7/30/2019 Understanding Floods
1/38
AnswersQuestions
Understanding
foods:
7/30/2019 Understanding Floods
2/38
7/30/2019 Understanding Floods
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.
7/30/2019 Understanding Floods
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
7/30/2019 Understanding Floods
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
7/30/2019 Understanding Floods
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.
5 Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
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.
6Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
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
7 Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
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
8Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
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/).
9 Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
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
10Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
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.
11 Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
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.
12Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
14/38
Photo: Floods can have positive eects on the environment, particularly in wetlands. Photographer: Briony Masters
13 Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
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.
14Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
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
15 Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
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?
16Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
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
17 Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
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
18Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
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
19 Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
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
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.
20Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
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
21 Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
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?
22Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
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.
23 Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
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.
24Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
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.
25 Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
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.
26Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
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.
27 Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
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
28Understanding Floods: Questions & Answers
7/30/2019 Understanding Floods
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
Top Related