FLOOD01798

• County: Statewide
• Title: Australian Rainfall and Runoff 1998, Revision of Book VI - Estimation of Large to Extreme Floods
• Date: 11/28/1998
• Prepared By: Rory Nathan, Sinclair Knight Merz
• Floodplain - Doc Type: Educational/Technical/Reference Information

I
I
I
I
I "
I
I
I
I
I
I
I
I
I
I
I
I
I
I

The procedures outlined in this Book are intended to
CDver the brDad spectrum of applications listed in Section
1.2 and the flood ranges described in Section 1.3.
However, when deriving flood estimates for a specific
application, the practitioner's interest may be focused on a
more limited range of procedures. This section provides
some guidance on the applicability of different parts of the
guidelines to specific investigation or design tasks,
Figure 2 gives a qualitative indication Df the range of
flood magnitudes and the relative degree of reliability
required for different applicatiDns. An extension of the flood
range of interest is associated with a greater degree. of
extrapolation and thus larger uncertainty (lower reliability).
The level of expertise and effort required for deriving design
floods increases with increasing level of reliability required.
For the following four groups of applications,
approximate or simplified flood estimation procedures may
be applicable. However, the practitioner has to apply
hislher own judgement in deciding on the degree of detail
PMF as a benchmark value and accuracy required for a specific application.

The PMF is traditionally defined as the limiting value of t(i) Planning and feasibility studies, initial screening of
flood that could reasonably be expected to occur. Its AEP is options, preliminary designs:
un~nown because there is no gene~IIY accepted way o! For these types of applications, where decisions
~stimating It, ~nd because the phrase reasonably possl~le based on the fiood estimates are only moderately
IS not numencally defin~d. The d:gree, of conseryatosm sensitive to estimation uncertainties, approximate
used to select and combine the various Inputs that Impact design flood estimates can be derived by a "quick"
on flood magnitude varies from one practitioner to another, method (see Sections 3 11 and 4 7)
and therefore its estimation does not provide a consistent ' . .
level of risk for design applications. Clearly, the AEP of the (ii) Performance checks or preliminary designs of
PMF is less than that of the PMP. structures where the lowest AEP olinterestis 1 in 500
For most design s~uations ~ is recommended that the and where no regional design rainfalls or detailed
PMF estimate should be replaced by an estimate of the PMP esllmates are available:
PMP Design Flood (that is, the flood derived from the PMP Many design codes require safety checks for
using AEP-neutral assumptions). These guidelines provide conditions exceeding the design objective.
specific recommendations for the derivation of the PMP Approximate estimates of floods with AEPs from 1 in
Design Flood, 100 to an absolute limit of 1 in 500 can be obtained by
However, it is recognised that estimates of the PMF extrapolatio,n of either flood or rainfall fr~quency
may still provide a useful benchmark for comparison with curves, uSIng the procedures described In, ot~er
the results of earlier studies and with risk-based estimates. Books. In general, rainfall-based flODd estomation
Given the limitations of the available design information it is methods should be calibrated WIth flood dat~ from the
expected that the greatest difference between PMFs and site. ConSIderably larger uncertainty. IS associated WIth
PMP Design FloDds will most commonly arise in the floods In. thiS range estimated WIthout the use of
derivatiDn of reservoir outflow floods (see Section 5 2) regional information, and correspondIngly greater
. . responsibility then rests with the designer to ensure
that the estimates are consistent with any rare flood
estimates for the region. For rainfall-based
procedures, the preliminary PMP estimates from
Section 3.11, combined with the interpolation
procedure in Section 3.6.1, can be used to reduce the
uncertainty in design rainfall extrapolation.

(iii) Preliminary designs or performance checks of new
structures, and adequacy assessments of major
existing structures with low to moderate failure
consequences, where the AEP range of interest
extends from Large to Rare floods and where regional

UKAt-1 U

AEP-neu&a/approach

To derive a design flood with an AEP equivalent to that
of the design rainfall employed it is necessary to adopt an
AEP-neutral approach. This invDlves selecting model inputs
and parameter values such that the 1 in Y AEP design
rainfalls are converted to the correSponding 1 in Y AE:P
floods. For the least sensitive model parameters it is
usually sufficient to adopt a single representative value
from the central range of observations; often e~her the
mean or median is adopted, althDugh the most appropriate
value depends on the degree of non-linearity in the
transformation between rainfall and runoff. For the more
sensitive model parameters it is desirable to adopt a joint
probability approach, where the inputs are characterised by
their probability distributions rather than by a single value.

The adoption of an AEP-neutral approach is accepted
practice for frequent to Large floods, where ~ is usually
possible to derive independent estimates of the design
floods to check that no bias has been introduced into the
transformation between rainfall and runoff, It should be
recognised that our understanding concerning the
distributions of, and correlations between, many design
inputs is limited, and that beyond the credible limit of flood
extrapolation it is not possible to assure the achievement of
the AEP-neutral objective. Thus, while the objectives of
AEP-neutral methods are clear, in many design situations
there is no way of knowing whether or not they have been
achieved. There is considerable uncertainty in the
magnitude of flood estimates beyond the credible limit of
flood extrapolation, and this uncertainty should be
recognised and incorporated into management decisions.

(b)

(c)

Impact of climate change

Estimates of Large to Extreme rainfalls (and the
resulting floods) are subject to change as our
understanding of the governing physical processes
increases, and as more data becomes available for
analysis. The estimates are also subject to change due to
long-term climatic variations, such as would result from
changes in atmospheric concentration of greenhouse
gases. However, it is considered that climatic trends
progress SD slowly that their influence on the assessment
of annual flood risks over the design life of infrastructure

(d)

t:iOOK VI M estimatIon or Large to t=xtreme t"looas

w<<ks is small oompared to other tlnceotainties. M time of
writing (1998) the Bureau of Meteorology do not consider
the influence of climatic trends when preparing PMP
estimates, and at present there is no known procedure for
the reliable incorporation of climate change scenarios into
flood estimates. Until better information becomes available
any lIS&essment of the irnpaol Df climate change on /lQod
risks is thus likely to be speculative and not su~ed for
general design considerations.

2.3 Relevance of Procedures to Specific
Applications

*