Bushfire effects on soil properties and post-fire slope stability: the case of the 2015 Wye River-Jamieson Track bushfire

Li, Yuanying; Wakai, Akihiko; Costa, Susanga

doi:https://doi.org/10.5194/nhess-2024-132

Preprints

https://doi.org/10.5194/nhess-2024-132

Preprints

18 Sep 2024

| 18 Sep 2024

Status: this preprint is currently under review for the journal NHESS.

Bushfire effects on soil properties and post-fire slope stability: the case of the 2015 Wye River-Jamieson Track bushfire

Yuanying Li, Akihiko Wakai, and Susanga Costa

Abstract. Bushfire is a destructive natural disaster that leads to vegetation loss and increased soil infiltration. Over a long post-fire period, root death and reduced reinforcement decrease soil shear strength. During rainfall, shallow landslides in burned areas become more frequent and widespread. This study focused on Wye River and Separation Creek in Australia, affected by the 2015 Wye River-Jamieson Track bushfire. Ten months after the bushfire, multiple slope failures, including the Paddy’s Path landslide, occurred during heavy rains from 12 to 14 September 2016, disrupting the Great Ocean Road connecting towns. This study aims to assess changes in slope stability during rainfall before and after the bushfire. Controlled laboratory burning tests simulated bushfire effects on soil, resulting in changed soil properties after the fire: increased permeability due to soil particle coarsening and reduced soil shear strength, especially cohesion. Considering the changes in soil properties before and after the fire, a simplified hydrological numerical model for infiltration calculation was employed to analyze time-dependent changes in groundwater level depth, surface water depth, and safety factor during rainfall. Comparing pre- and post-fire results indicated higher susceptibility to shallow slope failures in burned areas, with rapid rises in groundwater level and surface water acting as triggers. These findings enhance the understanding of landslide triggering mechanisms in post-fire slopes and provide insights for mapping landslide susceptibility in bushfire-prone regions.

Received: 18 Jul 2024 – Discussion started: 18 Sep 2024

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

Yuanying Li, Akihiko Wakai, and Susanga Costa

Status: final response (author comments only)

RC1:
'Comment on nhess-2024-132', Anonymous Referee #1, 21 Dec 2024
This manuscript presents a case study for a modeling framework that combines numerical models and laboratory experiments to predict slope factor-of-safety values under pre- and post-fire conditions for a landscape in Victoria, Australia, burned by the 2016 Wye River-Jamieson Track wildfire. This location experienced shallow landslides in response to rainfall approximately 10 months after the fire was contained. The authors predict factors of safety using a hydrological model that simulates subsurface and overland flow given input rainfall, and they use controlled laboratory burn experiments on similar soils collected outside of the burn area to parameterize certain model values. The results of their modeling experiments demonstrate a widespread increase in slope instability as indicated by factors of safety less than 1 after the fire due to increased soil saturation and diminished soil cohesion that overlaps with the location of observed landslides. The focus of this study and the predictive nature of their methods are of scientific value and would be of interest to the community.
However, this reviewer has multiple concerns that need to be addressed before publication can be recommended. These include model selection in the pre-fire versus post-fire cases, choice of model parameter values, details of the controlled laboratory burn experiments, a need for expanded literature review, and insufficient consideration of uncertainty. Each of these topics is described in greater detail below. If these can be satisfactorily addressed, as well as the line-by-line comments at the end of this comment, then this reviewer could recommend publication to the editor.
Model selection: two distinct subcategories of the model are chosen for the pre-fire and post-fire cases, namely the “fine sand” model and the “medium-coarse sand" model. These selections are made on the basis of measured grain size distributions and hydraulic conductivity in the controlled burn laboratory experiment. However, it is unclear the impact that this choice alone has on the modeling results, as different equations are used to calculate the factor of safety in these models (i.e., equation 17 for the post-fire case, and equation 18 in the pre-fire case). The post-fire factor of safety equation depends on the depth to groundwater and the distance from the groundwater level to the slip surface. On the other hand, the pre-fire factor of safety equation depends on the depth of the wetting front from the surface. All else held equal, this reviewer wonders how this difference in physical process representation impacts the modeling results.

