Understanding Floods

7/30/2019 Understanding Floods

1/38

AnswersQuestions

Understanding

foods:


2/38


3/38

From December 2010 to January 2011,

Western Australia, Victoria, New South Wales and

Queensland experienced widespread ooding. There

was extensive damage to both public and private

property, towns were evacuated and 37 lives were

lost, 35 o those in Queensland. Three quarters oQueensland was declared a disaster zone, an area

greater than France and Germany combined, and

the total cost to the Australian economy has been

estimated at more than $30 billion.

Since the beginning o 2011, oods have led to major

devastation and personal tragedy around the world.

At the same time as the Australian oods, more than

800 people died in oodwaters and mudslides in

Brazil and South Arica recorded 70 ood related

deaths. Many lives have also been lost due to

ooding in the Philippines, Pakistan and Sri Lanka.

Most recently, some areas o the Mississippi River

recorded the worst ooding since the 1930s.

The Queensland Floods Commission o Inquiry

was established to investigate the 201011 ood

disaster. To support this process, we have convened

a comprehensive panel o technical experts, rom

across Australia and internationally, to provide

scientifc and engineering perspectives around oods.

Our report does not examine the specifc events o

the recent Queensland oods, but rather ocuses on

a number o critical, underlying questions relevant

to oods generally. These questions include: what

do we mean by a ood; what are the causes and

consequences; how can we orecast and warn aboutthem; and how do we best plan or oods? We have

asked ourselves what processes and technologies are

well utilised, what is not being used eectively, and

where are the gaps in our understanding.

Finally, recognising that the science and engineering

raternity are oten criticised or somewhat inscrutable

language, this report has been written with more

general communication in mind. We have ocused on

clarity and brevity, ounded on rigour.

I am grateul or the time, enthusiasm and insights

rom our panel members and other specialists

we have consulted. Their diverse and extensive

expertiseand challenging yet collaborative style

has been vital to the integrity o this report.

Dr Geo Garrett AO

Queensland Chie ScientistChairman, Queensland Floods Science, Engineering

and Technology Panel

June 2011

Foreword

Page

Introduction .................................................... 3

Summary......................................................... 4

Q1: What is a ood?........................................ 5

Q2: What actors contribute to oods? ............ 9

Q3: What are the consequences o oods?...... 14

Q4: How do we orecast oods? ...................... 16

Q5: How do we communicate and warn

about oods? ........................................... 19

Q6: How do we estimate the chance o

a ood occurring?..................................... 22

Q7: How do we manage ood risks?................ 25

Q8: What does the uture look like?................. 29

Glossary .......................................................... 33

Acknowledgements ......................................... 35

Contents

Copies o this publication are available online

at www.chiescientist.qld.gov.au

The State o Queensland 2011. Published by the

Queensland Government, July 2011. Copyright protects

this publication. Excerpts may be reproduced withacknowledgement to the State o Queensland.


4/38

Australia has an extremely variable climate, and

is truly a land o droughts and ooding rains.

Throughout Australias long history, the ooddrought

cycle has been a natural part o lie, with periods

o severe drought ollowed by extensive ooding

playing an important and defning role in shaping theAustralian landscape and how we live.

Early Australians typically established settlements on

oodplains, along waterways and on coasts, where

ood and water were plentiul. As a result, oods have

had a proound eect on human lie and property. As

devastating as recent events have been, they are not

unique: 77 oods were recorded in Australia in the

past 35 years o the 20th century; eight major oods

were recorded in Australia in the 19th century; 23 in

the 20th century; and six in the frst decade o the

21st century. And nature will undoubtedly continue to

surprise us into the uture.

It is also important to recognise that oods can have

some benefcial consequences, or example through

replenishing water resources. Most o Australias

unique ora and auna have adapted to and depend

on ood cycles, relying on the oods to trigger

breeding, disperse seed, provide ood sources and

connect habitats.

In order to reduce the risk o oods to communities,

economies and environments into the uture, it

is important that lessons rom past oods, and

advancements in knowledge and technology, are

eectively communicated and applied.An important contribution the science and

engineering community can make is to help reduce

this risk, by minimising the chance that communities

and inrastructure will be ooded, and mitigating

the negative impacts when oods occur. We know

a lot about ood risk: more than 1000 Australian

ood studies have been conducted, and scientists

and engineers have developed a very sophisticated

armoury o methods to orecast and manage oods to

reduce risk. However, there is still uncertainty about

the many interacting actors that inuence such an

event, how these actors are changing in time, and the

consequences o a ood i it occurs. Moreover, natureis unpredictable, so no matter how detailed and

clever our calculations and management strategies

may be, there will always be a risk o ood.

O course, social science and government policy also

play pivotal roles in reducing the negative impacts

o oods, improving emergency responses and

optimising recovery o communities ollowing a ood.

Improvements in this regard rely not just on social

science research, but also on government leadership

and community awareness and engagement.

Given the science and engineering outlook o this

document, ood emergency responses and recovery,

which are primarily rooted in social science and

policy, will not be addressed. This report concentrateson oods caused by rainall and on three key

themes to understanding oods. The three themes

are oods and their consequences, ood orecastsand warnings, and managing oods. The ollowing

paragraphs expand on these three themes, posethe questions we have sought to respond to, and

summarise the answers.

Introduction

Flood peaks in eastern Australia over the period 26November 2010 20 January 2011.

Courtesy o the Bureau o Meteorology

3 Understanding Floods: Questions & Answers


5/38

Floods and their consequences

Q1: What is a ood?

When water inundates land that is normally dry, this

is called a ood. Floods can be caused by a number

o processes, but the dominant cause in Australia israinall. Floods are a natural process, but mankinds

activities aect ooding. Floods occur at irregular

intervals and vary in size, area o extent, and duration.

Q2: What actors contribute to oods?

Rainall is the most important actor in creating a

ood, but there are many other contributing actors.

When rain alls on a catchment, the amount o

rainwater that reaches the waterways depends on

the characteristics o the catchment, particularly its

size, shape and land use. Some rainall is captured

by soil and vegetation, and the remainder enters

waterways as ow. River characteristics such as sizeand shape, the vegetation in and around the river,

and the presence o structures in and adjacent to the

waterway all aect the level o water in the waterway.

Q3: What are the consequences o oods?

Floods impact on both individuals an d communities,

and have social, economic, and environmental

consequences. The consequences o oods, both

negative and positive, vary greatly depending

on the location and extent o ooding, and the

vulnerability and value o the natural and constructed

environments they aect.

Flood orecasts and warnings

Q4: How do we orecast oods?

Weather orecasts can provide advance warning

o a ood, and seasonal orecasts can alert o a

heightened chance o ooding in the coming months.However, orecasting river levels and ood extent is

a complex process that is continually being improved.

Q5: How do we communicate and warn

about oods?

Flood warning systems turn orecasts into messages

designed to reduce the negative impacts o oods.

Warning systems should be accurate, timely and

reliable. Prior community awareness o ood risk can

make warnings more eective. Improving our warning

systems could reduce social losses rom oods.

Managing oods

Q6: How do we estimate the chance o

a ood occurring?

Understanding the chance o dierent sized oods

occurring is important or managing ood risk. The

chance o a ood event can be described using

a variety o terms, but the preerred method is

the Annual Exceedance Probability (AEP). A ood

with a one per cent AEP has a one in a hundred

chance o being exceeded in any year. Currently,

the one per cent AEP event is designated as having

an acceptable risk or planning purposes nearlyeverywhere in Australia. However, good planning

needs to consider more than just the one per cent

AEP ood.

Q7: How do we manage ood risks?

Flood risk includes both the chance o an event taking

place and its potential impact. Land use planning

inormed by oodplain management plans can reduce

risk or new development areas. Flood risk is harder

to manage in existing developed areas; howevermodifcation measures such as dams or levees can

change the behaviour o oodwaters. Similarly,

property modifcation measures can protect against

harm caused by oods to individual buildings, and

response modifcation measures help communities

deal with oods.

Q8: What does the uture look like?

