1
Natural Hazards
„Nature to be commanded, must be obeyed“ (Francis Bacon, 1561-1626)
W. Eberhard Falck [email protected]
2
Objectives
• awareness of specific natural and man-enhanced processes that may pose risks
• understanding philosophical and methodological basics for risk management
• understanding the social choice options for facing such risks
3
Administrative matters
• Oral exam on 5 April 2013, 10h-13h and 14h-17h
• 10 mins preparation and 10 mins exam
• Exact schedule will be distributed
4
Introduction
What are hazards and risks ?
5
Useful link
• A multi-lingual glossary of technical and administrative terms provided by the European Environment Agency
• http://www.eionet.europa.eu/gemet/
6
Hazard
• a hazard is a situation that poses a level of threat to life, health, property, or the environment
• i.e. it is defined with respect to something that - from a human point of view - needs to be protected
7
Risk • Risk is a concept that denotes the precise probability of
specific events
• The notion of risk is independent from the notion of value
Risk = probability of an event x impact of the event
8
Hazard classification • by project objective
• by root cause
• by effect caused
• scenario-based
• taxonomy/phenomenology-based
• by state of threat
• The same phenomenon may have different root causes, e.g. a landslide may be caused by an earthquake, a rainstorm, or human action
• Classification allows to compare alternative paths of action
9
Hazard classification by state of threat
• Dormant - a situation has the potential to be hazardous, but no people, property, or environment are currently affected by it
• Potential - a situation wherein the hazard can affect persons, property, or environment
• Active - the hazard is certain to cause harm, as no intervention is possible before the incident occurs.
• Mitigated - A potential hazard has been identified, but actions have been taken in order to ensure it does not become an incident
10
‚Natural‘ hazards
• Natural processes have occured at all times and will continue to occur
• Most natural processes cannot be stopped
• Some natural processes are gradual (e.g. climate change), while others are sudden events (e.g. rock falls)
• Natural processes per se are not a ‚hazard‘
• It is their influence on our lives, on our environment that makes us consider them a ‚hazard‘
• Thus, hazards are societal constructs
11
Humans and (natural) hazards
12
Humans and nature
• Since the human species evolved, they had to live with and adapt to their environment
• Humans developed hazard and risk mitigation strategies
• Humans learned to cope with natural hazards
• There are different strategies for coping with slow, long-term changes and for sudden events
13
Personal heuristics and objective risk
• Often there is a counter-intuitive risk perception
• Perceived probabilities may not tally with real probabilities
• Need to develop strategies to communicate scientific insight
• There appears to be a considerable difference in the heuristics and acceptability between natural and anthropogenic risks
14
Risk and hazard perception
• Risk is a societal construction rather than an absolute concept • Perceptions may vary considerably between stakeholders:
Real hazards and risks - experts‘ opinion vs.
Perceived risks and hazards - other stakeholders
Objective and realistic risk categorisation vs.