Choice of model parameters: a key choice is made by the authors in determining the depth of soil and the initial position of the groundwater table. Whereas the maximum soil depth is taken to be 2m in the pre-fire case, the maximum soil depth is 0.3m in the post-fire case. This is likely appropriate when considering the depth of mobilizable material as fire modifications to the near-surface soil decrease soil cohesion, as the authors describe on lines 354-357. However, in the hydrological modeling component of the study, the authors state on lines 363-364 that “…the initial groundwater level is assumed to be at the bottom of the soil depth.” That is, the initial groundwater level is taken to be 2.0m below the surface in the pre-fire case and 0.3m below the surface in the post-fire case. Considering that the depth to groundwater and the distance between the groundwater level and the potential landslide slip surface are key inputs to the post-fire factor of safety equation, it is unclear what impact this difference in groundwater level assumption has on the study results.

Controlled burn laboratory experiments: the authors used a laboratory furnace to simulate burn conditions using soil samples collected from a nearby unburned area with similar bioclimate and pedogenic characteristics. There are multiple points:
Why were burn experiments conducted at such an extreme temperature (800 °C), and how dependent are the measured soil properties on this temperature? Most controlled laboratory burn experiments go up to ~500-600 °C at the surface, even for the high-severity burn cases (e.g., Doerr et al., 2004; Massman, 2015; Moody et al., 2005; Moody et al., 2009; Wieting et al., 2017); even when surface soil temperatures exceed 800 °C, subsurface temperatures are unlikely to approach this value (Robichaud and Hungerford, 2000).

With regard to the minidisk infiltrometers, did the authors use the equation of Vandervaere et al. (2000) to compute hydraulic conductivity, or a different method? And, were only 5 measurements taken using the minidisk infiltrometer for the unburned soil? The fit of the curve to data in Figure 8a shows considerable scatter, and estimates of hydraulic conductivity from minidisk infiltrometers tend to improve with longer measurement times due to its scaling with (t^0.5)².

Expanded literature review on post-fire infiltration rates: a throughline of the manuscript is that infiltration increases after fire. For instance, the authors state in the first line of the manuscript that “Brushfire… leads to… increased soil infiltration.” As a general statement, this is not true. The response of soil hydraulic properties and infiltration rates to fire is complex (e.g., Shakesby and Doerr, 2006), but it has most commonly been found to substantially decrease after fire due to hydrophobicity effects and return to pre-fire levels over a number of years (e.g., Cerdà, 1998; Ebel, 2019; Ebel et al., 2022; Larson-Nash et al., 2018; Lucas-Borja et al., 2022; Moody et al., 2015; Nyman et al., 2014; Robichaud, 2000; see McGuire et al. (2024) and references therein for a thorough review). The authors allude to this fact towards the end of the manuscript on lines 314-318. This is not to say that infiltration rates can be higher in the first year after fire (e.g., Hoch et al., 2021; Perkins et al., 2022), but the issue is more nuanced than currently stated. The manuscript would be strengthened by a more thorough description of the literature on this topic in the Introduction and would provide the authors a base from which to argue that soil hydrophobicity had diminished at their study site at the time of the landslides, less than 1 year after fire containment, resulting in higher infiltration rates than before the fire. The authors are encouraged to use the citations included here in addition to others they may wish to include.

Uncertainty of the results: as alluded to in points 1 & 2, the study currently does not consider the uncertainty involved with model selection and parameter choices. This makes it difficult for this reviewer to place confidence in the general predictive ability of the method. The manuscript would benefit from one of a sensitivity analysis of models and model parameters, incorporation of uncertainty through e.g., a Monte Carlo sampling strategy, or inclusion of additional observations in order to separate the data into calibration and validation datasets. However, this reviewer acknowledges that this may difficult or not possible to accomplish due to the limited availability of target data.

Lines 104-106: how deep-seated were these landslides?
Line 111: please zoom out on the context map in Figure 1a, readers unfamiliar with Australia will be unable to discern the location of the study area within the continent
Lines 137-139: Can the burn severity map be added as an overlay to Figure 1? This would help the reader to understand the burn severity within the study area. This sentence would likely fit better in Section 2.
Line 177: please include how more information about the infiltrometer measurements (e.g., the suction head used, the duration of time/volume that measurements were taken for) and what equations were used to calculate hydraulic conductivity.
Line 189: please include the equation(s) used to fit c and φ to the data
Line 223: by “lateral infiltration of groundwater”, do the authors mean lateral motion?
Line 266: relating to general comment 1 above, why is the horizontal component of groundwater motion not accounted for in the pre-fire case?
Lines 294-364: Sections 4.5, 5.1, and 5.2 seem to be closely related to each other; the authors may disagree, but they could be placed together in their own separate section
Line 371: Figure 12 is not referenced in the text
Line 372: how were the initial saturation degrees determined? Please provide details in the text on why they are different for each case
Line 373: Please move the Results and Discussion from Section 5.3 to a stand-alone section, e.g. Section 6