Australias growing population and changing climate

patterns imply that the characteristics o the oods

we experience will change in the uture. Better uture

land use planning and oodplain management can

mitigate the impacts o ooding. Appropriate urban

design can reduce the severity o ood impacts.

Catchment and waterway revegetation can reduce

the impact o ooding. Emerging technologies can

improve our ability to predict and manage oods.

Summary

4Understanding Floods: Questions & Answers


6/38

When water inundates land that is normally dry

this is called a ood

Every ood is dierent. They can occur suddenly and

recede quickly, or may take days or even months

to build and then discharge. They occur at irregular

intervals, and many decades can pass between

signifcant oods. On the other hand, there are many

examples o several large oods occurring within

short periods o time.

In Australia, many people live on land that is subject

to occasional ooding, known as oodplains. Cities

and other settlements have been constructed on

oodplains to take advantage o access to water and

good quality armland.

Floods in Australia are usually caused by rainall

Floods can be caused in a number o dierent ways;however the dominant cause o ooding in Australia

is rainall.

When rain alls over an area o land, some is absorbed

by the soil, while the rest becomes runoand owsdownhill. The area o land that contributes runo to a

particular point is called the catchment.

The amount o rainall, the intensity o the rainall over

time (the temporal pattern) and the distribution o therainall over an area o land (thespatial pattern) can

all vary widely. The oods that are produced by this

rainall are thereore equally variable, that is, every

ood is dierent.

Floods can also be caused by other mechanisms.

Recent international events have highlighted the

risk o ooding rom tsunamis. Large tides and

storm surges can also ood coastal areas. Inland

earthquakes, volcanoes or land slips can also cause

ooding, as can breaches/releases in natural or man

made barriers to ooding such as dams and levees.

Floods can occur suddenly

Heavy, intense rainall can occur suddenly, and the

quickly rising oods caused by this in the minutes

or hours ater the rainall are known asash oods.Flash oods are typically associated with relatively

small catchment areas where there may be little or

no permanent ow o water. As there is little time to

react, ash oods are particularly difcult to predict

and manage in real time, and this is discussed urther

in Q4 and Q5.

Floods can occur slowly

In larger catchment areas, rainall can build up over

hours, days or weeks. The runo rom this rainall

ows across land and then down gutters, drains,

gullies, creeks and rivers and may create signifcant

oods that inundate large areas o land or days,

weeks or months.

With more time to react, ood warning is more

eective or these types o oods, as described in Q4

and Q5.

Sometimes, a ash ood in the upper reaches o a

river system can evolve into a more general river ood

as it joins with other inows an d spreads out as it

travels downstream.

Many locations can be aected by both ash oods

and the more general river ooding. For example, a

particular residence might be potentially aected by a

ash ood through the local gully, or a river ood rom

the nearby major river, or a combination o the two.

Q1: What is a ood?

Photo: Flash ooding in Toowoomba, 10 January 2011. Image courtesy o Nicole Hammermeister.



7/38

Figure 1. Characteristics o oods. Conceptual diagram developed using the Integration and Application Network (IAN) tool (www.ian.umces.edu/symbols/). The lower sixth representsa subsurace crosssection.



8/38

Floods are a natural process

Floods are a natural process, and our ecosystems,

river systems and estuaries have adapted over long

periods o time to depend on an irregular pattern o

large oods. Many species, such as River Red Gums,

rely on Australias pattern o dry periods separated by

periods o intense rain and overbank ooding.

Mankinds activities aect ooding

We make a lot o changes to our catchments including

land clearing, urban development and dams, that

can change the impact o a ood on the natural

environment.

Floods can vary in size

The size o a ood event, or its magnitude, can beexpressed in many ways. The peak level o the water

at a particular location in a waterway is the most

unambiguous way, as it is relatively easy to measure

and is the principal driver o ood impact. The ood

magnitudes are usually classifed by their height,

Figure 2. Highest annualood heights at the Brisbane

City gauge, 18402011.Minor, moderate and major

ooding levels or theBrisbane City stream gauge

are highlighted.

Sources: Bureau oMeteorology, State Library

o Queensland (API033010016courtesy o the

State Library o Queensland;181874 Ross Webster;

2780300010248

State Library oQueensland).

Charlotte St 1893 Charlotte St 1974 Charlotte and Albert St 2011



9/38

and the Bureau o Meteorology uses three general

categories o ooding related to water level:

Major: This causes inundation o large areas,

isolating towns and cities. Major disruptions occur

to road and rail links. Evacuation o many houses

and business premises may be required. In ruralareas, widespread ooding o armland is likely.

Moderate: This causes the inundation o low lying

areas requiring the removal o stock and/or the

evacuation o some houses. Main trafc bridges

may be closed by oodwaters.

Minor: This causes inconvenience such as closingo minor roads and the submergence o lowlevel

bridges and makes the removal o pumps located

adjacent to the river necessary.

Figure 2 shows these general categories as they

pertain to the main Brisbane stream, or river, gauge.A stream gauge is a device used to measure the

height o water in the river, and there is a network

o gauges throughout the Brisbane River catchment.

The values on this chart are in metres above the

reerence level defned or this gauge. It is important

to note that the river heights or minor, moderate and

major ooding are dierent or dierent waterways,

depending on their individual characteristics.

Other important characteristics o oods that

contribute to their severity include:

The total amount o water in the ood, or the ood

volume. The ood volume contributes both to thelevel and duration o ooding. The ood mitigation

ability o dams and detention basins are less or

large volume oods.

How ast the ood rises, orrate o rise. A oodthat rises quickly obviously provides less time or

warning and evacuation.

How ast the water is owing, i.e. the ow velocity.

Faster ow causes a higher risk to human lie,

a higher risk o erosion, and more damage to

inrastructure.

The duration o ooding. A ood that lasts or a

longer time provides a greater impact owing to the

increased duration o the disruption to transport,

business and personal networks.

The areal extento ooding. Flooding that aectsa larger area, either within a river basin or across

multiple basins, provides greater impacts.

Floods occur at irregular intervals

Figure 2 also illustrates the sporadic nature

o ooding, showing the historic record o river levels

at the Brisbane City gauging station since 1840.

As illustrated, six major oods occurred in Brisbane

between 1885 and 1910, ollowed by more than

60 years without a major ood.

The chance o a ood o a certain level occurring is

usually reerred to in terms o the likelihood o that

level being exceeded in a par ticular year, orAnnualExceedance Probability (AEP). Q6 provides urther

inormation on what AEP means, and how it is

estimated and used.

Photo: Flooding o the Fitzroy River in Rockhampton, January 2011. Photographer: Michael Marston



10/38

Rainall is the most important actor in creating

a ood

Put simply, oods occur when the amount o water

owing rom a catchment exceeds the capacity o

its drains, creeks and rivers. This process begins

with rainall, but is aected by many other actors

(Figure 3).

In Australia, ooding is heavily inuenced by our

naturally high rainall variability which, relative to

other parts o the world, leads to a much higher

variability o the amount o water owing through

our waterways. A major actor in this variability

is the El NioSouthern Oscillation (ENSO) eect

(see Figure 4).

In Queensland, average annual rainall ranges rom

very low values in the southwest, to very high values

exceeding 2000 mm per year along the coast (Figure

5). However, even in those areas with generally low

rainall, relatively heavy rainall will occur in some

years, causing ooding (Figure 6).

Longterm climate change and variability may also be

having an inuence on rainall (a matter addressed

in Q8).

Catchments convert rainall into owing water

When rain alls on a catchment, the amount o

rainwater that is converted into ow down rivers and

other waterways depends on the characteristics o the

catchment.

Some rainall is captured:A portion o the rainthat alls on a catchment is captured by soil and

vegetation. Generally, the more rain that alls in a

particular area in a given period o time, the lower the

proportion that can seep into the ground or be stored

on the surace.

Q2: What actors contribute to oods?

Figure 3. An illustration o the actors that contribute to oods. These actors vary between locations and times, meaning thatno two oods are the same. Conceptual diagram developed using the Integration and Application Network (IAN) tool(www.ian.umces.edu/symbols/).



11/38

The atmosphere and the oceans interact strongly to

inuence our weather.

