Understanding Climate Risk

Science, policy and decision-making

Northern Victorian Flood Review Part II

with 2 comments

This post examines the hydrological influences on past and present flooding leading to the January 2011 floods in northern Victoria. The floods themselves are discussed n Part I here, and affects on the family farm in Kerang are shown here. Part III on climatic influences is here.

Local data used are daily flows at Laanecoorie Reservoir on the mid Loddon River, and daily rainfall from the Newstead and Cairn Curran rain gauges upstream. Hydrological issues concern long-term catchment change and flow data. These affect the assessment of historical floods on the Loddon River.

If there is no real ability to model the likely affect of floodwaters on a floodplain, planning will focus on past floods. This is  problematic because catchments have been modified over time for flood mitigation, irrigation management, drainage and the protection of property. The following affects the Loddon River in particular but will be relevant to all the rivers affected by flooding last summer.

The flood of record on the Loddon River is August 1909. A flash flood led to the dam at Laanecoorie giving way, washing out the gauge, so the level and peak flow was estimated. The peak downstream at Kerang was higher than the second largest flood on record in spring 1975 – reflecting the amount of water coming from the dam failure. The flood hydrograph is dominated by this peak but the preceding rainfall is unexceptional. Twenty years ago I made a submission to the Kerang flood study suggesting that the runoff co-efficient assumed for the 1909 flood was unrealistic, its peak was influenced not by rainfall but by dam failure and that 1975 should be used as the flood of record as is the case for most other rivers in the region. This advice was ignored.

The catchment has also changed over time. In the past, the catchment was flashier than it is today. That is, rainfall very quickly became runoff and the resulting floodwater travelled very quickly down the river in high peaks that passed very quickly. This is due to two factors: gold mining, and water storages and river management. Gold mining led to many of the upper catchments in central Victoria being cleared for firewood to run stamping mills to crush the gold ore. Firewood collection for towns and bakeries especially was also a factor. These led to many of the box-ironbark and wetter forests in the upper catchments being cleared. Subsequent recovery after gold mining declined and electricity replaced wood fuel led to the forest recovery. Catchment management to prevent erosion via the Soil Conservation Authority also slowed runoff from cleared slopes in the mid catchment.

Water storages have allowed flow regulation; full-scale flooding does not occur until the storages are full and subsequent runoff has to be released. Even then, storages do slow down floodwaters. There are two reservoirs of significance on the Loddon: Cairn Curran and Laanecoorie. Farm dams on small tributaries in the mid catchments will also slow down runoff,  preventing levels of low flow that would otherwise occur.

These changes over time can be illustrated by the August 1909, September–October 1916, December 1933 and October 1975 floods. Shown are daily flows at Laanecoorie and rainfall at Newstead. Using a single station provides a guide rather than full analysis which would require total rainfall on the upper catchment. The 1909 floods clearly stand out as having a peak on the two vertical  scales: flow on the left and rainfall on the right. 1916 has a greater rainfall event with lower flows. 1933 was a big flood occurring on a wet catchment. The 1975 event is the largest modern flood and was the best guide to what could happen in terms of flood behaviour. Later, lesser floods such as those in 1981 and 1993 also provided experience for locals and professional managers. These graphs also show that rainfall and flow occurred on the same day in the earlier floods but by 1975 there was a day’s delay between rain events and peak flow. This is evidence of a healthier catchment but greater regulation will also be having an effect.

Four floods at Laanecoorie

Rainfall (mm) and daily flow in megalitres at Laanecoorie Reservoir in 1909, 1916, 1933 and 1975.

Floods further downstream have become even less peaked than at Laanecoorie in the mid catchment. Big floods are now large slow-moving bodies of water that move down the main trunk stream. The catchments have different flood behaviour in their upper-mid and lower reaches. Under natural conditions when water reached the floodplains it would move into prior (ancient) streamlines and anabranches (stream that flow out and back into a river) across the plains. The whole mid Murray River system in northern Victoria and southern NSW behaves like this. In Victoria this behaviour can be seen in a whole system of surface- and groundwater-controlled stream and lakes. In NSW, large wetlands have been formed along major rivers such as the Lachlan and Murrumbidgee.

In the modern systems, most of the floodplain away from the main streams is now farmland protected from flooding. In NSW, many of the elongated wetlands are farmed and cropped. Much of the recent expansion in these wetlands has taken place since the 1970s using low security irrigation water and is one reason why the Murray Darling Basin is over-allocated and unsustainable. This expansion is still taking place further northon the Darling River.

This pattern of understanding then suggests we have a typology of floods along the river systems of the Murray Basin:

  1. Small floods within single streamlines that are largely beneficial.
  2. Moderate floods in single streamlines that require some level of protection for adjoining property (if that is current policy).
  3. Large floods that affect multiple rivers and want to join across the floodplain utilising prior streams, anabranches and lakes.

The floods in January 2011 were clearly of type 3. Type 3 floods though, are largely being managed within single river systems and are informed by past records of flood heights. These, as I have already pointed out, are suspect for the first half of the 20th century. We can learn a lot from rainfall events during those periods, but less from the flooding behaviour. While the response to the recent floods was massive they could have been better managed. Understanding the climatic influences on major floods is a key part of this better management.

This has turned into at least a three-part posting. In Part III, I will cover the climatic influences that lead to the third type of  flooding and link it to the recent floods.

Advertisements

Written by Roger Jones

May 22, 2011 at 11:17 am

2 Responses

Subscribe to comments with RSS.

  1. Roger

    I respond with a little encouragement from your Mother

    I agree in general with your views on “Northern Victorian Flood Review Part 11” not having the technical data at hand to argue but more from a local observation aspect.

    There are, however, several other aspects that perhaps should be considered.

    All floods are different , and 2011 was no exception, and this makes it difficult to draw up a computer model that doesn’t need some localized anecdotal fine tuning. In this case, it was unprecedented to have 3 to 400 MM of rain in Northern Victoria that created run-off to the drainage lines before floods from the hills arrived.
    On the other hand, after 13 years of drought the water table overall, was at an unusual low – I estimate the Avoca Marshes would have absorbed 5 feet of water in the soil profile – a similar rain event next year and we will have a different kettle of fish !!

    This was the first year that CMA and SES were so intimately involved and with such supreme power. From a Chief Flood Wardens perspective, their actions were generally counter-productive and disgraceful, even treasonable with the waste of the nations resources on actions that indicated they had no idea of the Geography or the Hydrology of the terrain and so many opportunities of better flood mitigation were lost

    In concluding, as the debate now begins as to whether the Northern Flood Plain should be reconstituted for farming or left as a flood plain, I believe, that as this world heads for inevitable over population and starvation, food production will become the essence at, unfortunately, the expense of the environment and other considerations,

    I can see much soul searching taking place in the future

    Stuart

    Stuart Simms

    May 30, 2011 at 1:34 am

  2. Stuart,

    thanks very much for those comments. As for your extra points – I thought that but wanted some-one who had experienced the whole schemozzle first-hand to point that out.

    My view is that flood modelling gets the big picture right (for instance better estimates of flood heights), but locals are best for dealing with floodwaters at the local level. As you say, each flood is different and it takes some skill and experience to know how to best manage that.

    Roger Jones

    June 1, 2011 at 6:34 pm


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: