the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 24 Sep 2025
The 15 September 2022 floods in northern Marche (Central Italy): disaster analysis, case studies and mitigation strategies for geomorphological- hydraulic risk
Abstract. On September 15th and 16th, 2022, a large area of the Marche region in central Italy experienced an exceptionally heavy rainfall event, with nearly 420 mm/m² of rain falling in just six hours. The intense rains, in addition to causing 13 fatalities, triggered a large number of landslides in the mountain areas and flood events, mainly distributed along the valleys of the hydrographic basins of Metauro, Cesano, Misa, and Esino Rivers. The physiographic setting of the territory and the poor maintenance of both the main and secondary hydrographic network, often insufficient or entirely absent, exacerbated an already exceptional event. Although extraordinary, the natural event affected an area already hit by intense meteoric events in the past, the most recent of which occurred just eight years earlier, in 2014.
This study presents the results of the systematic and detailed surveys conducted in several sites affected by the storm, also providing detailed case studies. These surveys highlighted the critical issues detected during the disaster and identified appropriate intervention measures for reducing hydraulic risk in the Region. Many of these measures are innovative and will serve as guidelines for future land-use planning and for improving public education and awareness in flood-prone areas.
- Preprint
(10671 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-4405', Anonymous Referee #1, 09 Oct 2025
Overall AssessmentThis manuscript is a highly valuable, systematic disaster analysis of the catastrophic September 2022 floods in the Marche region, Central Italy, which caused thirteen fatalities and extensive damage. The study rigorously documents the event, identifying critical geomorphological and hydraulic issues through extensive field surveys, historical cartography, and collaboration with regional authorities.The central strength lies in demonstrating how extreme meteorological conditions (classified as an "outlier event" with over 1000-year return times in some locations) interacted disastrously with long-term anthropogenic modifications (e.g., channelizing, burying rivers, and neglecting maintenance). The paper provides a clear path forward for immediate local fixes and critical long-term systemic changes.Strengths1. Detailed Empirical Data Collection: The study is grounded in systematic and detailed on-field surveys across multiple affected towns, including Cantiano, Sassoferrato, and Senigallia. This generated maps of critical issues and specific suggested interventions.2. Multidisciplinary Approach: The methodology effectively employed a multi-source approach, drawing on meteorological analysis, field geomorphological mapping, hydrological/hydraulic analysis, and historical records, such as the Gregorian historical cadastre (1835).3. Hydrogeomorphological Insight: The paper expertly differentiates the critical issues between mountain environments (where high slope and narrow riverbeds led to intense erosive phenomena and landslides) and foothill contexts (where channelized rivers and damaged earthen embankments caused widespread flooding across broad alluvial plains).4. Policy Relevance: The research was directly functional, emerging from an agreement with the Marche Regional Administration to identify priority risk mitigation interventions, making the findings immediately applicable to reconstruction efforts and risk planning.Points for Improvement and Discussion1. Framing of Mitigation Strategies: The authors clearly state that hydraulic restoration alone (reinforcing banks, removing sediment, rebuilding bridges) will not be sufficient to prevent future flooding. They identify the definitive long-term solution as the relocation of structures from flood areas and the creation of natural expansion/retention basins (requiring expropriations). This necessity for avoidance (protecting natural flood areas) and elimination (relocation/managed retreat), should be explicitly highlighted in the Discussion using established concepts like the Flood Adaptation Hierarchy (FAH). The structural fixes proposed (e.g., bridge enlargement, gabions) should be framed explicitly as lower-priority accommodations/defenses (Tiers 3–6 of the FAH) that buy time, rather than long-term solutions, as acknowledged by the authors (see Peck et al., 2022)Suggestion: Directly compare the proposed interventions (relocation vs. restoration) to the FAH to strengthen the paper’s contribution to modern risk management theory.2. Hydrological Modeling Details: While the text references hydrologic-hydraulic modeling results (e.g., calculating maximum flood discharge using Giandotti's method) and mentions that future work will include numerical modeling, the Methods section detailing the analytical sequence is primarily descriptive of data inventory. To enhance the methodological rigor for an engineering/geoscience audience, the authors could clarify how the quantitative hydrological-hydraulic models (often involving software like HEC-HMS and HEC-RAS, used in similar hydrogeomorphological studies - see Lombana et al., 2024) were specifically utilized to reach conclusions, such as the insufficiency of local fixes.3. Integration of Geomorphological Processes in Mitigation: The text emphasizes that the geomorphological effects must be monitored, noting that erosion/deposition processes (e.g., debris flow deposits, river point bar migration) were widespread. Future discussion could link the need for "natural expansion" areas not just to water attenuation, but also to managing sediment transport and geomorphic evolution, which is a critical concern when