Much o Australias rainall variation rom year to

year is caused by the natural climate phenomenon

known as ENSO, the El Nio Southern Oscillation.

ENSOs seesaw variations are intimately related to

variations in the atmospheric vertical circulation along

the equator over the Pacifc. This circulation, known

as the Walker Circulation, is caused by dierences in

sea surace temperatures between the eastern and

western Pacifc along the equator.

During normal circulation warm, moist air travels

west across the Pacifc and rises over Indonesia,

producing cloud and rain. The airstream then

becomes comparatively dry

and moves east at high altitude (approximately

12,000 m) and sinks over the normally cold waters

near the South American coast.

There are various measures o the El Nio Southern

Oscillation. One o these, the Southern Oscillation

Index (or SOI), measures the dierence in air pressure

between the eastern Pacifc Ocean (measured at

Tahiti) and the equatorial area around northern

Australia and Indonesia (measured at Darwin).

When the equatorial ocean surace o the coast

o South America is abnormally cool, the Walker

Circulation is strengthened. In this situation the SOI

is strongly positive, and the trade winds blow strongly

across the warm Pacifc, picking up plenty o moisture

(Figure 4a). This increases the likelihood o eastern

Australia experiencing above average rainall, and iscalled a La Nia event.

On the other hand, when the ocean surace o the

coast o South America is abnormally warm, the air

pressure between the eastern and western Pacifc

equalises or becomes a negative value, weakening

or reversing the trade winds. This situation, which

is a weaker than normal Walker Circulation

(Figure 4b), is accompanied by a strongly negative

Southern Oscillation Index and is called an El Nio.

In Australia this usually results in below average

rainall, and i this trend persists we can slip into

drought. The SOI helps tell us how strong a La Nia

or El Nio event is. For example, when the SOI

is consistently strongly positive (i.e. La Nia and

above average rain) we may experience ooding.

When the SOI is consistently strongly negative we

risk entering into periods o drought (Figure 4c).

Figure 4a

12 000 m

HL

Positive SOI - La Nia

Trade Winds increase

across the Pacific

Cooler waterWarmer water

H

E

G

E

N

D

Negative SOI - El Nio12 000 m

LLH L

Trade Winds

decrease

Cooler water Warmer water

H

Figure 4b

Figure 4. El Nio Southern Oscillation

Figure 4c

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

20

25

30

35

SouthernOscillationIn

dex

1970

1972

1974

1976

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

Monthly SOI

Graphics Library

Australian



12/38

Figure 5. Australian annual average rainall or the climate period 19611990. Theperiod 19611990 is the current climate standard (climate normal) used by the Bureau

o Meteorology or climate comparisons through time. Figure courtesy o the Bureau oMeteorology.

Copyright Commonwealth of Australia, 2010

Based on a standard

30-year climatology

(1961-1990)

Rainfall

(Millimetres)

3000

2000

1500

1000

600

400

300

200

100

50

0

Geraldton

PERTH

Albany

Carnarvon

PortHedland

Wiluna

Telfer

Broome

Kalumburu

Halls Creek

Giles

Kalgoorlie-Boulder

Esperance PortLincoln

PortAugustaCeduna

ADELAIDE

Marree

Oodnadatta

Alice Springs

TennantCreek

Katherine

DARWIN

Cook

Weipa

Kowanyama

Cairns

Townsville

Mackay

Normanton

Mount Isa

Birdsville

Mildura

Horsham

MELBOURNEWarrnambool Orbost

CANBERRASYDNEY

Dubbo

CoffsHarbour

BRISBANE

Charleville

RockhamptonLongreach

Bourke

Cape GrimSt. Helens

HOBART

Strahan

Newman

Projection: Lambert conformal

with standard parallels 10 S, 40 Soo

0

100

200

300

400

500

600

1961

1963

1965

1967

1969

1971

1973

1975

1977

1979

1981

1983

1985

1987

1989

Rainfall(mm)

Year

Total Annual Rainfall

Average Annual Rainfall

Figure 6. Annual rainall variability as measured against the longterm average orBirdsville, western Queensland or the period 19611990. The period 19611990 is the

current climate standard (climate normal) used by the Bureau o Meteorology or climatecomparisons through time. Data courtesy o the Bureau o Meteorology.



13/38

The greater the rainall intensity, the greater the

potential or runo. How long it rains, and the area

covered by the rain, are also important.

The more vegetation there is in an area, the greater

the amount o rainall that is captured and the less

water there is available to ow over the surace.Natural and artifcial storages like arm dams and

rainwater tanks have a similar eect in reducing

runo.

The soil types in a catchment, land use and weather

conditions prior to a rainall event are also important

as they control the amount o rainall that can infltrate

into the soil, and hence the amount o rainall which

becomes ow. I a large storm is preceded by a period

o wet weather, then the g round has little capacity

to absorb urther rainall, and a higher proportion o

the rainall will ow across the land surace and into

waterways. The construction o areas that cannotabsorb water, such as roos and roads, will also result

in reduced infltration and more rainall being turned

into runo.

Rainall that is not captured enters the waterways:Once water begins owing in a catchment, various

actors determine how much ows downhill into

successively larger waterways, and how quickly it

moves.

Typically, larger catchments result in greater

streamow i widespread rainall occurs or a long

time. The steeper the catchment area, the aster the

runo will ow.

Floods are also aected by the roughness o the

terrain being passed over. Dense vegetation and

artifcial obstacles such as ences and houses will

slow down water ow, oten leading to lower ood

levels downstream.

Swamps and natural ponds or lakes have the capacity

to store oodwater and release it slowly.Artifcial

structures such as dams or detention basins (small

reservoirs) can also store water or a period o time,

and reduce the peak o downstream ows while

extending the duration o an event. All such structureshave a fnite capacity and there is a limit to the

volume o catchment ow that can be stored.

River characteristics aect water levels

The capacity o drains, creeks and rivers within a

catchment to carry ows depends on a number o

actors:

Size and nature o the river: Put simply, the bigger,

straighter and smoother a river, creek or other

channel, the greater its capacity to carry water and the

less prone it is to ooding. Any process that reduces

this capacity, such as the placement o structures inthe channel, encroachment by development or build

up o sediment, contributes to increased ooding.

Vegetation in and around the river: Plants in a riveror on its banks slow the speed o the water owing in

it. The slower the water moves, the higher the water

level, and the greater extent to which the oodplain

surrounding the river will be inundated. This can

reduce downstream ood levels and ows. Plants

also reinorce river banks, decreasing erosion and

increasing the deposition o sediment.

Once a river overtops its banks, the maximum

ood level reached depends greatly on the natureo the adjacent oodplain. For example, wide, at

oodplains can store a greater volume o oodwater

than steepsided valleys, and the resulting oods

move more slowly. Modifcations to oodplains

such as clearing o vegetation or the construction o

embankments (or example, or a ood ree road or

rail corridor) can impact natural drainage patterns and

processes on river oodplains.

Structures: Structures that are placed in a creek or

waterway, or example culverts in an urban drainage

system or bridges in a river, reduce the watercarrying

capacity o the waterway and may contribute toooding. Debris can also become entangled on these

structures, worsening this process.

Levees along a waterway are designed to protect areas

behind the levee rom oods up to a certain level,

but their constraining inuence on ood ows can

cause upstream ood levels to be higher than they

otherwise would be. Road and railway embankments,

with insufcient crossdrainage capacity (or example,

use o culverts), can block o parts o the oodplain

with a similar eect. Once levees or embankments are

overtopped or breached, the way oodwaters spread

over a oodplain can alter signifcantly and the impacto ooding is oten severe.

Downstream water levels: The capacity o waterwayscan also be aected by the water level in the ocean or

lake they are owing into. For example, a king tide or

storm surge can hamper the release o water rom a

river into the ocean. A similar eect can occur near the

junction o creeks with rivers, where backwater eects

rom river ooding can extend a signifcant distance

up the creek.



14/38

Photo: Floods can have positive eects on the environment, particularly in wetlands. Photographer: Briony Masters



15/38

Q3: What are the consequences o oods?

