Modeling seismic site response to improve lacustrine paleoseismic records

Yang, Huitong; Wils, Katleen; Molenaar, Ariana; Moernaut, Jasper; Wu, Lei; Urrutia, Roberto; Pino, Mario; Van Daele, Maarten

doi:10.5194/egusphere-2026-2493

Preprints

https://doi.org/10.5194/egusphere-2026-2493

Preprints

01 Jun 2026

| 01 Jun 2026

Status: this preprint is open for discussion and under review for Natural Hazards and Earth System Sciences (NHESS).

Modeling seismic site response to improve lacustrine paleoseismic records

Huitong Yang, Katleen Wils, Ariana Molenaar, Jasper Moernaut, Lei Wu, Roberto Urrutia, Mario Pino, and Maarten Van Daele

Abstract. Linking earthquake-triggered sedimentary imprints in lakes to ground motion parameters is essential for quantitative paleoseismology. However, current approaches rely on empirical ground motion prediction equations (GMPEs) and use a single time-averaged shear-wave velocity (V_s30) as a simplified site response proxy for a whole lake. We established a 3D shear velocity model to compute site-specific GMPE predictions and applied 2D numerical site response simulations for Lake Riñihue, Chile, to evaluate local ground motions for the 1960 M_w 9.5 Valdivia and 2010 M_w 8.8 Maule earthquakes. Even with site-specific V_s30 inputs, 2D simulations predict peak ground accelerations (PGA) and peak ground velocities (PGV) that exceed GMPE estimates by more than a factor of two. By stepwise modification of model properties and testing additional flat-layered reference models, we demonstrate that impedance contrasts between stratigraphic units influence overall ground motion amplification, whereas multi-scale basin geometry controls its spatial distribution, generating localized ground motion spikes. Earthquake shaking in lakes can produce surficial sediment remobilization (SSR) and soft-sediment deformation structures (SSDS) in-situ. Comparison of site-specific ground motions with sedimentary records from lake cores shows that SSR and SSDS are independent processes controlled by different ground motion components. SSR depth is primarily controlled by slope angle and PGV, with its patchy spatial occurrence reflecting frequency-dependent site response, whereas SSDS is controlled by PGA, with different thresholds for progressively increasing deformation types, while also predicts deformation thickness. Our findings highlight that site-specific ground motion reconstruction is essential to accurately link ground motion parameters to lacustrine sedimentary imprints.

Received: 30 Apr 2026 – Discussion started: 01 Jun 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

Preprint (PDF, 28353 KB)

Supplement (38647 KB)

Download & links

Preprint (28353 KB)
Metadata XML
Supplement (38647 KB)
BibTeX
EndNote

Huitong Yang, Katleen Wils, Ariana Molenaar, Jasper Moernaut, Lei Wu, Roberto Urrutia, Mario Pino, and Maarten Van Daele

Status: open (until 13 Jul 2026)

Post a comment Subscribe to comment alert

RC1:
'Comment on egusphere-2026-2493', Anonymous Referee #1, 08 Jun 2026 reply
The manuscript “Modeling seismic site response to improve lacustrine paleoseismic records” by Yang et al. investigates the link between the seismic site response, ground motion parameters, and earthquake-triggered sedimentary imprints in lakes. The idea of using lake sediments as “lacustrine paleoseismographs” is not new but the authors approach the topic in a more advanced way by considering stratigraphic parameters and basin structure, i.e., spatial variations, influencing ground motion in the lake and comparing empirical GMPEs with physics-based 2D numerical simulations. Additionally, the authors evaluate how the local ground motions relate to surficial remobilization and SSDS characteristics.
The manuscript is well-written and contributes to community’s understanding of earthquake-triggered subaqueous failures and use of event deposits for paleoseismology. Also, it advances already available research works by incorporating numerical simulations and comparing different modelling approaches.
There are several points that need more explanations and discussion that might improve the significance of the work. Therefore, I suggest a moderate revision of the manuscript prior to its acceptance for publication.

Major comments:
In “Settings”, elaborate on the age of sediments: you use the names of Holocene Lacustrine sediments and glaciolacustrine deposits. But glaciolacustrine deposits are also of Holocene age, right? In a few sentences, address the age of the sediments and then tell which nomenclature you are using through the whole manuscript. Also, at some point you refer to “Holocene sub-basins on the slopes and glaciolacustrine sub-basins” – how were these assigned?

