Stričević et al. (2026) Flood frequency analysis of the Rasina River in Serbia

Identification

• Journal: Geografie
• Year: 2026
• Date: 2026-01-01
• Authors: Ljiljana Stričević, Nataša Martić-Bursać, Milena Gocić, Nikola Milentijević, Marko Ivanović
• DOI: 10.37040/geografie.2026.002

Research Groups

• University of Niš, Faculty of Sciences and Mathematics, Department of Geography, Niš, Serbia
• University of Priština in Kosovska Mitrovica, Faculty of Sciences and Mathematics, Department of Geography, Kosovska Mitrovica, Serbia

Short Summary

This study performed flood frequency analysis for the Rasina River in Serbia using L-moments and annual maximum series data from 1961–2020, identifying Log-Pearson Type III as the best-fit distribution for the region and revealing a significant decreasing discharge trend at one station.

Objective

• To perform flood frequency analysis of the Rasina River in Serbia using L-moments and annual maximum series data to identify best-fit probability distributions, calculate theoretical flood quantiles, and assess discharge trends.

Study Configuration

• Spatial Scale: Rasina River basin, Serbia, specifically at two hydrological stations: Brus and Bivolje.
• Temporal Scale: 60-year period from 1961 to 2020.

Methodology and Data

• Models used: L-moments method, Normal, Log-Normal, Gumbel, Pearson Type III, and Log-Pearson Type III probability distributions. Goodness-of-fit tests (Kolmogorov-Smirnov, Cramér-von Mises, and χ² tests). Homogeneity testing (V_i statistic). Mann–Kendall trend analysis.
• Data sources: Annual maximum series (AMS) discharge data from two hydrological stations (Brus and Bivolje) on the Rasina River.

Main Results

• The Rasina River basin was confirmed to be hydrologically homogeneous (V_i = 0.38 < 1).
• All five tested probability distributions (Normal, Log-Normal, Gumbel, Pearson Type III, Log-Pearson Type III) satisfied L-moment ratio diagram and Z-statistics criteria.
• Log-Pearson Type III showed the best overall fit for the region based on Z-statistics (Z = 0.09).
• Goodness-of-fit tests confirmed Log-Pearson Type III as the best fit for the Brus station and Pearson Type III for the Bivolje station.
• Theoretical flood quantiles were calculated for various return periods.
• A significant decreasing discharge trend was identified at the Bivolje station using Mann–Kendall analysis, with a rate of -1.28 cubic meters per second per year.

Contributions

• Application of L-moments for flood frequency analysis in the Rasina River basin, Serbia, providing a robust statistical approach.
• Identification of the best-fit probability distributions (Log-Pearson Type III and Pearson Type III) for specific stations within the basin, crucial for accurate flood risk assessment.
• Calculation of theoretical flood quantiles for the region, offering essential information for flood protection and water resource management.
• Detection of a significant decreasing discharge trend at one of the hydrological stations, highlighting potential hydrological changes and their implications for future water availability and flood patterns.

Funding

• Ministry of Education, Science and Technological Development of the Republic of Serbia (Agreement No 451−03-137/2025–03/ 200124)
• Faculty of Sciences and Mathematics, University of Niš

Citation

@article{Stričević2026Flood,
  author = {Stričević, Ljiljana and Martić-Bursać, Nataša and Gocić, Milena and Milentijević, Nikola and Ivanović, Marko},
  title = {Flood frequency analysis of the Rasina River in Serbia},
  journal = {Geografie},
  year = {2026},
  doi = {10.37040/geografie.2026.002},
  url = {https://doi.org/10.37040/geografie.2026.002}
}