Sahraei et al. (2026) Numerical simulation of hydrological response to extreme rainfall in Atacama Desert

Identification

• Journal: Natural Hazards
• Year: 2026
• Date: 2026-03-31
• Authors: Seyedamirali Sahraei, Cong Jiang, Mehdi Hamidi, Yaping Shao
• DOI: 10.1007/s11069-026-08013-z

Research Groups

• Faculty of Environment, University of Northern British Columbia, Prince George, Canada
• Institute for Geophysics and Meteorology, University of Cologne, Cologne, Germany
• Faculty of Civil Engineering, Babol Noshirvani University of Technology, Babol, Iran

Short Summary

This study investigates the hydrological response of the hyper-arid Salado River Basin in the Atacama Desert to the extreme March 2015 rainfall event using a fully coupled Atmospheric and Hydrological Modeling System (AHMS), finding that the event produced peak discharges of approximately 1000 m³ s⁻¹ and exceeded the 99.9th percentile of an 8-year modeled streamflow distribution.

Objective

• Reconstruct the March 2015 extreme rainfall and flood event in the Salado River Basin using the fully coupled AHMS framework.
• Quantify the resulting hydrological response, including spatial precipitation patterns and flood hydrographs at key locations.
• Place the 2015 event within a longer-term (8-year) hydrological context using offline simulations and probability density functions of streamflow.
• Assess the potential benefits and limitations of fully coupled modeling for flood reconstruction in data-sparse hyper-arid regions.

Study Configuration

• Spatial Scale: Salado River Basin (approximately 7500 km²) in northern Chile. Computational domain covers northern Chile and adjacent regions, with an outer domain at 60 km horizontal resolution and an inner domain at 20 km resolution.
• Temporal Scale:
  • Online (fully coupled) simulation: 13 days (18 to 31 March 2015) for the extreme event.
  • Offline continuous simulation: 8 years (January 2010 to December 2017) for long-term hydrological context.

Methodology and Data

• Models used:
  • Atmospheric and Hydrological Modeling System (AHMS)
  • Weather Research and Forecasting (WRF) model (version 3.7)
  • Distributed Hydrological Modeling System (HMS)
  • Noah-MP land surface model
• Data sources:
  • ERA5 reanalysis (initial and lateral boundary conditions for WRF, atmospheric forcing for offline simulations, sea surface temperature)
  • SRTM digital elevation model (topography)
  • MODIS-based USGS classifications (land cover)
  • FAO/STATSGO global soil database (soil texture)
  • Las Vegas rain gauge (observed precipitation for qualitative evaluation)

Main Results

• The online AHMS simulation reproduced basin-averaged rainfall of approximately 120 mm over 48 hours (24–26 March 2015) in the Salado River Basin.
• Peak discharges of approximately 1000 m³ s⁻¹ at El Salado and about 930 m³ s⁻¹ at Chañaral were generated, consistent with previous independent estimates.
• The offline simulation for 2010–2017 showed that the 2015 event's peak discharges exceeded the 99.9th percentile of the modeled daily streamflow distribution, indicating its exceptional nature.
• Simulated hydrographs suggest rapid, infiltration-excess-dominated runoff generation with short response times (6–12 hours) and significant transmission losses along ephemeral channels.
• Both online and offline simulations produced comparable flood magnitudes (1000–1100 m³ s⁻¹), but differences highlighted sensitivity to precipitation forcing and land-atmosphere interactions.

Contributions

• This study represents one of the first applications of a fully coupled atmospheric-hydrological framework (AHMS) to an extreme flood event in a hyper-arid Andean catchment.
• It provides a process-based numerical reconstruction of runoff generation in a data-sparse hyper-arid environment by dynamically simulating precipitation and land-atmosphere feedbacks.
• The research offers a physically coherent reconstruction of event-scale processes in a basin with virtually no discharge observations.
• It systematically places the extreme 2015 flood event within a multi-year (8-year) hydrological context using continuous simulations and probability density functions.

Funding

• Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) – Projektnummer 268236062–SFB1211

Citation

@article{Sahraei2026Numerical,
  author = {Sahraei, Seyedamirali and Jiang, Cong and Hamidi, Mehdi and Shao, Yaping},
  title = {Numerical simulation of hydrological response to extreme rainfall in Atacama Desert},
  journal = {Natural Hazards},
  year = {2026},
  doi = {10.1007/s11069-026-08013-z},
  url = {https://doi.org/10.1007/s11069-026-08013-z}
}