the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 22 Feb 2024
Geomorphic indices for unveiling fault segmentation and tectono-geomorphic evolution with insights into the impact of inherited topography, Ulsan Fault Zone, Korea
Abstract. Quantifying present topography can provide insights into landscape evolution and its controls, as the present topography is a cumulative expression of the types, distributions, and intensities of past and present processes. The Ulsan Fault Zone (UFZ) is an active fault zone on the southeastern Korean Peninsula that has been reactivated as a reverse fault around 5 Ma. This NNW–SSE-trending fault zone exhibits a predominantly reverse sense of movement today and dips towards the east. This study investigates the history of tectonic activity along the UFZ and the landscape evolution of the hanging wall side of the UFZ, focusing on neotectonic perturbations using 10Be-derived catchment-wide denudation rate and bedrock incision rates, geomorphic indices, and a landscape evolution model. We evaluated the spatial variation in the relative tectonic intensity from the variation in geomorphic indices along the UFZ. Five geological segments were identified along the fault based on the relative tectonic intensity and fault geometry. We then simulated four cases of landscape evolution using modelling to investigate the geomorphic processes and topographic changes in the study area in response to fault slip. The model results reveal that the geomorphic processes and the patterns of geomorphic indices (e.g., χ anomalies) depend on the inherited topography (i.e., the topography that existed prior to reverse faulting on the UFZ). On the basis of this important finding, we interpret the tectono-geomorphic history of the study area as follows: (1) the northern part of the UFZ has been in a transient state and is in topographic and geometric disequilibrium, as this part underwent asymmetric uplift (westward tilting) prior to reverse faulting on the UFZ around 5 Ma; and (2) its southern part was negligibly influenced by the asymmetric uplift before reverse faulting. Our study demonstrates geomorphic indices as reliable criteria for dividing faults into segments and, together with landscape evolution modelling, to investigate the influence of inherited topography on present topography and to help determine tectono-geomorphic histories.
- Preprint
(5151 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-198', Anonymous Referee #1, 19 Mar 2024
Review of “Geomorphic indices for unveiling fault segmentation and tectono-geomorphic evolution with insights into the impact of inherited topography, Ulsan Fault Zone, Korea”
This is a very interesting paper which is overall well written and clear. Such study coupling geomorphic analyses with neotectonic activities is very enlightening to constrain the geomorphic evolution along a regional major fault belt. In general, the dataset presented is credible and would attract the attention of researchers who work on tectonic and geomorphic evolution of southeastern margin of Korea, as limited data are available from the study area. However, as I will discuss in the following, some middle-minor revisions would be necessary before this manuscript becomes a suitable contribution to the journal.
Major comments:
(1) For geomorphic modelling of cases B1 and B2, the uplift rates of eastern end were set at 18 mm/kyr and 42 mm/kyr, respectively. The uplift rate of 18 mm/kyr for the northern part of the block was calculated based on a relationship between the incision rate and the distance. The authors should give more explanation for its validity, because such an uplift rate is smaller than the CARD value. Similarly, the uplift rate of 42 mm/kyr was obtained based on a relationship between the average CARD and the distance in the southern part.
(2) The UFZ has been divided into five segments based on geomorphology analyses alone. I understand how difficult it will be to obtain some data in an urbanized area. However, it will be more convincing if the authors can provide some other data, for example, the GPS slipping rates, stress accumulation, InSAR deformations…
(3) Base on the modelling results, segment 1 was considered to migrate westward, while segments 2-5 has migrated eastward. However, such a discrepancy was not explained in detail.
Minor comments:
(1) The geomorphic indices should be italic.
(2) For Figures 1a and 1b, I suggest to add the movement properties of the major faults (strike, normal, or thrust) if possible. Can the active faults and ancient faults be marked by different colors (Red and Black) in Figure 1b? I suggest to add the names beside the major fault, e.g. Ulsan Fault. I also have a question. There are three moderate earthquakes shown in the Figure 1a, but why most of them do not occur along the major fault belts?
(3) I suggest to add the methodology description of the students t-test.
(4) The channel incision rate was calculated based on cosmogenic nuclide. Thus, I suggest to add the outcropping and sampling description. What is the kind of the rock? What is the thickness of the sample?
(5) I suggest to add the Ulsan Fault in Figure 3a.
(6) Channels 5b and 5c should be clearly shown on 5a.
(7) Figure 9 was started to cite in chapter Discussions, behind the Figure 10.
(8) Figure 12c should be clearly shown on Figure 2a.
(9) The chapter Conclusions is too much lengthy. In fact, some of the content are not the conclusions.
-
RC2: 'Comment on egusphere-2024-198', Anonymous Referee #2, 22 Mar 2024
General comments
Lee et al. present an intriguing study that combines quantitative topographic analysis, landscape evolution modelling, and cosmogenic nuclide analysis to explore the dynamics of landscape evolution in a reactivated reverse fault zone on the southeastern Korean Peninsula. Their empirical data, including new catchment-averaged denudation rates derived from cosmogenic 10Be in detrital sediments and surface exposure ages from bedrock strath terraces, represent significant additions to the field of geomorphology. Despite these strengths, the manuscript requires substantial revision to meet the publication standards of Earth Surface Dynamics.
My primary concerns are as follows:
- The landscape evolution model's setup and parameterisation, particularly regarding stream-power erosion, feel somewhat like an arbitrary choice of conditions. The manuscript would benefit from additional sensitivity analyses, such as exploring a range of values for stream-power parameters, to document how different parameterisations might affect the results.
- The methodological framework for the quantitative topographic analysis is unclear and needs a comprehensive overhaul for clarity and reproducibility. This includes adding a more detailed description of what was done (e.g., knickpoint extraction), thus ensuring reproducibility, as well as modifying some of what was done. For instance, there is no need to quantify four topographic metrics with the same function (i.e., to document cross-divide steepness asymmetry) under the umbrella of 'Gilbert Metrics'.
- The manuscript's text, especially within the Methods, Results, and Discussion sections, requires extensive revision for clarity and detail. For example, the topographic analysis results should be quantitatively detailed to better support the study's conclusions.
