the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 19 Feb 2026
Spatially variable wildfire impacts on sediment mobilization: event-scale evidence from eastern Australian catchments
Abstract. Wildfires may profoundly affect forested catchments by altering hydrological processes, soil properties, and sediment dynamics. While post-fire increases in water quality constituents are well documented, the underlying divers for such changes remain poorly understood. This study examines how fire and short-term hydrologic conditions interact to shape sediment dynamics using a novel Bayesian Hierarchical modelling framework. The analysis is at event-scale based on multi-year, high-frequency turbidity and streamflow data collected from 14 forested catchments in eastern Australia, which were burned to varying severities during the 2019/2020 Black Summer fires and affected by subsequent floods. Further, we explored how these effects vary across catchments with the extent of burning and other catchment characteristics.
Model results show that severe burning had a clear steepening effect on the slopes of event concentration-discharge (C-Q) in 5 out of the 10 severely burnt catchments studied, indicating enhanced sediment mobilization following extreme burning. The influence of short-term hydrologic conditions on C-Q slopes was comparatively minor, suggesting that fire effects dominated post-disturbance sediment responses. The magnitude of fire effects (as post-fire changes in the C-Q slopes) did not always scale directly with the proportion of catchment burnt, which also seems to be driven by the location of extreme burning and forest types. These findings provide a large-scale, multi-catchment understanding of post-fire sediment mobilization mechanisms, which can inform future improvement of modelling and management of sediment in fire-affected forested landscapes.
- Preprint
(1818 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2025-6244', Anonymous Referee #1, 08 May 2026
-
RC2: 'Comment on egusphere-2025-6244', Anonymous Referee #2, 24 Aug 2026
Review Summary
The reviewed study presents the results of an investigation of fire impacts on water and fluvial sediment dynamics in 14 watersheds impacted by the 2019/20 Black Summer fires in Australia. The authors analyzed high-resolution turbidity and water discharge data collected before and after the fire using Bayesian Hierarchical modelling to investigate the potential role of fire, watershed characteristics, and hydrologic conditions at the stormflow event scale. Overall, this work is novel and of potential interest to the broad international readership of HESS, and the scientific approach appears to be generally sound. I have reservations, however, regarding some of the authors’ choices when structuring their analysis, a general disregard of the relationship between turbidity, fluvial suspended sediment concentration, and flux, as well as the lack of in-depth discussion of their results in terms of processes and the vast body of literature on this topic produced in the last 30+ years, and some of their interpretations. More minor quibbles included some structural choices, such as introductory material in the already sparse discussion section. For these reasons I recommend major revisions. See detailed commentary below.
Detailed Comments
Line Comment
23 ‘location’ is too vague – be more specific
30 This sentence is poorly structured – revise
35 This appears to be your only mention of sediment flux in the entire paper. Why do we care about turbidity – only as a water quality parameter or as an analog for suspended sediment and the mass flux of solids through watershed systems? You appear to be framing this work as a post-fire water quality paper, but much of the post-fire fluvial sediment dynamics work has considered this issue through the lens of earth surface processes, and hence with a focus on sediment flux.
38 This statement is incorrect and also a very cursory treatment of the topic of fire-recovery timescales. Although perhaps most watersheds will return toward ‘baseline’ hydrologic and fluvial constituent dynamics within 5 years after fire, there are many exceptions. I suggest you revisit the literature.
57-58 There have been more of these studies, particularly over the past 10 years. Revisit the literature to setup your study more appropriately here in the introduction, and to consider in the context of your results in the discussion.
72 Must be a typo – a proportion of 0.1 or 10% rather than 0.1% of catchment area, right?
84 Okay, so not a typo? How is 0.1% of catchment burned at high severity a relevant threshold? Seems vanishingly small.
Figure 1 It would be more informative to have a color-shade map illustrating the distribution of burn severities, rather than just extreme severity. An inset map of the continent/region would be helpful to orient the international audience of this journal. Also consider adding graticules.
Table 1 This is pretty confusing. The catchments that you have classified as extremely burnt have 43.1-99.2% unburned area. This seems like a poorly structured classification system.