The consequences o oods, both negative and

positive, vary greatly depending on their location,

duration, depth and speed, as well as the vulnerability

and value o the aected natural and constructed

environments. Floods impact both individuals and

communities, and have social, economic, andenvironmental consequences (Table 1).

Floods have large social consequences or

communities and individuals

As most people are well aware, the immediate

impacts o ooding include loss o human lie,

damage to property, destruction o crops, loss o

livestock, and deterioration o health conditions

owing to waterborne diseases. As communication

links and inrastructure such as power plants, roads

and bridges are damaged and disrupted, some

economic activities may come to a standstill,people are orced to leave their homes and normal

lie is disrupted.

Similarly, disruption to industry can lead to loss o

livelihoods. Damage to inrastructure also causes

longterm impacts, such as disruptions to supplies

o clean water, wastewater treatment, electricity,

transport, communication, education and health care.

Loss o livelihoods, reduction in purchasing powerand loss o land value in the oodplains can leave

communities economically vulnerable.

Floods can also traumatise victims and their amilies

or long periods o time. The loss o loved ones has

deep impacts, especially on children. Displacement

rom ones home, loss o property and disruption

to business and social aairs can cause continuing

stress. For some people the psychological impacts

can be long lasting.

Photo: Atermath o a ood: ood damaged household itemsawaiting collection and cleanup ater the Brisbane oods,

January 2011. Photographer: Michael Marston

Can the lost item be bought andsold or dollars?

Direct loss:Loss rom contact with ood water

Indirect loss:No contact loss as a consequenceo ood water

Yes monetary (tangible) e.g. Buildings and contents,

vehicles, livestock, crops,inrastructure

e.g. Disruption to transport, loss

o value added in commerce andbusiness interruption, legal costs

associated with lawsuits

No nonmonetary ( intangible) e.g. L ives and injur ies, loss o memorabilia, damage to cultural or

heritage sites, ecological damage

e.g. Stress and anxiety, disruptionto living , loss o community, loss

o cultural and environmental sites,

ecosystem resource loss

Table 1. Types o loss rom oods. Modifed rom Disaster Loss Assessment Guidelines, Emergency Management Australia, 2002.



16/38

In Australia oods are the most expensivenatural disasters

In Australia, oods are the most expensive type o

natural disaster with direct costs estimated over the

period 19672005 averaging at $377 million per year

(calculated in 2008 Australian dollars).Until recently, the most costly year or oods in

Australia was 1974, when oods aecting New South

Wales, Victoria and Queensland resulted in a total

cost o $2.9 billion. The Queensland Government

estimates costs or the 2011 oods will exceed this

fgure or Queensland alone; with the damage to local

government inrastructure estimated at $2 billion, and

the total damage to public inrastructure across the

state at between $5 and $6 billion.

Flooding in key agricultural production areas can

lead to widespread damage to crops and encing and

loss o livestock. Crop losses through rain damage,waterlogged soils, and delays in harvesting are urther

intensifed by transport problems due to ooded

roads and damaged inrastructure. The owon eects

o reduced agricultural production can oten impact

well outside the production area as ood prices

increase due to shortages in supply. On the other

hand, ood events can result in longterm benefts to

agricultural production by recharging water resource

storages, especially in drier, inland areas, and by

rejuvenating soil ertility by silt deposition.

Damage to public inrastructure aects a ar greater

proportion o the population than those whose

homes or businesses are directly inundated by

the ood. In particular, ood damage to roads, rail

networks and key transport hubs, such as shipping

ports, can have signifcant impacts on regional and

national economies.

Shortterm downturns in regional tourism are oten

experienced ater a ooding event.

While the impact on tourism inrastructure and the

time needed to return to ull operating capacity may

be minimal, images o ood aected areas oten

lead to cancellations in bookings and a signifcant

reduction in tourist numbers.

Flooding o urban areas can result in signifcantdamage to private property, including homes and

businesses. Losses occur due to damage to both the

structure and contents o buildings. Insurance o the

structure and its contents against ooding can reduce

the impacts o oods on individuals or companies.

Floods have signifcant consequences orthe environment

In many natural systems, oods play an important

role in maintaining key ecosystem unctions and

biodiversity. They link the river with the land

surrounding it, recharge groundwater systems,

fll wetlands, increase the connectivity between

aquatic habitats, and move both sediment and

nutrients around the landscape, and into the marine

environment. For many species, oods trigger

breeding events, migration, and dispersal. These

natural systems are resilient to the eects o all but

the largest oods.

The environmental benefts o ooding can also help

the economy through things such as increased fsh

production, recharge o groundwater resources, and

maintenance o recreational environments.

Areas that have been highly modifed by humanactivity tend to suer more deleterious eects rom

ooding. Floods tend to urther degrade already

degraded systems. Removal o vegetation in and

around rivers, increased channel size, dams, levee

bank and catchment clearing all work to degrade the

hillslopes, rivers and oodplains, and increase the

erosion and transer o both sediment and nutrients.

While cycling o sediments and nutrients is essential

to a healthy system, too much sediment and nutrient

entering a waterway has negative impacts on

downstream water quality. Other negative eects

include loss o habitat, dispersal o weed species, the

release o pollutants, lower fsh production, loss o

wetlands unction, and loss o recreational areas.

Many o our coastal resources, including fsh an d

other orms o marine production, are dependent on

the nutrients supplied rom the land during oods.

The negative eects o oodwaters on coastal marine

environments are mainly due to the introduction o

excess sediment and nutrients, and pollutants such

as chemicals, heavy metals and debris. These can

degrade aquatic habitats, lower water quality,

reduce coastal production, and contaminate coastal

ood resources.

Photo: Flood damage ater the Toowoomba ood disaster,January 2011. APN Regional Media



17/38

The Bureau o Meteorology (BoM) is the lead national

agency or ood orecasting and warning services

in Australia, working in partnership with agencies

at the state and local government levels. The BoM

provides orecasts o the water level in rivers at critical

locations during ood events. Local governments andemergency agencies may urther interpret the river

water level orecasts and provide advice on ood

inundation extent. The BoM also provides severe

weather warnings that include risk o ash ooding.

In addition, the BoM provides orecasts o rainall

and river ow or the coming three months. These

seasonal orecasts may help alert agencies, and the

public, o entering a period o heightened chance

o ooding.

Weather orecasts provide advance warning

o a oodReliable orecasts o weather, in particular rainall,

can allow advance warning and orecasting o oods.

Weather orecasts or the next one to seven days rely

on increasingly accurate computer models o the

atmosphere and ocean/atmosphere interactions.

Dramatic improvements in the data available to such

models (rom satellite observations) and in computing

power have contributed to this increased accuracy. In

some parts o the world, threedayahead orecasts

o heavy rain are now as accurate as onedayahead

orecasts were in the mid1990s (www.hpc.ncep.

noaa.gov/images/hpcvr/hpc20yr.gi).

Radar (and sometimes satellite) images can be useul

or tracking areas o heavy rain and their movement

(Figure 7). Rainall in the next one to our hours may

be orecast based on these images in combination

with computer models. However, such orecasts give

only a very short lead time (the time between when aorecast is made and the orecasted event occurs)

or response.

The accuracy o climate and weather orecasts varies

with lead time, spatial scale (or size) o the region

o interest, the weather or climate variable being

orecast (or example, rain, thunderstorm), as well

as with latitude. Generally, temperature orecasts

are more accurate than rainall orecasts. The midlatitudes (in Australia 30S to 60S) are easier to

orecast than the tropics (so Melbourne has more

accurate orecasts than Darwin). It is generally easier

to orecast when the lead time o the orecast is

relatively shortso a sevenday orecast is usually

less accurate than a orecast o tomorrows weather.

Finally, it is generally easier to orecast rainall over a

large area (or example, a large catchment) than local

rainall (or example, a reservoir). This is because the

intensity o any rain system varies on small spatial

scales, but the variation is somewhat averaged out

over a large area.

Rainall orecasts can be used to extend the lead timeor ood orecasts. However, because orecasts o

rainall or specifc locations and timing are not ully

accurate, ood orecasts based on rainall orecasts

are oten subject to signifcant uncertainty.

Forecasting river levels and ood extent

is complex

Flood orecasts are critical to emergency responses to

limit property damage and avoid loss o lives.

Flood orecasters rely heavily on realtime data

about rainall and river water levels as well as rainallorecasts. A network o rain gauges (sometimes

combined with radar images) are used to monitor

rain that has allen on the catchment. Water levels

(i.e. river height) at stream gauging stations along the

river are also measured. A simple river height recorder

is shown in the picture. The orecasters then use

hydrological computer models to work out how much

rainall will run o dierent parts o the catchment,

how long it will take or runo to reach the river, how

long that water will take to travel rom upstream to

downstream, and how water rom dierent tributaries

converges in the river network.

Flood orecasters estimate the river ow rate at

various key locations and lead times and convert theestimates to river water level orecasts (Figure 8).

Flood orecasts by the BoM are issued to emergency

management agencies and the public through the

media and the BoMs website. The orecasters

regularly update their orecasts as new observations

are made o rainall and river water level, and as

rainall orecasts become available.

Because rain that has allen on a catchment takes

time to travel to the outlet o the catchment, river ow

downstream o the catchment within a certain period

will largely be inuenced by rain that has already

allen on the catchment and been observed. Thismeans that the river ow orecast or this period will

be reasonably accurate. River ow orecasts beyond

this period will be less accurate as it is necessary to

use rainall orecasts.

I a critical dam operation is involved in a ood event,

the orecasters communicate with dam operators.

Decisions about releasing water rom dams take into

account orecasts about how much water will ow

into the dam an d assessments o how water releases

may aect water levels downstream. In turn, ood

orecasts or downstream areas take into account

water release decisions.

Forecasts o river water level are most useul when

interpreted in terms o where the water is likely to

spread beyond the river. Such interpretations may

be provided to the public by local governments and

emergency agencies, usually based on preprepared

ood maps using historical ood data and in some

cases oodplain hydraulic models (see Q6 or

more inormation).

Q4: How do we orecast oods?



18/38

Figure 7. Radar image showing rainall over the SouthEast Queensland and New South Wales coast, 9 January 2011. The intensityo the rain (interpreted rom droplet size) is indicated by the changing colour scale. Courtesy o the Bureau o Meteorology.

Photo: A manual stream gauging station on the Fitzroy River,Rockhampton, January 2011. Note the historical reerence

to heights o previous oods: 1918, 1954 and 1991.

Photographer: Michael Marston



19/38

New technologies are available, but not yet widely

used in Australia, or providing near realtime

mapping and delivery o orecast ood inundation

extent on the internet. These technologies use

accurate groundelevation data, robust oodplain

hydraulic models, new spatial inormation technologyand internet mapserving sotware. Adoption o the

technologies would signifcantly enhance the value o

ood orecasts in Australia.

Flash oods are difcult to orecast, although

technologies are available and used operationally

overseas or ash ood orecasting. These

technologies are generally based on monitoring o

rainall using rain gauges and radar images, high

resolution rainall orecasts or the next ew hours,

understanding o the catchment condition (how

much rainall will run o) and understanding o the

local drainage systems (how much water is needed

to cause a ood). As ash ood orecasts improve inaccuracy and are integrated with communication and

response systems over time, they can become highly

valuable.

Seasonal orecasts alert o a heightened chance oooding in the coming months

The BoM regularly issues orecasts o rainall and

temperature or the coming three months (www.

bom.gov.au/climate/ahead/). The prospect o a wet

season would lead to an increased chance o ooding,

so orecasting seasonal rainall can help alert us o

ood risk. The high variability o rainall across mosto Australia rom one year to another is largely the

result o the El NioSouthern Oscillation (see Q2

or more inormation). Computer models are used to

predict the development o the La Nia events that are

oten associated with heavy rains. However, seasonal

orecasts are not highly accurate, and are expressed

only in probabilistic terms, i.e. percentage chance o

occurrence. For example, Queensland may have low

rainall on some occasions even in very strong

La Nia events. Nevertheless, such orecasts have

been demonstrated to be useul or industries such

as agriculture, water resources and fnance, indicating

that the rainall orecasts may also be useul ororecasting seasons with an increased chance

o ooding.

In December 2010, the BoM commenced a seasonal

streamow orecasting service. This service provides

orecasts o probabilities o total ow exceeding

various volumes in the coming three months

(www.bom.gov.au/water/ss/). The orecasts are

also based on computer models, taking into account

how wet and dry the catchments are at the start o

the season as well as the climate. Although orecasts

o oods are not directly made at the seasonal time

scale, it is reasonable to expect an increased chanceo ooding when total streamow volume in the next

season is orecast to be high.

In addition to the services provided by the BoM, some

state governments and universities also provide

seasonal orecast outputs including probabilistic

streamow orecasts in some instances.

Figure 8. Modelled versus observed water heights or the Mary River at Gympie, Queensland, together withobserved rainall. While there is reasonably good agreement in this instance, as can be seen rom the dierences

between the red and green lines, actual river height can vary rom what we expect based on modelled results due

to data and model limitations. (This model red line was run on 18 January 2011; no urther observations green line were taken ater 20 January 2011 or the above diagram.) Data courtesy o the Bureau o Meteorology.

-1

0

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

0

1

2

3

4

5

6

7

8

9

4/01/1110:00

6/01/1110:00

8/01/1110:00

10/01/1110:00

12/01/1110:00

14/01/1110:00

16/01/1110:00

18/01/1110:00

20/01/1110:00

22/01/1110:00

24/01/1110:00

26/01/1110:00

Rivergaugeheight(m)

Rainfall(mm/30minutes)

Date/Time

Rainfall and modelled and observed gauge height for the Mary River at Gympie,

Queensland 04/01/2011 to 26/01/2011

Rainfall

Modelled River Height

Measured River Height

Minor

Moderate

Major



20/38

Flood warning systems turn orecasts into

messages designed to reduce losses

A warning system consists o a number o key steps:

monitoring rainall and river ow rate; making

orecasts about river water levels and ood extent;

interpreting orecasts or their meaning in terms oimpact on those at risk; composing and disseminating

warning messages; response by those at risk and

emergency services; and review and improvement

(Figure 9).

In Australia, ood warning systems or rivers involve

the BoM, which provides orecasts o river water

heights at specifed locations to relevant authorities,

and to the public via broadcast media and the BoMs

website. Local government or State Emergency

Services then interpret the BoMs orecasts to provide

local inormation on areas likely to be aected,

potential impacts and action advice to those atrisk. Messages are also disseminated throughout

the community, particularly by personal or inormal

networks. Social media (or example, Facebook,

Twitter) will almost certainly play an increasingly

important role in the uture. This will also raise

challenges due to the potential risk o

misinormation.

Flash oods account or most ooding atalities in

Australia and currently present the most challenges

due to the limited warning time. While the BoM

provides severe weather warnings, which can include

the risk o ash ooding, specifc ash ood orecasts

and warnings (i.e. including specifc location and

timing inormation) are not generally provided.

However, some local governments have warning

systems or these events.

Q5: How do we communicate and warn about oods?

Figure 9. Components o a ood warning system, noting that the precise arrangements vary between states.While ash oods are not covered in this fgure, the Bureau o Meteorology also provides severe weather

warnings, which can include general warnings o potential ash ooding, via its website. Compiled by the

Science, Engineering and Technology Panel.

Forecast

interpretation

Flood forecasts

Monitoring

Warning

communication

Further

dissemination

Response

Radio, websites, press conferences, television, text and phonemessages, emails, social media, sirens/alarms and speakerphone, doorknocking, noticeboards, letterbox drop

Households and individuals, businesses, local and stategovernments, State Emergency Services, Police

Bureau ofMeteorology

Monitoring andmodelling rainfall and

river water levels

Bureau ofMeteorology

Forecasts of riverheights

Warning

construction

Local Government

Flood predictions:areas, times and

heights of flooding(Lead agency in

Qld)

State Government /State EmergencyServices / Police

Flood predictions;infrastructure and

people at risk

Local Government/State Government/State EmergencyServices / Police

Risk and responseadvice

Dam operators andother infrastructureowners/operators

Impacts on assets;operational decisions

Review

andIm

provement

Unofficial or personal forms of communication:word of mouth, social media, print media



21/38

Warning systems should satisy a number o

technical and communication attributes

They should be inormative: Warning messagesshould indicate what the threat is, what action should

be taken, by whom and when, in understandable,

unambiguous and consistent language. A warningalso needs to have personal meaning or those at

risk (whether individuals, agencies or businesses).

