Laserfiche WebLink
<br />averaging technique should be used to estimate the skew coefficient. <br />The prediction equation technique as described above is preferable <br />technically and should provide a more accurate value. It should therefore <br />be developed first; only if it is not statistically significant should <br />the averaging technique be used. If the variability of the runoff <br />regime in the averaging technique is so large as to preclude obtaining <br />40 stations with reasonably homogeneous hydrology, the arithmetic mean <br />of about 20 stations (except where data are scarce) with representative <br />drainage areas and meteorologic, topographic, and geologic characteristics <br />may be used to estimate the generalized skew cofficient if there are at <br />least 25 years of record. <br />3. Broken Record--Annual peaks for certain years may be missing <br />because of conditions not related to flood magnitude, such as gage <br />removal. In this case, the different record segments are analyzed as a <br />continuous record with length equal to the sum of both records, unless <br />there is some physical change in the watershed between segments which <br />make the total record nonhomogeneous. <br />4. Incomplete Record--An incomplete record refers to a streamflow <br />record in which some peak flows are missing because they were too low or <br />too high to record, or the gage was out of operation for a short period <br />because of flooding. Missing high and low data require different <br />treatment. <br />When one or more high annual peaks during the period of systematic <br />record have not been recorded, there is usually information available <br />from which the peak discharge can be estimated. In most instances the <br />data collecting agency routinely provides such estimates. If not, and <br />such an estimate is made as part of the flood frequency analysis, it <br />should be documented and the data collection agency advised. <br />At some crest gage sites the bottom of the gage is not reached in <br />some years. For this situation use of combination of probability calcula- <br />tions is recommended as described in Appendix 4. <br />5. Zero Flood Years--$ome streams in arid regions have no flow for <br />the entire water year. Thus, the annual flood series for these streams <br />will have one or more zero flood values. This precludes the normal <br />statistical analysis of the data using the recommended log-Pearson Type <br /> <br />14 <br />