Have you validated sediment properties (vp, vs, density) from the literature by measurements on the samples? What are the uncertainties and did you do any sensitivity analysis of the influence of uncertainties on the modelling results? If not, how would you expect them to influence the result and how big uncertainties do you expect? Please add this (and how one could address these uncertainties) in the Discussion of other parts of the text. Did you also need to include some damping/quality factors or other sediment parameters into the model?

You used two Vp settings (L.146-151), what was their impact and which one do you trust more? How was it included in different models (Table 2)? Address the difference in the outcome of 2D numerical simulations using both models. Why are there no alternative models for Vs, for example?

Add to the discussion or where suitable, how the properties of investigated lake sediments are expected to influence its seismic response. Does it correspond to what you obtained from the modelling?

182: address how nonlinear site amplifications can influence observed seismic impact in comparison to linear case. Are there magnitude/distance thresholds at which you expect switch from linear case to non-linear? Or how is this distinction used?

What are the limitations of 2D modeling you used (e.g., in comparison to 3D)? Address in Discussion.

Modelling: how did you chose the “pressure-independent multi-yield plasticity (PIMY) model”? Did you consider other constitutive models?

261-262: “recording sites with Vs30 > 700 m s-1 to represent basement conditions” – why did you choose vs30>700 m/s as a criterion for basement? Provide a reference for such a decision. Why not using the records from the sites/depths with Vs>800 m/s (engineering bedrock)?

Section 3.4, L. 279-280: could you show seismic profile/core photos illustrating what you mean by “surficial sediment remobilization (SSR) and soft-sediment deformation structures (SSDS)” and discuss how you identify them?

You use Kriging to interpolate sediment properties. Which Kriging algorithm do you use specifically? Add name and reference(s).

Sediment cores you mention in the study (in particular, in Fig. 4 and section 4.4.1) – how deep are they? Is their full length shown in Fig. S4? Could they be used to validate the properties of your model and properties interpolation? Could you provide a summary of core depths and events identified in them in the manuscript? Either as a table or a map where maybe depth can be shown with color and event type with symbol shape.

Why do you have higher PGA at sites with thinner and stiffer sediments? This needs an explanation. Is this due to damping of energy at high frequencies by soft sediments? Can this also be the effect of non-linearity?

372 “the highest ground motions among all BMs are still achieved by BM01” – address why this is the case.

You mentioned several times contrasts in sediment properties. Address their influence on the amplification in the discussion. Section 4.3.2: BM05 shows quite a flat amplification. Does it mean that the contract between Unit 0 and 1 has the strongest impact on the amplification? BM03 includes the highest acoustic impedance contrast among all the models. Interestingly, the amplification is not that high but you see localized peaks on the slope. Do you have an idea why?

How does the observed amplification in the lake compare with the amplifications expected in the literature?

Section 4.4.2 – usually for the regressions, slope angle is used on log scale (i.e., use log(slope) for the regressions) since the slope normally has lognormal distribution. Please show the histogram of slope angle and if it is lognormal, try using the log(slope) for the regressions. Could you also show correlation coefficients between different variables used for regressions? Also, with log(slope). Also, check the distribution of PGV, PGA, SSR depth, and SSDS.

Have you compared your event-triggering PGA and PGV values from GMPEs and 2D simulations with the ones from literature? It would be interesting to see the discussion of such a comparison.

Discuss the influence of non-linear site response that can be triggered by strong earthquakes like the ones used in the manuscript and possible effects of basin shape on your modelling results (e.g., 2D effects) – how can they alter the ground motion amplitude and frequency content?

Based on your results and “sensitivity analysis” with different “BM” models, can you conclude which components of the model are the most crucial for reliable result – precise velocities? Good stratigraphic/morphological model? Or what? Where should the focus be in case of limited resources for data mining and what can be sacrificed?

In Figs. S5-S9, there are fluctuations of amplification on the sides. Is it an artifact due to boundary conditions or how would you explain this?

Regressions: please elaborate how you chose the form of the equations and dimensions of predictors and predicted value.

Minor comments:
109-110 “The studied subbasin has no direct river inflows, and therefore sedimentary event deposits can be attributed to seismic triggering.” Add references confirming your statement (that it is not a spontaneous failure or human-triggered).

119-120 what is the penetration depth of sparker data?

1b: typo in the legend – should be sparker instead of “spaker”.

1b, Fig. 5: typo in the legend – should be bathymetry instead of “bathmetry”.