- The paper should address how lithological variations might influence the cosmogenic nuclide results. This addition is crucial for interpreting the data accurately.
- The calculation of the integral metric χ should be revised to accurately reflect non-uniform spatial variations in background rock uplift, a critical factor for interpreting spatial patterns in χ in this context.
Detailed line-by-line comments will follow, aimed at addressing these and other specific points of concern.
Specific comments
Lines 34-35: I suggest using 'topography' or 'topographic data' instead of 'geomorphic characteristics' throughout the manuscript.
Line 35: I suggest using 'topographic metrics' instead of 'geomorphic indices'. The former depicts the output of such methods much better, and changing it consistently throughout the text would be beneficial.
Line 37: You should include the reference of Wobus et al. (2006) when citing ksn.
Lines 35-43: The sentences describing progress in extracting quantitative information from topographic data focusing on tectonic geomorphology starting at "The classic approach…" and following to "between tectonic forcing and river incision (…)" can use some rephrasing to depict better decades of theoretical, numerical and empirical advances in quantitative topographic analysis. Here are some suggestions of papers that have compellingly done a similar task (for inspiration): Wobus et al., 2006; Kirby and Whipple, 2012; Whittaker, 2012; Lague, 2014; Demoulin et al., 2017; Mudd et al., 2018.
Line 38: What causes transience is not the response of rivers to tectonic activity but rather the spatial or temporal change in tectonic forcing itself. So, it would be nice to rephrase the text.
Lines 40-41: You can elaborate better on why the integral analysis was introduced to the geomorphic community. For example, you could argue that it does not require deriving slope from elevation data to extract ksn or instead focus on the many benefits it presents (e.g., the slope of rivers' long profile in elevation-chi space is equal to ks or ksn, whereas the normal long profile is just the local channel slope).
Lines 42-43: "enabled the determination of the dynamic evolution of a fluvial system" is awkward and difficult to follow. I guess you want to say that it can be used to determine whether a landscape is in a steady or transient state, given its boundary conditions, and that it can be further used to assess long-term drainage divide instability.
Line 45: The "site-specific parameters" are constrained by empirical data, not simulations, so the phrasing here is strange. With the modelling, you could determine a range of reasonable values given feasible parameters from empirical data.
Line 49: "(steady state or transient state, and equilibrium or disequilibrium)" is awkward as they mean the same thing. It may be beneficial to define what you mean by those terms.
Lines 53-54: Most of these studies focus on extracting quantitative information from topographic data to identify spatial and temporal variations in tectonics, climate conditions, and lithology. So, I suggest rephrasing.
Line 57: I suggest: "… inherited topography is non-negligible because (1) the present…"
Lines 57-60: Although I can understand the rationale for the three points why inherited topography is influential, the writing is challenging to follow. Please rephrase it to make it concise. The following sentence starting with "Therefore" presents the same idea more concisely and clearly.
Line 56: The structure of this paragraph is strange. You start with "We show" before stating what you will do. My point is the simple structure with "In this study, we …" should come earlier than 'we show'.
Line 65: The figure placement is off here, making reading more difficult. I will address figures/captions in the end.
Line 71: The start of this paragraph is strange: "We target an area… as study area." You should first explain to the reader what you want to do, for example, using something like: "To explore the role of inherited topography …, this study …"
Line 72: Suggestion: "… studying relationships between geology, tectonics and geomorphology".
Line 73: I am not sure 'along' is the right word here. Maybe across?
Lines 80-89: Okay, so here is the paragraph I expected. As I mentioned in my previous comment, this should come earlier. I would make it the introduction's third paragraph, merging it with the formerly third paragraph. You could use here what is written at the beginning of your conclusion: "The Ulsan Fault Zone (UFZ) has been one of the most active fault zones on the Korean Peninsula since its reactivation ~ 5 Ma. Our study area, the eastern, mountainous, hanging wall block of the UFZ, has undergone regional uplift under an ENE–WSW oriented neotectonic maximum horizontal stress after 5 Ma." This addition would make the introduction more compelling and help give the reader context. Those sentences alone could replace the former third paragraph, making the information more concise and precise.
Lines 83-85: I suggest you rephrase the sentences: "Evaluation of the relative tectonic intensity using geomorphic indices is particularly valuable in the study area. It is challenging to find surface deformation caused by neotectonic faulting in Korea due to low slip rates, rapid physical 85 and chemical erosion, and vast urbanisation." I follow the point, but how it is written makes it difficult to understand.
Lines 95-98: These sentences can use some rephrasing to improve clarity and readability: "Early studies proposed that the main strand of the UFZ is located within the incised valley (Kim, 1973; Kim et al., 1976; Kang, 1979a, b). However, subsequent studies have suggested that the UFZ is located either in and around the incised valley, or that it lies along the mountain front to the east of the incised valley, or possibly in both locations (Okada et al., 1998; Ryoo et al., 2002; Choi, 2003; Choi et al., 2006; Ryoo, 2009; Kee et al., 2019; Naik et al., 2022)." It was difficult to understand, particularly the last sentence, which started with 'however'.
Line 115: I prefer 'rivers draining the TMR' to 'channels on the TMR' because rivers flow away from it in opposing directions. Using 'rivers' would be preferable to 'channels' throughout.
Lines 116-118: You could merge the two sentences to make them clearer, like "whereas those on the western flank form a more complex drainage system flowing north or southward from a low-elevation valley floor drainage divide."
Lines 126-132: The contrasting western/eastern landscape morphology is very interesting, and the hypotheses you list are even more interesting. They could make a more compelling framing of the narrow problem you will solve with your study. You could potentially map out the empirical consequences of these hypotheses and use your data to test them.
Lines 137-138: Can you elaborate more on how you did the calculations?
Line 150: 'Topographic analysis' instead of 'Morphometric analysis'.
Line 151: Suggestion: "Previous studies of tectonic geomorphology used a variety of topographic metrics to infer relative magnitudes of tectonic forcing…" I am unsure if 'intensity' is the best word. Maybe 'magnitude'?
