the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 02 Mar 2026
Forecasting coastal dune mobility: A logistic regression model driven by meteorological data and climate indices
Abstract. Predicting dune mobility under changing climatic conditions remains a challenge in aeolian geomorphology, particularly in data-scarce regions. This study presents a novel application of binomial logistic regression to forecast dune activation and migration using readily available meteorological data. We combine established dune mobility indices (Tsoar and Lancaster) into a new integrated index (TsoLa) and evaluate its performance against observed dune migration rates derived from satellite imagery. The model incorporates wind speed, precipitation, and the Southern Annular Mode (SAM) as predictors, achieving robust predictive accuracy (AUC > 0.75) for two distinct coastal dune fields in NE Patagonia, Argentina. Our results demonstrate that even with standard climatic inputs, logistic regression can effectively identify periods of dune activity, offering a low-cost tool for coastal management. The approach is transferable to other aeolian systems, providing a framework for assessing dune dynamics under current and future climate scenarios.
- Preprint
(1969 KB) - Metadata XML
-
Supplement
(41 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
CC1: 'RC Comment on egusphere-2026-377', Graziela Miot da Silva, 31 Mar 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2026/egusphere-2026-377/egusphere-2026-377-CC1-supplement.pdfCitation: https://doi.org/
10.5194/egusphere-2026-377-CC1 -
AC1: 'Reply on CC1', Mauricio Toffani, 07 Apr 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2026/egusphere-2026-377/egusphere-2026-377-AC1-supplement.pdf
-
AC1: 'Reply on CC1', Mauricio Toffani, 07 Apr 2026
-
CC2: 'Comment on egusphere-2026-377', Thomas Smyth, 27 Apr 2026
Publisher’s note: this comment is a copy of RC1 and its content was therefore removed on 30 April 2026.
Citation: https://doi.org/10.5194/egusphere-2026-377-CC2 -
RC1: 'Comment on egusphere-2026-377', Thomas Smyth, 29 Apr 2026
This is an interesting and relevant study, however, several areas require clarification to improve robustness and reproducibility. More detail is needed on the processing and analysis of aerial imagery (e.g. data sources, spatial resolution, and dune delineation). The manuscript would also benefit from the use of subsection headings to improve structure and readability. Finally, some key concepts and interpretations (e.g. model transferability and the distinction between wind intensity and Resultant Drift Potential) would benefit from clearer definition and supporting evidence.
L21: There is not enough evidence within the article to support the claim that the model is “transferable to other aeolian systems,” particularly as the manuscript concludes by suggesting that future research should test model performance in “diverse climatic and geomorphic settings.” Model transferability typically requires explicit testing, and this statement may therefore be overstated.
L36: A brief explanation of binomial logistic regression would be helpful, as this methodology underpins much of the analysis.
L184 and throughout: The use of subsection headings is recommended to improve readability and overall structure.
L187: Please refer to Figure 1, where the meteorological stations are mapped.
Methods: Further detail is required regarding the processing and analysis of aerial imagery. For example:
- L238: Please clarify which datasets correspond to satellite imagery and what is included under “aerial” data.
- L239: How was the spatial resolution of the Google Earth imagery defined or estimated?
- L238: Was migration assessed only for unvegetated dunes? How were dune areas delineated? Further detail would improve clarity and reproducibility.
- L245: How was a “dune front” defined? An illustrative figure showing how dune fronts were identified and how migration rates were calculated would greatly assist the reader in understanding and reproducing the methodology.
L309: Please provide a reference for these “established” procedures.
L356 to Table 2: A brief reintroduction of TsoLa would be helpful here, as it was last defined in L237.
Figures 9 and 10: Please clarify what values of 0 and 1 represent in the classification, particularly as the term “classification” is not clearly introduced in the Methods.
L458: No results relating to vegetation change or landscape “greening” are presented. Additional evidence or justification would be helpful to support this statement.
L462: A fuller description of the “worm-like” dunes is needed. An illustrative figure would also help the reader interpret and apply this concept to other sites.
L462–L464: Sand supply does not appear to have been included in the analysis. Additional evidence or justification would be helpful to support this interpretation.
