International Science Community Association Research Journal of Physical Sciences 13 2 2025 08 4 Exploring Multiscale variability of monthly runoff time series of Diani River using fractal theory in Southern Guinea 1 5 EN Agbazo Noukpo Médard 1 Département d’Hydrologie, Université de N’zérékoré, BP 50, N’zérékoré, Guinée Keita Oumar 2 Département d’Hydrologie, Université de N’zérékoré, BP 50, N’zérékoré, Guinée Camara Lonsenigbe 3 Département d’Hydrologie, Université de N’zérékoré, BP 50, N’zérékoré, Guinée Sylla Abdoulaye 4 Département d’Hydrologie, Université de N’zérékoré, BP 50, N’zérékoré, Guinée > Research Paper 2025 5 27 2025 08 4 Runoff time series modeling is necessary for hydrological applications, including understanding the evolution of river regimes and forecasting and controlling floods. However, in the Guinea republic, West Africa's water tower, the intrinsic characteristics involved in hydrological variables dynamics remain unknown. This preliminary work aims to explore, for the first time in Guinea, the multifractality and complexity properties of monthly runoff time series measured from 2000 to 2019 on the Diani River, which is one of the largest rivers in Guinea. To this end, the following parameters have been computed: Lyapunov exponent, Hurst exponent, Higuchi fractal dimension, width of the multifractal spectrum and spectrum asymmetry index. Numerical results indicate: i. a clear footprint of persistence and multifractality in the runoff time series irrespective of the time period. However, the persistence and multifractality degree depend on the time period considered ii. a sign of chaotic dynamic systems and predictive instability in runoff variations. iii. the predictive scheme based on the multifractality and persistence could be adapted for Diani river runoff prediction. The conclusions drawn from these results should prove useful for the validation of global and regional climate models. Copyright@ International Science Community Association