7. Conclusions

Sea-level rise is a potential effect of climate change that will affect coastal areas, but other factors such as changes in storminess and windiness could alter the likelihood of inundation and erosion inland. These need to be understood in the context of response to natural climate cycles, such as El Niño, to ensure planning for communities is consistent with the ever-changing risk profile. The human response to the effects of climate change can conveniently be grouped into three main categories-protect, adapt or retreat. However, the implementation of any of these planning responses is fraught with environmental and socio-economic difficulties, particularly in already-settled areas. The impacts of the 2002 "Weather Bomb" event highlights the complex nature of the physical, economic and social issues that need to be managed.

Approximately 35 % of affected properties (about 300 within the 'high impact' settlements) had flooded land, 37 % had flooding of the basement and/or outbuildings, and 28 % suffered house flooding above the floors of the living areas. The exact nature of the flooding and losses (other than value) within homes was not centrally recorded and not covered within the scope of the questionnaire.

The depth to which a house is flooded affects the number and types of items that are damaged. Household mean losses are disproportionately low when mean flooding in that house is 5 cm or less deep (cleanup of silt/debris may be possible in many cases, reducing permanent loss of contents). Average losses rise roughly proportionately to mean flooding depth for depths from 5 to 50 cm (through the ranges of height of beds, lounge furniture, TVs etc.). Household mean losses are disproportionately high when mean flooding in that house is more than 50 cm deep (reaching the heights for table and bench tops and anything stored on top of or above them).

Vehicle mean losses jump upward on properties where the mean household flooding was above 5 cm deep, and then remain relatively constant, even if the vehicle sits on a property with mean flooding over the height of the vehicle roof (the control is probably the height of the base of the vehicle floor).

The largest combined overall losses in the Weather Bomb event were overwhelmingly for flooding in the 10-50 cm depth bracket, but this is partly because this was the most common mean depth of flooding for a flooded house. If mean household flooding depths > 50 cm had been more common, the losses could have been an order of magnitude greater (other than for vehicle losses, which should have remained at similar levels). Conversely, if mean flooding had rarely exceeded 5 cm depth, losses would have been an order of magnitude less (including vehicle losses).

To put the economy of the affected area into context, over the past 6 years the Thames-Coromandel economy has been relatively small, contributing less than 7 % of the total Waikato employment; in contrast 10.4 % of the Waikato population lives in Thames-Coromandel. Putaruru and Tirau contribute an even smaller proportion of the Waikato employment and population than the Thames-Coromandel area. Analyses of the economic impacts of New Zealand natural hazard events have been relatively rare compared to those undertaken overseas. Based on overseas recommendations designed to lessen variations amongst individual impact studies, impacts have been assessed under the categories of (1) direct and (2) indirect impacts. Due to difficulties in determining intangible economic impacts, these have not been modelled for the Weather Bomb event. Injury resulting from the event, which would produce an intangible economic effect, was judged by TCDC and Civil Defence to be particularly low (no injuries were recorded, other than one fatality).

Direct costs were limited almost entirely to property damage. The estimated insurance claims made as a result of the event were $21.5 million, with around $8 million related to the Thames-Coromandel area. Data from the survey suggests that this was split 0.84/0.16 between households and businesses ($6.7 and $1.3 million respectively). Total uninsured losses are estimated at $2.1 million, based on the survey data. In addition, TCDC estimated agency response costs at $3.1 million, with the much of that the cost of labour. The total direct costs are thus estimated to have been $13.2 million for the TCDC area. This is around 0.6 % of the area's asset base, which is estimated to be around $1750 million. If we assume that only the uninsured losses are borne by the area's asset base, then the loss to the area is only around 0.1 % of its asset base. The true loss to the area likely is somewhere between these 0.1 - 0.6 % loss extremes. This estimate is clearly crude, but hopefully provides an indication of the relative severity of the event.

To better illustrate the spatial and loss-type breakdown of insurance claims Appendices 3 and 4 present a summary of the Earthquake Commission (EQC) and AMI insurance claims related to the Weather Bomb event. The EQC claims totalled about $1.5 million, with about $886,000 in the Thames-Coromandel district. In comparison the EQC claims in the South Waikato district were about $73,000. The AMI claims totalled $662,000 at Thames township office, and $161,000 at Coromandel township office These were attributed to "house", "contents" and "vehicle" losses, in decreasing order of value.

Indirect "flow-on effect" costs take on three main forms: (1) losses due to business disruption, (2) potential impacts on insurance premiums, and (3) second- and subsequent-round effects, typically on those dependent in some way on businesses that were directly affected. The business survey results suggest that the net impact of the weather bomb on business sales was positive (around 30 % more revenue from increased business than the value of lost business). Note that this does not represent a true positive net impact to businesses in the Thames-Coromandel area because many negative impacts, particularly damage to property (a direct loss, quantified above), are missing. The cost borne by insured TCDC households and businesses via insurance excess payments is estimated to total around $0.45 million.

Excess payments are not the only insurance-related indirect loss. Losses as a result of the loss of a no-claims bonus, premium increases and, in some cases, cancellation of policies also occurred but are extremely difficult to quantify. These losses are also only partly related to the Weather Bomb event alone. Adopting a longitudinal framework involving assessment of the incremental impact of successive hazard events is likely to provide a fairer picture of the longer-term cost of repeated events.

