Northern Victorian Flood Review I
This post reviews some of the events that led to the northern Victorian floods of January 2011. Part II looks at the hydrological influences and Part III on the climatic influences on the floods. Recommendations for the flood review are made in Part III.
There are two flood enquiries happening in Australia at the moment. The one in Queensland is understandably much higher profile because of the loss of life and damage caused. It has the powers of a Royal Commission, so can subpoena, call witnesses and cross examine.
Victoria’s enquiry is a review. Former Police Chief Neil Comrie has been appointed to examine matters extending from flood prediction and warning all the way through to access to emergency funding and government services. Public meetings are being held in affected areas and written submissions are being taken until May 27.
The biggest difference is the Review is more reflective and will address the information delivered, whereas the Commission is investigative and is charged to assess contributing cause to the disaster and its management. It can interrogate.
Because the family farm is up in Kerang, I’ve kept a watch on the situation in northern Victoria as the drought conditions eased during 2010 and a la Niña delivered old-fashioned average rains. This rain felt like wet weather to most people, because it had been so dry. In August the weather became wetter with decent falls. Large rains in September caused a moderate flood and the Kerang back swamp filled up for the first time in years. At this stage, it was all good news – there was water flooding into parched wetlands and after all the rain, spring temperatures were fuelling huge amounts of growth.
My mother is a local flood warden and they had a meeting in Kerang in September to update people on the flood situation, one technical and one more public. She and I had discussed the situation, she warned the meeting that more floods might be on the way, using past experience as an example. I remember thinking maybe a 25% chance of a big one at that stage (that turned out to be very conservative). Some of the older heads in the district also felt the spring would be a wet one. We were going on previous years’ experience, particularly 1975, where the water reached the top of the levee bank. It was the town levee bank at that stage and we ended up with people from town building up the bank with shovels full of dirt as the water lapped over. Town and property both saved. November ’75, it was, so a late spring flood was on the cards. The latest record of a big flood was December 1933.
Because the drought had kept storages empty and wetlands on the decline, all the local storages had been filled up. The Kerang lakes, which can take water in a flood, and many other storages had been filled. Full storages provide insurance if it’s been dry. These lakes supply irrigation water during the irrigation season, so are usually pretty full in spring and summer, supply permitting. But full lakes also cuts flood mitigation capacity. So at that stage, the odds between flood vs drought management were probably sitting on a knife’s edge.
The local flood wardens didn’t hear back from regional authorities again in 2010 after the September meeting. There was another big flood in December. It got close to the top of the bank, nearing the 1975 level, but didn’t go over. Perhaps at that stage everyone thought that they’d dodged a bullet and flood season was over. That’s what history would suggest. But the northern catchments were very wet, with full mid and lower catchment storages, soil moisture capacity was full and only the rivers had emptied. Irrigation water was still being delivered, so the channel system was running at delivery capacity. Not full, but not empty, either.
In January this year, central northern Victoria got some of its largest falls ever. Falls in the week to Jan 16 exceeded 200 mm. Large parts of Victoria received their wettest January falls ever.
All that water had to go somewhere. The northern flowing rivers from the Kiewa to the Wimmera were bank-full or over-bank. The Murray had a flood peak from the Kiewa, King, Ovens and Upper Murray Rovers. The Goulburn and Campaspe arrived in Echuca, flooding out housing estates, and adding to Murray flows. The Loddon had a big slow-moving peak bearing down on Kerang. The Avoca was shooting through Boort and wanted to spread out over the floodplain between Kangaroo Lake, the Marshes and Benjeroop, joining the lower Loddon. The Wimmera flowed through Horsham, inundating houses, then north to Lake Hindmarsh, eventually with small amounts of water flowing into Lake Albacutya, last flooded in the 1970s. Swan Hill was threatened by the potential for Murray and Loddon-Avoca flows to join and flood the town.
Satellite photos of the floods are below, six days apart. The large black flows are on the Avoca, Loddon and Pyramid Creek systems, including anabranches and prior streams. Floodwaters in the Gunbower Forest along the Murray can be seen on January 26. The rivers joining to the north are the Murray and Murrumbidgee. The large lakes to the west are Lakes Buloke and Tyrell.
In the history of Victoria, these were big floods. Many people were affected. Valiant community efforts managed to save some areas and salvage others. But circumstances have changed from those in the past. In old days, a solid timber house would dry out. Many of the older houses this time around have since been condemned because the timber was no longer up to it. Composite materials cannot take any soaking at all, so newer doors and furniture all have to be replaced. Patterns of land-use have also changed. Many newer property owners had never before experienced major floods and didn’t know what to expect or where floodwaters were likely to go. Many properties have absentee landlords, so the old days of expecting that next door’s levee bank will be maintained and monitored have passed. Many banks have weakened due to drought and rabbits. Landforms have been modified by dozer operators who don’t know too much about local flood-ways and drainage lines.
Changes in the water management cycle over the past few decades have let certain elements – aspects of flood management being one – fall through the gaps.
Water management has become much more centralised – the last time major floods occurred, State Rivers offices were in most regional centres. Government amalgamations have merged organisations and corporatized parts of the water management cycle. Irrigation and supply are managed by corporate such as Goulburn Murray Water. Catchments are managed by catchment management authorities. They have responsibility for catchment management, particularly those relating to long-term condition, such as maintaining environmental flow targets.
The Bureau of Met is responsible for flood hydrology and manage hydrometeorological data as well. They’re very good at estimating how much rainfall is likely to turn into runoff. Recent changes to the BoM’s website where real-time monitoring gives rainfall, flood status and river levels in some places is really useful. The biggest gap is when the water gets onto the floodplain and is charging downstream. Understanding how high peaks are likely to get and exactly where the water will go is a guessing game. There isn’t a good flood modelling capacity in Victoria and flood management is based mainly on the statistics of past floods. The January rains delivered record volumes of water to the Loddon and Avoca at least, therefore there was no precedent in terms of past floods (the one they used, 1909, is faulty as I’ll go into in the next post).
Emergency management has also changed. It’s much more centralised, so the flood is being managed from remote locations. Emergency management operates local hubs but may be controlled by people who work in general emergency management rather than flood management. Individuals may not have experience with flooding let alone local flooding. When floodwaters were rising, if we wanted sand-bags to build up weak spots in the levees we’d have to call a central number. The legitimacy of the request would be checked against a database to ensure against needless waste. And they’d get back to us. At some stage.
When the floodwaters were overtopping the farm banks, this system was still in place. Which is no good if you want access to critical resources yesterday. And because there was no flood modelling capacity, we had no information on how high the peak was likely to get, so it wasn’t clear as to how much needed to be done. In the end, we lost it and the flood peak exceeded the previous peak by at least 10 cm. And then sat there for over a day.
After the flood peak passed Kerang, people on the lower Loddon felt they’d been given up, because the town had been saved and those downstream faced record inundation with limited resources. A few people took it upon themselves to do all they could, clocking up thousand dollar mobile phone bills as they coordinated evacuation, bank works and general community support. In the face of these and other great acts, the review will be asking, could the whole situation have been better managed, and what are the lessons for the future? The short answers are “Yes”, and “There are many lessons to be learnt”. Even if average conditions remain dry, there will be more big floods. Not a question of if, but when.