This means going well beyond a specifed river height

to indicating the area likely to be covered by the

ood, its depth and speed in terms o locally relevant

landmarks. In turn, this requires understanding o the

relevant river systems.

They should be accurate: As warnings are predictionsabout the uture, there is inevitably some uncertainty.

Uncertainties can also arise rom the construction and

wording o warning messages themselves.Engaging

with the community and business groups aected hasbeen shown to improve understanding o the issues

and reduce the chance that u ture warnings will be

ignored.

They should be timely: Warnings need to allow

enough time or appropriate action. This is particularly

a challenge or ash oods.

They should be trustworthy: Warnings are more likely

to be heeded i they come rom multiple trusted

sources.

They should reach the appropriate audiences:The audience or a warning will typically consist

o many subgroups, each with its own needs and

expectations, preerred way o receiving warnings,

and own ways o interpreting messages.No one

warning source will reach, or be un derstood by,

everyone.Warning systems work best when designed

with the needs and expectations o the ultimate users

in mindsomething best achieved with their input.

The capacity o individuals to receive or respond to

Direct communication; chance to ask questions; high credibility

Resource intensive; requires access to flooded area

Direct, specific communication

Requires access to flooded area; difficult to hear

Electricity not required; widest reach home, work, travelling

Variable accuracy; requires public to be listening

Quick dissemination; becoming very widespread; capacity for images

Electricity/internet required; variable accuracy

Landlines becoming less common; people often not at home/indoors

Quick dissemination, but usually has to be actively accessed; power

and telecommunication infrastructure needed; internet required

Useful for roads, infrastructure and location-specific information; can

be controlled remotely

Electricity required; variable accuracy; limited reach; requires public

to be listening

Ability to reach almost all audiences, but may miss youth

Slow; requires access to flooded area

Informative/detailed; ability to reach wide audience

Time needed; variable accuracy

Uses info from multiple sources; persuasive

Variable accuracy

Sirens/alarms

Quick; reliable; limited information and reach, but becoming more

versatile with voice and remote capabilities

Doorknocking

Speaker phone

Text message

Can reach wide audience very quickly; no power needed

Less reliable for areas with poor mobile phone coverage

Radio message

Websites/

social media

Automated

telephone

Email

Noticeboards

Television

Letterbox drop

Print media

Word of mouth

Informative

Accurate/trustworthiness

Timeliness

Audiencereach

Varyingaudienceca

pacities

Reliable/resilient

Littlelabourrequired

Works well for this aspect

Satisfactory for this aspect

Limited use for this aspect

Does not support this aspect

Variable for this aspect

Table 2. Pros and cons o dierent ood warning communication methods. Compiled by the Science, Engineeringand Technology Panel.



22/38

warnings may be reduced because o disabilities, age,

language, or other commitments.

They should be reliable:Warnings need to work

under extreme conditions (or example, in inundated

areas, in the absence o electricity), as this is when

warnings are most needed. A variety o warningsources increases the likelihood that warnings will be

maintained throughout a ood event.

Prior awareness o risk can make warnings

more eective

For warnings to achieve their aims, people need to

know about their communitys ood risk, what actions

will improve their saety, and how they will receive a

warning to implement those actions. Pr ior awareness

and preparation are key to success.

Those issuing warnings and advice need to know

what sorts o actions are easible or those who are

being warned. This requires mutual learning between

government agencies with ormal responsibility or

the warnings, and the people who need to respond to

warnings. People need to see the personal relevance

o the warnings to their situationsthis is a major gap

in many communities.

Community education programs can use a wide range

o approaches, including community and industry

acilitated climate, weather and ood workshops

that identiy community and industry risk and key

management decisions. Another option is to run ood

rehearsals or drills. A challenge isto maintain oodawareness and preparedness during the periods

between oods.

Improving our warning systems couldreduce losses

Improvements in warning systems, in particular in

community response, is one o the most costeective

means by which we could reduce the economic and

social losses rom oods and save lives.

Areas where current systems could be improved

include:

Persuading people to take eective protective

action once a warning has been issued. We know

how to raise awareness, but not how to ensure

action. This is a signifcant knowledge gap.

Using more locally relevant inormation so that

people relate personally to warning messages, and

know what to do or their own saety.

Reducing uncertainty in predictions while providing

enough time or eective action, by harnessing

the promise o advances in ood modelling

and communication technologies. To date, the

advances in modelling and modern inormation

and communication technologies have had limited

impact on overall warning system eectiveness.

Communicating the uncertainties in warnings,

possibly through a mix o verbal qualifers like

may, could, be at least and probabilities.

However, public appreciation o numerical

probability statements is understandably limited.

Providing eective warnings or ash ooding. Thisis currently a major gap across Australia. Technical

advances may now make ash ood warnings

easible, but the issues o rapid decision making

by all the agencies involved and by those at risk

would need to be addressed.

Improving methods to evaluate warning perormance.

Photo: Warnings should be accessible to the whole community. Here, government and emergency personnel provide warningsor the community during a live television broadcast, January 2011. An Auslan interpreter translates these warnings or the dea

community. Photographer: Hugh OBrien



23/38

Understanding the chance o dierent sized oods

occurring is important or managing ood risk

The best method or calculating the chance o

dierent sized oods occurring is statistical analysis

o longterm ood records rom stream gauging

stations. Where a longterm ood record exists,and no signifcant changes have occurred to the

catchment, a statistical technique known asoodrequency analysiscan be used to determine the

likelihood o oods o dierent sizes occurring at

a specifc site in the uture (Figure 10). However,

Australias ood records do not extend ar into the

past, and ood events are highly variable, meaning

there is still a level o uncertainty in defning such

ood estimates. Climate change may also aect how

much we can rely on past ood records.

Where sufcient ood records do not exist, or a very

rare ood needs to be estimated,rainall basedtechniques are used. These use statistical analyses orainall records, together with computer models based

on the geographical characteristics (or example,

catchment area, waterway length) o the region being

studied, to determine the chance o dierent sized

oods occurring. These models can be set up to take

account o changes that aect runo, such as new

dams and urbanisation, but the computer models

used to convert rainall to runo are not perect,

making rainall techniques generally less reliable than

the use o longterm ood records.

Both o these techniques result in predictions or

peak water ows at key locations in rivers. These

predictions are translated into ood levels at any

point o interest in the oodplain, through the use

o urther computer models known as oodplain

hydraulic models.

Figure 10. Estimating the chance o oods o dierent size based on ood requency analysis o historical oodrecords at Bellingen, NSW. There is always a level o uncertainty inherent in such analyses. For example, the chance

o a ood with a stream ow o 2,200 m3/s (as arrowed, let hand axis) in any year is estimated to be between 1

in 50 (2%) and 1 in 10 (10%). This is said to be within 90% confdence limits, i.e. we are 90% sure that it will bein this range with a 10% chance we will be wrong, and it will be outside this range, higher or lower. The more

confdence there is in the data the closer the confdence limits (red dashed lines) will be to the estimate (black line).Courtesy o WMAwater.

Q6: How do we estimate the chance o a ood occurring?



24/38

Floodplainhydraulic models are virtual

representations o the river and its surrounding

land, or oodplain. They incorporate things such as

ground levels, roads, embankments and r iver sizes

to estimate predicted ood ows. The output o the

models includes representations o predicted ood

levels and the predicted speed o water ow.

It should be noted that any actual ood event will

vary in some manner rom the theoretical events

rom oodplain computer models (Figure 11). Such

variations are primarily due to dierences in the

rainall pattern, along with other actors such as how

wet the catchment was beore the event, and whether

the centre o the storm was over the lower or central

reaches o the catchment. (See Q2 or more on these

variations.)

The chance o a ood event can be described usinga variety o terms

Floods are oten defned according to their likelihood

o occurring in any given year. The most commonly

used defnition in planning is the 1 in 100year

ood. This reers to a ood level or peak that has a

one in a hundred, or one per cent, chance o being

equalled or exceeded in any year. Similarly, a 1 in

200year ood has a one in two hundred, or 0.5 per

cent, chance o being equalled or exceeded in any one

year.

Other terms that express the same idea, such as

one per cent annual exceedance probability(or oneper cent AEP), are preerred because they avoid the

common misconception that a 1 in 100year ood,

or example, can only occur once every 100 years;

or that you are sae or another 100 years ater you

experience such an event. For example, in Kempsey,

NSW, major oods approaching the one per cent AEP

level occurred in 1949 and then again a year later

in 1950.

In reality, the chance o experiencing dierent

sized ood events in a given period o time can be

estimated mathematically (see Table 3). I you lived

or 70 years in a location that had a one per cent

chance o ooding in any one year (that is, it would

only ood i a 1 in 100year ood occurred), then

there would actually be a 50 per cent chance, or one

in two odds, o you experiencing at least one ood

during that 70 year period. The way we calculate

this is: 100 per cent minus the chance o a ood not

happening 70 times in a row, i.e. 0.5 = 1 0.9970.

It is also important to remember that the chance that

you will be aected by a ood is not only dependent

on the likelihood o your own property ooding.

Floods can disrupt transport networks, impact tourist

destinations and prevent ood rom reaching markets.

With more than 100 rivers and creeks in Queensland

the chances are good, when ooding occurs, that

many people will be aected, either directly orindirectly (see Q3 or more inormation).

Good planning needs to consider more than just

the one per cent AEP ood

In nearly all but a ew locations, even the best plans

are not able to ully eliminate the chances o a ood.

Nevertheless, or planning purposes, it is importantto decide what level o ood risk is acceptable or

individuals and the community. This should take into

consideration both the chance o a ood happening

and the consequences o a ood (see Q7 or more

inormation).

Until about 30 years ago, it was common to use

the largest historical ood in an area or planning

purposes, and this is still used in some rural

locations.

However, currently nearly everywhere in Australia

the one per cent AEP event, or 1 in 100year ood,

with an appropriate additional height (or reeboard)or buildings is designated as having an acceptable

risk or planning purposes, regardless o the potential

consequences o the ood.

Chance o a ood o a particular sizebeing exceeded in any one year

Chance o experiencing a ood in a 70 year period

at least once at least twice

10% (1 in 10 odds) 99.9% 99.3%

5% (1 in 20 odds) 97.0% 86.4%

2% (1 in 50 odds) 75.3% 40.8%

1% (1 in 100 odds) 50.3% 15.6%

0.5% (1 in 200 odds) 29.5% 4.9%

Table 3. Probabilities o experiencing a given size ood once or more in a lietime. Modifed rom Floodplain DevelopmentManual: the management o food l iable land, NSW Government, 2005.



25/38

There are oten strong social and economic reasons

or considering a higher standard than the one per

cent AEP ood. For example, in some locations ood

levels associated with rarer oods are signifcantly

higher and are likely to cause signifcant devastation;

inundation o a particular location may have

signifcant economic and social consequences or

a much wider region.

For example, London is moving to a planning level

above the 1 in 500year ood (0.2 per cent AEP) or

land adjoining the Thames estuary. Also, many parts

o the Netherlands use planning levels above the 1

in 1000year coastal ood event (0.1 per cent AEP),

because inundation o large, low lying areas would

have major consequences.

Figure 11. Post event comparison o ood extent modelled (predicted) by a oodplain hydraulic model (blue) and an actualood event (red line) in Wagga Wagga, NSW, 1974. While predictions are mostly very good, so me variations can be observed

between predicted and actual (observed) ooding, e.g. in the right hand side o the image. Courtesy o WMAwater.



26/38

Flood risk includes both the chance o an event

and its potential impact

Flood risk is a combination o the chance o a ood

occurring and the consequences o the ood or

people, property and inrastructure (Figure 12). The

consequences o a ood depend upon how exposedthe community is to ooding and how vulnerable its

people, property and inrastructure are to the oods

impacts.

Managing risks rom oods may involve altering

the chance o ooding aecting a community;

and/or reducing the impacts o ooding by reducing

the communitys vulnerability and exposure to

ooding. The methods that are eective in reducing

ood risk are very location specifc. There is no one

sizeftsall solution and a variety o measures are

generally necessary to reduce risk.

Flood risk management is a partnership betweengovernment and the community using a range o

measures to reduce the risks to people, property

and inrastructure. Decisions on managing ood risk

should be made in consultation with the community

that may be impacted by oods.

Preparing a oodplain management plan that outlines

how ood risk to existing and uture development can

be managed or a particular location can inorm these

decisions. This process involves more than simply

ensuring that building oors are above a particular

ood level. It also considers how ood behaviour and

hazard may vary in dierent parts o the oodplain

and how dierent sized ood events might have an

impact on people, property and inrastructure.

Floodplain management plans can reduce risk or

new development areas

Managing ood risk is generally simpler in new

development areas. Preparing a oodplain

management plan enables strategic decisions about

where, what and how to develop the oodplain

while reducing residual ood risk (i.e. the risk let

ater management measures are put in place) to an

acceptable level.

Local councils can use local planning instruments to

inuence the longterm development o an area in

consideration o ooding, by restricting the location

o development (zonings) and placing conditions

(controls) on development.

Zonings can limit the impacts o new development

on ood behaviour in other areas and the exposure

o people and property to risk by locating new

development away rom areas where:

The main oodwaters ow. Development in these

areas may alter ood behaviour aecting otherproperties. Maintaining these owpaths can also

provide green corridors through cities.

The speed and depth o water make it hazardous

to people, property and inrastructure.

It is not possible to evacuate people to ood ree

areas and there is no practical alternative.

Zonings can also limit the development o the

remaining available land by considering:

The use o community acilities during a ood.

Critical acilities, such as emergency hospitals,

should ideally be located in areas where they will

not ood and can operate during a ood event(see Figure 13).

The vulnerability o occupants to ooding.Aged care and disabled acilities should generally

be located in areas where they can be readily

evacuated to dry land.

The vulnerability o buildings and contents.

Homes and their contents are generally more

vulnerable to ooding than industrial and

commercial buildings and thereore should be

located in less vulnerable areas.

Conditions on development can include: minimum flllevels or land and minimum oor levels or buildings

(to reduce how oten people and property are exposed

to ooding); building regulations (that reduce the

potential or structural building damage); and the

ability to evacuate people to ood ree areas (which

may aect the way land is developed).

Flooding can also have signifcant impacts on

inrastructure, which needs to be considered

when designing inrastructure. Appropriate design

standards or inrastructure exposed to ood risk can

reduce its vulnerability to ooding.

Q7: How do we manage ood risks?

Flood risk

Chance

of a flood

Consequences

of a flood

Exposure

of a property/communities

to flood waters

Vulnerability

of a property/communities

exposed to flood waters

Figure 12. Components o ood risk.



27/38

Flood risk is harder to manage in existingdeveloped areas

Flood risk is harder to manage where development,

or the right to develop, already exists. Flood risk

to existing inrastructure is usually altered through

improvements to protection as part o any upgrade.

However, or people and property there are basically

three ways o managing ood risk to reduce the

consequences o ooding: by modiying oodbehaviour, property, or community response.

None o these measures is a standalone solution or

addressing ooding issues. The preerred option is

oten a combination o ood, response and property

modifcation measures to reduce risk to an acceptable

level and to manage this residual risk appropriately.

Flood modifcation measures change the behaviouro oodwaters

Flood modifcation measures aim to reduce ood

levels, velocities or ows, or exclude oodwaters rom

areas under threat or events up to their designed

capacity. They are a common and proven means o

reducing damage to existing properties under threat

rom ooding. They tend to be more expensive than

response or property modifcation measures but willoten protect a larger number o properties.

