One of the values I want to find out is the 2-year 24-h rainfall. From what I have learned so far, first the rainfall data should be separated into events of different durations. For this, I need to determine a "minimum interevent time". How do I find the most appropriate value for this?
Once I have decided on the interevent time and have obtained separate rainfall events, I guess I somehow need to discretise them to be able to calculate probabilities for different precipitation durations and depths and to create an IDF curve. How would I do this?
Is there any programme that could help? Could I do the whole process in Excel?
Any help would be much appreciated!
Hi Juliane,
I suggest you follow these articles especially the first one where they have used to calculate the "minimum interevent time (MIT)" in a very simple way.
Article Relationship between minimum inter-event time and the number...
https://www.engr.colostate.edu/~ramirez/ce_old/classes/cive322-Ramirez/IDF-Procedure.pdf
Article The identification of an appropriate Minimum Inter-event Tim...
https://www.mahahp.gov.in/old_mahahp.org/resource/modules/12%20How%20to%20analyse%20rainfall%20data.pdf
You can use the RainEMT tool for determining MIT and IDF described in the following article:
Article Analysis of rainfall events over Peninsular Malaysia
Thanks
Ankur
Juliane G., you also could consider using CumFreq (Wageningen University). Frequency and regression analysis of hydrologic data, R.J. Oosterbaan, 1994, www.waterlog.info/cumfreq.htm, chapter 6 in: H.P.Ritzema (Ed.), Drainage Principles and Applications, Publication 16, second revised edition, 1994, International Institute for Land Reclamation and Improvement (ILRI), Wageningen, The Netherlands. ISBN 90 70754 3 39.
I used the CumFreq program (Gumbel distribution) for developing DDF (Depth Duration Frequency) curves for the Greater New Orleans Urban Water Plan. See paragrapgh 3.2 of the Water System Analysis report; https://www.dropbox.com/s/c14h3k4tbvvlocu/GNO%20Urban%20Water%20Plan_Water%20System%20Analysis_13Oct2013.pdf
This can be done by deriving DDF
curves (Depth Duration Frequency) relationships.
The process includes fitting an extreme value
distribution to a time series of annual maximum
recorded rainfall depths. The process of fitting
the probability distribution is generally carried
out separately for each duration.
Thank you very much for your help!
Nanco Dolman : I have read the book chapter you mentioned and used the CumFreq programme. Oosterbaan does not mention an interevent time to separate precipitation data, instead he describes three methods to obtain durations longer than the measurement intervals: successive totals, moving totals and maximum totals. So what I did is calculating the 24 h moving totals for every year and then I used the highest of these values of each year for the frequency analysis in CumFreq. Is this the right way to do it? It seems very reasonable to me, but I am confused because I read that you should separate the rainfall data first (see citation below). Maybe these are two different methods, but then I am wondering in which case to use which...
Chapter 6 in Bedient, Huber and Vieux (2013): Hydrology and Floodplain Analysis:
“Return periods can be assigned to rainfall events on the basis of several different parameters (…). To work with any of these parameters, the rainfall time series must first be separated into a series of discrete, independent events. (…) For ease of computation, a statistical measure is usually employed to separate independent storm events. A minimum interevent time (MIT) is defined such that rainfall pulses separated by a time less than this value are considered part of the same event.”
Ankur Srivastava : The first article just says “From rainfall measurements, for every year of record, determine the annual maximum rainfall intensity for specific durations (or the annual maximum rainfall depth over the specific durations).” It is unclear to me how this should be done exactly.
The second article says “The entire rainfall record in a year is analysed to find the maximum intensities for various durations. Thus each storm gives one value of maximum intensity for a given duration.” But I cannot see how it was calculated either. They use HYMOS and I tried to find that programme online, so I could maybe figure it out myself, but it seems like it is not available anymore.
However, regarding the analysis of one storm event, they say “Now for any zone the maximum average depth of rainfall for various durations of 4, 8, 12, 16 and 20 h can be obtained from Table 8.4 by sliding a window of width equal to the required duration over the table columns with steps of 2 hours.” That would be the same as the “moving totals” described by Oosterbaan, I guess.
The third article I have not yet read, but I requested it.
- Badges
- Science topic
- Similar topics
- Analytical Chemistry
- Column