Reconstructing post-Jurassic overburden in Central Europe: New insights from mudstone compaction and thermal history analyses of the Franconian Alb, SE Germany

The Franconian Alb of SE Germany is characterized by large-scale exposures of Jurassic shallow marine limestones and dolostones which are frequently considered as outcrop analogues for deep geothermal reservoir rocks in the North Alpine Foreland Basin farther south. However, the burial history of the Franconian Alb Jurassic 15 strata is not well known as they were affected by emersion, leading to extensive erosion and karstification with only remnants of the original Cretaceous and Cenozoic cover rocks preserved. To estimate the original thicknesses of the post-Jurassic overburden we investigated the petrophysical properties and the thermal history of Lower and Middle Jurassic mudstones to constrain their burial history in the Franconian Alb area. We measured mudstone porosities, densities, and maturities of organic material and collected interval velocities from seismic refraction 20 and logging data in shallow mudstone-rich strata. Mudstone porosities and P-wave velocities vertical to bedding were then related to a normal compaction trend that was calibrated on stratigraphic equivalent units in the North Alpine Foreland Basin. Our results suggest maximum burial depths of 900 - 1700 m of which 300 - 1100 m are attributed to Cretaceous and younger sedimentary rocks overlying the Franconian Alb Jurassic units. Compared to previous considerations this implies a more widespread distribution and increased thicknesses of up to ~900

Following long-lasting denudation, Cenozoic subsidence of the North Alpine Foreland Basin towards the south, contemporaneous to ongoing uplift of basement areas towards the east, led to erosional retreat of incised valleys 1995; Zweigel et al., 1998). Nevertheless, the post-Jurassic burial history of the Franconian Alb area is rather uncertain, as only a few remnants of Cretaceous and Cenozoic sediments are preserved locally (Dill, 1995;Peterek and Schröder, 2010), .  (Bachmann et al., 2002). Subtracting reported regional Middle/Upper Keuper and Jurassic sediment thicknesses did not quantify the maximum post-Jurassic sediment overburden, we aim to tackle this question by combining several methodological approaches that rely on independent data sets.

Study aim 130
In this study we combine mudstone porosity and density data from helium and mercury porosimetry with vitrinite reflectance data and mudstone velocity data from downhole sonic velocity, downhole geophone and seismic refraction field surveys to gain independent insights on the maximum burial of the Franconian Alb. The results will be compared with and discussed in the context of previous studies (Bader, 2001;Hejl et al., 1997;Peterek and Schröder, 2010;Schröder, 1987;von Eynatten et al., 2021).

135
Our results shed new light on the evolution of the Franconian Alb area and the original distribution and thicknesses of Cretaceous and Cenozoic sediments in Central Europe. They are also of great relevance for an improved understanding of diagenetic pathways and hydraulic properties of the Permo-Triassic clastics and Late Jurassic carbonate rocks in the Franconian Alb. The latter serve as important outcrop analogue for the most important deep geothermal (Malm) aquifer in the North Alpine Foreland Basin (Kröner et al., 2017;Mraz et al., 2018), whose 140 petrophysical properties are known to strongly depend on burial depth (Bohnsack et al., 2020(Bohnsack et al., , 2021Homuth et al., 2014;Steiner et al., 2014). Finally, the integration of different parameters and measurement types provides an important reference data set (Table A-2) for future studies, aiming to use petrophysical properties of exhumed and near-surface located mudstones for burial history studies.  Table 1 summarizes all sample locations, sample sources, sample types, sample depth below ground, and stratigraphic positions in addition to applied methods and number of measurements per sample.

150
Measured and calculated values for each sample are shown in Appendix Table A-1. Macroscopically "pure" Jurassic clay-/mudstones (minimum sample size 10 x 10 x 10 cm) were selectively sampled at 0.5 m minimum depth (to avoid alteration/weathering) from nine active and closed claystone pits and from five newly drilled shallow drill cores (up to 12 m below ground level). Except for core samples from Velburg and Zankschlag all samples were packed and stored in an air-evacuated light-, water-and air-proof aluminium barrier foil directly 155 after extraction to preserve the best possible in-situ conditions. Interval velocity data of Lias and Dogger clay-/mudstones from a shallow seismic refraction survey for low velocity layers in the course of this study (see Figure   1 for locations), published borehole geophone data of Buness and Bram (2001) and sonic log velocity data from a shallow wellbore (Zapfendorf) in the NW part of the study area (Welz, 1994) were also integrated.   Abbreviations: ABDNB = Autobahndirektion Nordbayern; GSC = grain size classification; ρt = true (skeletal) density; ρb = Bulk density; ØHg = Mercury intrusion porosimetry-derived porosity; Vp = P-wave velocity (in situ); VR = Vitrinite reflectance; XRD = X-ray diffraction.

Reference data from the North Alpine Foreland Basin
Density and sonic log data of 9 deep wells in the North Alpine Foreland Basin ( Figure 3) have been filtered for appropriate mudstone intervals using gamma-ray (mudstone cut-off at 60-120 API) and/or resistivity values 165 (mudstone cut-off at 4-8 Ωm) as a mudstone discriminator and log values were subsequently averaged over 150 m depth intervals. The data were used to validate the normal compaction trend (NCT) determined by Drews et al. (2018) with regard to mudstone density data.
Mudstone compaction has been intensively studied in the past (e.g. Aplin et al., 2006;Dewhurst et al., 1998; 185 Vasseur et al., 1995) and is mainly controlled by grain size (Fawad et al., 2010;Mondol et al., 2007;Yang and Aplin, 2004), mineralogical composition (Fawad et al., 2010;Marion et al., 1992;Mondol et al., 2007), and texture (Fawad et al., 2010;Marion et al., 1992;Mondol et al., 2007). Strongly increased rock strength and velocity was observed for mudstones with high sand content and <40% clay (Marion et al., 1992) as well as with elevated cement content (Horpibulsuk et al., 2010). These issues were considered in this study by measuring the mudstones' 190 mineralogical composition and grain size distribution. As the mudstones' compaction behaviour is thought to be almost irreversible even after unloading they are particularly well suited to record maximum burial, respectively overburden (e.g. Corcoran andDoré, 2005, Hillis, 1995;Menpes and Hillis, 1995) and have therefore frequently been applied in various studies (e.g. Baig et al., 2019;Henk, 1992;Issler, 1992). The degree of mudstone compaction is thereby best reflected in the rocks' (bulk and true) density, porosity, and its ability to conduct 195 acoustic pulse signals. All three parameters were determined or used in this study. Another source of information for maximum burial of mudstones is given by vitrinite reflectance, a measure of the increasing thermal maturation of organic matter contained in mudstones (Hertle and Littke, 2000;Liu et al., 2020;Sweeney and Burnham, 1990).

Mineralogy
For XRD-based whole rock mineralogical classifications the dried mudstone samples were crushed and grinded 200 with the McCrone XRD mill and analysed by a X-ray diffractometer D5000 (Siemens). A qualitative Rietveld analysis of the resulting signal was then done with the DIFFRAC.SUITE software EVA and thereafter, semiquantitatively with the DIFFRAC.SUITE software TOPAS 4.2 (both by Bruker).

Grain size analysis
Full disaggregation of the solid samples was achieved by applying the "saturation-freeze-thaw" method of Yang and Aplin (1997). Particle size analysis by sedimentation was done applying a SediGraph III Plus by Micromeritics. The grain size classes are differentiated according to the geotechnical grain size classification scheme for soils (Deutsches Institut für Normung, 1987), where the clay fraction comprises particles <2 μm, the silt fraction particles of 2-63 μm, and sand particles are >63 μm. The grain size classification scheme follows Potter et al. (1980).