Lines 151-156: This list of arguably 'old' topographic metrics can be removed from the text as you do not use them in your paper.
Lines 158-159: The sentences "Further, the study area likely involves low fault slip rates and high rates of physical and chemical erosion, making it difficult to observe the vertical displacement by neotectonic faulting on the surface" should have come earlier in the methods section and been more extensively elaborated. Moreover, the climate context was not described before.
Lines 159-160: "As a result, we adopted and used alternative morphometries, including" should be rephrased. This is a strange way to introduce some of the most successful topographic metrics, perhaps the 'gold standard' of tectonic geomorphology. In contrast, the "old" topographic metrics you point to at the beginning of the paragraph, such as Hack's SL index, are barely used in quantitative topographic studies anymore. Some of the papers I have suggested discuss the reasons for this. So, I suggest you rework this whole paragraph.
Line 161: Change 'morphometries' to 'topographic metrics' and be consistent throughout the manuscript. I am unsure about the word 'intensity'. Maybe "relative magnitudes of tectonic forcing"?
Line 166: What does 'elevation' of a swath profile mean? A swath profile presents maximum minimum and mean elevation values. I could not follow.
Line 164: Add 'normalised' to channel steepness and be consistent.
Lines 165-166: The phrasing in "Although the elevation of the swath profile and topographic relief are not the same as cumulative vertical displacement, these two morphometries can reasonably be used as a proxy to infer the latter" is challenging to follow. Please rephrase. Moreover, how and why can these topographic metrics be used to infer cumulative vertical displacement? This is not obvious. Finally, 'morphometries' must be replaced by 'topographic metrics' throughout the manuscript.
Lines 168-169: How and why are these topographic metrics used to identify topographic transience? What does steady or transient state mean here? It should be clearly defined.
Lines 150-175: The entire Methods section needs rework. Suggestion: a three-subsection structure with 1) Topographic analysis, 2) Cosmogenic nuclide analysis, and 3) Landscape evolution modelling.
For the topographic analysis section, instead of these two long paragraphs, I suggest one single paragraph starting with something like: "We used a 5-m-resolution digital elevation model (DEM) to extract the following topographic metrics: …." Next sentence: "These metrics have been commonly used to reveal the pattern and style of landscape adjustment due to … " "The DEM was generated using digital contours provided by the National Geographic Information Institute (NGII) of the Republic of Korea (https://www.ngii.go.kr/kor/main.do) and was projected to WGS84 UTM coordinates" or something along those lines.
The following paragraph will describe each topographic metric used in the paper. I would start with the normalised channel steepness, then the metrics for assessing long- and short-term drainage divide stability, and finally, the swath profile. Alternatively, you could still have each metric described in its own section, but the initial paragraph should follow something similar to what I described above.
Line 176-180: Suggestion: "Swath profiles quantify how minimum, mean, and maximum elevation varies across a region along a profile" or something similar instead of the present phrasing.
The following sentence is very confusing and should be reworked: "Swath profiles can be used to investigate and understand the relationship between surface topography and associated or causative variables, such as dynamic topography, which is a topographic change caused by mantle convection (Stephenson et al., 2014), precipitation (Bookhagen and Burbank, 2006), and uplift and exhumation rates (Taylor et al., 2021)." Maybe adding 'or' before "precipitation" does the trick. Additionally, perhaps adding "or spatial and temporal patterns on precipitation" could be helpful.
Line 180: Why use a line centred on the MDD to produce swath profiles instead of the MDD itself, like Fonte-Boa et al. (2022)?
Line 182: Normalised 'channel' steepness index (ksn)
Line 183: Use italics for every variable in the main text.
Line 186: Instead of "… a dimensional coefficient of erosion," use "a dimensional coefficient of fluvial erosion efficiency." Instead of "… and includes", use "encapsulating different controls on erosion, such as …"
Lines 188-189: If you added units for K, do the same for other variables, such as A.
Line 206: 1) You need to justify the choice of reference concavity here. Okay, you could say that you used the same value as most previous studies and that the chosen value is within the range of feasible empirical values. However, there are several studies with a somewhat similar setup as ours (using topographic data to extract quantitative information on the rates of tectonic processes) showing systematic variability in the concavity of bedrock river profiles due to spatial variations in rates of tectonic processes such as Kirby and Whipple (2001), and Clubb et al., (2020). Furthermore, if you have variations in concavity across the UFZ, then you are likely to misinterpret patterns of channel steepness and knickpoints (e.g., Mudd et al., 2018; Gailleton et al., 2021). Thus, I strongly recommend you investigate how concavity varies across the study area. The topographic software you use in the paper (LSDTopoTools and TopoToolbox) have algorithms that can be used to carry out this task readily.
2) You must provide proper information on how you computed ksn from topographic data. For example, how was the DEM hydrologically corrected to ensure channel bed elevation decreases monotonically as you move along the river profile? What was the threshold for channel initiation? What was the flow routing method? Have you computed ksn as the derivative of χ and elevation, which I suppose to be the case given the reference cited (e.g. Mudd et al., 2014)?
Line 209: The phrasing needs rework. It is not the 'law of divides' that makes a topographic elevation separating two adjacent hillslopes downslope to opposite sides.
Lines 208-211: The phrasing can use some rework. Start explaining divide stability/instability as a function of contrasts in erosion rates in adjacent river basins separated by a drainage divide. Then state that because erosion rates depend on topography, we can use topographic metrics to infer the degree of instability of divides. A helpful reference for your rephrasing here is He et al. (2024). I suggest you describe how one can use topographic data to predict drainage divide migration direction similarly to He et al. (2024).
