From Burial to Barrier: How burial history controls the hydraulic conductivity in argillaceous formations

Burchartz, Raphael; Mbui, Brian Mutuma; Achtziger-Zupančič, Peter; Gaus, Garri; Seemann, Timo; Winhausen, Lisa; Spychala, Yvonne; Jalali, Mohammadreza; Littke, Ralf; Amann, Florian

doi:10.5194/egusphere-2026-964

Preprints

https://doi.org/10.5194/egusphere-2026-964

Preprints

17 Mar 2026

| 17 Mar 2026

Status: this preprint is open for discussion and under review for Solid Earth (SE).

From Burial to Barrier: How burial history controls the hydraulic conductivity in argillaceous formations

Raphael Burchartz, Brian Mutuma Mbui, Peter Achtziger-Zupančič, Garri Gaus, Timo Seemann, Lisa Winhausen, Yvonne Spychala, Mohammadreza Jalali, Ralf Littke, and Florian Amann

Abstract. Deep geological repositories for high-level radioactive waste (HLW) rely to a large extend on the long-term hydraulic integrity of host rocks to limit fluid flow and radionuclide migration. Low hydraulic conductivity (K < 10^-10 m/s) is a key factor for effective long-term barrier performance, and argillaceous formations are promising candidates due to their strong aquitard characteristics. However, predicting their bulk hydraulic behaviour across temporal and spatial scales remains difficult, as it reflects the combined effects of intrinsic material properties and post-depositional evolution. This study compiles 782 hydraulic conductivity measurements from six European argillaceous formations, including laboratory and field scales. By integrating petrophysical, mineralogical, and reconstructed burial history data, we identify systematic links between burial evolution and hydraulic behaviour. Results show that maximum burial depth and associated stress and temperature conditions exert a first-order control on matrix-scale hydraulic conductivity, which is largely retained after uplift. In contrast, bulk hydraulic behaviour at the rock-mass scale reflects interactions between maximum burial depth and present-day depth, defining processes such as decompaction, fracturing, and self-sealing processes. Three evolutionary trends emerge from the compiled data: (1) Shallowly buried (<400 m), poorly indurated formations show limited hydraulic variability and scale independence; (2) Moderately buried (~800 m – 2,000 m), overconsolidated formations retain low matrix hydraulic conductivity after uplift, but exhibit gradually (partly significantly) enhanced hydraulic conductivity at depths <100 m due to the evolution of a pronounced decompaction zone. When devoted to less pronounced uplift and at greater present-day depths (>250 m) matrix and bulk hydraulic conductivities converge and predominantly range within a natural variability between 10^-14 to 10^-12 m/s, indicating effective self-sealing processes; (3) deeply buried formations (>2,000 m) become increasingly indurated and brittle, with reduced self-sealing capacity due to the loss of swellable clay mineral phases and fracture-dominated bulk hydraulic behaviour. Matrix and rock-mass hydraulic conductivities may diverge by several orders of magnitude. These trends provide predictive insights into the long-term barrier performance of argillaceous host rocks in HLW repositories.

Received: 19 Feb 2026 – Discussion started: 17 Mar 2026

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Raphael Burchartz, Brian Mutuma Mbui, Peter Achtziger-Zupančič, Garri Gaus, Timo Seemann, Lisa Winhausen, Yvonne Spychala, Mohammadreza Jalali, Ralf Littke, and Florian Amann

Status: open (until 17 Jun 2026)

Post a comment Subscribe to comment alert

RC1:
'Comment on egusphere-2026-964', Anonymous Referee #1, 04 May 2026 reply
General Comments
The authors have conducted extensive research to understand the role of burial history in self-sealing capacity of several argillaceous host rocks. The authors have focused on hydraulic conductivity as a key characteristic for the sealing integrity of host rocks in the context of high-level nuclear waste disposal in Europe (specifically in Germany, Switzerland, Belgium, France, and Hungary). They have compiled a large dataset including geological, petrophysical, mechanical, and mineralogical data from OPA, COx, To-Do, AMA, BCF, and Boom formations at laboratory and field scales. The manuscript narrates well from Introduction to Methodology, Results, and Discussion; however, I see room for improvement. These mainly involve language enhancement, reference placement, literature improvement, terminology correction, and results-discussion orientation.
Specific Comments
General
The manuscript uses a mix of American and British English (e.g., "behaviour"/"behavior"). Please choose one and maintain consistency throughout. Additionally, proofread the manuscript thoroughly for grammar, structure, and fluency.

The manuscript is filled with abbreviations (e.g., CCC, URL, LSB, OCR). Use abbreviations only if they are well-understood, repeated multiple times, or core to the paper. If abbreviations are used, define each one once in the text and use it thereafter.

Log returns no unit; Log₁₀ K is dimensionless.

Abstract
Line 13: "Hydraulic integrity" is uncommon. "Sealing integrity" is recommended.

Introduction
Lines 35-40, 42-47: These lines require references.

Lines 41-47: Please provide the equation for hydraulic conductivity.

Lines 62-63: Regarding "stress and temperature," specify which stress is being referred to. Pore pressure plays a key role in burial history too as it controls effective stress.

Lines 65-75: The critical depth separating mechanical and chemical compaction (e.g., cementation) is typically 2-3 km. Temperatures greater than 80-100 °C are commonly used for chemical compaction (e.g., see https://doi.org/10.1144/GSL.SP.2006.253.01.19). Please revise or provide a reference if using "> 70 °C."

Lines 108-111: The contents within the brackets are inconsistent. Please adhere to one format, either location or sample ID.

Materials and Methods
Lines 128-130: Boom is missing from the caption.

Lines 167-168: The character "x" has been used instead of a multiplication symbol in values (×). This also applies to Table 1. Ensure consistency throughout.

Line 288: Please clarify the nature of the permeability (hydraulic conductivity) and porosity data at both laboratory and field scales. Are these matrix permeability/porosity, fracture permeability/porosity, or both?

Figure 2: The legend of Figure 2 lists "ToDo," while the text repeats "To-Do." Ensure consistency throughout.

Table 2: Permeameters are typically used with samples as received or dried; "saturated" can be misleading. Researchers commonly employ steady-state or unsteady-state methods to determine permeability under single-phase flow conditions. For samples that do not contain swelling clays, saturation with water and subsequent measurement of permeability using that water might be feasible, as might be the case for sandstones. However, for low- to ultra-low permeability samples (e.g., mudrocks), gas is typically used to determine permeability (slip-corrected permeability). Saturating such samples with water is not recommended because it creates two-phase flow, and the resulting measurement would not represent gas permeability but rather diffusivity, or at best, gas effective permeability. Furthermore, the samples used are commonly core plugs, not entire cores.

Lines 331-345:

Clarify whether "Burial estimates" refers to maximum burial depth or present-day burial depth. Table 4 suggests it is maximum burial depth; better to list both depths in Table 4.

The estimation of maximum burial depth is provided for OPA and AMA but not for the other formations. Please provide these estimates for all formations studied.

Several sentences in this section do not end with a period (.). Please add them.

"T_max" is commonly used, not "T_peak."

Basin modelling, maximum burial temperature, and vitrinite reflectance are commonly employed to estimate maximum burial depth (e.g., see https://doi.org/10.1016/j.energy.2023.129966). It is suggested to highlight the methodologies used in this study and specify which method was employed to determine the maximum burial depth for each formation.

Results
Table 5: Please specify what "mbgs" and "Geom." stand for in the caption.

Figure 3: The in-situ K data were determined under a specific in-situ effective stress. Were the laboratory K data obtained under the same in-situ effective stress? If not, the observed differences in K could potentially be attributed to variations in effective stress.

Lines 395-410:

The Y-axis in Figure 4 denotes "Depth." To avoid confusion between present-day burial depth and maximum burial depth, please be more specific about the Y-axis in Figures 4 and 6, similar to how it is done in Figure 11.

The K medians for lab and field data converge more closely if the filtered depth is increased. Please justify the selection of present-day burial depth > 100 m.

Significant reductions in conductivity or porosity are typically observed within the first 100 m. Excluding this 100-m interval raises concerns, as the sealing integrity of host rocks needs to be proven for the entire depth, even if they have undergone weathering and decompaction at shallower depths. You may consider highlighting this as a limitation of the study.

