Warrie, Mogul, the issue of pre-release from a dam to potentially reduce flooding just before a flood event is a treacherous matter if there are significant catchments downstream of the dam.
The issue is not just about wasting water but equally, or more importantly, about significant potential to make flooding worse downstream of the dam compared to if no pre-releases occur.
Flooding downstream of Warragamba dam is not just from the dam outflows but also other catchments joining downstream as well.
Pre-releases free up dam space but also the released water is still travelling down the river when the rain comes - so if rain falls downstream of the dam on top of a river that has raised levels because of the pre-releases the flooding can be made worse.
Below is the Access G forecast 12Z for 3 day rain from 2200 AEST 18 March to 21 March with the Warragamba Dam catchment shown - it shows a significant area of the heaviest predicted rain being downstream of the dams - it would be potentially dangerous to pre-release in that situation.
But also I find such single images of forecast rainfall can be quite misleading, because the forecasts are often not accurate for where the heaviest rainfall will be.
The second animated image below is the same time period and same area (image is smaller to save file size) but instead shows each of fifty the EC ensemble members to demonstrate the uncertainty in the location and amount of the heaviest rainfall. When that is taken into account it is very difficult to be sure of rainfall upstream versus downstream of the dam.
Gotta say though you guys in NSW seem to have better consistency in forecasts with less location error than forecasts we get for SEQ which often resemble a "bingo board" (wide displacement error) for where heavy rain will be.
Anyway a lot of people think dams should pre-release based on forecasts, but unless the rain positioning is accurate it is a difficult matter to judge with the possibilities of worse flooding outcomes.