Storm tide statistics

Document Sample
Storm tide statistics Powered By Docstoc
					Storm tide statistics
The first image most Australians have of cyclone damage is     Ideally, the way to obtain statistical information on storm
Darwin after cyclone Tracy. Many Queensland coastal            tide occurrences is to analyse actual water levels recorded
residents also would have images of local damage, flooding     along the coast over many years. Unfortunately, water level
and beach erosion. However, worldwide, the most                data for sufficiently long periods will not be available in
devastating effect of cyclones in terms of human lives is      the foreseeable future. Yet it is possible to estimate storm
the catastrophic flooding of coastal lands caused by the       tide occurrence statistics using tide predictions and storm
associated very high sea levels or storm tides.                surges calculated from cyclone occurrence statistics.
                                                               Several procedures can be used to obtain such estimates
In places like Bangladesh, storm tide flooding has resulted
                                                               and the one considered most effective is described below.
in thousands of deaths. We are fortunate in Queensland to
have very few recorded instances where a storm tide has
been the main cause of significant damage or loss of life.
The highest storm tide reported in Queensland occurred at
Bathurst Bay on Cape York Peninsula in March 1899. A sea
level increase of 14 metres was reported, with the sea
penetrating inland about 5 kilometres.

Most storm tides are much less severe than the one at
Bathurst Bay. Yet they still contribute to damage in
Queensland coastal areas by flooding low-lying land,
preventing access to and from cyclone-damaged areas,
contaminating water supplies, and allowing higher-than-
normal wave action to cause beach erosion and attack
coastal structures. A good understanding of storm tides and
their frequency of occurrence is required to ensure that all
associated adverse effects are taken into consideration
when designing and locating coastal subdivisions and works.

Storm tide levels
A storm tide is a higher-than-normal sea level that occurs
due to the presence of a storm or cyclone. The height of
the storm tide is obtained by adding the effect the storm
has on the sea level (the storm surge) to the normal tide
level at the time.

The Beach Protection Authority has installed 21 special-
purpose storm tide recorders along the coast to supplement
the tide gauges already operated by the Maritime
                                                               The Beach Protection Authority storm tide recorder for the
Operations Division of Queensland Transport and various
                                                               Port Douglas area is located near the mouth of the
Harbour Boards. The Authority’s gauges have been designed      Mossman River.
to operate under the most extreme conditions and are
equipped with a telephone interrogation system to provide
storm tide information directly to the Bureau of Meteorology
and the State Emergency Service. They provide data used to
calibrate mathematical models of storm surge events.

                                                               Reference: Pattearson, C. Storm Tide Statistics. IN: “Beach Conservation,”
                                                               Issue No. 62, January 1986, Beach Protection Authority of Queensland.

    Prediction procedure                                             Storm tide statistics for Queensland
    While the storm tide data recorded for the coast of              In 1984 the Beach Protection Authority commissioned
    Queensland are not extensive at present, much more               consulting engineers Blain Brenner and Williams Pty Ltd to
    information is available on cyclone occurrences in other         carry out a study of this type to estimate storm tide
    regions. This data is accurate enough to give a usable           statistics for a number of sites along the Queensland coast.
    picture of cyclone occurrences and allow average cyclone         The sites were Surfers Paradise, Yeppoon, Mackay, Bowen,
    return intervals to be estimated. This is possible because,      Townsville, Lucinda, Mourilyan Harbour, Cooktown and
    unlike water levels which are site-specific, cyclone data can    Weipa (see Figure 1). Statistics are now available for all of
    be collected on a regional basis and applied to all areas        these coast sites but no report was produced on Weipa in
    within the region. The regional collection of data means         the Gulf of Carpentaria.
    that enough cyclone occurrences can be recorded to
                                                                     The situation in the Gulf of Carpenteria is made complex
    provide representative information on percentage
                                                                     by the shallow, semi-enclosed nature of the Gulf. Further
    occurrences of cyclone characteristics such as central
                                                                     research is required to develop suitable storm tide
    pressure, forward speed, radius of maximum wind, and
                                                                     prediction techniques for this area.
    direction of approach to the coast.
                                                                     Storm tide level information produced in this study is
    By assuming that future cyclones will follow patterns
                                                                     available in report form for all east coast sites. The data is
    similar to those of recorded cyclones, it is possible to
                                                                     presented as plots and tabulations of water level return
    create an artificial future history of cyclones for an area
                                                                     period, and plots of the likelihood that, once exceeded, a
    such that the average number, size and behaviour of the
                                                                     given water level will last for a certain period of time
    predicted cyclones is in the same balance as the recorded
                                                                     (persistence). Return interval plots are produced for both
    cyclones. This procedure is ideally suited to computer
                                                                     the central site in the report and a number of secondary
    analysis and its use enables the computer to assign dates
                                                                     sites in the region of the central site.
    and times to each predicted cyclone. The procedure is not
    intended to predict actual events and each time the
                                                                     Figure 1
    computer carries out a prediction a different future history
                                                                     Coastal locations for which storm tide studies were
    is obtained. However, each history has the same average          commissioned.
    distribution of cyclones and cyclone behaviour.

