[Edit 17 Oct 2016. Some web links have changed. See Where is ARR?]
Australian Rainfall and Runoff (ARR) was launched at the Hydrology and Water Resources Symposium in Hobart on 7 Dec 2015. It’s not finished yet but there is a draft of much of the document, running to over 2000 pages. You can download a copy here or read it on the web.
As part of the launch there was a one day workshop and 36 ARR related papers were presented at the symposium. I’m still looking through presentations, papers, reports and draft guideline but wanted to document first impressions and provide pointers to more information:
- A one-page summary of the new ARR is available here.
- The new approach to at-site flood frequency analysis is settled and Flike software is now available.
- Regional Flood Frequency Estimation (RFFE) for ungauged catchments has been tested and revised. Software is now available. Comments at the conference suggested this is still a beta version, and more testing is being undertaken, but it looks good.
- Regional and at-site methods can be combined in many cases to produce better flood estimates by incorporating parameters from RFFE as priors in Flike. This is a really important and useful outcome of these ARR projects.
- The latest ideas on temporal rainfall patterns were presented. The recommended procedure is likely to involve the use of an ensemble of 10 to 20 patterns rather than the single patterns that were documented in ARR1987. These patterns are being developed for 12 regions around Australia.
- Work on continuous rainfall sequences is at an advanced stage and is probably useable now.
- All methods are being thoroughly tested.
- Practitioners are being encouraged to use new approaches and provide feedback.
- Areal reduction factors are settled with recent work extending to shorter durations.
- There was an excellent presentation on method selection by Rory Nathan (check out figure 1.3.2 in the draft guideline)
Some issues and limitations
- Temporal rainfall patterns are not yet publically available.
- Work on design losses is well advanced but information on pre-burst rainfall is required to use these losses in practice. These pre-burst rainfall values are still being developed.
- Rainfall IFD values are likely to be revised. This may mean that the RFFE analysis and software will need to be updated.
- Rainfall based methods of flood estimation require selection of design values for a range of inputs including three that are still being worked on:
- Rainfall depth
- Rainfall temporal pattern
- Pre-burst rainfall to use with design losses
There will need to be testing of these inputs and results.
For more information:
- Check the slides of the talks presented at the ARR workshop.
- Read those parts of the draft guideline that are of interest
- Check the project reports available at the ARR web page. There were some updated reports provided at the workshop which are available here.
- Look through the conference papers. These have not been released yet but have been promised. They will probably be available through informit but that may take sometime. All the papers are listed on the conference program.