L463–L464: The distinction between “wind intensity” and Resultant Drift Potential (RDP) would benefit from further clarification. As RDP inherently incorporates wind speed (i.e. intensity) alongside directional variability, treating “wind intensity” as a separate control introduces some ambiguity. It would be helpful to clarify whether this refers to aspects of the wind regime not captured by RDP or reflects overlap between these terms.
L466–L468: It may be helpful to frame this discussion in the context of Haim Tsoar (2005), who describes physical–biological interactions in dune systems using the concept of hysteresis, whereby dunes become stabilised when wind power is sufficiently low.
Tsoar, H. (2005). Sand dunes mobility and stability in relation to climate. Physica A: Statistical Mechanics and its Applications, 357(1), 50-56.
Citation: https://doi.org/10.5194/egusphere-2026-377-RC1 -
RC2: 'Comment on egusphere-2026-377', Graziela Miot da Silva, 04 May 2026
The paper presents a study that combines two well stablished dune mobility indexes to create a new one (TsoLa), and relates this to dune mobility in Patagonia. The paper is well written and addresses the predictability of dune mobility which is an important research gap, particularly in face of climate change.
In my view it remains unclear what limitations of the existing Tsoar and Lancaster indices this new formulation is intended to address, and what additional insight it provides beyond these established approaches. Tsoar and Lancaster’s indexes are based on similar parameters included in the model proposed here. A more explicit justification is needed, ideally supported by a direct comparison between the existing indices and the proposed one, to demonstrate whether and how the new index improves the assessment of dune mobility.
A few other specific comments:
Figure 1 – the wind roses suggest that NW and S/SW winds are dominant, but the RDD indicates net dune migration towards the E. There seems to be a lack of W in the wind roses to reflect an easterly net direction of dune migration. A legend showing which speed classes are represented by the colours could be useful here.
Table 1 – did the authors consider veg and non veg parts in the calculation of the dunefield dimensions?
Line 316 – Only 30% of the winds exceeded the threshold, but Figure 5 seems to suggest that winds above ~6m/s are quite frequent in the record. This needs clarification, or perhaps change the colours in the wind roses as changes in shades of blue are too subtle and hard to see. More information about the wind data would be helpful – e.g. was the met station data provided classified in any way?
Figure 6 – The very high values in the early 1990s seem a bit inconsistent with the rest of the record. Sometimes older datasets can have quality or methodological issues, have the authors checked whether these values are real, or could they be artefacts? A bit more detail on data quality and consistency over time would help here.
Figure 8 – The meaning of this figure is a bit unclear, could the authors clarify what the axes represent? A more intuitive explanation in the caption or text would really help the reader.
Similarly, Figures 9 and 10 need clearer legends, as it is currently difficult to interpret what is being shown
Line 454 – DPs are only a function of wind and grain size, they do not reflect longer daylight hours etc.
Line 429 – the authors claim that there is a progressive greening of the landscape associated with a decline in dune migration rates, but no data is offered to verify this.
Line 489 – could the authors clarify how the index was able to particularly identify stable dune conditions if the dunefields have been active/migrating over the years? Figure 11 suggests that the dunefields have been migrating over time, although at decreasing rates. It’s not really clear what extra insight the new index is adding here—would we see a similar pattern just using RDPs? Could the index be used to predict what will happen in the long term, e.g. with changes in the SAM? After reading the introduction, I thought that was the direction that the paper was going to take. Similarly, a more complete discussion about the applicability of this index in other areas of the world would be useful as this was one of the study’s objectives.
Dos the TsoLa index define a threshold above which dunes become mobile (such as in Fig 7)? And does it have similar limitations to existing indices (e.g. only working when rainfall is above ~50 mm)
Table 3 - legend must include that dunefield migration are rates in m/year. What are “Dune 1 to 5” and “Dune 1 to 6”?
-
EC1: 'Comment on egusphere-2026-377', Andreas Baas, 15 May 2026
Dear Authors,
We have now received two comprehensive reviews of the manuscript and we invite you to submit a revised version that addresses the questions and concerns raised (please refer to the instructions sent to you via email).
Kind regards,
Andreas Baas (handling editor)
Citation: https://doi.org/10.5194/egusphere-2026-377-EC1
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 763 | 282 | 82 | 1,127 | 114 | 59 | 101 |
- HTML: 763
- PDF: 282
- XML: 82
- Total: 1,127
- Supplement: 114
- BibTeX: 59
- EndNote: 101
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1