Figure 11 displays K and porosity versus maximum burial depth. It is suggested to include a plot of porosity versus present-day burial depth (laboratory and in-situ) in Figure 4, similar to the K versus Depth plot, if such data are available. This would help maintain the link between present-day and maximum burial depths.

Figure 5: Replace "log₁₀ Hydraulic Conductivity (m/s)" on the X-axis with "log₁₀ K". It is suggested to use symbols rather than full terminology on the axes of figures throughout the manuscript for clarity.

Lines 477-489: Please clarify what is meant by "effective confining stress." Do you intend to refer to the (modified Terzaghi) effective stress, calculated as , where is the (modified Terzagi) effective stress, is the confining stress, is the Biot coefficient, and is the pore fluid pressure? It is assumed that you mean to discuss hydraulic conductivity (K) behaviour as a function of "effective stress." The subsection uses "effective confining stress," "effective stress," "confining stress," and "stress," while Figure 8 labels the X-axis as "effective confinement stress." This lack of consistency is confusing. Please be more specific in your terminology.

Line 502: It is assumed that "stress conditions" should be "effective stress conditions." The same applies to line 517.

Lines 554-557: "However, if porosity alone would determine hydraulic conductivity, COx would be expected to show higher K values than the OPA and AMA, both of which have porosity values approximately 3–4 % lower than COx. Instead, COx generally exhibits lower hydraulic conductivity." This observation is natural, as porosity alone does not solely control permeability or hydraulic conductivity (k or K). Porosity is a single metric that cannot fully represent the entire pore structure; pore size distribution (PSD) is a more informative parameter. Moreover, fractures, while not significantly contributing to total porosity, largely control permeability. Figure 11 supports this observation. Although you focus on fractures, consider discussing the role of PSD in controlling K in the Discussion section.

Discussion
The section discusses the self-sealing capacity of the host rocks, which is related to their hydraulic conductivity. It indicates that this capacity is controlled by swelling clay minerals and fractures. However, it is unclear what type of hydraulic conductivity data (matrix, fracture, or both) was used in the Materials and Methods section. This ambiguity has not been resolved in the Results section either. Furthermore, the swelling clay mineralogy and fracture properties of the formations (laboratory or in-situ) have not been specified in the Materials and Methods or described in the Results. The discussion appears to be based on metrics that were not previously measured or described. Please clarify.
Conclusions
Line 818: Pore size distribution is also a key characteristic that should be mentioned.

Technical Corrections
Line 12: "to a large extend" is incorrect. Use "to a large extent."

Line 366: "ssamples" is a typo; it should be "samples."

Line 837: “depth” should be replaced with “depths.”

Reply
Citation: https://doi.org/10.5194/egusphere-2026-964-RC1
RC2:
'Comment on egusphere-2026-964', Anonymous Referee #2, 08 May 2026 reply
General Comments
This manuscript presents substantial work in compiling and integrating a comprehensive dataset of hydraulic conductivity (K) measurements derived from both laboratory and in situ tests for six potential host rocks in Europe: COX, OPA, BC, To-Do, AMA, and BDF. The authors aim to identify a systematic links between the maximum burial depth and K by investigating burial history, uplift-related decompaction, cementation, mineral alteration and scale effect on K values of six host rocks.
The text and analysis are generally clear and thorough. The manuscript is supported by an extensive bibliography, comprising 1,250 references, which reflects the considerable effort invested in this work. The study appears to be primarily a synthesis of existing knowledge, and it does not introduce particularly novel findings. Given the distinct characteristics of the six formations considered and their heterogeneity, which make it challenging to derive well-validated general conclusions.

Specific Comments
General:
Figure 12 is a key graph used to support the main conclusions, where “maximum burial depth” is used for the y-axis. It assumes that K measured from the same region/borehole share a single value of maximum burial depth. However, maximum burial depth is a parameter with considerable uncertainty and is often difficult to quantify accurately. Table 4 provides approximate values for each region/borehole, some of which are derived from 1D basin modelling.

one of the key conclusions of the authors is that shallowly buried (<400 m), poorly indurated formations exhibit limited hydraulic variability and scale independence (L24). However, among the six host rocks investigated, only Boom Clay falls within the <400 m maximum burial depth category. Boom Clay is a relatively homogeneous clay formation, characterized by strong self-sealing capacity, relatively uniform hydraulic conductivity (K), and limited scale dependence, which is one of the reasons it has been selected as a potential host rock for HLW disposal.