1b: label “core locations” is too high with respect to the pink circle.

1b: it is hard to differentiate the lines showing sparker, pinger, and bathymetry. In B&W, it would be impossible. Consider using lines with different styles (continuous, dashed, …).

1b: there is no red line marking the location of the seismic profile shown in Figure 2, as suggested by figure caption.

104 – should there be a reference to Fig. 1a instead of 1b?

108 – typo, “Holcene”

Homogenize the use of comma before “and” in the lists.

354 – space missing in “from2D”. Same for the captions of Figures S5, S6, S7, S8.

358 – “Dashed lines indicate GMPE predictions.” – redundancy.

8a: Why did you select such steps for the slope angle legend?

8c-e: please add depth axis for the cores.

Reply
Citation: https://doi.org/10.5194/egusphere-2026-2493-RC1
RC2: 'Comment on egusphere-2026-2493', Anonymous Referee #2, 08 Jun 2026 reply

The topic is original and addresses an important gap between paleoseismology and ground-motion modeling. The use of two well-known major earthquakes, the 1960 Mw 9.5 Valdivia and the 2010 Mw 8.8 Maule events, strengthens the study and provides valuable benchmark cases. The manuscript effectively demonstrates that Vs30 alone may be insufficient to explain deformation patterns and that basin geometry and impedance contrast can play a significant role. Similar observations have also been emphasized in other studies, for example by Pamuk et al. (2026).
The interpretation that SSR is more strongly related to PGV and slope, whereas SSDS is more closely associated with PGA, is scientifically meaningful and contributes to our understanding of earthquake-induced soft-sediment deformation. However, it requires following revisions.
- The velocity model for Lake Riñihue relies largely on published studies and analog data from similar lacustrine environments. This limitation reduces the robustness of the results, and the associated uncertainties should be quantified and discussed more explicitly.
- The finding that the 2D simulations produce PGA and PGV values up to twice those predicted by the GMPEs is important. However, since these results are not validated using local strong-motion recordings or independent geophysical observations, the conclusions should be presented more cautiously.
- The 1960 earthquake lies outside the calibration range of the GMPEs. Although the authors acknowledge this limitation, the implications of this extrapolation and its potential influence on the results should be discussed more thoroughly.
- In particular, the number of sediment cores used for SSDS analysis is relatively small. Consequently, the proposed PGA threshold values should be presented as preliminary findings rather than definitive conclusions.
- Several grammatical errors and awkward sentence constructions remain in the manuscript. Examples include phrases such as “SSR initiate” and “while also predicts deformation thickness,” which should be revised for clarity and correctness.
The manuscript has clear scientific merit and addresses an important research question. However, the uncertainties associated with the velocity models, the lack of validation for the numerical simulations, and the limitations of the statistical analyses need to be discussed more comprehensively before the manuscript can be considered for acceptance. It needs MAJOR REVISION.

Reply

Citation: https://doi.org/10.5194/egusphere-2026-2493-RC2

Huitong Yang, Katleen Wils, Ariana Molenaar, Jasper Moernaut, Lei Wu, Roberto Urrutia, Mario Pino, and Maarten Van Daele

Supplement

https://doi.org/10.5194/egusphere-2026-2493-supplement

Huitong Yang, Katleen Wils, Ariana Molenaar, Jasper Moernaut, Lei Wu, Roberto Urrutia, Mario Pino, and Maarten Van Daele

Viewed

Total article views: 68 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
53	11	4	68	13	6	4

HTML: 53
PDF: 11
XML: 4
Total: 68
Supplement: 13
BibTeX: 6
EndNote: 4

Views and downloads (calculated since 01 Jun 2026)

Month	HTML	PDF	XML	Total
Jun 2026	53	11	4	68

Cumulative views and downloads (calculated since 01 Jun 2026)

Month	HTML	PDF	XML	Total
Jun 2026	53	11	4	68

Viewed (geographical distribution)

Total article views: 60 (including HTML, PDF, and XML) Thereof 60 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 20 Jun 2026

Short summary

Lake sediments preserve traces of past earthquakes and can help reveal how strong the shaking was. We studied Lake Riñihue, Chile, affected by earthquakes in 1960 and 2010, by modelling its basin structure and simulating its response to earthquake shaking. Lake floor shaking can be about twice as strong as empirical estimations, because sediment layers and basin geometry amplify shaking unevenly. Our results further show that different sediment imprints record different shaking components.


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