Flood mitigation dams can reduce downstream oodlevels by temporarily storing and later releasing

oodwaters. Most dams are used to supply water to

the community, but they can, when purpose built,

also provide some ood mitigation or events up

to their ood storage capacity. In larger oods, this

mitigation capacity can be exceeded and oods

pass through with little, i any, reduction. On the

negative side, dams can cause disruption to existing

communities, loss o valuable land and negative

environmental impacts, and good sites or dams are

difcult to locate. Detention basins act like dams but

at a much smaller scale and are most suitable or

green feld developments, where sizing constraints

tend not to exist.

Leveesare generally raised embankments built to

eliminate inundation o the areas protected by the

levee up to a certain size event. In larger oods,

levees can be overtopped with water ooding into

and inundating areas protected in the smaller

events. Levees can trap local stormwater, causing

damage unless ood gates and pumps are provided.

However, levees, whether temporary or permanent,

can increase ood levels in areas not protected by thelevee (as noted in Q2).

Waterway or oodplain modifcations such aswidening, deepening, realigning or cleaning rivers and

owpaths can improve the transport o oodwaters

downstream and reduce the likelihood o blockage,

but can increase velocities and erosion and cause

negative environmental impacts. The benefts o

cleaning and clearing are only temporary unless these

continue to be maintained.

Other structures such as roads, railways and

embankments also have an impact on ood risk

management because they can alter ood ows andbehaviour. Floodgates can also be used to prevent

backow rom river systems into drains.

Figure 13. Land use planning should consider ood behaviour and risk. Probable Maximum Flood (PMF) is anestimation o the largest possible ood that could occur at a particular location. Critical public inrastructure such as

hospitals and emergency management centres are ideally located outside the inuence o the PMF. From Q6 we notethat a 10% ood is a 1 in 10year ood and a 1% ood is a 1 in 100year ood. Courtesy o WMAwater.



28/38

Figure 14. Property modifcation measures to manage ood risk in new and existing development areas. Thisschematic was developed utilising the Integration and Application Network (IAN) tool (www.ian.umces.edu/

symbols/).

Property modifcation measures can protect

against harm caused by oods

In addition to the zoning and development controls

or new and redevelopments mentioned above,

modifcations to existing property are also essential i

the growth in uture ood damage is to be contained.

Land flling involves building up lowlying areas

and can improve the ood immunity o structures

constructed on that land, but can adversely aect

ood behaviour elsewhere and thereore is generally

limited to the ringes o the oodplain.

Flood proofng involves the sealing o entrances,

windows, vents, etc. to prevent or limit the ingress

o oodwater. Generally it is only suitable or brick

commercial buildings with concrete oors and it

can prevent ingress or outside water depths up to

approximately one metre. Ideally, new developments

would use ood resilient designs and materials, asaddressed in Q8.

House raising is widely used to reduce the requency

o inundation o habitable oors, thereby reducing

ood damage. This approach provides more exibility

in planning, unding and implementation than

removal o development. However, its application

is limited as it is not suitable or all building types

and only becomes economically viable when above

oor inundation occurs requently (or example, on

average at least once in every 10 years). It also does

not remove the risk to people who occupy the house,

particularly in larger ood events.

Removal o developmentis generally only considered

where there is signifcant potential or atalities to

residents and/or potential rescuers due to ooding,

and where other measures are not able to reduce

these risks. There are instances where a largeproportion o, or an entire town has been relocated

due to ooding. For example Clermont, Queensland,

was relocated to higher ground ater the ood o

1916. This approach generally involves voluntary

purchase and demolition o the residence to remove

it rom the oodplain. Voluntary purchase has no

environmental impacts, although the economic cost

and social impacts can be high. Communities oten

oppose such schemes due to the impact on their

community, surrounding property values and way o lie.



29/38

Response modifcation measures help

communities deal with oods

Measures to modiy the response o the community to

a ood are essential to deal with residual ood risk,

because development controls and ood mitigation

works generally cannot deal with all possible oods.Response modifcation measures can include: ood

warnings, upgrading ood evacuation routes, ood

evacuation planning, ood emergency response and

ood education programs.

Implementing eective ood response within the

community can reduce the danger and damage

associated with oods. Flood warning and evacuation

plans can be very cost eective and may, in some

cases, be the only economically justifable risk

management measures. However, like all mitigation

measures, they require ongoing maintenance and

support. This is discussed urther in Q5.

Photo: In ood prone areas, houses with habitable oors built above the potential ood height can reduce the loss and damageassociated with oods. Photographer: Michael Marston



30/38

Climate change will aect ooding patterns

Because ood events are inuenced by a number o

actors, based on the current science it is difcult to

confdently state that, overall, extreme ood events in

Queensland will increase in intensity or requency as

a result o climate change. However, increased coastalinundation rom sea level rise and increased chance

o ash ooding because o an increase in shortterm

heavy rainall events both seem highly likely, based

on current assessments o projected climate change.

There are our specifc questions about climate

change that have relevance to uture ooding in

Queensland:

Will the systems that drive our regional climate beaected?This question is relevant because o the

strong tendency or widespread ooding to occur

during La Nia events, the wetter extreme o the El

NioSouthern Oscillation, ENSO (see Q2 or moreinormation). I La Nia events, or their eects on

Queensland rainall, became more requent or more

intense because o global warming, we can expect

more requent ooding. Currently it is projected that,

in the uture, ENSO variations may be dierent rom

those in the recent past. However, we are not currently

able to project confdently what those changes will be.

Will average rainall and short-period rainall events

change?Average rainalls in SouthEast Queenslandare projected to increase in summer and decrease

in winter. Regarding shortperiod (e.g. less than 24

hour) rainall events, the Intergovernmental Panel onClimate Change recently concluded that it was likely

such heavy precipitation events would become more

requent over most land areas. This could lead to

increased ood risk, especially or ash oods.

A reevaluation o probable maximum precipitation

could be required, along with ood resilience capacity

or critical inrastructure such as dams and bridges.

Will the sea level rise?Global warming is expected to

lead to sea level rise, increasing the risk o ooding

near the coast, including the lower reaches o coastal

rivers.

Will changes in storminess aect coastal ooding?

Any increase in the requency or intensity o stormscould lead to increased storm surge risks, and this

would exacerbate the increased likelihood o coastal

inundation arising rom projected sea level rise.

However, it is very difcult at present to predict how

storms might change. Current predictions are or a

decrease in the global number o tropical cyclones,

but with a possible increase in the intensity o the

strongest cyclones and more intense rainall.

Better land use planning and oodplainmanagement can mitigate the impacts o ooding

The uture will see Australias population continue togrow, placing increased pressures on our waterways,

many o which already experience high levels o ood

risk. A growing population will result in increased

development on the oodplain and the temptation

to build in ood corridors. Rising land prices and a

resulting move to smaller block sizes are expected to

result in our cities becoming more densely populated,

increasing the chance o ooding in the cities. More

houses built closer together increases the number o

houses potentially exposed to ood damage.

Better strategic land use planning will be

essential to limit the growth o ood riskThis will require improved ood studies and

oodplain management plans to enhance our

understanding o how oods will behave under

changing climate and catchment conditions.

Implementing the fndings o these studies and

management plans will help constrain development

rom areas where it would increase the negative

impacts o oods on other properties, including

the delineation o ow corridors to support the

sae passage o oodwaters through urban areas.

Improved ood studies and management plans would

also help restrict development rom areas where

there would be an intolerable ood hazard to people

or property, or where people could not be readily

evacuated to dry land, or a practical alternative, in the

case o a ood.

In areas that are deemed suitable or development,

these improved ood studies and management plans

can help identiy the types o developments that will

be suitable or specifc locations and the development

controls necessary to reduce any residual ood

risk in these areas to an acceptable level. These

controls could include minimum fll and oor levels,

assessing and delineating ood evacuation routes

to ood ree land, designing subdivision layouts toacilitate staged evacuation and providing innovative

building designs. Inormation rom these studies

can also inorm regional land use planning, local

environmental plans and development control plans.

Future development should also be guided by

detailed ood models (Figure 15). These models

would enable all impacts o proposed new urban land

uses to be quantifed, and or issues such as ood

evacuation strategies, the im

Understanding Floods

Documents

Transcript of Understanding Floods