these boreholes is therefore not on par with the very recent data acquisition in the Bözberg-1-1 and Bözberg-2-1, Bachs-1-1, Stadel-2-1 and Stadel-3-1, Bülach-1-1, Marthalen-1-1, Rheinau-1-1 and Trüllikon-1-1 boreholes.
In the more recent boreholes, the interpreted mineral data correlate very well with the respective laboratory measurements (high correlation coefficients, usually > 0.9). Several laboratorymeasurements of porosity exist, each using a different geophysical principle and hence each measuring a slightly different water content. Still, the correlation of laboratory and multimineralporosity is high (> 0.8). Data trends from the laboratory mineralogy are also reproduced very well so that the continuous profiles interpreted from the geophysical logs using additional data can confidently be used for further geostatistical analyses and site characterisation.
Due to reduced data quality, correlation coefficients between interpreted multimineral data and laboratory measurements are lower in the older Benken, Riniken and Weiach-1 boreholes. The mineralogical data from the Benken borehole, which provided the highest data quality of the older boreholes, still shows correlation coefficients of ~ 0.8. However, correlation of porosity is much lower and ranges around 0.5. Hence, interpretation results from these boreholes should not be directly compared to the more recent boreholes and should only be qualitatively used as additional information.