A regional version of NZIER's computable general equilibrium (CGE) model was used to estimate flow-on effects to those dependent on directly affected parties. A 9-sector, 9-commodity model and social accounting matrix was used. These effects were found to be negligible, which is perhaps not particularly surprising, given the relatively low direct losses. Industry output, factor demand and household welfare were found to be affected by less than 1 %, probably close to the margin of error.

The flow-on effects of the weather bomb are likely to have been minimal because: (a) the duration of the event was relatively short, (b) the severity of the weather bomb in terms of its direct economic costs were also relatively mild and (c) Thames-Coromandel is in effect a small, borderless economy.

The table below presents a summary of the value of the total economic impact of the weather bomb on the TCDC economy.

Source: NZIER/GNS

In summary, while hardship was extreme for the individuals who lost their entire house, for the vast majority of damaged households the insurance cover was apparently very high and so uninsured loss was very low. Locally borne (i.e., uninsured) losses at community, and even suburb or street level appear to have been quite low. The data and economic modelling, summarised above, points toward a low community-level economic impact, even in the 'high-impact' areas.

Anticipating the intensity of the storm, MetService issued media releases and in those deliberately used the term "weather bomb" to maximise public attention to the potential severity of the event. There was a moderate to high reported awareness of MetService warnings that had been issued prior to the Weather Bomb event. This ranged from 81 % in the low impact Coromandel area to 46 % in South Waikato.

MetService issues Heavy Rain and other Warnings according to specific criteria (see Appendix 1); territorial authorities decide on the basis of this and other information if flood and other warnings are necessary within their jurisdiction. Because the local catchments are small and have short response times, no river monitoring systems existed along the western Coromandel (there are, in contrast, monitoring systems on many rivers in larger catchments in the wider Waikato region). Two questions were asked to explore the public's perceptions of warnings, and to see if people made any link between the extreme weather/Heavy Rain Warnings and the subsequent flooding. However, the questions asked in this survey do not fully address this complex issue and further work is need to explore the links (real or perceived) between weather and flood warnings, the public's understanding of them, and how best to improve the effectiveness of both. The need for improved tools for responding to floods in New Zealand has also been identified in recent research. The data from this study provides a sound baseline for research currently underway. This research incorporates new information from the ongoing community consultation process that has been underway since the time that the survey data presented here was conducted (see further work below).

The community survey measured a range of attitudes, beliefs and understandings of flood risk, the Weather Bomb event and the community in general. From previous social research we are aware that people's understanding of risk and their response to risk are determined not only by scientific information or direct physical consequences, but also by the interaction of psychological, social, cultural, institutional and political processes. Changing people's perceptions of risk alone will not necessarily bring about changes in their behaviour or increased action to address a particular risk. People may not be motivated to prepare if they do not perceive or accept their risk status or perceive the hazards as salient. Irrespective of the level of perceived threat, people's actions will be limited if they perceive hazard effects as being difficult or impossible to mitigate through personal action (low outcome expectancy). Even if people believe that personal actions may reduce adverse impacts, people may not implement these actions if they do not think they are competent to carry them out (low self efficacy). Acknowledgement of a threat may not guide actions if people lack resources (e.g., time, skill, need for cooperative actions etc) required to implement reduction actions (low response efficacy), transfer responsibility for their safety to others (low perceived responsibility), lack of trust in information sources, or because of uncertainty regarding the likely timing of hazard occurrence (Paton 2003). The community consultation over future risk management options, undertaken over the past six months, is continuing to address a range of issues relating to acceptable risk, appropriate mitigation options and the community's willingness to pay. Follow-up research is needed to see how the perceptions documented in the study influence this process.

A key issue resulting from this study is the need to explore in more detail the links between weather warnings and appropriate responses of individuals in high risk areas. There is considerable research pointing to the value of simple alert schemes to help individuals, organisations and communities respond to developing crises (Mileti 1999, Sorensen 2000, Hander 2002). The need for improved tools for responding to floods in New Zealand has been identified in recent research on the Waikanae floodplain (Johnston et al. 2002) and suggested by this study. Despite the above suggestions, it is still not clear, given the short flood response time in the catchments and the lack of flood warnings systems how a system would work, but it is an issue worth exploring. Environment Waikato does operate an automated flood warning system for the region in many watercourses with larger catchments. 

Decision making to mitigate the effects of climate-change impacts for vulnerable localities needs to be a sequential process, taking into account scientific uncertainty, public perceptions, and risk exposure. It is essential that the climate-change planning process start now, given the longevity of housing developments and infrastructure in increasingly at-risk communities.

The data for this report was collected over a very limited time period (August/September 2003). Since this study was undertaken, both the local and regional councils have been consulting extensively with the affected communities to identify and discuss a range of risk management options. Ongoing research will explore how the understanding and attitudes documented in this report have shaped the outcomes of this process. Further work is also ongoing in the Waikato and other regions to explore issues related to weather warnings and links to flood response. The nature and effects of possible 'warning saturation' also requires further research. Another ongoing area of work is the development of better flood depth-to-damage ratios. This is needed to build better loss modelling capabilities for future scenarios.