Flood analysis procedure was performed on annual discharge data for River Oyan in Oyo State Nigeria for the period of 1988 to 2010 utilizing five probability distribution models namely: Normal, Log-Normal, Gumbel or Extreme Value Type I, Log-Pearson Type III, Gamma or Pearson Type III and New Model. The models were used to predict flood discharge estimates at 2, 5, 10, 25, 100, and 200 years return periods. The results indicated that the Normal distribution predicted discharge values ranging from 8633.35 m³/s for two years to 17768.07 m³/s for 200 years return period, the Log- Normal distribution predicted discharge values ranging from 7894.05 m³/s for two years to 25113.08 m³/s for 200 years return period, the Gumbel or Extreme value Type I distribution predicted discharge values ranging from 8056.25 m³/s for two years to 21557.55 m³/s for 200 years return period, the Log–Pearson Type III distribution predicted discharge values ranging from 8194.08 m³/s for two years to 20351.67 m³/s for 200 years return period, the Gamma or Pearson Type III distribution predicted discharge values ranging from 8285.53 m³/s for two years to 19317.46 m³/s  for 200 years return period. From the result, Gamma or Pearson Type III, was found most suitable for flood estimation of River Oyan, based on goodness of Fit test using Chi-Square analysis. Future researchers should include more probability distributions; also precision of flood data should be improved by taking data for a longer period of time.

Key words: River Oyan, Flood Discharge, Frequency Distribution, Goodness of Fit.