Rombeek et al. (2025) Torrential rainfall in Valencia, Spain, recorded by personal weather stations preceding and during the 29 October 2024 floods

Identification

• Journal: Hydrology and earth system sciences
• Year: 2025
• Date: 2025-11-26
• Authors: Nathalie Rombeek, Markus Hrachowitz, R. Uijlenhoet
• DOI: 10.5194/hess-29-6715-2025

Research Groups

• Department of Water Management, Delft University of Technology, the Netherlands

Short Summary

This study quantifies the spatial and temporal structure of the torrential rainfall event in Valencia, Spain, on 29 October 2024, using high-density personal weather station (PWS) data. It demonstrates the significant potential of PWSs for real-time rainfall monitoring and flood early warning systems, complementing traditional rain gauge networks.

Objective

• To quantify and describe the spatial and temporal structure of the torrential rainfall event in Valencia, Spain, on 29 October 2024, using personal weather stations (PWSs).
• To evaluate the potential and limitations of PWSs for real-time heavy rainfall monitoring and enhancing flood early warning systems.

Study Configuration

• Spatial Scale: Province of Valencia, Spain, focusing on the Magro (1661 km²), Rambla de Poyo, Turia, and Júcar river catchments. Areal rainfall maps covered areas exceeding 2500 km².
• Temporal Scale: The study primarily focuses on the 29 October 2024 rainfall event (within less than 24 hours), utilizing PWS data at 5-minute resolution. Return period estimations used historical data up to 87 years.

Methodology and Data

• Models used:
  • Ordinary Kriging (OK) with an isotropic spherical variogram model for interpolating point rainfall measurements to create areal rainfall maps.
  • Generalized Extreme Value (GEV) distributions fitted to annual maximum rainfall and discharge values for return period estimations.
• Data sources:
  • Personal Weather Stations (PWSs): Approximately 225 Netatmo brand PWSs, providing near real-time rainfall observations at 5-minute temporal resolution via API.
  • Spanish Meteorological Agency (AEMET) rain gauges: 40 automatic rain gauges with 5-minute temporal resolution, providing daily and 6-hourly precipitation totals.
  • Júcar Hydrographic Confederation: Daily average discharge measurements for 28–30 October 2024.
  • Hydrosheds: Catchment delineations.
  • E-OBS dataset: Gridded observational precipitation data for catchment-average rainfall time series (74 years).
  • EStreams dataset: Discharge time series for three catchments.
  • Digital Elevation Model (DEM): From the European Space Agency.
  • River Network Database: From Copernicus Land Monitoring Service.

Main Results

• Torrential rainfall locally exceeded 300 mm within less than 24 hours on 29 October 2024 in Valencia, Spain.
• The PWS network provided a density approximately 7 times higher than the dedicated AEMET network, enabling a more detailed analysis of rainfall dynamics.
• PWS daily rainfall sums showed a high correlation (r = 0.94) and a low bias (4% underestimation after a 1.24 bias correction factor) when compared to AEMET data.
• The Magro catchment experienced a first extreme rainfall peak (up to 180 mm in a few hours) in the morning, followed by a second peak that moved downstream, coinciding with the flood wave and likely exacerbating the flood magnitude.
• Areal rainfall maps based on interpolated PWS data indicated catchment-average rainfall exceeding 150 mm/day across an area greater than 2500 km².
• Estimated return periods for 300 mm/day rainfall at individual AEMET gauges ranged from approximately 2000 to 8600 years.
• The catchment-average rainfall for the Magro catchment (ID2, 706 km²) on 29 October 2024 had an estimated return period exceeding 10,000 years.
• The highest daily average discharge for the Júcar catchment on 30 October 2024 corresponded to an estimated return period of approximately 900 years.
• PWS measurements were frequently interrupted in affected areas due to power outages or damage, suggesting the reported rainfall sums are lower limits.
• PWS data could have provided early indications of extreme rainfall and potential flooding in the Magro catchment early on the morning of 29 October.

Contributions

• Provides the first detailed spatio-temporal analysis of the devastating 29 October 2024 Valencia floods using a high-density network of personal weather stations (PWSs).
• Quantifies the extreme nature of the rainfall event, estimating return periods exceeding 10,000 years for catchment-average rainfall and 900 years for peak discharge in affected areas.
• Demonstrates the significant value of crowdsourced PWS data in complementing sparse official rain gauge networks, improving spatial resolution, and reducing uncertainty in areal rainfall estimates for hydrological applications.
• Highlights the potential of PWSs for real-time rainfall monitoring and enhancing flood early warning systems, particularly in fast-responding catchments where official data may be delayed or insufficient.
• Offers insights into the complex spatio-temporal evolution of the storm, including the exacerbating effect of sequential rainfall peaks coinciding with flood wave propagation.

Funding

• Joint Cooperation programme for Applied scientific Research on flood and drought risk management in regional river basins (JCAR ATRACE)
• Perspectief research programme Future flood risk management technologies for rivers and coasts (project no. P21-23), financed by Domain Applied and Engineering Sciences of the Dutch Research Council (NWO).

Citation

@article{Rombeek2025Torrential,
  author = {Rombeek, Nathalie and Hrachowitz, Markus and Uijlenhoet, R.},
  title = {Torrential rainfall in Valencia, Spain, recorded by personal weather stations preceding and during the 29 October 2024 floods},
  journal = {Hydrology and earth system sciences},
  year = {2025},
  doi = {10.5194/hess-29-6715-2025},
  url = {https://doi.org/10.5194/hess-29-6715-2025}
}