The correlation matrix for maximum stage and maximum discharge revealed that there are eighteen regression equations with correlation coefficients of over 50% from maximum stage while only ten can be obtained from the maximum discharge matrix for the seven stations. Analysis of variance (ANOVA) was used to evaluate whether there are differences between the average value, and mean, across several population groups. The appropriate equations were derived for the maximum stage (H) and maximum discharge (Q), with the coefficient of correlation (r) of the ranging between 50% and 99% and coefficient of determination (r2) ranging between 25% - 99%. Jt was observed that four equations out of the eighteen derived for the maximum stage (H) have negative slopes while three out of ten equations derived for maximum discharge (Q) have negative slopes. All the equations with negative slope are from the stations upstream of the basin. It was also observed that station 35 was highly correlated to stations 52 and 64 for both the maximum stage and maximum discharge and the four equations derived with station 35 as dependent variable are all having negative slope. It could be concluded that the derived equations for the maximum stage (H) and maximum discharge (Q) can be adopted for mitigation of natural disasters and could be used for warnings for short-term events like flash floods to seasonal outlooks of the potential water supply for irrigation and even design of small hydraulic structures such as culverts.