Moreover, I cannot understand why you use three different metrics (i—mean upstream relief, ii—mean upstream gradient, and iii—channel head elevation) throughout the manuscript that yields a similar measurement (i.e., cross-divide steepness asymmetry). You need a single metric instead. I suggest you use the across-divide difference in hillslope relief (ΔHR) normalised by the across-divide sum in hillslope relief (∑HR), referred to as the divide asymmetry index (DAI) introduced by Scherler and Schwanghart (2020) and readily implemented in TopoToolbox. Alternatively, you could compute variations in mean upstream relief using TAK. Channel head elevation and mean upstream gradient are unnecessary here. For example, you do not have accurate information about channel heads and have not extracted them using a more sophisticated algorithm (e.g., Clubb et al., 2014).
Lines 213: Add 'long-term' to the phrase "evaluate long-term divide stability".
Line 218: Start with "where x is the distance upstream from an arbitrary baselevel zb (at x = xb)". Remove the "x' dummy variable" part.
Line 220: Delete "by multiplying by A_0 as a coefficient". This sentence can be integrated with the previous one for conciseness.
Lines 220-221: The sentence "Equation (4a) illustrates the linear relationship between elevation and the χ index for a steady-state channel" needs some rephrasing to make clear that this is the case when rock uplift, bedrock erodibility, and climate conditions are invariant along-profile, provided that the reference concavity is adequate. As such, spatial variations in these boundary conditions will be expressed by non-linear shapes in elevation-chi profiles. Moreover, I suggest using "establishes" instead of "illustrates".
Lines 222-223: I would move the sentence "Because the χ index is sensitive to the base-level elevation (zb; Forte and Whipple, 2018), we analysed the χ index with two different base-level elevations" to the end of the paragraph, using "Finally, because … we calculated the χ metric assuming two different …"
Lines 223-228: I could not follow the text or the rationale. The most straightforward baselevel elevation for extracting the drainage network in your case should be 0 m a.s.l. It is hard to imagine that using a baselevel elevation of less than 50 m, you extract fewer river segments than using a higher baselevel elevation. For instance, your Fig. 6 shows the opposite, with more river segments extracted using baselevel = 50 m than when baselevel = 200 m. Therefore, I strongly recommend you extract the drainage network using baselevel 0 m. This would allow the extraction of complete drainage networks, facilitating the visualisation of patterns in ksn and knickpoints. For example, it is challenging to grasp spatial patterns in ksn, chi, or flow directions, in Fig. 6.
Lines 226-227: It is necessary to describe how you extracted Gilbert metrics and performed the chi-transformation with sufficient detail to ensure the reproducibility of the results. There is nearly no information about the parameters or algorithms used here.
Line 229: Suggestion: "River profile analysis and knickpoint extraction"
Lines 230-234: Why have you framed the analysis using log S-log A profiles? You are not analysing logS-logA profiles. So, I suggest focusing on the shape of river profiles on elevation-distance or elevation-chi spaces.
Line 234: Rephrase the sentence, "The boundary between adjacent piecewise lines can be identified physically as knickpoints." Delete 'physically'. Moreover, defining what you mean by knickpoint from the first use is necessary.
Line 235: Suggestion: "or exposure of a previously buried rock-type".
Lines 236-237: More detail is needed here to describe how you extracted long profiles and performed the integral transformation of the x coordinate. Otherwise, one could not reproduce any of your results. Did you use carving or filling procedures to hydrologically correct the DEM? Did you use a threshold for channel initiation?
Lines 239-242: 1) Much more detail is needed here. If you used the method Gailleton et al. (2019) introduced to extract knickpoints, then you need to describe the user-defined parametrisation. Otherwise, your approach is not reproducible, which is precisely the point the method introduced by Gailleton et al. (2019) addressed. 2) These sentences highlight why you should not frame the beginning of this sentence based on patterns of slope-area data.
Line 244: I suggest you define what you mean by a steady state in the first usage. Otherwise, the text gets confusing.
Line 249: Instead of "represents", use "can be interpreted as …". It is also necessary to add a few citations after this sentence.
Lines 251-252: This was a good example of using 'steady-state'.
Line 253: Change one of the two 'during' in the sentence for a synonym.
Line 258: The placement of the figures makes it difficult to read the manuscript.
Lines 269-273: Did the sample strategy target catchments with comparable upstream drainage areas? This is not clear.
Line 270: 1) 'Document' instead of 'trace'. 2) I could not follow. Maybe explain that you want to compare across and along-MDD variations in CADRs instead of having the complicated sentence "…to trace variations in the CADR along the MDD and to compare the CADRs of the western and eastern flanks of the TMR".
Line 271: Suggestion "topographic" instead of "morphometric".
Lines 273-274: Rephrase the sentence "We avoided collecting samples from: (1) catchments containing golf courses and (2) downstream areas where alluvial fans are located, and faults occur (Fig. 2) to avoid possible contamination by anthropogenic debris" to "To avoid possible contamination by anthropogenic debris, we …"
Lines 275-280: 1) Rework the sentences: "The basins W1 and E1 contain rhyolite and dacite bedrock. The basins W2, W3, E2, and E3 contain rhyolite, dacite, and granite bedrock. The other basins (W4–W8 and E4–E8; eight basins) contain sedimentary, volcanoclastic, and granite bedrock." You can combine them into a single sentence starting with "… For example, …" 2) It was unclear from reading these sentences how you explored potential lithological variations in the results. This seems important given that you do have lithological variations along the MDD and are assessing how erosion varies along the MDD. There is no subsequent table or figure with any lithological information (e.g., distribution of rock types per sampled catchment or the areal contribution of quartz-bearing lithologies). Please add the sampled catchments in Fig.1b.
Line 281: Suggestion: "following a standard protocol" instead of "following the standard protocol".
Line 291: Later in the text (Line 314), you state you used the "CRONUS-Earth online calculator (Balco et al., 2008; version 3), applying the LSDn scaling scheme (Lifton et al., 2014)." Why use a different approach to estimate erosion rates from cosmogenic 10Be abundances here? I suggest you be consistent throughout the study.
Furthermore, if you had access to Mudd et al.'s (2016) CAIRN program, which uses the same software you used to calculate topographic metrics (i.e., LSDTopoTools), you could have used it to calculate catchment-averaged atmospheric pressure using CAIRN, which could then be fed into CRONUS-Earth to estimate catchment-averaged denudation rates.
