In this issue:

Professor Russell Blong

NHRC is kindly sponsored by:
Swiss Re Australia
QBE Insurance
Benfield Greig Australia
Guy Carpenter & Co



Probably the most severe direct impacts of the June 1998 East Gippsland (Victoria) floods were to the rural sector. Two weeks after the floods, the toll was put at about 20,000 sheep; 3,000 cattle; 330 km of destroyed fencing  and 415 km of damaged fencing. One vegetable farmer estimated losses at $300,000 to $400,000 when 70 per cent of the cabbage and cauliflower crop was filled with silt and rendered unsaleable. Particularly distressing to the farming community was the timing of the floods - right on the heels of a two-year drought. Heavy silting of riverside pastures effectively means an extension of the agricultural drought. The loss of topsoil through erosion of the drought-affected slopes has long-term consequences.

In its urban impacts, the East Gippsland event was no Nyngan (1990) or Katherine (1998) event.  Losses were diffuse - in Bairnsdale less than 10 commercial and residential buildings were flooded from the Mitchell River; probably a few more buildings were flooded in Lakes Entrance as the lakes overflowed; some houses were flooded near Swift's Creek, Ensay (further downstream) and Bruthen as the Tambo River and its tributaries rose above the banks. Although over-floor depths were not often more than 30 cm (1ft), some tourist lodges sustained large damages to carpets and chipboard cupboards, and owners of some houses faced large bills for the reconstruction of jetties and for restumping. Clean-up costs - including labour hire - were a substantial expense for some commercial premises.

Power failure attributed to strong winds led to the loss of frozen goods in several businesses that were not flooded, and a general store at Ensay was unable to sell petrol for two days. The collapse of phone lines meant that EFTPOS was unavailable. Attempts to use a generator at one supermarket in Orbost led to a destructive fire.

A far-reaching indirect effect of the floods was the loss of income due to the cancellation of holidays. This affected tourist sites inundated or isolated by the floods, but also the entire region, perhaps fostered by inaccurate media reporting. Lakes and Wilderness Tourism (responsible for the promotion of tourism in East Gippsland) estimated a 50 per cent downturn from the equivalent school holiday season in 1997.

The Victorian Government has been prominent in relief, with the establishment of flood recovery centres for financial and personal counselling; grants for boundary fencing; grants for emergency accommodation and loans at 4 per cent for farmers, small businesses and households. Tourism in the region is to be promoted by a $300,000 advertising campaign. The State Government has also presented a cheque for $500,000 to East Gippsland Shire Council towards the estimated $6 million required to repair roads and bridges.

As usual after a natural disaster, the affected communities have sought explanation. The obvious control on the near-record levels reached in the Tambo and Bemm River catchments and the major floods generated in the Mitchell and Lower Snowy River catchments is the intense rain produced by an east coast cyclone - 285 mm in 24 hours at one site near the Bemm River - falling on a drought-hardened surface. On-shore winds and high tides exacerbated lake levels. Several people in Lakes Entrance blamed an inadequate clearance of drains and a build up of sand near the lake exit.

Perhaps the final word of explanation can be left to Victorian Premier Jeff Kennett, who was reported to have said, "The bottom line is, a lot of people are in trouble now, especially some vegetable growers, because they are growing their produce on flood-prone land, so they shouldn't be surprised" (The Weekly Times, 1 Jul 1998, p.6). Those who utilise a floodplain or lakeside resource - including many caravan parks - will from time to time experience nature's tax. Floodplains tend to flood, sometimes with little warning.

Map Source: East Gippsland Regional Catchment and Land Protection Board (1997) Regional Catchment Strategy.

For further information contact:
Stephen Yeo
Phone: +61 2 9850 9685



As part of a larger project, four weather-related natural disasters were examined to help gain an understanding of the distribution of damage costs caused by natural hazards in Australia.   The proportion of the damage costs borne by the three societal mechanisms of insurance, government relief and charity, and the residual proportion borne by individual victims (including businesses) was estimated for each case study.  Similarly, the proportions of total government assistance funded by the three levels of government were also estimated for each case study.  The case studies were:  

  •  the 1974 Brisbane floods;
  •  the 1983 South Australian bushfires;
  •  the Nyngan flood; and
  •  the 1994 New South Wales bushfires.

Data for the first three studies were drawn from published reports on the social and economic impacts of the events.  Data for the fourth study were drawn from sources such as government departments, media reports and the Insurance Council of Australia.



Figure 1:  Relative Damage Costs Borne by Insurance, Government, Charities and the Affected Parties for Four Natural Disasters

Figure 1 shows the relative damage costs borne by insurance, government, charities and affected parties for each event.  The relatively high proportion of damage borne by insurance for the Brisbane flood may be explained by over two-thirds of total damage being in the categories of commercial/industrial, motor vehicles and boats; all of which were insurable for flood damage.  Also, in 1974 approximately 1,800 householder and homeowner policies in the Brisbane area had flood cover with the State Government Insurance Office.  Nevertheless the affected parties (individuals and businesses) were left to bear 37 percent of the damage.  By comparison, 9 percent of the total damage at Nyngan was covered by insurance and affected parties were left to bear only 9 percent of the total damage, the bulk of the damage being borne by government (65 percent).It is interesting to consider the varying constraints under which the different parties operate.  The amount of charity relief tends to be affected by the amount of media exposure.  Government sources of relief tend to be sensitive to the amount of media coverage.  Public pressure and the risk of electoral feedback sometimes leads to governmental guidelines being modified or waived.  Insurers generally adhere to well-defined conditions in their contracts, although public pressure can lead to exceptions. For the case studies examined, the relative level of government relief assistance from the three tiers of government were fairly consistent for three of the four events; the federal government contributing between 70 and 75%, State governments contributing between 25 and 30% and local governments contributing less than 1%.  The proportion of federal government assistance for the Brisbane floods was slightly higher because the Queensland Flood Relief Act (1974) made provision for additional federal government assistance to residents beyond that available under the Natural Disaster Relief Arrangements (NDRA).

