Achite et al. (2026) Bivariate Characterization of Long-Term Hydrological Drought Risks Using SRI and Archimedean Copulas

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Hydrology
• Year: 2026
• Date: 2026-03-30
• Authors: Mohammed Achite, Tolga Barış Terzi, Osman Üçüncü, Kusum Pandey, Tommaso Caloiero
• DOI: 10.3390/hydrology13040104

## Research Groups -

Short Summary

This study develops a bivariate probabilistic framework to characterize long-term hydrological drought risk in the Wadi Sahouat basin, northwestern Algeria, using the 12-month Standardized Runoff Index (SRI-12). It demonstrates that multivariate return periods significantly differ from univariate estimates, especially for extreme events, underscoring the compounded risk of prolonged and severe droughts.

Objective

• To characterize long-term hydrological drought risk in the Wadi Sahouat basin (northwestern Algeria) using a bivariate probabilistic framework.

Study Configuration

• Spatial Scale: Wadi Sahouat basin, northwestern Algeria.
• Temporal Scale: 42-year period (1973/74–2014/15) for long-term drought analysis; return periods estimated for 10 to 200 years.

Methodology and Data

• Models used: Bivariate probabilistic framework, run theory (for drought event identification with SRI ≤ −1.0), Weibull distribution (for marginal distributions of drought duration and severity), Archimedean copulas (specifically Gumbel copula for dependence structure).
• Data sources: 12-month Standardized Runoff Index (SRI-12) derived from runoff data at hydrometric stations.

Main Results

• The Weibull distribution was selected as the best fit for the marginal distributions of both drought duration and severity.
• The Gumbel copula provided the best fit for modeling the dependence structure between duration and severity at both hydrometric stations, indicating significant upper-tail dependence.
• Multivariate return periods for hydrological droughts substantially differ from univariate estimates, particularly for extreme events, revealing a compounded risk for prolonged and severe droughts.
• Univariate and bivariate return periods were estimated for target intervals ranging from 10 to 200 years.

Contributions

• Presents a novel bivariate probabilistic framework for assessing long-term hydrological drought risk in semi-arid regions.
• Quantitatively demonstrates the significant difference between multivariate and univariate drought return periods, highlighting the underestimation of extreme drought risk by univariate methods.
• Provides a robust methodology for characterizing the joint probability of drought duration and severity, crucial for water resource management in vulnerable regions.

## Funding -

Citation

@article{Achite2026Bivariate,
  author = {Achite, Mohammed and Terzi, Tolga Barış and Üçüncü, Osman and Pandey, Kusum and Caloiero, Tommaso},
  title = {Bivariate Characterization of Long-Term Hydrological Drought Risks Using SRI and Archimedean Copulas},
  journal = {Hydrology},
  year = {2026},
  doi = {10.3390/hydrology13040104},
  url = {https://doi.org/10.3390/hydrology13040104}
}