References
Cerdà, A. (1998). Changes in overland flow and infiltration after a rangeland fire in a Mediterranean scrubland. Hydrological processes, 12(7), 1031-1042.
Doerr, S. H., Blake, W. H., Shakesby, R. A., Stagnitti, F., Vuurens, S. H., Humphreys, G. S., & Wallbrink, P. (2004). Heating effects on water repellency in Australian eucalypt forest soils and their value in estimating wildfire soil temperatures. International Journal of Wildland Fire, 13(2), 157-163, https://doi.org/10.1071/WF03051.
Ebel, B. A. (2019) Measurement Method Has a Larger Impact Than Spatial Scale For Plot-Scale Field-Saturated Hydraulic Conductivity (K_fs) After Wildfire and Prescribed Fire in Forests. Earth Surf. Process. Landforms, 44: 1945–1956. https://doi.org/10.1002/esp.4621.
Ebel, B. A., Moody, J. A., & Martin, D. A. (2022). Post-fire temporal trends in soil-physical and-hydraulic properties and simulated runoff generation: Insights from different burn severities in the 2013 Black Forest Fire, CO, USA. Science of the Total Environment, 802, 149847, https://doi.org/10.1016/j.scitotenv.2021.149847.
Hoch, O. J., McGuire, L. A., Youberg, A. M., & Rengers, F. K. (2021). Hydrogeomorphic recovery and temporal changes in rainfall thresholds for debris flows following wildfire. Journal of Geophysical Research: Earth Surface, 126, e2021JF006374. https://doi.org/10.1029/2021JF006374
Larson-Nash, Sierra S., et al. "Recovery of small-scale infiltration and erosion after wildfires" Journal of Hydrology and Hydromechanics, vol. 66, no. 3, Slovak Academy of Sciences, 2018, pp. 261-270. https://doi.org/10.1515/johh-2017-0056
Lucas-Borja, M. E., Fernández, C., Plaza-Alvarez, P. A., & Zema, D. A. (2022). Variability of hydraulic conductivity and water repellency of soils with fire severity in pine forests and reforested areas under Mediterranean conditions. Ecohydrology, 15(8), e2472. https://doi.org/10.1002/eco.2472
Massman, W. J.: A non-equilibrium model for soil heating and moisture transport during extreme surface heating: the soil (heat–moisture–vapor) HMV-Model Version 1, Geosci. Model Dev., 8, 3659–3680, https://doi.org/10.5194/gmd-8-3659-2015, 2015.
McGuire, L.A., Ebel, B.A., Rengers, F.K. et al. Fire effects on geomorphic processes. Nat Rev Earth Environ 5, 486–503 (2024). https://doi.org/10.1038/s43017-024-00557-7
Moody, J. A., J. D. Smith, and B. W. Ragan (2005), Critical shear stress for erosion of cohesive soils subjected to temperatures typical of wildfires, J. Geophys. Res., 110, F01004, doi:10.1029/2004JF000141.
Moody, J. A., Kinner, D. A., & Úbeda, X. (2009). Linking hydraulic properties of fire-affected soils to infiltration and water repellency. Journal of Hydrology, 379(3-4), 291-303, https://doi.org/10.1016/j.jhydrol.2009.10.015.
Moody John A., Ebel Brian A., Nyman Petter, Martin Deborah A., Stoof Cathelijne, McKinley Randy (2015) Relations between soil hydraulic properties and burn severity. International Journal of Wildland Fire 25, 279-293, https://doi.org/10.1071/WF14062
Nyman, P., Sheridan, G. J., Smith, H. G., & Lane, P. N. (2014). Modeling the effects of surface storage, macropore flow and water repellency on infiltration after wildfire. Journal of Hydrology, 513, 301-313, https://doi.org/10.1016/j.jhydrol.2014.02.044
Perkins, J. P., Diaz, C., Corbett, S. C., Cerovski-Darriau, C., Stock, J. D., Prancevic, J. P., et al. (2022). Multi-stage soil-hydraulic recovery and limited ravel accumulations following the 2017 Nuns and Tubbs wildfires in Northern California. Journal of Geophysical Research: Earth Surface, 127, e2022JF006591. https://doi.org/10.1029/2022JF006591
P.R Robichaud, Fire effects on infiltration rates after prescribed fire in Northern Rocky Mountain forests, USA, Journal of Hydrology, 231–232, 2000, 220-229, https://doi.org/10.1016/S0022-1694(00)00196-7.
Robichaud, P. R., & Hungerford, R. D. (2000). Water repellency by laboratory burning of four northern Rocky Mountain forest soils. Journal of Hydrology, 231, 207-219, https://doi.org/10.1016/S0022-1694(00)00195-5.
Shakesby, R. A., & Doerr, S. H. (2006). Wildfire as a hydrological and geomorphological agent. Earth-Science Reviews, 74(3-4), 269-307, https://doi.org/10.1016/j.earscirev.2005.10.006.
Vandervaere, J., Vauclin, M. and Elrick, D.E. (2000), Transient Flow from Tension Infiltrometers I. The Two-Parameter Equation. Soil Sci. Soc. Am. J., 64: 1263-1272. https://doi.org/10.2136/sssaj2000.6441263x
Wieting, C., Ebel, B. A., & Singha, K. (2017). Quantifying the effects of wildfire on changes in soil properties by surface burning of soils from the Boulder Creek Critical Zone Observatory. Journal of Hydrology: Regional Studies, 13, 43-57, https://doi.org/10.1016/j.ejrh.2017.07.006.
Citation: https://doi.org/10.5194/nhess-2024-132-RC1
- AC2: 'Reply on RC1', Yuanying Li, 04 Jun 2025
  
