Roux et al. (2026) Hydrological regime shifts in Sahelian watersheds: an investigation with a simple dynamical model driven by annual precipitation

Identification

• Journal: Hydrology and earth system sciences
• Year: 2026
• Date: 2026-02-17
• Authors: Erwan Le Roux, Valentin Wendling, Gérémy Panthou, Océane Dubas, Jean-Pierre Vandervaere, Basile Hector, Guillaume Favreau, Jean-Martial Cohard, Caroline Pierre, Luc Descroix, Eric Mougin, Basile Hector, Caroline Pierre, Jérôme Demarty, Nathalie Rouché, Jordi Etchanchu, Christophe Peugeot
• DOI: 10.5194/hess-30-929-2026

Research Groups

• IMT Atlantique, Lab-STICC, UMR CNRS 6285, Brest, France
• HydroSciences Montpellier (Univ. Montpellier, IMT Mines Ales, CNRS, IRD), Ales and Montpellier, France
• Institut des Géosciences de l’Environnement (Univ. Grenoble Alpes, INRAE, CNRS, IRD, Grenoble INP), Grenoble, France
• Institut d’Ecologie et des Sciences de l’Environnement de Paris (CNRS, Sorbonne Univ., Univ Paris Est Creteil, IRD, INRAE, Univ. de Paris), Paris, France
• Patrimoines locaux, Environnement et Globalisation (MNHN, IRD, CNRS), Paris, France
• Géosciences Environnement Toulouse (CNRS, IRD, UPS, CNES), Toulouse, France

Short Summary

This study investigates hydrological regime shifts in Sahelian watersheds using a simple dynamical model driven by annual precipitation. It finds that four studied watersheds (Gorouol, Nakanbé, Dargol, Sirba) experienced regime shifts during the droughts of the 1970s–1980s, transitioning from a low to a high runoff coefficient regime.

Objective

• To investigate the timing of hydrological regime shifts in Sahelian watersheds, specifically asking when these shifts occurred.
• To develop a parsimonious dynamical model capable of reproducing and characterizing these regime shifts, assuming the observed hydrological changes correspond to a regime shift from a low to a high runoff coefficient regime.

Study Configuration

• Spatial Scale: Four watersheds in the Central Sahel: Gorouol (23 200 km²), Dargol (7350 km²), Sirba (40 300 km²), and Nakanbé (21 800 km²). The model operates at the watershed scale.
• Temporal Scale: Pluri-decennial observations (1950s–2010s). Simulations cover the period 1956–2014, with an annual time step.

Methodology and Data

• Models used:
  • A novel lumped dynamical model was developed, simulating the annual runoff coefficient (K) with water holding strength (S) as the single state variable and annual precipitation (P) as the unique external forcing. This model incorporates feedback mechanisms between soil, water, and vegetation.
  • The GR1A lumped bucket model was used for benchmark comparison.
• Data sources:
  • Annual precipitation: Computed for each watershed based on gauge station data (1956–2014), using kriging methods for spatial interpolation.
  • Annual runoff coefficient: Calculated from annual watershed outflow (V, in cubic meters) per unit of watershed area (A, in square meters), normalized by annual precipitation (P, in millimeters). Daily discharge data (in cubic meters per second) were obtained from SIEREM and ADHI databases, with Nakanbé discharges corrected for dam effects.

Main Results

• The developed simple dynamical model successfully reproduces the increasing trend of observed annual runoff coefficients (and corresponding decreasing trend of water holding strength) for all four Sahelian watersheds over the 1956–2014 period.
• For all watersheds, over 90% of the selected best 1000 parameterizations of the model were bistable, indicating the potential for two alternative runoff coefficient regimes (low and high).
• Most selected parameterizations underwent a regime shift between 1965 and 2014, transitioning from a low to a high runoff coefficient regime.
• The year of the regime shift, defined as the year with the greatest number of shifts, was identified as: Gorouol (1971), Nakanbé (1972), Dargol (1973), and Sirba (1983). These shifts occurred during the severe droughts of the 1970s–1980s.
• The model, despite its parsimony, performed better than a classical hydrological model (GR1A) in capturing the observed trends in runoff coefficients.

Contributions

• Development of a novel, parsimonious lumped dynamical model for Sahelian watersheds that explicitly accounts for eco-hydrological feedback mechanisms and can simulate hydrological regime shifts, requiring only precipitation and runoff data.
• Introduction of a new, practical definition of hydrological regimes based on a threshold in the bifurcation diagram, enabling the identification of regime shifts in the context of transient dynamics.
• Quantitative characterization of the timing of hydrological regime shifts in four Central Sahelian watersheds, attributing them to the droughts of the 1970s–1980s.
• Demonstration that a simple model incorporating feedback mechanisms can effectively capture long-term trends and regime shifts in data-scarce regions, outperforming traditional hydrological models.

Funding

• Agence Nationale de la Recherche (ANR, France) under contract no. ANR-20-CE01-0014-01 (TipHyc project).

Citation

@article{Roux2026Hydrological,
  author = {Roux, Erwan Le and Wendling, Valentin and Panthou, Gérémy and Dubas, Océane and Vandervaere, Jean-Pierre and Hector, Basile and Favreau, Guillaume and Cohard, Jean-Martial and Pierre, Caroline and Descroix, Luc and Mougin, Eric and Hector, Basile and Pierre, Caroline and Demarty, Jérôme and Rouché, Nathalie and Etchanchu, Jordi and Peugeot, Christophe},
  title = {Hydrological regime shifts in Sahelian watersheds: an investigation with a simple dynamical model driven by annual precipitation},
  journal = {Hydrology and earth system sciences},
  year = {2026},
  doi = {10.5194/hess-30-929-2026},
  url = {https://doi.org/10.5194/hess-30-929-2026}
}