Watershed

Oral

366685 - Detailed analysis of the variability of Hershfield-type rainfall sampling adjustment factors

Wednesday, June 6
2:00 PM - 3:30 PM
Location: Greenway IJ
Co-Authors: Patricio Muñoz-Proboste, Memphis, Tennessee – University of Memphis; Peter Molnár, Zürich, Switzerland – ETH-Zürich

Even “continuously-gaged” precipitation records are actually discrete, as rainfall depths are aggregated over measurement periods typically ranging from 10 to 60 minutes. This invariably introduces negative biases when estimating extremes over durations similar to the totalization period. Starting with Hershfield’s work in the 1950’s, rainfall sampling adjustment factors (SAFs) have been used in order to convert fixed (”clock”) - time maxima into continuous (“sliding”) maxima. A problem in the literature though, is that there are multiple definitions for rainfall SAFs.; more surprisingly, nobody has systematically looked at the sources of variability in SAFs.

We analized 35-yr long rainfall records at 52 stations distributed over Switzerland (15,900 mi2), with data every 10 min. All records are concurrent, and were collected with the same methodology. Independent storms were identified and assigned to fixed, warm or cold seasons, as well as to convective or non-convective types (based on co-occurrence of lightning). For a series of durations (≥20 min), we extracted the “true” maxima from our “continuous” (10 min) data, as well as for data totalized over a range of different periods, for 8 combinations of season and storm type, using all possible ways of totalizing. We then applied partial duration analysis to obtain rainfall depth-duration-frequency (DDF) values at each station, for each totalization period, way of totalizing, and season/storm type. SAFs were subsequently computed as the ratios between DDF values obtained for fixed and continuous time windows.

There is ample variability in SAFs within each station, due to the different ways in which the data can be temporally totalized, as well as across stations. For example, the Hershfield factor - the 1-h to 60-min SAF - has a mean value of 1.12, but ranges from 1.04 to 1.22. Both season of the year and storm type have strong effects on rainfall SAFs.

Claudio I. Meier

Associate Professor
University of Memphis

Presentation(s):

Send Email for Claudio Meier


Assets

366685 - Detailed analysis of the variability of Hershfield-type rainfall sampling adjustment factors



Attendees who have favorited this

Please enter your access key

The asset you are trying to access is locked. Please enter your access key to unlock.

Send Email for Detailed analysis of the variability of Hershfield-type rainfall sampling adjustment factors