Finally, you should not use topographic shielding to compute catchment-averaged denudation rates for your study area (see DiBiase, 2018). I suggest you recalculate your rates.
Line 311: Rephrase "following laboratory protocol…" to something like "following the same laboratory protocol described above…"
Lines 314-315: 1) Again, I suggest you calculate cosmogenically-derived erosion/exposure ages in a consistent manner. 2) Suggestion: Use 'uncertainty' instead of 'error' here.
Line 317: Suggestion: "Landscape evolution modelling" instead of "Modelling landscape evolution".
Line 318: 1) Delete 'next'. 2) Use "landscape evolution model toolkit …" instead of "landscape evolution model".
Line 319: Rephrase. You will investigate the evolution of specific model setups to get insights about the evolution of the uplifted eastern hanging wall block of the UFZ rather than "comprehensively investigating" its evolution.
Lines 319-320: Suggestion: "… These simulations were compared to results from topographic analysis and 10Be measurements …". Delete "in conjunction with measured geomorphic indices".
Line 324: This parametrisation for stream-power river incision is awkward and distant from the commonly used values in modelling studies or reported values from empirical studies. First, the K value seems to be way too low, especially given the tectonic context of the study area (compare, for example, with values reported by Stock and Montgomery, 1999; Whipple et al., 2000; Kirby and Whipple, 2001; Zondervan et al., 2020; and Peifer et al., 2021). This will have significant implications for your simulations involving perturbation phases related to changes in the tectonic field. More reasonable values would be between 10-5 to 10-6. Furthermore, you have m and n values that are also awkward. The most straightforward parametrisation would have n = 1, while m could vary between 0.4 to 0.6. In this n = 1 case, the river response to the perturbation does not depend on the channel slope. In contrast, when n > 1, the river response depends on channel slope, leading to complexity that we do not understand fully. So, while it would be okay to have scenarios using a set of stream-power parameters such as m = 0.6 and n = 1.5, it should not be your only parametrisation scenario.
Suggestions: 1) use your catchment-averaged denudation rates to parametrise reasonable values for K (e.g., Gallen (2018)); 2) perform a sensitivity analysis with different parametrisations using empirically feasible values, preferably with values higher and lower than the obtained in 1.
Line 326: Why did you use an incision threshold? This is not clear, and even if you have a compelling explanation, it should not be your only parametrisation. Most modelling studies do not account for incision thresholds. So, why is it important here? If you want to include it, I suggest you perform a sensitivity analysis to determine its influence on simulations.
Line 330: Add units for the diffusivity coefficient.
Line 331: While the Kd value does seem reasonable, a sensitivity analysis with lower and higher values is also necessary. One scenario could perhaps scale K/Kd similarly to Whipple et al. (2017), with K/Kd = 0.002.
Line 335: Change the signal for the hillslope erosion to plus in the equation.
Lines 355-358: The rationale for the modelling setup, with the two phases, should be better introduced earlier in the text. This would improve the readability of the text considerably.
Lines 356-357: Although reasonable, as it was configured based on the constraints for the study area, having the first stage running for only 3 Myr feels somewhat awkward, given that landscape equilibration concerning phase 1 will be important later in the paper. This feels particularly important, given the usage of such a low K value (K = 5.56E-07). From this K alone, I'd expect that reaching a steady state configuration could take two orders of magnitude longer than 3 Myr. So, my first suggestion would be to run additional scenarios with different durations for phases 1 and 2. Because river response timescales depend on K (and S in case n is more than 1), the sensitivity tests will be critical to evaluate simulation outputs.
Lines 367-370: 1) This sort of key information: "This assumption is based on the overall tendency of high-east and low-west topography of the Korean Peninsula, supported by the long-term, regional westward tilting that was initiated during the Middle Miocene when the East Sea started to widen, and since which time the strongly asymmetric (high-east) Taebaek Mountain Range has been rapidly uplifted (Min et al., 2010; Kim et al., 2020)" should have come way earlier in the text. It would help frame the narrow problem this study is addressing and give the reader more context. 2) The Taebaek Mountain Range was not mentioned earlier in the text and was not identified in previous figures. As such, it is difficult to follow.
Lines 370-373: Again, this critical information should have come earlier, as it would greatly help framing the narrow question addressed by the paper: "In addition, the shore platform on the western coast of the peninsula (0 m a.s.l.; Choi et al., 2012a; Jeong et al., 2021) and marine terraces along the eastern coast (18–45 m a.s.l.; Choi et al., 2003a, b; Kim et al., 2007; Heo et al., 2014; Lee et al., 2015), formed at the same time (i.e., during MIS 5), indicate that this regional differential uplift has lasted until very recently." Please rephrase to clarify why this indicates that regional differential uplift lasted until recently. Also, define 'MIS 5'.
Lines 373-375: Why have you not created a single model domain with a wider 'uplifted' region in the north and a narrower uplift region toward the south? Manipulating the uplift field could potentially achieve this. Having one single model domain, including both the northern and southern parts of the block, would make the paper easier to read and the results more clearly interpretable.
Lines 380-383: While I understand the setup used for the UFZ, it would be beneficial to perform a sensitivity analysis for the distance between the MDD and the UFZ (e.g., having the UFZ initially closer to the MDD).
Lines 382-385: As discussed above, sensitivity analysis is essential for these parameters (K, Kd, m, and n).
Lines 389-390: This key information needs to come earlier in the text (introduction/study area). Please elaborate more. Until this point, we had no information about the 'backbone' mountain range of the Korean Peninsula, for which long-term exhumation rates are available. This is important.
Lines 394-396: While the parametrisation for U in phase 1 does seem reasonable, a sensitivity analysis is necessary.
Lines 398-416: 1) Rework the text as it is slightly confusing and difficult to follow. 2) The chosen rates for the perturbation phase seem somewhat too specific, and I got slightly confused with the 'ratio of west/east channel incision' part. I suggest performing a sensitivity analysis again as you need to test how other parametrisation affects simulations. 3) Have the 'terrace uplift rates' calculated in the Study Area section played any role in the parametrisation of the models? If not, why not? What was the role of these calculated terrace uplift rates in the manuscript?