Relativistic perspective of risks being subjective reflections of power and interests
15
Risk awareness and its consequences
• Due to extensive risk management activities by governments etc. the individual awareness of natural risks has significantly decreased
• Low risk awareness leads to complacency and e.g. settlement of risk zones
• Economic pressure combined with low risk awareness leads to use/settlement in risk zones
• Often there are no adequate resources to provide protection in high risk zones
• or, the societal cost would be too high and it would be against the interest of society at large to provide protection for some individuals in high risk zones
16
Acceptability of risk
• Risk perception and risk acceptance changed over history
• Due to more sophisticated infrastructure etc. the risk level increased for the same type of hazard
• With increasing level of integration (e.g. transport networks, economic interrelations) the level of risk increases for the same type of hazard
• Modern society has a lower tolerance of risk due to overall level of risks to humans being reduced
17
Coping strategies
• Acceptance - e.g. ‚budget‘ in or the Job-syndrom
• Avoidance - e.g. do not settle in a flood-plain
• Mitigation - e.g. have emergency plans in place
• Transfer - e.g. take out insurance
• Management - e.g. build dykes
• In the course of history, coping strategies evolved from risk avoidance to risk management
18
The cost of risk management
• Managing and avoiding risks has societal and economic costs = opportunity costs
• Opportunity costs are the costs of not pursuing other gainful economic opportunities
• Opportunity costs include e.g. – not using arable land in a flood-plain – expenditure for dyking that could be spend on profit-making
infrastructure – insurance premiums
• However: concerted efforts in risk management had also benefical effects on socio-economic developments, e.g. the Egyption Civilisation, the Dutch Golden Age
19
The Netherlands 1300 - 1960
• Large-scale land-losses led to a continuing battle against the sea • Changes in the morphology of the Rhine - Meuse - Scheldt delta
due to human coastal defence measures • Dyked-in land created new economic opportunities • The societal and physical infrastructure developed was one
contributing factor for the Dutch Golden Age (17th century)
20
Costs vs. Benefits
• Costs of riks management have to be off-set by adequate benefits
• Modern risk management works on the ALARA principle: As Low As Reasonably Achievable
• Benefits are often deferred benefits - reaped only by succeeding generations (e.g. the Dutch Golden Age)
• Some risk management projects are very long-term projects
• Costs and risk reduction may not have a linear relationship
21
Vulnerability and resilience
• The vulnerability to natural hazards depends on the level of socio-economic development
• Highly-developed and -integrated societies are more vulnerable to hazards that disable their infrastructure
• Less-developed and -integrated societies are more resilient and require less resources to recover
• This also shapes the risk perception
• Resilience means the capacity of socio-ecological systems to absorb recurrent disturbances brought about by natural events
• Note: of course, a desaster is a desaster to those affected.
22
Vulnerability
• The International Panel for Climate Change (IPCC) defined vulnerability as „ ... the degree of incapability to cope with the consequences of climate change and accelerated sea-level-rise.“
• It is largely a matter of definition and sometimes of value judgement what constitutes a change or what would be considered an adjustment within a system
• Such value judgements imply the desirability of a certain state of natural or societal systems
• The ‚desirable state’ frequently is either a personal preference or negotiated between the respective actors in a society
23
Example: Vulnerability of buildings
Sum mer 1998 / 99 15
sary route modelling analysis can beundertaken.
Sustenance. Modern urban communitiesare highly reliant on their utility andservice infrastructures such as watersupply, sewerage, power supply and tele-communications. These so-called lifelinesare significantly dependent on each otherand on other logistic resources such as fuelsupply (see, for example, Granger, 1997).
The community is also dependent on theavailability of food supplies, clothing,medical supplies and other personal items.Information is being accumulated on all ofthese, as well as on the enterprises thatwholesale, distribute and service thesesectors (such as transport, material hand-ling equipment and storage). All of the keyfacilities in Cairns have been identified inthe building database and basic data onpower and water supply infrastructure areavailable.
Security. The security of the communitycan be measured in terms of its health andwealth and by the forms of protection thatare provided. Physically, these may beassessed by the availability of facilities suchas hospitals, nursing homes, industries,commercial premises, agricultural landuse, ambulance stations, fire stations,
up the community. Extensive use is beingmade of the detailed data from the 1996National Census to flesh out our under-standing of the socio-demographic andeconomic dimensions of vulnerabilityunder both the ‘security’ and ‘society’components.
Whilst other approaches to comprehen-sive risk assessment, such as HAZUS(NIBS, 1997), may segment the elementsat risk differently, there is very close accordon their overall and individual significance.
Synthesis and modelling. Clearly, therange and variety of information neededto fuel a comprehensive risk analysis isenormous. Whilst there are many sourcesnow available from which such inform-ation can be captured or derived, much ofit with the essential spatial and temporalattributes needed, there remain importantgaps. Our knowledge of hazard phenomenaand the processes that drive them, forexample, are far from perfect. It is neces-sary, therefore, to develop appropriatemodels (process, spatial and temporal) tofill the knowledge gaps. The behaviour ofsome hazards, such as bushfires and floods,have an established body of modellingresearch behind them, whilst others, suchas cyclones and earthquakes, especially in
case study 85 to 90% of the informationused has some form of spatial content.Similarly, the relationships that are mostsignificant in risk analysis and risk assess-ment are largely spatial. To accommodatethis spatial emphasis, the Cities Projectmakes extensive use of GIS tools andtechnologies.