Given this, it is questionable whether the properties of Boom Clay can be generalized to all shallowly buried clays. At a minimum, additional evidence from other shallowly buried formations would be required to support this conclusion.
data from Lausen is excluded from Figure 12 due to the lack of burial depth information. If the Lausen dataset, characterized by K values spanning approximately 7 orders of magnitude (purple triangles near the surface in figure 4b) , were included in Figure12a, the conclusion that “for depths < 2,000 m, K_{_lab} and K__{in situ} show a close agreement” would no longer be valid.

In Figure 12a , five boreholes/regions are included for depths greater than 2000 m, namely To-Do, BO3-BO5, and BCF. The To-Do formation includes 8 laboratory and 6 in situ measurements (Table 5). BDF has only 3 laboratory results and no specific in situ data. BO3, BO4, and BO5 include only laboratory tests.

In total, only 17 K measurements (8+6+3) for K_{_lab} and K_{_in situ} coming from To-Do and BCF. Such a limited samples raises the question of whether it is sufficient to support the conclusion that K_{_lab} and K_{_in situ} diverge for max burial depths greater than 2000 m.
The points raised above collectively challenge the robustness of the manuscript’s final conclusion (Line20) that maximum burial depth and the associated stress and temperature conditions exert a first-order control on matrix-scale hydraulic conductivity (K).

This conclusion appears oversimplified. Hydraulic conductivity (K) is influenced by a range of factors. History of burial, stress and T are only part of the controlling processes. Mode first-hand data and rigorous analysis are needed for characterizing them as “first-order” controls.

Abstract
L14: (K<10^-10 m/s):
Please provide an appropriate reference to support this value. In the context of safety assessments for HLW, values higher than 10⁻¹¹ m/s are generally considered difficult to justify.
L15: strong aquitard characteristics:
Please specify explicitly what characteristics define a strong aquitard in this context. For a rock formation to be considered a promising host for HLW, not only K, many other properties must be taken into account.

Introduction

L53-54: the potentially strong spatial anisotropy of hydraulic properties, which can vary across different scales, partly by several orders of magnitude…..:
The discussion here relates to the difficulty of measuring hydraulic conductivity (K) at larger scales due to variability. Anisotropy itself does not necessarily make hydraulic conductivity (K) measurements more challenging. The term heterogeneity would be more appropriate than anisotropy in this context.
L55: anisotropy arises from …orientation of the bedding… , …fractures, and heterogeneities in sedimentological:
Please clarify between anisotropy and heterogeneity. Heterogeneity refers to spatial variability of properties and is a broader concept, whereas anisotropy specifically describes directional dependence of those properties (i.e., variation with the direction of measurement).
Paragraph A ( Ln78-89 )and paragraph ( Ln 90-98):
There is some overlap between these two paragraphs. Paragraph A introduces the idea that the rock may undergo either ductile or brittle deformation (L84) and provides explanations for both mechanisms. Paragraph B restates the mechanism for fracture formation in the context of uplift for host rocks.
To improve clarity and avoid redundancy, consider consolidating these sections and presenting a unified explanation of fracture formation mechanisms.

Studied Formations and Sites

L38: the formation isconsidered homogeneous
Typo error
L145: using pulse or constant-head permeameter tests
For Boom Clay, it is constant-head permeability test that is used to derive K value in the lab. Pulse is for the subsequent diffusion test.
L157: pronounced overconsolidation
Please add a range of over-consolidation ratio (OCR) for AMA, To-Do, COX, and BCF.
L160: variable depths (formation tops between 2 and 74 mbgs)
Table 1:
“bgs” may not be clear for all readers. Please clarify it.

Methods

Table 2
Aertsens et al., (2004, 2008, 2013); reported Lab results for permeameter tests, not oedometer tests.
In permeameter test, it is constant pressure gradient that is used for K measurement.

Results

L 398: in situ data shows a pronounced decrease in hydraulic conductivity within the upper 100 m of nine orders of magnitude:
The wording is unclear and misleading.
L399 and Figure 4: laboratory data from the AMA and OPA lie in a distinctively narrower range between
On Figure 4a, data for AMA and Lausen are represented by two dots. Please draw all K_lab points available on the figure. Also, please clarify the sample size of data from Lausen borehole.
L 400: Tests on AMA samples were conducted at effective stresses between 1 and 20 MPa, whereas OPA samples from Lausen were tested under effective stress conditions ranging from 0 to 50 MPa.
Please state explicitly the implications of stress on the differences in measured hydraulic conductivity (K) between in situ and laboratory tests.
L405: The less consolidated Boom Clay
The term “less consolidated” is not appropriate in a geotechnical context. Given that the overconsolidation ratio (OCR) of Boom Clay is 2.4 (Ln141), it should be described as a normally to slightly overconsolidated clay.
L428: OPA data from shallow boreholes in the Swabian Alb (Germany) do not include continuous reports of interval lengths
What does it mean continuous reports? Are the data from boreholes in the Swabian Alb excluded from Figure5?
Figure 5: How many samples from Lausen because only two symbols are visible for Lausen. It would be helpful to add the sample size for each borehole.

L 431: two clear interval length-related trends appear in BO2.0 and BO4.0
For BO2.0, there is only one data point corresponding to a larger packer interval (y=47 m), while five scatter data points are derived for y=10 m. The two data points with hydraulic conductivity values on the order of 10⁻⁹ and 10⁻⁶ m/s, are omitted from the trend without justification, which undermines the reliability of the observed trend. Additional data obtained from longer packer intervals are needed to robustly support this interpretation.
For BO4.2, getting a trend based on 3 points is not convincing.
L 432: in the deeper borehole sections between 45 – 95 m
The word deeper is misleading, as this refers to the length of the packer interval rather than depth. Please rephrase to avoid confusion.
Fig.6 & line 461: Fig. 6 is derived from Δ𝑙𝑜𝑔10(𝐾) = 𝑙𝑜𝑔10𝐾_𝑖𝑛_{𝑠𝑖𝑡𝑢}- 𝑙𝑜𝑔10𝐾_𝑙𝑎_b(line 328) ), with data grouped into 10 m depth bins (L326).
It appears that only data points for which both laboratory and in situ measurements are available within the same depth bin have been selected. I am wondering how many samples are included in the two bins for the To-Do formation. Please report or each formation, the sample size in each depth bin. Also please clarify the data selection in the text. This information is important for assessing the robustness of the statistical analysis.
L446: Lausen borehole at depths >30 m
Where can the reader find the depth range for samples taken from Lausen borehole?
L453: the interquartile ratios
Where is the value for the interquartile ratios?
L419: interquartile range
L435: dashed vertical lines are IQR
Please clarify what is IQR and how to calculate interquartile ratios.
Section 4.3.2
L465: Laboratory measurements for which the flow direction with respect to bedding orientation was reported (n=367) were separated into tests performed parallel (KP; n=135) and perpendicular (KS; n=232)
Please state explicitly the sample size for each clay formation within each category (parallel and perpendicular).
Each category must have different numbers of samples for each type of clay. According to Table 5, there are 269 laboratory data points for Boom Clay (BC). If BC constitutes the largest dataset in each category, then the calculated geometric mean—where each sample contributes equally—is likely to be dominated by BC.
Figure 7:
4.3.3 Stress dependency of laboratory data Figure 7 presents the anisotropy ratio for each formation (COX:Kp/Ks= 1/0.8 = 1.25; OPA: 1/0.9 = 1.11; AMA: 1/0.6 = 1.67; BC: 1/0.4 = 2.5), the overall geometric mean anisotropy ratio of approximately 3 seems to be dominated by BC.
4.3.3 Stress dependency of laboratory data
The analysis is very vague.
OPA: no trend between K and σ’ has been identified because only a few data points originate from the same sampling location (lines 481).
The data supporting the statement that “additional results from the Lausen borehole… show that hydraulic conductivity decreases systematically with increasing effective stress” are not presented in the manuscript.
AMA: The statement that the decreasing trend “vanishes at confining stresses above ~5 MPa (10–20 MPa range)” (L485) is not clearly supported. From Figure 8a, the decreasing trend appears to remain valid at least within the 0–10 MPa range,
L506: The compiled data shows a median KP/KS ratio of approximately 3, in good agreement with this range, indicating that anisotropy introduces only a minor influence on the overall comparability of hydraulic conductivity values.
This sentence is hard to understand. Anisotropy for the four formations, varying between 1.11-2.5, which is not far from the range of 1.5-5 (line 506). The poenitentia influences of this level of anisotropy on the “overall comparability of hydraulic conductivity values” needs to be clarified.
Regarding KP/KS ratio of approximately 3, please see comments for line 465.