Line 419: Suggestion: "initial topography (i.e., topography achieved after stage 1)" or something similar to improve clarity on those lines.
Line 422: Substitute "geomorphic indices analysis" for "topographic analysis" and change "verify" to "compare."
Lines 425-426: Rephrase these sentences. Suggestion: "… quantitatively compare the simulated topography generated in the four cases". Delete "to compare the modelled topographies with the observed topography in the study area" as you mention this on lines 422-423.
Line 430: The placement of figures makes reading the text more difficult.
Lines 439-445: This paragraph needs significant reworking. Its current form offers little quantitative description of the patterns of normalised channel steepness or knickpoints. In fact, from lines 440-445, you provide more interpretation than results. I suggest you elaborate on a similar presentation of results as the one done for the CARD. For instance, you could write something like: "... We find that normalised channel steepness varies from X to X m0.9, with a regional mean of X and a standard deviation of X. Low values (ksn < X) are observed in. High values (ksn > XX) are ..."
*Please check my comments about the figures.
Lines 441-442: It is necessary to detail how this excluding of artefact knickpoints was conducted within the Methods section with enough detail to ensure reproducibility. In addition, please provide quantitative statistics for the presence of 'artefact' and lithological knickpoints. How many lithological knickpoints did you extract? What is their spatial distribution? At each lithological transition, do you observe knickpoint clustering? Does every river crossing such lithological transitions perpendicularly show knickpoints? Do they show similar magnitudes?
Lines 443-445: The finding from previous literature that major knickpoints in the study area cannot be driven by eustatic changes should have been presented and explained earlier in the text.
Lines 455-460: Please rework this paragraph. First, it is better to describe the observed pattern per topographic metric. For instance, when you mention that you have a high value, it is necessary to be explicit. How high? Moreover, I am confused about which ways Fig. 7 is different from Fig. 8. Having many variables that arguably serve the same function (i.e., Gilbert metrics to gauge cross-divide relief asymmetry) makes the visualisation and description of the results worse.
Line 478: 'Significant' here means statistically significant?
Lines 484-486: Higher by how much?
Lines 509-510: Please elaborate more on the sentence: "This pattern contrasts with the main spatial trend of CADR but corresponds to the patterns shown by the other geomorphic indices (Figs. 7 and 8)."
Lines 510-511: You state that the CADRs on the western flank river basins are generally higher than those on the eastern flank. How much higher?
Lines 512-516: These sentences need reworking. No previous investigation or results showed potential influences of lithology on CADRs. In addition, Fig. 1b does not show a clear potential explanation, as suggested here. As I mentioned before, I suggest adding some form of analysis of catchment lithology to the paper.
Line 546: Suggestion: Use 'model domain' instead of 'modelled areas'.
Lines 545-552: It is important to realise that the integral metric χ should be calculated differently if the background rock uplift is not spatially uniform in a formulation accounting for the spatial gradient in uplift. This is explained in detail in Willett et al. (2014). Additionally, there is a blog entry in the TopoToolbox blog that elaborates on that, introducing an algorithm tailored to perform such a calculation using TopoToolbox's dependencies:
(https://topotoolbox.wordpress.com/2020/11/13/use-of-chi-analysis-in-experimental-landscapes-dulab/)
Lines 550-552: This sentence can use rephrasing to improve clarity and readability.
Lines 562-565: I was slightly confused with the phrasing here. What do you mean? These two sentences can use some reworking.
Lines 585-586: Please elaborate more on this sentence: "higher sensitivity of MDD to fault slip in Case B2 may be attributable to its shorter channels compared with Case B1". Why?
Line 627: 'Areas with lower swath profile' is awkward. What does a lower swath profile mean? A swath profile shows mean, maximum, and minimum elevation values. How lower? In addition, it would be beneficial to be explicit about what a 'lower degree of tectonic intensity' means. To parametrise your model scenarios, you estimated some average long-term surface uplift rates at the fault values. How about being more precise in analysing the results and comparing them with those?
Lines 654-656: Again, it is necessary to introduce these key results from previous studies early in the text. This feels important. They should not appear out of the blue in the discussions.
Line 656: Delete 'as follows'.
Lines 656-664: These sentences are confusing and should be reworked. It would be more beneficial to discuss the reasoning behind Cheon et al.'s (2013) segmentation of the UFZ in relation to your findings.
Lines 667-670: Please rephrase the sentences "The χ index represents the longer-term view for topography owing to its reliance on the integral method from the far downstream to the channel head (Forte and Whipple, 2018; Zhou et al., 2022). Other geomorphic indices, such as mean upstream gradient and relief, respond sensitively to". This can be better elaborated. For instance, the phrasing of 'longer-term view for topography' is poor. Perhaps explain that the χ method is well suited to assess the long-term stability of drainage divides, while cross-divide differences in steepness are better suited to evaluate short-term divide stability.
Lines 670: I disagree for the reasons I explained when discussing the Methods section. I find the modelling exercise's parametrisation somewhat arbitrary, and other parameter values need to be tested.
Lines 672-673: I feel that it is necessary to elaborate more extensively on uncertainties associated with the modelling exercise.
Lines 673-675: I could not follow these sentences: "Comparing geomorphic indices that are sensitive to minor variations in boundary conditions could lead to a misinterpretation of the geomorphic evolution. For these reasons, we chose to focus on a comparison of the pattern of χ indices."
Lines 676-678: I could not follow the ideas expressed in these sentences. Why do the variations in morphology along the MDD make comparing the morphology on the easter-western flanks of the MDD difficult?
Line 678: Maybe I am missing it, but I do not recall the mean values per each topographic metric presented earlier in the text. Because of this, I find it hard to follow this statement.
Lines 676-682: This paragraph was challenging to follow (and I have not fully understood it). This needs rework.