mitigating floods with structures like leaky dams in headwater systems. (see Wolstenholme et al., 2025)4. Systemic Deficiencies and Vulnerability: The discussion correctly identifies the need to upgrade monitoring networks (hydrometric instrumentation was washed away) and improve public awareness/early warning systems. These deficiencies underscore the systemic challenges in disaster-prone regions and align with systematic reviews identifying resource provision and technology gaps (EWS, communication networks) as critical pillars of flood mitigation for critical infrastructure. (see Arefi et al., 2025).The recommendation for this manuscript is for Minor to moderate revision focusing on integrating the policy recommendations into modern risk management frameworks.ReferencesPeck, Andrew J., Stevie L. Adams, Andrea Armstrong, et al. “A New Framework for Flood Adaptation: Introducing the Flood Adaptation Hierarchy.” Ecology and Society 27, no. 4 (2022). https://doi.org/10.5751/ES-13544-270405.Lombana, Lorena, Biswa Bhattacharya, Leonardo Alfonso, and Antonio Martínez-Graña. “Hydrogeomorphological Approach for Flood Analyses at High- Detailed Scale: Narrow Rivers with Broad Complex Alluvial Plains.” CATENA 242 (July 2024): 108081. https://doi.org/10.1016/j.catena.2024.108081.Wolstenholme, Joshua M., Christopher J. Skinner, David Milan, Robert E. Thomas, and Daniel R. Parsons. “Hydro-Geomorphological Modelling of Leaky Wooden Dam Efficacy from Reach to Catchment Scale with CAESAR-Lisflood 1.9j.” Geoscientific Model Development 18, no. 5 (2025): 1395–411. https://doi.org/10.5194/gmd-18-1395-2025.Arefi, Farhad, Asghar Tavan, Seyed Mobin Moradi, Salman Daneshi, and Hojjat Farahmandnia. “Identifying Challenges and Future Directions of Flood Hazards Mitigation Strategies in Health Facilities: A Systematic Literature Review.” BMC Emergency Medicine 25, no. 1 (2025): 174. https://doi.org/10.1186/s12873-025-01339-0.Citation: https://doi.org/
10.5194/egusphere-2025-4405-RC1 -
AC1: 'Reply on RC1', Fabrizio Bendia, 20 Oct 2025
1) Your suggestion is appropriate and linking the article to the Flood Adaptation Hierarchy (FAH) would certainly have improved the quality of the work by helping to define intervention strategies from a qualitative point of view. We could also have referred to the United Nations Sustainable Development Goals.
In any case, we provide a possible integration proposal to be inserted around line 827:
The structural fixes proposed can be considered a traditional approach to handling flood risk (which, for instance, includes bridge enlargement, gabions, reinforcement of levees, etc.). These short-term solutions are no longer sufficient to handle the increasing rainfall, sea-level rise, and continued development in floodplains that are making flooding worse. Furthermore, this strategy may perpetuate social inequities, leaving those who are the most vulnerable at greatest risk. The Flood Adaptation Hierarchy “FAH” (Peck et al., 2022) proposed a list of solutions examining decades of research and practice from around the United States and developing a way to make decisions about flood risk solutions.
Protecting and restoring natural floodplains means not building on them, removing human alterations and restoring water flows (tier 1 of FAH).
It is obvious that, to do this, in many cases it will be necessary to intervene through expropriations, which may sometimes include the loss of crops or the inability to use the real estate above the affected land parcels. Tier 2 of FAH identifies as positive and encourages the adoption of such practices.
The other structural solutions proposed (e.g., bridge widening, gabions) are classified as low-priority adaptation/defence measures (FAH levels 3-6) that allow time to be gained, rather than as long-term solutions (Peck et al., 2022).
2) In this work, the results of the hydrological-hydraulic modelling have been cited without providing detailed information on the methodology and procedures adopted. The project document referenced by the manuscript was prepared in 2016 to define possible intervention strategies in response to a flood event. Immediately after the event of 15 September 2022, new studies were conducted using a combination of hydrological and hydraulic models (e.g., HEC-HMS and HEC-RAS) to reconstruct the actual event and plan initial environmental remediation interventions. These ongoing studies are redefining the maximum flood discharge for specific return times and will enable us, based on hydraulic simulations appropriately calibrated with actual event data, to delineate the potentially floodable areas and assess the effectiveness of existing infrastructure (bridges, weirs, etc.) and/or proposed interventions. Detailed information on these aspects will be included in this manuscript.
3) The management of sediments produced and released along slopes and watercourses has been a much-debated issue, which has seen our research group liaise with public administrations and other technicians. While the enormous quantity of debris produced may represent an obstacle (landslide deposits, obstruction of river sections, and the possibility of its remobilisation and re-triggering of processes), on the other hand, it can also represent a resource, as the material (mainly limestone) can be an important asset. It is available in large quantities, already “cut” and ready to be used in the reprofiling or reinforcement of embankments, in construction processes or sold on the market. Instead of having to dig it out in mining plants.
We have not dealt with this topic to remain relevant to geomorphological processes without straying into economic and administrative issues. In fact, several bureaucratic problems have been encountered which, in some situations, have slowed down or inhibited the removal of this material and its reuse.
