Eq. (6) ensures that the pth quantile of the T-year return level matches
the T-year return level estimated from the observed sample. After inserting
Eq. (6) into Eq. (5), numerically c and subsequently b can be
determined. For each given a these parameters allow to determine
the log-likelihood function (Eq. 4). While varying a, we can eventually
determine the value a that maximizes Eq. 4. We varied a in steps
of 0.005 and interpolated between these steps. For various confidence
levels, we will present the results for the peak flow at Point of Rocks,
Potomac River for the 1000-year return level in the Result section