Lines 683-688: These sentences were complex to follow and felt repetitive (perhaps that should belong in the Results section). What does 'inconsistent' here mean? "In contrast with all other geomorphic indices, differences between the western-flank and eastern-lank χ index values are inconsistent."
Line 688: What do you mean by inconsistent pattern in χ? And why is it 'decoupled' (is this the best word here?) from catchment-averaged denudation rates?
Line 690: I suggest using 'agrees' or 'consistent' instead of 'coupled' or 'decoupled' throughout the text.
Lines 683-694: This paragraph needs rework to improve clarity and readability.
Line 695: "To facilitate the investigation of the geomorphic evolution of the study area" reads awkwardly here. I suggest some rephrasing.
Lines 695-701: I am confused. You classified the UFZ into two segments (north and south). Isn't this the precise classification you criticised in the paragraph starting at line 654?
Lines 722-723: By simply multiplying the integral metric by K, one can estimate channel response timescales (e.g., Gallen, 2018). As such, you could bracket reasonable channel response timescales for the UFZ, effectively testing this hypothesis.
Lines 724-726: A complete sensitivity test for different parametrisations for rock uplift is necessary for supporting this statement.
Lines 729-734: Considering the very low K value used in the simulation, has the landscape achieved a steady state in this modelled scenario with only 3 Myr of model run? What was the criteria for defining steady state here? This feels awkward as my simulations take me much more time (hundreds of Myr) to achieve a steady state if I use K values similar to yours. It would have been nice to have snapshots of erosion rates presented to the reader after phase 2 of the simulation.
Lines 751-793: You are missing a big opportunity to use your paired catchment-averaged denudation rates to estimate drainage divide retreat rates using an approach similar to Hu et al. (2021) and Stokes et al. (2023).
Lines 761-762: Use 'indicates eastward divide migration' instead of 'is related to the...'
Lines 779-781: This is difficult to follow; please elaborate more: "Therefore, we interpret that the streams flowing within the drainage in the vicinity of the MDD and the elevated ridge on the western flank of segment 1 are the results of antecedent streams."
Lines 794-826: The conclusion is way too long. Rework is necessary.
Figure comments
Fig. 1: Panel A should prioritise showing topography rather than satellite imagery. In Panel B, including sampling sites and sampled catchments is necessary. Additionally, it would be important to show the Taebaek Mountain Range somehow.
Fig. 2: The river network in panel A feels strange. Instead, I suggest extracting (and showing) all rivers starting at the baselevel elevation of 0 m. Including sampling sites and sampled catchments is necessary. Why are the marine terrace uplift rates exhibited here? I do not recall them being discussed further in the text. I would also add the swath profile centre lines here.
Fig. 4: While the concept is promising, the figure's complexity makes it challenging to grasp. Simplifying the visualisation, possibly by presenting the setup in a plan view, would improve clarity and comprehension.
Fig. 5: 1) It is challenging to visualise patterns in channel morphology in panel a due to the drainage networks' incompleteness. I suggest extracting rivers assuming baselevel = 0 m, ensuring river networks are complete, extending downstream until the ocean. Including sampling sites and sampled catchments is necessary.
2) The interpretation of knickpoints in river profiles appears flawed. First, you are likely considering concavities in the long-profile as knickpoints (i.e., points identified by a downstream along-profile decrease in ksn). 'Concave' knickpoints should not be identified here. Additionally, there are many instances in panel D of significant along-profile breaks in channel slope that were not identified as knickpoints (e.g., around 200 m of elevation at a distance slightly below ten and slightly above five), and they should be. I guess these results are caused by the parametrisation used to extract knickpoints from topographic data. Furthermore, the exclusion of knickpoint at artefacts and lithological boundaries also appears arbitrary. In summary, rework is necessary to ensure that knickpoints are identified accurately and consistently along the profiles. Finally, change the caption for the 'X-ksn plot'. Ksn is the slope of the profile in elevation-chi space.
Fig. 6: This figure's current format is not helpful for the manuscript. I suggest that the sensitivity tests on how changing base level elevation affects chi patterns are presented as supplemental material. If you want to show spatial patterns on other channel morphology metrics, I suggest you depict complete river networks. I strongly recommend quantifying relief for each pixel for your DEM rather than 'channel relief', given that ksn is already a robust measure of local channel slope normalised by upstream drainage area.
Fig. 7 and 8: Combine these two figures into a single figure, starting with the swath profile in panel a, followed by variations in chi, ksn, and upslope hillslope relief. Consider omitting unnecessary metrics to streamline the presentation.
Fig. 9: In Panel A, prioritise topography over satellite imagery for better visualisations of landscape features. For clarity, Panel B should focus on chi and mean upstream relief only.
Fig. 10-12: I assume these figures will undergo significant changes after revision.
References
Clubb, F.J., Mudd, S.M., Milodowski, D.T., Hurst, M.D. and Slater, L.J., 2014. Objective extraction of channel heads from high‐resolution topographic data. Water Resources Research, 50(5). DOI: 10.1002/2013WR015167.
Clubb, F.J., Mudd, S.M., Hurst, M.D. and Grieve, S.W., 2020. Differences in channel and hillslope geometry record a migrating uplift wave at the Mendocino triple junction, California, USA. Geology, 48(2). DOI: 10.1130/G46939.1.
Demoulin, A., Mather, A. and Whittaker, A., 2017. Fluvial archives, a valuable record of vertical crustal deformation. Quaternary Science Reviews, 166. DOI: 10.1016/j.quascirev.2016.11.011.
DiBiase, R.A., 2018. Increasing vertical attenuation length of cosmogenic nuclide production on steep slopes negates topographic shielding corrections for catchment erosion rates. Earth Surface Dynamics, 6(4). DOI: 10.5194/esurf-6-923-2018.
Fonte-Boa, T.M., Peifer, D., Fonseca, A. and Novo, T.A., 2022. The southeast Brazilian rifted continental margin is not a single, continuous upwarp: variations in morphology and denudation patterns along the continental drainage divide. Earth-Science Reviews, 231. DOI: 10.1016/j.earscirev.2022.104091.