4) The presence of an early warning system along the vulnerated watercourses would have greatly reduced the risk. The Marche region is characterised by a mountainous area and a foothill area. Almost all of the rainfall fell in the mountainous area, and the flood reached the areas near the coast (where almost all of the deaths occurred) approximately three hours later. The real question is whether there could have been time to close access to critical hydraulic crossings and, in general, to the vulnerable areas. It is hoped that decision-makers will take these shortcomings into account and that the hydraulic risk management system will be strengthened, starting with these watercourses. For our part, we have participated and continue to participate in public meetings, conferences, seminars, activities in schools and educational on-field activities with schoolchildren, professionals, local administrators and citizens precisely because we believe in the importance of communicating risk and raising awareness among the population and administrators.
The authors are willing to make the requested changes.
Citation: https://doi.org/10.5194/egusphere-2025-4405-AC1
-
AC1: 'Reply on RC1', Fabrizio Bendia, 20 Oct 2025
-
RC2: 'Comment on egusphere-2025-4405', Anonymous Referee #2, 29 Oct 2025
The paper deal with the study of the 15 September 2022 floods in northern Marche (Italy). Although the authors demonstrated that there was a lot of work behind this paper, It is opinion of this reviewer that the manuscript need a very strong revision process before to be suitable for publication to NHESS.
It seems that the paper at the moment lacks of originality and it is mostly based on geomorphological (conventional) surveys and the proposal of mitigation measures. The most interesting and innovative parts should have been the analysis of LiDAR and satellite data e numerical simulations. However, none of these techniques have been described and it is not clear how (and if) have been used. There are not images and/or GIS analysis of LiDAR data. The numerical modelling is not even described. It is not possible to understand what parameters have been used/included in the modelling, how they have been calibrated and how the results validated. Also the use of satellite data is not clear. How the authors defined damaged areas and extend of the event?
The results of the work (including the suggestion of mitigation measures) should have been based on the outputs gathered from these “more innovative” techniques (as stated by authors) and geomorphological survey. However, at the moment, only on field geomorphological surveys seems to have been used.
Furthermore, the use of “risk” in the title is not really appropriate. There is not mention in the manuscript about the parameters necessary for defining the “risk”.
I suggest to re-structure the paper focusing on all these aspects. This would make the paper strong and of interest for the scientific community. More detailed comments of the manuscript can be find below:
Table 3 is impossible to be read.
Fig 12. Please provide a legend explaining what we are looking at in the QGis screen. The layers in the QGIS are in Italian and often they are acronym (with no explanation).
Line 285-290. It is not very clear how satellites data have been used. How the extend of the event and the expect level of damage were extracted?
Fig 14. The two insets are low resolution.
Chapter 2.3.4.1 and 2.3.4.2 are both geomorphological setting?
Line 690 – 695. This is not clear. Is this the difference between two profiles (pre and post-event) or just an interpretation of the LiDAR data? Furthermore, why LiDAR data is never shown? The author mention LiDAR as one of the innovative techniques but there are not Figures or analysis based on LiDAR data. For example why the authors didn’t perform an analysis of differences between the recent LiDAR (post event) and older ones (pre event)?
Line 755: we have here material and methods again??
Line 745. The authors mention about hydrologic-hydraulic modelling in the introduction and here. But there is not evidence in the paper about the modelling. Again, modelling should have been one of the innovative part of the research (as stated by the authors in the introduction) but there are not evidence of the modelling or of the development of the models. If the modelling has been done, what parameters have been used? How they have been calibrated? How the results validated? The is no mention at all about all this aspects… jus a sentence in all the paper (The results of the hydrologic-hydraulic modelling showed that the city of Senigallia presents important criticalities represented by the bridges in the historical centre.).
Line 775. Here (and in the conclusion) the authors mention again about numerical modelling: “Thanks to hydrological-hydraulic modelling, it is now possible to know detailed flow rates and hydrometric levels in each section of the investigated catchment areas (aimed at dimensioning the reconstruction of bridges and other risk reduction interventions).” However, we can’t see any proof of this result in the paper. It is not even clear if the numerical simulation have been done.
Citation: https://doi.org/10.5194/egusphere-2025-4405-RC2 -
AC2: 'Reply on RC2', Fabrizio Bendia, 31 Oct 2025
Dear reviewer,
Thank you for your appreciation and for considering the final publication of this article after strong revisions.
We agree with your comments; the innovative part could be further developed, compatibly with the time allocated for the review phase.
In particular, we can supplement the article by specifying the precise use made of the lidar data. This data was useful for assessing the variation in the sinuosity of certain sections of the watercourses concerned; for understanding the geometric changes undergone by the embankments after the event; and, above all, for calculating the volumes of material to be removed to restore the hydraulic section of the final stretch of the Misa River, close to the city of Senigallia.
With regard to satellite images, extensive use was made of post-event maps from the European Copernicus programme in order to ascertain the actual flooded areas.
With regard to numerical modelling, we can supplement the article with the data in our possession.Citation: https://doi.org/10.5194/egusphere-2025-4405-AC2
-
AC2: 'Reply on RC2', Fabrizio Bendia, 31 Oct 2025
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 690 | 148 | 13 | 851 | 5 | 6 |
- HTML: 690
- PDF: 148
- XML: 13
- Total: 851
- BibTeX: 5
- EndNote: 6
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1