Mechanical compaction deduced from porosity-velocity relationships
Due to the mudstones' largely irreversible elastoplastic compaction behaviour, the degree of mechanical mudstone compaction provides a good first-order estimate of the maximum mean effective stress (e.g. Corcoran and Doré, 2005;Goulty, 1998;Hillis, 1995), hence the maximum burial depth, thereby assuming that the vertical stress represents the largest principal stress and the vertical effective stress gradient is known.

215
Mechanical compaction in terms of porosity decrease and velocity increase of both Mesozoic and Cenozoic mudstones from the North Alpine Foreland Basin have been previously investigated as a function of vertical effective stress by Drews et al. (2018). The North Alpine Foreland basin is situated directly south of the study area ( Figure 1 and Figure 3) and uplift since maximum basin subsidence is estimated to have not exceeded more than ~500 m there (Baran et al., 2014;Drews et al., 2018;Kuhlemann and Kempf, 2002;Zweigel et al., 1998). Thus 220 the depth-related increase in mudstone compaction in the North Alpine Foreland Basin (NAFB) is likely a good analogue for our study area. Drews et al. (2018) determined a mudstone compaction trend which utilizes porosity decay as a function of vertical effective stress, based on the exponential compaction law of Athy (1930) (eq. 1): Heppard et al. (1998), Rubey and Hubbert (1959), and Scott and Thomsen (1993). Øsh is the mudstone porosity at a particular depth. Following Drews et al. (2018) the mudstone porosity at the surface Ø 0_sh was set to 0.4 (dimensionless) and the compaction coefficient C to 31 MPa -1 .
The porosity-velocity relationship proposed by Raiga-Clemenceau et al. (1986) can then be used to derive a velocity vs. vertical effective stress relationship: Equation 2 is the mudstone porosity-velocity relationship of Raiga-Clemenceau et al. (1986) where Vp is the pwave velocity in mudstones. For the NAFB, Drews et al. (2018) set the matrix velocity of mudstones Vpshm to 5076 m/s and x to 2. Alternatively, Ø can be substituted by the water-saturated mudstone bulk density ρb_sat using the following relationship: Where ρt is the true or skeletal density of the mudstone and ρf is the density of the pore-filling fluid with 1.0 g/cm³ for water. The maximum burial depth TVDmax can then be estimated from VES: with the vertical effective stress gradient VESgrad typically varying between 10-16 MPa/km in hydrostatically 240 pressured sedimentary basins, derived from a vertical stress gradient of 20-26 MPa/km and a hydrostatic pore pressure of 10 MPa/km (Bjørlykke, 2015). For the NAFB, Drews et al. (2018Drews et al. ( , 2020) determined a vertical effective stress gradient of 13 MPa/km, which will also be used for depth calculations in this study.

Porosity and density
Dry bulk densities ρb_dry and porosities ØHg of 72 clay-/mudstone samples have been measured with a mercury 245 intrusion porosimeter ("Poremaster 60" by Quantachrome) which analyzes pore diameters in the range of 0.0036 -950 μm under pressures of up to 60000 psia. Prior to measurements, samples were dried at 65°C until no change in mass could be determined for 24 hours. Thereby, cracks may have formed during sample preparation and dehydration (Klaver et al., 2012). In turn this might result in the intrusion of mercury into these cracks at low pressures, but associated data excursions are rather obvious and were removed prior to further analysis as proposed 250 by Klaver et al. (2015). True (skeletal) densities ρt were determined for a subset of 41 samples by applying Helium pycnometry ("Accupyk II 1345" by Micromeritics), which enables analysis of even smaller pores (0.22 nm) than mercury (3.6 nm) (Hedenblad, 1997;Krus et al., 1997). For samples lacking direct ρt measurements, the mean true density ρt_mean was used for further calculations. Using bulk density ρb_dry and true density ρt, respectively ρt_mean the (effective) porosity Øcalc was calculated:

Velocity modeling based on density/porosity measurements
Applying the porosity-velocity relationship (c.f., eq. 2) proposed by Raiga-Clemenceau et al. (1986), respectively the velocity-density relationship by using density instead of porosity values (c.f., eq. 5) then allows for the calculation of mudstone velocities. Calculating mudstone velocities from Øcalc yields Vpcalc, while mudstone velocities based on measured ØHg values are labelled Vpcalc-Hg.

Mudstone velocity
In situ mudstone velocities Vp were derived from near surface (15-45 m TVD, see Table 1) seismic refraction data acquired in the course of this study (see locations in Figure 1), published borehole geophone measurements (Buness and Bram, 2001), and downhole sonic log readings (Welz, 1994).

Vitrinite reflectance
Random vitrinite reflectance in oil (VR) was determined for 11 selected samples (Table 1) using a magnification of 100× in non-polarized light at a wavelength of 546 nm (Taylor et al., 1998). Yttrium-Aluminium-Garnet (R=0.899%) and Gadolinium-Gallium-Garnet (R=1.699%) standards were used for calibration. As the vitrinite maturation is mainly affected by temperature as well as by the duration of maximum burial (Nöth et al., 2001) and

270
only to a minor degree by pressure (Hunt, 1979), these measurements are strongly dependent on the evolving heat flow and therefore the geothermal gradient within a sedimentary basin (Suggate, 1998).
. A VR-depth-trend was constructed, based on published vitrinite reflectance data (Gusterhuber et al., 2012) and partly unpublished data for Cretaceous mudstones in the northern part of the NAFB in Austria, where the samples' burial depths were known to allow calibration ( Figure 3). From the correlation between the measured sample vitrinite reflectance and the VR-depth-trend, the burial depth of Franconian Alb clay-/mudstones was inferred. As the Mesozoic burial history of the northern part of the Upper Austrian Molasse Basin (Nachtmann 280 and Wagner, 1987) is rather similar to the Franconian Alb area Schröder, 1987), a comparison between our samples and the developed VR-depth-trend is considered as reasonable.

Mudstone composition
41 clay-/mudstone samples were analyzed in terms of their grain size classification ( Figure 4A) and 37 regarding 285 their mineralogical composition ( Figure 4B) to ensure that we base our study on a rather homogeneous sample set in terms of grain size and mineralogical composition.

Grain size classification
Most of the claystone pit samples contain <10% of grains >63 μm (sand fraction), 40-60% of grains in the range 2-63 μm (silt fraction), and 40-60% of grains <2 μm (clay fraction). Therefore the majority of samples classifies 290 as "mudstones" or "claystones" ( Figure 4A). Exceptionally high clay fraction percentages were observed for few samples from the claystone pit Großheirath as well as for core samples from Mistelgau and Zankschlag ( Figure   4A). The fact that cores from one well location were sampled at various depth levels, explains the large spread in grain size classifications, particularly for the Zankschlag well samples, where several meters of cores were analysed. Two Zankschlag core samples with increased sand and decreased clay contents ( Figure 4A) were classification scheme of Potter et al. (1980). This is because major deviations in petrophysical properties (e.g. porosity and p-wave velocity) of mudstones and compaction behaviour are reported for samples with increasing sand admixture and <40% clay content (Marion et al., 1992).

Mineralogical composition 300
Clay mineralogical studies of marine Jurassic clays and marls in our study area by Krumm (1965)

Integrating mudstone porosity and velocity data
Dry bulk densities ρb_dry and porosities ØHg were analyzed from 72 samples by Hg-intrusion porosimetry and true (skeletal) densities ρt with an average value ρt_mean of 2.73 ± 0.06 g/cm³ ( Figure 6A) of 34 clay pit and shallow drill core samples (Table 1) were determined by He-pycnometry. Mudstone porosities were also calculated (Øcalc), based on bulk densities ρb_dry and true (skeletal) densities ρt_mean (eq. 5).