It is apparent that the local government sector has a small role in terms of providing financial disaster relief assistance, although it can sustain substantial damages.  This is due to the structure of the NDRA whereby local government authorities need only pay for reparation costs to an upper limit of $25,000 per local government area, the remainder being covered by grants from the State government.  Thus the residual costs to local government are generally small.

There are inherent uncertainties associated with estimating the costs of natural disasters due to lack of data, the fragmented nature of data, the wide range of data sources and the urgency associated with the relief effort.  Although the case studies reported here are based on the best available data, it may be prudent to note Dacy and Kunreuther's wise words when presenting damage estimates for natural disasters: "...the damage figures unfortunately cannot be accepted as an exact mathematical calculation of the evil effects."

Reference: Dacy D.C. and Kunreuther H. 1969, The Economics of Natural Disasters: Implications for Public Policy, The Free Press, New York, p.5.

Based on "Adaption of the Insurance Industry to Climate Change and Consequent Implications", a report of research undertaken by Roy Leigh while employed at the Climatic Impacts Centre, Macquarie University.  The research was supervised by Dr Roslyn Taplin and Dr George Walker and was funded by the Commonwealth Department of Environment, Sports and Territories.

Roy Leigh is currently employed by the Natural Hazards Research Centre and may be contacted on:
Phone: +61 2 9850 8118



Consider the following scenario:

An El Niño-Southern Oscillation [ENSO]-interested layperson may see the above scenario as totally inconsistent with the basic ENSO characteristics perceived or remembered by the broader community (that consistently negative values of the Southern Oscillation Index [SOI] can be associated with drought conditions in these areas). A climatologist might label it as unexpected.

Were the above-normal rainfalls and localised floods due to "Murphy's law" or was the outcome within the ENSO variability thresholds (as we understand them nowadays)? Will these events undermine the scientific efforts to further the level of general ENSO awareness amongst the broader community? Will such an event (perceived in the affected regions as "science got it wrong") result in a call for more intensive ENSO research? Or will the market judgment (with the current emphasis on short-term strategies) prevail and the first hesitant exposure to the ENSO by various end-users (such as the primary production or insurance sector) be abandoned?

To put things into perspective, it has to be mentioned that the mostly experimental ENSO forecasts for the September to December 1997 period did not entirely exclude the possibility of above-normal rainfall in eastern Australia: however, they all allocated the highest probabilities with the drier-than-normal scenario. The International Research Institute for climate prediction, for example, gave a 60/25/15 percent chance of below-normal/normal/above-normal rainfall for eastern Australia for that particular period. Similarly, the statistically based seasonal outlook of the Australian Bureau of Meteorology consistently mentioned "significant bias towards drier-than-normal conditions throughout much of eastern Australia". Thus, while from a scientific point of view the low-odds outcome was not totally inconsistent with the above statements, it may not have necessarily furthered the trust of the affected end-users of this information in the scientific capability to predict the impacts of ENSO. This despite the fact that in other parts of the world, like Papua New Guinea for example, the latest ENSO was one of the strongest events in history.

The scientists may argue that, in order to establish a link with other potential ENSO stakeholders, they necessarily had to push a very basic version of the ENSO characteristics (which has been sometimes even further simplified by the media) during the first broad-based ENSO awareness campaign. The initial information given was mostly based on the comprehensive (one number) Southern Oscillation Index, which has been, for decades, used successfully in the assessment of large-scale atmospheric circulation in the Indian-Pacific ocean region. Unfortunately, the comprehensiveness of this index may have often overshadowed the fact that it is not always  a perfect indicator of regional climatic conditions. This secondary message appears to have come across less clearly.

This lack in understanding does not seem to be entirely the fault of the end-users. It is equally attributable to the reaction of scientists to the changing political and economic environment in which the success of research is being increasingly judged on the immediate applicability and marketability of its results. This leads to the paradoxical situation that above-normal rainfall in parts of eastern Australia during periods of negative SOI values is considered (although as an unlikely scenario) within the thresholds of the ENSO variability by the scientists but, due to the push towards the applicability and marketability of the results and a general inexperience of the scientific sector in public relationship matters, this message has not always been fully conveyed to the broad community.

Thus, from the scientific sector point of view the latest ENSO event exposed two points which merit urgent attention:

a) improvement of the predictive (SOI) index through incorporation of additional variables (possibly the Indian Ocean sea surface temperatures, or the introduction of an additional regional weighting of the individual index components)

b) refining the ENSO-information campaign aimed at the end-users of ENSO information

These points may play a crucial role in influencing the future of  ENSO-related research funding, determining the strength of the fragile link to the private sector and deciding the speed of the development and the success of regional ENSO mitigation strategies.

However, in pursuit of these goals the scientific sector cannot succeed without the support and feedback of the other ENSO stakeholders. Workable ENSO mitigation strategies benefiting all sectors of the society can only be achieved through a strong convergence of academic, governmental and business sector views of ENSO. Only such a broad-based approach can lead to the necessary narrowing of the currently wide physical and perception thresholds of this phenomenon. The latter goal should be seen as a gradual, interactive,  learning process including all sectors of the society.

This abstract contains some ideas described in I. Kuhnel's article entitled "ENSO's Moving Thresholds", which appeared in the June 1998 issue of the UNESCAP Water Resources Journal.

For further information contact:
Dr Ivan Kuhnel
Phone: +61 2 9850 9685



nhq homepage
nhrc homepage