Meeting Environmental Outcomes – A Planning Framework - Water and Agriculture: Sustainability, Markets and Policies - page 205

In state 1, corresponding to a year in which there was a successful high flow event in
the preceding season, the release threshold is zero. The release curves become
successively higher in states 2 through to 4. States 4 and 5 are similar, reflecting the way
in which the penalty function is structured. In state 5 there is a penalty for an event lasting
less than the full 14 days. No penalty is incurred in state 4 even if a high flow event lasts
only ten days, creating an additional incentive to make a release in that state. It is not
intended to infer that this is an appropriate environmental design, but it does highlight
how the structure of the penalty function affects the release strategy.
The penalty level generates a very reliable release strategy within the simple
hydrological specification used to predict flows. Under current conditions, the likelihood
of achieving a high flow event within five years is approximately 46 per cent. With the
supplementary releases, the likelihood of achieving a high flow event within five years
increases to 95 per cent. The shift in the interarrival pattern of events is shown in Figure
G. As well as reducing the likelihood of an extended period without a high flow event,
there is a marked increase in the frequency of events occurring in years 2 and 3. The
direct resource costs, excluding penalties, of the release strategy over the 100 year
simulation period are approximately $4 million a year. This reflects the cost of holding a
high security entitlement, the volume of water required to generate a continuous 14-day
high flow event, and the level of reliability achieved.
The release strategy is incorporated into the Murrumbidgee IQQM model to more
closely simulate actual flows at Wagga Wagga. The IQQM modeling of loss and
attenuation effects changed the profile of flows and the number of days at which the
target flow rate was maintained. That is, while on average the target was maintained,
fluctuations occurred between days that did not meet the relatively stringent definition of
a successful event. As a consequence, there was no increase in the predicted number of
14-day events at or above 30 gigalitres per day. Nevertheless, there is a substantial shift in
the flow regime with a large number of events being generated below the target threshold.
A comparison with Table 1 indicates that the number of high flow events of 25 gigalitres
a day restores about 50 per cent of the natural flow conditions for a 14-day event
(Table 2).