Whilst GIS have been used over the pastdecade as tools to address specific aspectsof the risk management problem, espec-ially in hazard mapping and the spatialmodelling of phenomena such as bushfires,or flood and storm tide inundation, thereare few examples of integrated risk man-agement applications. There are obviousadvantages in developing a fusion betweena philosophy of risk management and thepower of GIS as a decision support tool,hence Risk-GIS as it has been christened inthe Cities Project. As such Risk-GIS providesthe analytical ‘engine’ which drives theCities Project’s urban geohazard risk assess-ment process. Risk-GIS also provides amost potent form of risk communication(an aspect about which AS/NZS 4360:1995is unfortunately silent) through its capacityto provide a visual representation of risksituations.
police stations and works such as levees.Also important are socio-demographic andeconomic issues related to the elderly, thevery young, the disabled, household in-come, unemployment, home ownershipand the resources available at the fire andpolice stations. Emergency plans are also akey component of community security.
Society. Here we find most of the ‘warmand fuzzy’ measures such as language,ethnicity, religion, nationality, communityand welfare groups, education, awareness,meeting places, cultural activities and soon. Some of these may be measured interms of the facilities that they use, such aschurches, meeting halls, sporting clubs andso on; however, the more meaningfulmeasures relate specifically to the indiv-iduals, families and households that make
intraplate areas such as Australia, are as yet,less well served.
A key aspect of these models is anunderstanding of the probability of recur-rence of events of particular severity andthe levels of uncertainty that exist in boththe data employed and the models them-selves. Given these uncertainties we remaincautious about presenting most of ourfindings as predictions of risk; rather weprefer to caveat them as providing nothingmore than indications of what the futuremay hold.
The synthesis of data and the essentialmapping of the spatial relationships be-tween the hazard phenomena and theelements at risk requires the use of toolssuch as geographic information systems(GIS). In the work undertaken in the Cairns
Characteristic Flood Wind Hail Fire Quake
Figure 3: Relative contribution of building characterist ics to vulnerability
Building age *** ***** ** ***** *****Floor height or vertical regularity ***** * **** *****Wall material *** *** ***** **** ****Roof material **** ***** **** ***Roof pitch **** *** *Large unprotected windows ** ***** **** ***** **Unlined eaves *** *****Number of stories **** ** * *****Plan regularity ** ** *** *****Topography ***** **** **** ***
Risk assessment and prioritisationScenario analysis. This is an emergingtechnique that contributes to ‘futurememory’, an understanding of ‘what willhappen when …’. The output embracesforecasts or estimates of community riskincluding economic loss and potentialcasualties, or assessments of the impact ofsecondary or consequential hazards, suchas the spread of fire or the release ofhazardous materials following an earth-quake. It also provides essential input toboth the development of risk treatmentstrategies and to framing long-term fore-casts or estimates.
In an effort to address the diverse rangeof applications to which the output fromrisk scenarios may be put, we have adoptedthe practice of running a range of scen-arios, typically extending from the rela-tively small and more frequently occurringevents to those in the ‘maximum probable’or ‘maximum credible’ range. Figure 4, forexample, illustrates the cumulative rangeof risk associated with storm tide inun-dation scenarios for Cairns. This figure issimilar to the ‘risk curves’ employed by theinsurance industry and, indeed, the x-axiscould be scaled to dollars of loss or poten-tial fatalities, whilst the y-axis could bescaled (perhaps logarithmically) to eventprobability.