L508: Hydraulic conductivity from in situ tests, such as packer experiments, mostly conducted in vertical boreholes, predominantly reflect KP(perpendicular)
In an anisotropic formation, the measured K likely reflects a combination of effect of both Ks and Kp.

L552: Across the investigated formations, mean porosity values show a positive correlation with laboratory-measured hydraulic conductivity (Fig. 11a),
The four data points (excluding the Boom Clay point highlighted in yellow) are scattered. As a result, the inferred positive or negative correlations between porosity or bulk density and hydraulic conductivity (K) are not convincingly supported by the data.

Section 4.6 and Fig 12:
Given that Figure 12 is a key figure supporting the analysis and main conclusions, please include a detailed table reporting sample sizes used for each borehole, and state explicitly how the values used for graphing Figure 12 were derived. In the current manuscript, the data are scattered in different places, which makes it difficult for readers to locate them efficiently.
Boom Clay(BC): The maximum burial depth of BC shown in Figure 12 is 200 m, and the porosity is 0.4. However, the K data of Boom Clay, which composed data from 5 boreholes and cover a (present-day) depth range between 59-366 m (line 367 and Table 4). Over-consolidated ratio (OCR) for Boom Clay is 2.4 (Line 241). Based on these information, It is not clear what the procedure used by the authors to process raw data to get a maximum burial depth of 200 m, and a porosity of 0.4 for Boom Clay.
COX: The maximum burial depth is listed in Table 4 as 850 m, with Blaise et al. (2014) cited as the reference. However, the last paragraph on p. 83 of Blaise et al. (2014) states that “the maximum burial depth of the Callovian–Oxfordian claystones in the Andra URL area was estimated to be around 500–700 m ……” Please clarify the origin of the value of 850 m reported in Table 4, or how it was derived.
The OPA data are derived from multiple regions and boreholes, including:
Mont Terri(MT)

Rinken (Jira Ost)

Weiach (Nördlich Lägern)

Benken(Zürich Nordost)

Additional boreholes providing in situ data (as shown in Figure 5)

Shallow OPA data from Swabian Alb(excluded from Figure 12 due to lack of burial depth information)

Shallow OPA data from Lausen (excluded from Figure 12 due to lack of burial depth information)

Please provide a summarized table for each borehole/region, clearly reporting the sample sizes for K_{_lab} and K_{_in situ} for each location.
Additional comments are provided below for L588.

L569: depth between 1400 m and 3300 m:
The maximum burial depth in Table 3 for AMA is 1300-3300m. Please make the values consistent.
L571: The BCF …with burial depths of up to 4,500 m.
The max burial depth of BCF in Figure 12 is 4000m. Please clarify where 4000 comes from.

L588: For formations that experienced maximum burial depths < 2,000 m, laboratory-derived and in situ hydraulic conductivity show a close agreement, indicating consistent matrix and bulk hydraulic properties at depth. In contrast, a divergence between laboratory- and field-scale hydraulic conductivity values emerges in formations with burial depths ≥2,000 m, notably in the To-Do and BCF.
For data with maximum burial depths < 2000 m in Figure 12a, agreement between K_{_lab} and K__{in situ} is indeed observed for Boom Clay and MT, whereas the remaining regions/boreholes (e.g. JO, NL and COx) exhibit obvious discrepancies between K_{_lab} and K__{in situ}.
Moreover, data from Lausen is excluded from Figure 12 due to the lack of burial depth information. If the Lausen dataset, characterized by K values spanning approximately 7 orders of magnitude (purple triangles near the surface in figure 4b) , were included in Figure12a, the conclusion that “for depths < 2,000 m, K_{_lab} and K__{in situ} show a close agreement” would no longer be valid.
In Figure 12a , five boreholes/regions are included for depths greater than 2000 m, namely To-Do, BO3-BO5, and BCF. The To-Do formation includes 8 laboratory and 6 in situ measurements (Table 5). BDF has only 3 laboratory results and no specific in situ data. BO3, BO4, and BO5 include only laboratory tests.
Please provide a summarized table for each borehole, clearly reporting the sample sizes for each borehole regarding K_{_lab} and K_{_in situ} .
In total, only 17 K measurements (8+6+3) for K_{_lab} and K_{_in situ} coming from To-Do and BCF. Such a limited samples raises the question of whether it is sufficient to support the conclusion that K_{_lab} and K_{_in situ} diverge for max burial depths greater than 2000 m.

L608: The compiled porosity data clearly follows the expected burial trend, with progressive pore space reduction accompanying mechanical compaction and cementation. The hydraulic conductivity shows a broadly similar pattern, with a marked decrease between the Boom Clay and COx that faced maximum burial of around 850 m depth.
The conclusion of a “similar pattern” is based on only two data points for BC and COX (see Figure 12a), which is insufficient to support such a claim. Additional data points are needed to substantiate this observation.
L611: For mechanically compacted and lithified formations (e.g., COx, OPA), the vast majority of matrix hydraulic conductivity determined from laboratory tests fall within a range of 10^-14 to 10^-12 m/s, independent of site, formation, or test conditions. This range, previously defined as the natural variability envelope, likewise characterizes the hydraulic behavior at the rock mass scale in the absence of near615 surface perturbations such as weathering or decompaction.
Among the six types of host rock considered in this manuscript, please specify which formations are classified as “mechanically compacted and lithified formations” and definition of “mechanically compacted and lithified formations”.
Do the authors intent to classify the data in Figure 12a into 3 categories: 1) BC and Cox for depth <850m, 2) OPA & BO1 for depth between 850-2000m, and 3) AMA, BCF and TO-Do for >2000m? if so, this categorization should be explicitly stated in the text. Please clearly define the depth ranges for each category and specify which boreholes/regions are included in each group.
This manuscript distinguish between small scale (matrix) and large scale (rock-mass) (line 32) hydraulic properties. Should the lab tests referred to here represent K at the small scale(matrix)?
For the category with max burial depths of 850-2000m, please discuss the possible causes of the differences between K_{_lab} and K__{in situ} for Cox, JO and NL.
Line 616-626:
The conclusion that K__{in situ} >> K_{_lab} for the category with maximum burial depths >2000 m is based on a limited sample size from the To-Do and BCF formations, and therefore requires additional data to be adequately supported.

5.1 self-sealing characteristics of investigated formations
L661: In the moderately indurated formations (Clay, COx, OPA),
Please clarify what is meant by “Clay” , Boom Clay?
L664: indicating efficient sealing of natural or excavation-induced discontinuities below approximately 250 m depth..
Please clarify what is meant by “below approximately 250 m depth” ,
L673 Toarcian-Domerian argillites exhibit generally overlapping in situ and laboratory hydraulic conductivity,
For the To-do data in Figure 12 a : K_{_lab} and K_{_in situ} differ by at least one order of magnitude. Please clarify the definition of “overlapping” in this context. It should also be noted that the To-Do formation includes only 8 laboratory and 6 in situ measurements (Table 5), which may limit the robustness of any inferred overlap between the two datasets.
L684: In the case of the BCF, the highly indurated nature and general depletion of swellable clay minerals do not favour rapid self-sealing.
Please clarify the fraction of swellable clay minerals for each formation considered in this manuscript, since it is the controlling factor of self-sealing property.
Line 723 suggests that smectite may be the primary swellable clay mineral. In this context, it would be more informative to report the smectite fraction for each formation.

L685: In the case of the BCF, the highly indurated nature and general depletion of swellable clay minerals do not favour rapid self-sealing.
Please clarify what is meant by “below approximately 250 m depth” ,

5.2 The role of maximum burial depths

L700: An initial porosity loss exceeding 20 % is observed between the Boom Clay and the COx, corresponding to a differential burial of approximately 600 m.
For the category with max burial depth <600m, the porosity-maxi burial depth relationship is derived from BC and OPx,. Additional data points within this depth range are required to robustly support the proposed relationship.

L706: Neuzil, (1994) demonstrated the existence of a log-linear relationship between porosity and hydraulic conductivity across a broad range of argillaceous formations, a trend also reflected in the data compiled for this study.
Please add a porosity-K graph based on figure 12. Please include also the log-linear relationship proposed by Neuzil, (1994) in the figure for comparison.

L708: hydraulic conductivity would be expected to decrease further with continued pore-space reduction beyond approximately 850 m burial. This is generally 710 consistent across the investigated formations and observational scales (laboratory versus in situ) up to maximum burial depths of around 2,000 m
An explicit porosity-K curve based on figure 12 is necessary to prove this.
Figure 11 present a porosity-K graph for BC, Cox, and OPA, all with maximum burial depths < 2,000 m. The OPA data appears to deviate from the overall trend of decreasing K with decreasing porosity . This variability should be discussed in more detail, as it may indicate that the formation-specific effects play a more significant role than a consistent global trend.

L715: At maximum burial depths exceeding 2,000 m, a divergence emerges between laboratory-derived matrix hydraulic conductivity ….
This paragraph summarizes plausible explanations for the scale effect on hydraulic conductivity (K) in formations with maximum burial depths exceeding 2000 m. However, the discussion is primarily based on 17 hydraulic conductivity measurements from To-Do and BCF formations. Additional datasets are needed to robustly support this observation.

5.3 The role of present-day depth
L752: Mazurek et al. (2023) demonstrated that, aside from hydraulically active fractures, diffusion remains the dominant transport mechanism even in the weathered Opalinus Clay encountered in the Lausen borehole.
This sentence is oversimplified and it needs some context. It is suggested to improve it as “Aside from hydraulically active fractures, diffusive transport still remains the dominant transport mechanism governing solute exchange between fractures and matrix, even in the weathered Opalinus Clay encountered in the Lausen borehole.”
L758: However, in situ data from AMA borehole BO4 reveal a depth-dependent trend in hydraulic conductivity comparable to that observed in the OPA.
Where can the reader find the depth dependence of the in situ data for BO4? Please indicate clearly in the manuscript or figures where this relationship is presented.

Reply
Citation: https://doi.org/10.5194/egusphere-2026-964-RC2

Raphael Burchartz, Brian Mutuma Mbui, Peter Achtziger-Zupančič, Garri Gaus, Timo Seemann, Lisa Winhausen, Yvonne Spychala, Mohammadreza Jalali, Ralf Littke, and Florian Amann

Viewed

Total article views: 866 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
556	261	49	866	47	72

HTML: 556
PDF: 261
XML: 49
Total: 866
BibTeX: 47
EndNote: 72

Views and downloads (calculated since 17 Mar 2026)

Month	HTML	PDF	XML	Total
Mar 2026	440	189	43	672
Apr 2026	87	60	5	152
May 2026	29	12	1	42

Cumulative views and downloads (calculated since 17 Mar 2026)

Month	HTML	PDF	XML	Total
Mar 2026	440	189	43	672
Apr 2026	87	60	5	152
May 2026	29	12	1	42

Viewed (geographical distribution)

Total article views: 862 (including HTML, PDF, and XML) Thereof 862 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 19 May 2026

Short summary

We compile 782 hydraulic conductivity measurements from six European argillaceous formations relevant for high-level radioactive waste disposal. We show that maximum burial depth controls matrix-scale conductivity, while present-day depth, decompaction, fracturing, and self-sealing govern rock-mass behaviour, defining three depth-related evolutionary trends.


Total:	0
HTML:	0
PDF:	0
XML:	0