EagFirRsp

The DESIGN rainfall applied to the model Mitigate.xp appears to be the 10% AEP 1hr event. Relates to
BSM's preferred manner of running individual storms. Volunteered by Premise that it could be handled a
little easier by ticking the selected storms in the global storms database, but not an issue.

An initial depth of 0.01m has been applied to every node. This was a hangover from earlier versions of
XPStorm’s Extran engine in which it was used to counter model instability, which is greatly reduced
in Version 9 of XpStorm. Initial depths in the models have now been changed to zero.

Evaporation has been set to 0.1mm/day (typically assume 0mm/day). Default value, but insignificant in
the scheme of things – has been changed to zero in the remodelling.

Issues

The hydrology method hasn’t been fully explained within the report but it appears that a mixture of ARR2016
and ARR1987 methods has been applied. One version of ARR should be consistently followed throughout the
study. If ARR2016 rainfall is selected, the full ARR2016 method should be employed. Agreed with Premise
that it will be okay for BSM to use ARR87 rainfalls with ARR87 temporal patterns with zero initial loss.

The report states that the models have been calibrated to the Rational Method. QUDM v4 states:“Utilisation
of the Rational Method, as described within the Manual, may be considered appropriate…as a ‘checking tool’
(not calibration) of numerical models developed for small ungauged catchments (i.e. checking for potential
gross errors)”. The models were previously set-up with appropriate routing parameters and the initial losses
were then varied between the AEP's to achieve a close calibration to the Rational Method across the range of
AEP. Agreed with Premise that the initial losses will be changed to zero and the Rational Method will only be
used as a sanity/comparison check. The Manning’s n values for the subarea runoffs were set at 0.020 and
0.045 for impervious and pervious areas respectively in the original modeling, which we suggest are
appropriate values, and the remodelling has also used these values. The comparison check with the Rational
method, for the “developed case” flows at the model outlet, prior to the addition of the basin, is summarised in
Table 1 below:
Table 1: XPStorm Developed Scenario, Comparison With Rational Method
 Event                 Rational Method        XPStorm Model
 AEP (ARI)             (m3/s)                 (m3/s)
 AEP 1% (100 yr)       6.93                   6.76
 AEP 10% (10 yr)       4.01                   4.23
 AEP 63% (1 yr)        1.80                   2.90

Initial losses should be 0mm/hr. Applying no initial losses is consistent with the requirements of both
ARR1987 and ARR2016 that confirm the temporal patterns used for design purposes are the ‘burst’ elements
within greater duration events. ARR2016 further states that there is no evidence that infiltration losses change
with respect to the recurrence interval being modelled and that continuing losses can be applied equally to
frequent and rare events. As above, the initial losses have been reduced to zero for the remodelling. A
continuing loss rate of 2 mm/hr has been maintained in the remodelling, across the full range of AEP.

Relating to the XPSTORM model the routing method applied in the model is not endorsed by either version of
ARR. It is important that basis for the basin design enables confidence that the detention system will be
sufficient for the development. Noted by Premise that the SWMM "Runoff" routing method is a US-derived
method, whereas Premise would have preferred an Australian derived method (preferably, Laurenson's). BSM
have been using the SWMM Runoff routing method in XPStorm and in its predecessor UDD for almost 20



                                                                                                   Page2
BSM Doc File EagFirResp.Doc