Abstract
When the systematic observed hydrological record is not long enough, the historical flood information is useful to increase the precision of flood-quantile estimation. Historical flood data, which can be used in flood frequency analysis by adjusted-moment method or maximum likelihood method, often contain much larger error than systematic observed data. Thus it is important to evaluate the effect of the error on the precision of flood-quantile estimators. Monte Carlo simulation by using Gumbel distribution shows that flood-quantile estimator contains positive bias when the standard error exceeds a certain level (in this study, approximately 1/10 to 1/8 of the average for total sample). Moreover, when it exceeds 1/5 to 1/4 of the average, the improvement of the precision cannot be confirmed for the calculated flood-quantile estimators.