  Thank you for your thoughtful review and valuable comments on our manuscript. We have addressed each of your comments point by point. Please find our detailed responses attached for your review.
  
  Citation: https://doi.org/10.5194/nhess-2024-132-AC2
RC2:
'Comment on nhess-2024-132', Anonymous Referee #2, 23 Apr 2025
The manuscript presents a case study on the 2015 Wye River-Jamieson Trask fire. It aims to investigate changes in soil properties after a fire and apply the measurements obtained to inform a hydrological model. Soil samples were collected at unburned sites with similar characteristics to those at nearby burned sites. In the lab, several soil tests were performed in the unburned samples and then, after burning them in a muffle furnace, in the burned samples. The hydrologic model was used to evaluate slope stability during rainfall events under unburned and burned conditions. The study is interesting and highlights the importance of informing models with real measurements. Some of the laboratory test results are very interesting, and I suggest more discussion of them should be developed. However, at this time the manuscript is a bit confusing, as it doesn’t follow a continuous line of thought. I suggest a major revisions and restructuring of the manuscript before publication.
I would restructure the paper as follows:
introduction,

Materials and methods:
2.1. Scope of the study - Study area (move also section 5.2. here)

2.2. Soil sample collection

2.3. Laboratory burning test conditions: Laboratory burning tests

2.4. Numerical method: hydrological numerical method (name?) – model description and parametrization for this specific area.

Results

Discussion

Conclusion

Line 129-191: should be in the result section.

Line 194-293: I suggest moving most of the equations to the supplementary materials (they take up a lot of space and this article is not intended to present a new model) and reducing the model descriptions to the essential information in the materials and methods sections.

Comments:
Bushfire vs wildfire: wildfire is a general term that includes bushfires, and the literature you are citing is about wildfires. I would suggest changing the term since, as you mention at line 338, your study area is a large eucalypt open forest.

In general, you could use more citation.

Laboratory burning tests: you should be more clear about the number of samples you collected in the field and the amount of repetitions of each test you did. Did you measure properties only once per sample or did you do repetitions? Are the values in Table 1 the average?

The result and discussion section is poor. As suggested, after separating it, I would develop the discussion. Use a paragraph to explain future research needed to improve your laboratory experiments and the model.

Line 21: are prevalent natural disasters: prevalent to what?

Line 23: Mediterranean: this is a climate region, which include a large part of the Western United States.

Line 49 – 62: lot of information about laboratory simulations, I suggest reducing it.

Line 64 – 68: You don’t need to go in detail on the different models used by Stanley and Gartner. You can just refer to them as citation at the end of Line 65.

Line 86: simplified hydrological approach- based numerical method: does the method/model have a name? If so, mention the name.

Line 94: erupted: it is an uncommon word, I suggest using “started”.