350
We preferred the calculated porosity values rather than Hg-porosities because continued mercury intrusion even at the device's maximum injection pressure (see inset in Figure 6B) suggested that micropores <0.003 μm were not fully involved in the measurement. The cross-plot of calculated porosities Øcalc versus measured porosities ØHg reveals major discrepancies due to the incomplete involvement of micropores by using Hg-porosities ( Figure 6B).
The relation between downhole mudstone velocities and bulk densities is well captured by the NCT established  were measured roughly on a meter-scale and most likely also captured larger unloading structures due to the shallow present-day burial depth, the measured porosity data are derived from cm-sized samples, which most likely are not as much affected by unloading and if such effects were recognized they were removed from the analysis (see caption of Figure 6).
Applying an average vertical effective stress gradient of 13 MPa/km to field velocity data of mudstones Vpseis and

395
Vplog yields a maximum burial depth for Franconian Alb area samples of 0.0-1.8 km (0.9 ± 0.4 km mean), whereas Vpcalc and Vpcalc-Hg yield 1.0-3.6 km (1.8 ± 0.4 km mean) versus 1.7-6.9 km (2.8 ± 0.8 km mean) burial, respectively (   As no information on the paleo heat flow in this region is available, no vitrinite reflectance evolution with depth could be modelled for the study area. However, a comparable VR-depth-trend is derived from published (Sachsenhofer, 2001)

The Franconian Alb burial history in a regional context
Our burial depth calculations for the Early to Middle Jurassic mudstones of the Franconian Alb area suggest a burial depth of at least 900 m, based on downhole and shallow seismic refraction mudstone velocities, but rather ~1700 m as inferred from calculated porosities Øcalc and VR data as any unloading and drying effects can be ruled out in these data sets ( Figure 10). A strong overestimation of maximum burial depths derived from ØHg porosity 455 values is displayed in Figure 10C but has low reliability due to the incomplete micropore involvement ( Figure   6B). As the thicknesses of Early Jurassic strata (~20 m in the southern and ~100 m in the northern Franconian  Table A-1. Our Vitrinite Reflectance data ( Figure 10A and D), indicating burial depths of 0.8-2.2 km (mean 1.7 km), correlate very well with burial depth of ~1.7 km inferred from calculated porosities Øcalc applying He-pycnometry derived mean true densities ρt_mean and bulk densities ρb ( Figure 10B) (see Table A-1).

465
West of the Franconian Line, AFT-data  and field mapping-and literature-based interpretations (sedimentological studies, thermochronological data, radiometric age data, etc.) suggest deposition and subsequent removal of > 1000 m of Cretaceous and Cenozoic sediments (Peterek and Schröder, 2010;Schröder, 1987;Schröder et al., 1997) of which only c. 320 m of Upper Cretaceous strata are preserved (Dill, 1995). Hence, compared to the more distal western parts of the Franconian Alb, strongly increased depositional thicknesses along 470 the front of the Franconian Line can be considered due to the uplift and major exhumation of the Bohemian Massif to the east, combined with westward thrusting, and syntectonic deposition of the eroded material Peterek and Schröder, 2010).   (2021) estimated at only 30°C/km. This gradient contrasts to an elevated regional geothermal gradient of 38°C/km argue that no Cretaceous sediments were deposited in the western part of the Franconian Alb area. This conclusion can most likely be related to the more distal-to-source position of their study area, positioned between Tübingen and Würzburg, compared to ours (Franconian Alb area). As Cretaceous sediments in the Franconian Alb area were 505 most likely sourced from the Bohemian Massif towards the east (Niebuhr et al., 2011(Niebuhr et al., , 2012Schröder, 1987;Schröder et al., 1997;Voigt et al., 2008), a reduced sediment supply to positions more distal to the source can be expected. Westward decreasing Cretaceous sediment columns, as proposed by  and Peterek and Schröder (2010), support this interpretation.