146 I don’t agree with your conflation of turbidity and suspended sediment concentration. This may seem semantic, but although turbidity and SSC are often highly correlated, they are not the same thing. The relationship between turbidity and SSC also can change, for example after watershed disturbances such as wildfire
153 Why did you decide to only evaluate the role of your selected predictors on slope? What about offset (intercept)?
154 Intense wildfire, or severe?
161 I don’t think that you are actually evaluating ‘sediment export.’ Rather, your approach is investigating the effect of certain variables on the slope of event-scale turbidity-Q relationships.
165 Why wasn’t time since fire included, or number of events since fire, or some cumulative metric such as total water discharged since fire?
169-172 The QS,lastvolume & QS,thisvolume variables appear to be in terms of total water volume discharged for a given event, but you refer to their role as an indicator previous and current ‘sediment mobilization’, by which I think you mean mass flux? I agree that antecedent stormflow events can act to flush a system or prime it with more sediment depending on sediment supply and transport details Why preload this with specific interpretation? I suggest you present more on the nuanced role that previous and current stormflow event magnitude can play in suspended sediment concentration – discharge behavior.
Table 3. Your choice of watershed variables has some notable gaps – what about geology, geomorphic metrics (e.g., slope), total burn area, burn area distance from the channel…
Figure 3 Please change figure symbols to better illustrate both pre- and post-fire values. The current version obscures pre-fire values by overlaying dense post-fire data clouds.
Section 4.1 This overview of fire impacts on hillslope water and sediment dynamics is background material and more suitable for the introduction rather than the discussion section.
Section 4.2 The discussion is very brief and does not adequately explore the study findings in the context of the potential mechanisms at play and findings from previous literature.
Citation: https://doi.org/10.5194/egusphere-2025-6244-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 868 | 385 | 69 | 1,322 | 78 | 112 |
- HTML: 868
- PDF: 385
- XML: 69
- Total: 1,322
- BibTeX: 78
- EndNote: 112
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The manuscript titled “Spatially variable wildfire impacts on sediment mobilization: event-scale evidence from eastern Australian catchments” studies the impact of the fire on C-Q slope and hypnotize that the extreme burning and recent hydro-climatic conditions drive the sediment mobilization in recently burned watersheds. It is an interesting study and the authors have a valuable post-fire dataset. After reading it, however, I think the manuscript requires substantial revision between it can be accepted for publication. The main concerns are presented below, followed by detailed comments. I hope this revision will help guide a stronger analysis.
C-Q analysis – methodology used: The C-Q analysis, in particular on the selection of slope as explanatory variable, would benefit a more detailed methodological justification. Hysteresis index and flushing index are standard metrics used to compare C-Q relationships across sites and studies. The manuscript currently focuses on the slope of the relationship, but this choice is not clearly explained. The authors should justify why the slope is important and which are the mechanisms that could create a change, or they should consider adopting hysteresis and flushing indexes to facilitate comparison with the literature. The authors can Jing et al. (2024) (https:// doi.org/10.1029/2024WR037216) for reference.
Fire severity – soil burn severity: Fire severity (or soil burn severity) measures the magnitude of the impacts of fires on vegetation (and soil). Fire severity is easily obtained by remote sensing images, while soil burn severity needs in situ measurements. The threshold the authors use (0.1% extreme burn), which seems a very low threshold. They use Rhoades et al. (2011) and Marcotte (2024) which don’t mention this threshold in their study. I suggest using fire severity – in particular percentage of high severity – as a variable.
Results: The results section is currently confusing and should be reorganized. In addition, the authors make several interpretative claims (a variable “increase” or “play a critical role”) without presenting statistical evidence. A form of statistical analysis is necessary to support those statements. The authors should conduct appropriate statistical tests and clearly present the results, acknowledging any limitations of the approach.
Discussion: The discussion is currently short and presents only a handful of references. The authors should expand the discussion and interpret their findings with the existing literature on post-fire impacts on sediment yield, water balance, and C-Q dynamics (right now they are citing only three references). The Discussion should also address what the results reveal about the hydrological and erosion consequences of fire, how these compare with findings from similar catchments internationally, and what mechanistic explanations can be offered.
Specific comments: