Periodic Reporting for period 1 - MODKARST (Extreme flood events in western Mediterranean: integrating numerical MODelling and flood records in KARST systems)
Reporting period: 2023-09-16 to 2025-11-15
Changes in rainfall patterns are a direct consequence of the current climate change. Climate projections indicate an intensification of extreme rainfall events (and floods), which will directly affect social, ecological, and economic systems. Thus, one of the greatest challenges of climate science is to understand, model, and predict the variability of floods both in temporal and spatial terms in the context of the present-day global warming. Nevertheless, the uncertainties in projected rainfall at regional scale are still high, and even higher in Mediterranean areas where the climate is characterized by extreme and sudden rainfall events. The instrumental record is too short to correctly estimate return floods periods. Thus, geological records are required to better understand the long-term variability, at millennial to decadal scales, of natural extreme flood events. Among the most important flood archives, including riverine and lake sediments, stalagmites represent one of the best geological records (i.e. accurate chronology, wide temporal range and good preservation) to study past flood frequency. However, stalagmites have not yet been properly exploited for this purpose. MODKARST will provide a unique opportunity to develop a quantitative flood database for the Western Mediterranean realm based on speleothems. To do so, the action plans to infer past flood events from the last 18,000 years based on detrital layers (indicative of flood events) recorded in stalagmites from 5 different caves. This data in combination with karst hydraulic models and water-level monitoring will allow to quantify past extreme flood events. MODKARST will help to better understand the relation among flood recurrence and climate changes, and will shed light on how to better predict the variability of floods in the context of present-day global warming.
The MODKARST project aimed to reconstruct the long-term variability of extreme flood events in the Western Mediterranean by integrating numerical modelling and speleothem-based flood records in karst systems. Despite the early termination of the fellowship, substantial scientific progress was achieved, particularly within Work Package 1 (WP1).
In WP1, the team compiled and organized extensive environmental datasets from selected caves, including water level and discharge measurements, topographic and geological data. These were used to construct 3D cave models and implement hydrological simulations using the SWMM (Storm Water Management Model) software. Initial simulations revealed that current cave geometries and flow conditions could not reproduce the flood levels observed in speleothem records. This discrepancy led to the development of two distinct models: one reflecting the current cave configuration, and another incorporating geomorphological changes inferred from sedimentary evidence. To constrain the timing of these changes, OSL (Optically Stimulated Luminescence) dating was applied to clastic deposits.
A major achievement was the implementation of a flood alert system in the Las Güixas show cave, based on real-time environmental monitoring. This system significantly improved cave management and visitor safety, especially given the cave’s high sensitivity to intense rainfall events. The monitoring revealed that flash floods can occur within 8–12 hours after heavy precipitation, and that CO2 concentrations inside the cave can spike due to rapid degassing during such events. These findings were crucial for adjusting visitor access and planning emergency responses.
Although WP2 and WP3 were not fully implemented, preliminary work was conducted. In Ostolo Cave, flood layers were identified in seven stalagmites spanning the last 19,000 years. U-Th dating and layer counting revealed seven distinct periods of increased flood frequency. In B1 Cave, another stalagmite showed nine flood-rich periods over the last 4,200 years, with mineralogical and isotopic data suggesting a strong link between flood frequency and climatic humidity. These findings underscore the importance of geomorphological evolution in interpreting paleoflood records and highlight the need for careful calibration of hydrological models in karst environments.
In WP1, the team compiled and organized extensive environmental datasets from selected caves, including water level and discharge measurements, topographic and geological data. These were used to construct 3D cave models and implement hydrological simulations using the SWMM (Storm Water Management Model) software. Initial simulations revealed that current cave geometries and flow conditions could not reproduce the flood levels observed in speleothem records. This discrepancy led to the development of two distinct models: one reflecting the current cave configuration, and another incorporating geomorphological changes inferred from sedimentary evidence. To constrain the timing of these changes, OSL (Optically Stimulated Luminescence) dating was applied to clastic deposits.
A major achievement was the implementation of a flood alert system in the Las Güixas show cave, based on real-time environmental monitoring. This system significantly improved cave management and visitor safety, especially given the cave’s high sensitivity to intense rainfall events. The monitoring revealed that flash floods can occur within 8–12 hours after heavy precipitation, and that CO2 concentrations inside the cave can spike due to rapid degassing during such events. These findings were crucial for adjusting visitor access and planning emergency responses.
Although WP2 and WP3 were not fully implemented, preliminary work was conducted. In Ostolo Cave, flood layers were identified in seven stalagmites spanning the last 19,000 years. U-Th dating and layer counting revealed seven distinct periods of increased flood frequency. In B1 Cave, another stalagmite showed nine flood-rich periods over the last 4,200 years, with mineralogical and isotopic data suggesting a strong link between flood frequency and climatic humidity. These findings underscore the importance of geomorphological evolution in interpreting paleoflood records and highlight the need for careful calibration of hydrological models in karst environments.
MODKARST introduced a novel methodological framework by combining numerical hydraulic modelling with paleoflood reconstructions from speleothems in karst systems—an approach rarely applied in this context. This integration allowed for a more nuanced understanding of how cave morphology and sedimentation processes influence flood records.
One of the key innovations was the modelling approach, which accounted for both current and past cave configurations. This was essential to reconcile modelled water levels with the flood evidence preserved in speleothems. The use of OSL dating to temporally constrain geomorphological changes further enhanced the reliability of the interprations of flood layers. These methods represent a significant advancement in the field of karst hydrology and paleoclimate reconstruction.
The project also demonstrated the practical applicability of its findings, such as in the management of Las Güixas show cave is a direct outcome of the research and serves as a model for other tourist caves in flood-prone karst regions. The project’s results have been disseminated through peer-reviewed publications, conference presentations, and outreach activities, ensuring broad visibility and impact.
To ensure further uptake and success the following actions are recommended:
Further research: Expand tracer tests to define hydrological catchments and improve model calibration.
Demonstration: Apply the modelling framework to additional karst sites across Europe.
Commercialisation: Develop user-friendly tools for cave managers to assess flood risk.
IPR and standardisation: Explore protection of modelling protocols and contribute to guidelines for cave safety.
Internationalisation: Foster collaborations with karst research centres and speleological societies.
One of the key innovations was the modelling approach, which accounted for both current and past cave configurations. This was essential to reconcile modelled water levels with the flood evidence preserved in speleothems. The use of OSL dating to temporally constrain geomorphological changes further enhanced the reliability of the interprations of flood layers. These methods represent a significant advancement in the field of karst hydrology and paleoclimate reconstruction.
The project also demonstrated the practical applicability of its findings, such as in the management of Las Güixas show cave is a direct outcome of the research and serves as a model for other tourist caves in flood-prone karst regions. The project’s results have been disseminated through peer-reviewed publications, conference presentations, and outreach activities, ensuring broad visibility and impact.
To ensure further uptake and success the following actions are recommended:
Further research: Expand tracer tests to define hydrological catchments and improve model calibration.
Demonstration: Apply the modelling framework to additional karst sites across Europe.
Commercialisation: Develop user-friendly tools for cave managers to assess flood risk.
IPR and standardisation: Explore protection of modelling protocols and contribute to guidelines for cave safety.
Internationalisation: Foster collaborations with karst research centres and speleological societies.