FORECASTING OF FLOOD EVENTS FOR WATER RESOURCES MANAGEMENT IN GUCHA-MIGORI RIVER BASIN, KENYA

• DOI: 10.37017/jeae-volume10-no1.2024-1
• Publication Date: 2024-05-09T09:29:09Z
• Authors: Edith Onyango; Raphael Wambua; Hesbon Otieno
• Keywords: Gumbel distribution; Flood forecasting; Return period

Floods are devastating natural disasters frequently occurring in many river basins such as Gucha-Migori River Basin, Kenya. However, non-structural countermeasures hydrologic modelling, flood proofing, and continuous forecasting have not been fully explored implemented to reduce risk, damage, vulnerabilities of events. The major challenge is the selection relevant probability distribution. main objective study was forecast events Basin for integrated water resource management. Daily hydrological datasets between 1969 2015 were obtained from Water Resources Authority. After ranking independent had achieved, return periods frequencies computed. relationship magnitudes their respective through modelling using distribution functions (PDFs). evaluated PDFs include Normal Distribution, Log-Normal Gumbel Log-Person Type III Distribution. Coefficient Determination (R2), Goodness Fit test, Best-Fit Distribution Curve guided suitable model. model out tested four then applied predict P-percent annual exceedance probabilities corresponding specific magnitudes. From results, 132 that occurred period 2015, only 42 identified Based on frequency analysis, highest lowest, average standard deviation recorded 423.90, 240.8, 284.80 m3/s, 45.54 respectively. It revealed a event magnitude equal or exceeding lowest (240.8 m3/s) (423.90 particular year 0.98 0.02 curve selected be best fit Basin. According derived Gumbel's formula, estimated 1.05, 1.11, 1.25, 2, 5, 10, 25, 50, 100, 200 years 221.38, 231.92, 245.77, 278.01, 321.39, 350.11, 386.39, 413.31, 440.03, 466.65 m3/s. This research provides useful information enhanced determination occurrence thus proper resources planning management strategies basin.