Extreme Floods Estimation

[SPA]

Hi,

Is it possible to use RORB for design of extreme floods between 1:2000 AEP and up to the AEP of the PMP?

The currently functionality allows the input of design rainfalls up to 1:2000 AEP and a column for the PMP itself in the IFD table. However, it is possible to incorporate 1:5000, 1:10,000, 1:100,000 etc. AEP intervals? If so, (i) how would one go about it and (ii) is it possible to do a Monte Carlo analysis?

Thanks.

[Benson Liu]

Hi SPA,

It is possible to use RORB for extreme floods up to the AEP of the PMP. Currently you can alter the IFD data (through the “View/edit IFD data” button) such that you can input any AEP, the only limit is the total number of AEPs specified. The rainfall depths rarer than the 1 in 2,000 though you would need to calculate yourself as the BoM only supply up to the 1 in 2,000.

One thing to be careful when doing extreme floods through RORB is the use of ARFs. RORB does a simplified method of applying ARFs for events rarer than the 1 in 2,000, there is more detail on this in the release notes.

As for Monte Carlo analysis, it is possible to do Monte Carlo Analysis up to and including the AEP of the PMP but you would need to input rainfall depths for an AEP that is rarer than the AEP of the PMP so that RORB can sample either side of the AEP of the PMP.

Kind regards,

Benson

[SPA]

Hi Benson,

Thanks for your detailed response. I will attempt your suggestions and get back if there are issues.

A query on temporal patterns : the Datahub provides TPs up to the 1 in 2000 AEP only. But for the AEPs rarer than this, up the PMP, which TPs should be used? Do we just adopt GSAM/GSDM? If so, are we able to edit the temporal patterns file (*increments.csv) like the IFD file? Otherwise, how do we run the file for AEPs less than 1 in 2000?

Thanks.

[Benson Liu]

Hi SPA,

For AEPs rarer than the 1 in 2,000, you can edit the temporal patterns file like you have suggested but RORB will not sample specific temporal patterns for different AEPs unless you set up the file as the same format as point temporal patterns. The point temporal patterns from the data hub are separated into frequent, intermediate and rare categories for AEPs. If in the file you were to put ‘very rare’ for the AEP then RORB would use these temporal patterns for events rarer than and including the 1 in 100. If you would like to control which temporal patterns are used for specific AEPs then you would need to set up separate temporal pattern files and multiple simulations.

As for what temporal patterns to adopt, the GSAM/GSDM are options you can use but you would need to decide whether they are acceptable for your catchment.

Kind regards,

Benson

[roloko3965]

http://doramascoreanos.me/

[Author]

Posts