Copula-Based Modelling of Drought Severity-Duration-Frequency Relationship of Sokoto-Rima-River Basin, Nigeria

• Samuel E. CHUKWU

  Department of Agricultural & Bioresources Engineering, Federal University of Technology, Minna, Nigeria

  Author

• Martins Y. OTACHE

  Department of Agricultural & Bioresources Engineering, Federal University of Technology, Minna, Nigeria

  Author

• Precious O. ATEMOAGBO

  Department of Agricultural & Bioresources Engineering, Federal University of Technology, Minna, Nigeria

  Author

• Emmanuel O. AGBESE

  Department of Civil Engineering, Federal University of Technology, Minna, Nigeria

  Author

Effective drought management in vulnerable semi-arid regions necessitates a comprehensive understanding of drought characteristics and their interdependencies, crucial for proactive water resource planning and risk assessment. This study addresses the critical gap in bivariate drought modelling for the Sokoto-Rima River Basin, Northern Nigeria, by employing a copula-based approach to analyse the Severity-Duration-Frequency (SDF) relationship. Utilising historical monthly rainfall data (1945-2015), the Standardised Precipitation Index (SPI-6) was identified as the optimal timescale for hydrological drought characterisation. Drought severity and duration marginal distributions were best fitted by the Generalised Extreme Value (GEV) and Lognormal distributions, respectively. A strong positive correlation (Kendall's τ=0.7599) was established between drought severity and duration. The Survival (BB8) copula emerged as the most efficient dependence structure, demonstrating superior goodness-of-fit (lowest RMSE: 0.0023, AIC: -82.9876, BIC: -79.0850) and accurately capturing the joint probability distribution. Tail dependence analysis using the Capéraa, Fougères, and Genest (CFG) estimator revealed a weak upper tail dependence coefficient (0.001), indicating that while individual extreme events are common, the simultaneous occurrence of extremely severe and prolonged droughts is historically less probable. The derived bivariate SDF curves offer a robust tool for probabilistic risk assessment, quantifying the likelihood of specific drought severity-duration combinations for various return periods. This research provides invaluable insights for developing advanced drought early warning systems and informing sustainable water resource management and climate change adaptation strategies in data-scarce, drought-prone environments.

• FJET_12_44_37

Copyright (c) 2025 FUDMA Journal of Engineering and Technology

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.