Gailleton, B., Mudd, S.M., Clubb, F.J., Peifer, D. and Hurst, M.D., 2019. A segmentation approach for the reproducible extraction and quantification of knickpoints from river long profiles. Earth Surface Dynamics, 7(1). DOI: 10.5194/esurf-7-211-2019.
Gailleton, B., Mudd, S.M., Clubb, F.J., Grieve, S.W. and Hurst, M.D., 2021. Impact of changing concavity indices on channel steepness and divide migration metrics. Journal of Geophysical Research: Earth Surface, 126(10). DOI: 10.1029/2020JF006060.
Gallen, S.F., 2018. Lithologic controls on landscape dynamics and aquatic species evolution in post-orogenic mountains. Earth and Planetary Science Letters, 493. DOI: 10.1016/j.epsl.2018.04.029.
He, C., Braun, J., Tang, H., Yuan, X., Acevedo-Trejos, E., Ott, R.F. and Stucky de Quay, G., 2024. Drainage divide migration and implications for climate and biodiversity. Nature Reviews Earth & Environment. DOI: 10.1038/s43017-023-00511-z.
Hu, K., Fang, X., Ferrier, K.L., Granger, D.E., Zhao, Z. and Ruetenik, G.A., 2021. Covariation of cross-divide differences in denudation rate and χ: Implications for drainage basin reorganization in the Qilian Shan, northeast Tibet. Earth and Planetary Science Letters, 562. DOI: 10.1016/j.epsl.2021.116812.
Kirby, E. and Whipple, K., 2001. Quantifying differential rock-uplift rates via stream profile analysis. Geology, 29(5). DOI: 10.1130/0091-7613(2001)029<0415:QDRURV>2.0.CO;2.
Kirby, E. and Whipple, K.X., 2012. Expression of active tectonics in erosional landscapes. Journal of structural geology, 44. DOI: 10.1016/j.jsg.2012.07.009.
Lague, D., 2014. The stream power river incision model: Evidence, theory and beyond. Earth Surface Processes and Landforms, 39(1). DOI:
Mudd, S.M., Attal, M., Milodowski, D.T., Grieve, S.W. and Valters, D.A., 2014. A statistical framework to quantify spatial variation in channel gradients using the integral method of channel profile analysis. Journal of Geophysical Research: Earth Surface, 119(2). DOI: 10.1002/esp.3462.
Mudd, S.M., Harel, M.A., Hurst, M.D., Grieve, S.W. and Marrero, S.M., 2016. The CAIRN method: automated, reproducible calculation of catchment-averaged denudation rates from cosmogenic nuclide concentrations. Earth Surface Dynamics, 4(3). DOI: 10.1002/2013JF002981.
Mudd, S.M., Clubb, F.J., Gailleton, B. and Hurst, M.D., 2018. How concave are river channels?. Earth Surface Dynamics, 6(2). DOI: 10.5194/esurf-6-505-2018.
Peifer, D., Persano, C., Hurst, M.D., Bishop, P. and Fabel, D., 2020. Growing topography due to contrasting rock types in a tectonically dead landscape. Earth Surface Dynamics, 9. DOI: 10.5194/esurf-9-167-2021.
Scherler, D. and Schwanghart, W., 2020. Drainage divide networks–Part 1: Identification and ordering in digital elevation models. Earth Surface Dynamics, 8(2). DOI: 10.5194/esurf-8-245-2020.
Stock, J.D. and Montgomery, D.R., 1999. Geologic constraints on bedrock river incision using the stream power law. Journal of Geophysical Research: Solid Earth, 104(B3). DOI: 10.1029/98JB02139.
Stokes, M.F., Larsen, I.J., Goldberg, S.L., McCoy, S.W., Prince, P.P. and Perron, J.T., 2023. The erosional signature of drainage divide motion along the Blue Ridge Escarpment. Journal of Geophysical Research: Earth Surface, 128(1). DOI: 10.1029/2022JF006757.
Willett, S.D., McCoy, S.W., Perron, J.T., Goren, L. and Chen, C.Y., 2014. Dynamic reorganization of river basins. Science, 343(6175). DOI: 10.1126/science.1248765.
Whipple, K.X., Snyder, N.P. and Dollenmayer, K., 2000. Rates and processes of bedrock incision by the Upper Ukak River since the 1912 Novarupta ash flow in the Valley of Ten Thousand Smokes, Alaska. Geology, 28(9). DOI: 10.1130/0091-7613(2000)28<835:RAPOBI>2.0.CO;2.
Whipple, K.X., Forte, A.M., DiBiase, R.A., Gasparini, N.M. and Ouimet, W.B., 2017. Timescales of landscape response to divide migration and drainage capture: Implications for the role of divide mobility in landscape evolution. Journal of Geophysical Research: Earth Surface, 122(1). DOI: 10.1002/2016JF003973.
Whittaker, A.C., 2012. How do landscapes record tectonics and climate?. Lithosphere, 4(2). DOI: 10.1130/RF.L003.1.
Wobus, C., Whipple, K.X., Kirby, E., Snyder, N., Johnson, J., Spyropolou, K., Crosby, B. and Sheehan, D., 2006. Tectonics from topography: Procedures, promise, and pitfalls. In: Willett, S.D., Hovius, N., Brandon, M.T., Fisher, D.M. (Eds.), Tectonics, Climate, and Landscape Evolution: Geological Society of America Special Paper. Penrose Conference Series. DOI: 10.1130/2006.2398(04).
Zondervan, J.R., Whittaker, A.C., Bell, R.E., Watkins, S.E., Brooke, S.A. and Hann, M.G., 2020. New constraints on bedrock erodibility and landscape response times upstream of an active fault. Geomorphology, 351. DOI: 10.1016/j.geomorph.2019.106937.
Citation: https://doi.org/10.5194/egusphere-2024-198-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 222 | 47 | 12 | 281 | 3 | 5 |
- HTML: 222
- PDF: 47
- XML: 12
- Total: 281
- BibTeX: 3
- EndNote: 5
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1