510
The lateral variation of calculated burial depths derived from two independent data sets ( Figure 10A-D) is showing no regional trends nor are areas of increased or reduced burial depth noticable. Only in the case of the porosityderived burial depth estimations ( Figure 10B), a trend towards increased amounts of post Lower Jurassic paleothicknesses in the northwestern part of the Franconian Alb can be conjectured, though this impression is based on a sparse data density in the area of interest.

515
Additional information comes from published AFT-and measured VR-data. From the VR results, no distinct differential vertical movements between various parts of the Franconian Alb can be inferred. According to von the doming centre which is located further to the north of our study area. Hence, our study area was most likely less affected by doming-related processes. The AFT-results of Hejl et al. (1997) and the sedimentological observations of Schröder (1987) and Peterek and Schröder (2010) additionally suggest that higher sediment thicknesses (~2 km) were deposited directly west of the Franconian Line compared to the more distal-to-source parts. The more distal-to-source locations of the majority of our samples most likely explains these reduced burial depths. Reasons for reduced sediment removal in the southwestern part of the study area are given by Peterek and Schröder (2010). They suggest temporarily reduced erosion rates in this area due to the coverage by Neogene lake sediments that protected underlying Mesozoic sediments from erosion.
Information on the timing of sediment deposition and removal in the study area could not be inferred from our 530 data but this issue has been investigated by various authors (Peterek and Schröder, 2010;Schröder, 1987;Schröder et al., 1997;Ziegler, 1987). Early Cretaceous sediments (Late Valanginian to Barremian) constitute, if at all, only minor to neglegible ratios of the original sediment column and have most likely been removed already during the Late Valanginian to Cenomanian erosional event (Schröder, 1987). Sedimentation resumed during the Cenomanian/Turonian to Campanian (Ziegler, 1987;Meyer, 1981Meyer, , 1989a, and related deposits must have constituted the majority of the eroded sediments, as sedimentation in most parts of the study area ceased thereafter (Peterek and Schröder, 2010). This termination in sedimentation was superseded by the profound erosion of Cretaceous sediments, caused by the latest Cretaceous to Paleocene inversion (Schröder, 1987;Schröder et al., 1997). Probably uplift associated with thermal doming of the Bohemian Massif continued after the Miocene, resulting in the absence of a widespread sediment cover in the study area (Peterek and Schröder, 2010;Schröder, 540 1987;Schröder et al., 1997).
In summary our data suggest that considerable amounts of post-Jurassic sediments must have been removed from the investigated area. Having information on the paleo-stress conditions during burial of nowadays surfaceexposed sedimentary rocks is a key for relating their petrophysical properties to their deeply buried analogues. Our results indicate that the Upper Jurassic "Malm" carbonates, which are exposed in the Franconian Alb area and 545 plunge southwards to depths of up to 5500 m in the Alpine foreland (Bachmann et al., 1987), constitute suitable analogues for reservoirs drilled at equivalent burial depths of ~1050 m in the NAFB. This would directly apply to the geothermally productive Malm reservoirs in the proximal north of Munich and in the Moosburg-Landshut area ( Figure 3).

550
This study aimed to quantify eroded thicknesses of post-Jurassic sediments that were originally deposited in the  (Welz, 1994), and seismic refraction survey Vpseis). Additionally, the burial depth results from the correlation between the VR-depth-trend and measured VR are listed. From these results, also the amount of removed post-Jurassic (post-Jur) sediments was estimated. In case of buried samples where the mean burial depth is not equal to the total amount of eroded sediments, the amount of total sediment removal was additionally calculated. Values smaller than zero are excluded as they indicate unrealistically low burial depths, meaning that these samples were deposited later than the Middle Jurassic, although they are pre-Upper Jurassic sediments. Location abbreviations and associated locations and sampled stratigraphic units are listed in Table 1 and illustrated in Figure 1.