Line 115-116: seems part of the introduction.

Line 126-127: In caption Figure 3, information about the block soil sample measurements should also be added in the text. All soil sample measurement information should be included in the soil sample collection.

Line 138-139: description of severity should go to the study site description.

Line 58-159: should go in the discussion.

Line 178: increase the permeability: This result is interesting because there are several studies in literature that state the opposite. I suggest discussing this result in more detail in the discussion section.

Line 302 – 305: can you explain better the source of these equations and how you calculate the cohesion and the angle of shear resistance of soil.

Line 313-323: you are not presenting either methods or results. This should go in the introduction.

Line 318: clear ash within days: that ash that clogs the soil pores will not be cleared by wind (or if it does, please provide references), but it will move the ash that is in the top of the soil, which could create a permeable layer on top of the water repellent one that it is in the soil.

Line 334-364: part of this section is part of the study area description, some on a chapter regarding model parametrization.

Line 356 – 357: you reduce the soil depth in the burned areas based on root concentration? Why?

Table 2: give more information in the text on how you parametrized the model. Is it based on the laboratory analysis you did?

Line 416 – 418: you mention a burn severity map. Did you parametrize the model in the same way for moderate and high severity burned areas? The impact on the soil is very different between the two severities. In addition, which severity are you simulating in the lab experiment?

Line 434 – 438: the conclusions should be rewritten. You shouldn’t repeat part of the introduction and methods in the conclusion.

Figures
Figure 1: I suggest adding a map of Australia because not all the readers would identify the location.

Figure 1: I suggest adding a soil burn severity map (or the burn severity map from dNBR if you don’t have on site measurements).

Refer figure 12 and 13 in the text and be consistent with the word use: Fig. or Figure.
Citation: https://doi.org/10.5194/nhess-2024-132-RC2
- AC1: 'Reply on RC2', Yuanying Li, 04 Jun 2025
  
  Thank you for your thoughtful review and valuable comments on our manuscript. We have addressed each of your comments point by point. Please find our detailed responses attached for your review.
  
  Citation: https://doi.org/10.5194/nhess-2024-132-AC1

Yuanying Li, Akihiko Wakai, and Susanga Costa

Viewed

Total article views: 1,122 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
713	157	252	1,122	30	49

HTML: 713
PDF: 157
XML: 252
Total: 1,122
BibTeX: 30
EndNote: 49

Views and downloads (calculated since 18 Sep 2024)

Month	HTML	PDF	XML	Total
Sep 2024	54	13	26	93
Oct 2024	37	4	73	114
Nov 2024	32	8	48	88
Dec 2024	43	16	51	110
Jan 2025	21	20	40	81
Feb 2025	12	7	0	19
Mar 2025	15	11	1	27
Apr 2025	26	13	3	42
May 2025	16	8	1	25
Jun 2025	26	11	6	43
Jul 2025	25	6	0	31
Aug 2025	60	16	1	77
Sep 2025	332	12	2	346
Oct 2025	14	12	0	26

Cumulative views and downloads (calculated since 18 Sep 2024)

Month	HTML	PDF	XML	Total
Sep 2024	54	13	26	93
Oct 2024	37	4	73	114
Nov 2024	32	8	48	88
Dec 2024	43	16	51	110
Jan 2025	21	20	40	81
Feb 2025	12	7	0	19
Mar 2025	15	11	1	27
Apr 2025	26	13	3	42
May 2025	16	8	1	25
Jun 2025	26	11	6	43
Jul 2025	25	6	0	31
Aug 2025	60	16	1	77
Sep 2025	332	12	2	346
Oct 2025	14	12	0	26

Viewed (geographical distribution)

Total article views: 1,072 (including HTML, PDF, and XML) Thereof 1,072 with geography defined and 0 with unknown origin.

Country	#	Views	%

Cited

Latest update: 16 Oct 2025

Short summary

Landslides during rainfall in burned areas have become more frequent and widespread. This paper employs laboratory burning tests to evaluate the effects of bushfires on soil properties and to compare slope stability during rainfall before and after bushfires. These findings enhance the understanding of landslide-triggering mechanisms in post-fire slopes and provide insights for mapping landslide susceptibility in bushfire-prone regions.


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

Bushfire effects on soil properties and post-fire slope stability: the case of the 2015 Wye River-Jamieson Track bushfire

Viewed

Viewed (geographical distribution)

Cited

1 citations as recorded by crossref.