    Using a predicted cyclone history, it is possible to calculate
    the height and duration of the storm surge associated with
    each cyclone occurrence. This is done using a mathematical
    model of the area, with information on seabed levels and
    the properties and track of the cyclone being modelled.

    The tide level at the time of occurrence of the predicted
    cyclone is calculated from known tidal constants, in the
    same way as normal tidal predictions are made. In areas
    where there is little interdependence between tide and
    storm surge, the storm tide can be obtained by simply
    adding together the tide and storm surge level. This is the
    case for most of the east coast of Queensland.

    Using this procedure and projecting a cyclone history
    sufficiently far into the future, it is possible to generate
    enough storm tide occurrences at any site to calculate
    return periods for various storm tide levels. Typically, a
    prediction period of about 10 000 years is required to
    obtain sufficient return period data for planning and design
    of most coastal structures and developments.

    It should be remembered that, because daily weather
    patterns are not considered in this calculation, the
    procedure does not fully define all storm tide levels close
    to the highest tide level. This is not considered a major
    defect, as it is usually only storm events that are of
    interest to designers and planners.

       Table 1.
       Probability of occurrence of a storm tide of particular return period over various periods of time.

            Return period of                        Probability of occurrence (percent) for the period (years) shown
            storm tide (years)                25 years         50 years       100 years       200 years      500 years
                    50                          39               63              87              98            99·9
                    100                         22               39              63              87            99·3
                    200                         12               22              39              63            92
                    500                         5                9·5             18              33            63
                    1000                        2·5              5               9·5             18             39
                    5000                        0·5              1               2               4             9·5

This satellite photograph shows cyclone “Simon” as it moved along the coast near Yeppoon on February 24, 1980.
The diameter of the cloud formation shown is approximately 700 Kilometres.

    Warning                                                          Figure 2
    There is a popular misconception that an event can occur         Predicted storm tide level frequencies for Surfers Paradise.
    only once during its designated return period.For example,
    the 1 in 50 year event would be said to occur only once in
    any 50-year period. Unfortunately, this is not true.

    A return period represents the average number of events
    that occur in a very long period, but it is possible to
    experience the 1 in 50 year event in one year and again in
    the following year or even in the same year. It is better to
    look at the risk or probability of a storm tide level
    occurring over the design life of the development or works
    rather than to consider only the return period. Using this
    approach, both probability of occurrence and persistence
    can be combined to set realistic design storm tide levels.

    The risk or probability of occurrence of a storm tide level in
    any given period can be calculated from return period data.      Figure 3
    Table 1 gives an example of the results of such a                Predicted storm tide level frequencies for Mackay.
    calculation. It shows that there is a significant probability
    of relatively long return period events occurring in a
    relatively short design period. For example, there is
    approximately a 9·5 percent chance of a 1 in 500 year
    event occurring in any 50-year period. This is particularly
    important when designing coastal subdivisions as the
    design life of a township would be expected to be hundreds
    of years. There is a 98 percent chance of the 1 in 50 year
    storm tide occurring in any 200-year period, and a 4
    percent chance of the 1 in 5000 year storm tide occurring
    within the same period.

    Wave set up
    In addition to the storm tide, wave action causes a local
    rise in water level at the coast, known as wave set up. The
    magnitude of wave set up is dependent on the height of           Figure 4
    the waves. A suggested allowance for wave set up for each        Predicted water level persistence at Mackay.
    site is given with the storm tide return interval data
    presented in the study reports. This allowance is additional
    to the storm tide level given in the graphs and tables in
    these reports.

    The reports produced as a result of this study will benefit
    the Beach Protection Authority in its assessment of
    coastline behaviour, erosion-prone area widths and
    coastal management options. They will also benefit
    local governments, developers, and coastal planners
    and designers.

                                                                                                                    BP1114 July 1999

Shared By:
Description: Storm tide statistics