Acceptability. In the approach to riskassessment set out in AS/NZS 4360:1995, it
24
Resilience • Is the capacity of a socio-ecological system to absorb recurrent
disturbances brought about by environmental events so as to retain essentially the same function, structure, identity, processes and feedbacks
• It reflects the degree to which a complex adaptive system is capable of self-organisation, and
• the degree to which the system can build capacity for learning and adaptation
• These definitions tacitly assume that socio-economic systems must undergo a certain amount of change in order to be resilient
• Social and economic resilience is at community level rather than at the individuals’ level and it relates to the social capital of societies and communities
25
Resilience vs. Resistance
• Resilience should not be confused with resistance
• Resistance broadly is the ability to stop or resist change.
• Resilience is the capacity to absorb disturbances before a change occurs.
• A historical evolution from resilience towards resistance in many engineered approaches to hazard mitigation can be observed.
• There are limitations to strategies building on resistance so that in more recent times resilience building is being favoured.
26
The limits of resilience
• Loss of ecological and social resilience may be hidden, and resilience can be eroded or reinforced accidentally or deliberately through human action.
• Biodiversity, functional redundancy, and spatial patterns can all influence resilience of ecosystems.
• In social systems, governance and management frameworks can spread risk by diversifying patterns of resource use and by encouraging alternative activities and lifestyles.
• However, many communities are quite set in their rather limited portfolio of economic and social activities and nevertheless have been able to recover after many disastrous events over centuries
27
Humans and change • The concept of resilience and vulnerability does have a
considerable anthropocentric compound. • It tacitly implies the desirability of maintaining a certain status
quo. • This ignores the fact that over geological timescales natural
systems evolve, gradually or catastrophically, and that there is nothing good or bad about this.
• Value judgement are introduced by the way humans relate to these changes and whether they consider them as a threat.
• Such threats can be seen from a local perspective or in more recent times also from a global perspective.
28
Coping with change
• The concept of resilience implies a value statement, since ‚coping’ and ‚adaption’ are considered positive and tolerable, while ‚change’ is not
• Some authors use the concepts of ‚function’ or ‚service’ to describe the viability or otherwise of natural or socio-economic systems, but
• this requires a definition of what levels of ‚functioning’ and ‚services’ are required or are desirable by whom, for whom and why
29
Key features of resilience The U.S. INDIAN OCEAN TSUNAMI WARNING SYSTEM PROGRAMME (2007) developed a list of
eight key features: • Governance - Leadership, legal framework, and institutions provide enabling conditions for
resilience through community involvement with government; • Society and Economy - Communities are engaged in diverse and environmentally sustainable
livelihoods resistant to hazards; • Coastal Resource Management - Active management of coastal resources sustains
environmental services and livelihoods and reduces risks from coastal hazards; • Land Use and Structural Design - Effective land use and structural design that complement
environmental, economic, and community goals and reduce risks from hazards; • Risk Knowledge - Leadership and community members are aware of hazards and risk
information is utilized when making decisions; • Warning and Evacuation - Community is capable of receiving notifications and alerts of coastal
hazards, warning at-risk populations, and individuals acting on the alert; • Emergency Response - Mechanisms and networks are established and maintained to respond
quickly to coastal disasters and address emergency needs at the community level; • Disaster Recovery - Plans are in place prior to hazard events that accelerate disaster recovery,
engage communities in the recovery process, and minimize negative environmental, social, and economic impacts.
30
Strategies to increase resilience WARDEKKER et al. (2009) proposed a set set of strategies and features that is
of likely to increase resilience with mainly urban coastal areas in mind: • Homeostasis - multiple feedback loops counteract disturbances and
stabilise the system; • Omnivory - vulnerability is reduced by diversification of resources and
means; • High flux - a fast rate of movement of resources through the system
ensures fast mobilization of these resources to cope with perturbations; • Flatness - the hierarchical administrative levels relative to the base should
not be top-heavy; • Buffering - essential capacities are over-dimensioned such that critical
thresholds in capacities are less likely to be crossed; • Redundancy - overlapping functions; if one fails, others can take over.
31
Human induced natural risks
• Human interferences with nature change types of hazards, risk patterns and severity of impact
• For instance – river corrections increased the risks of flooding downstream – lack of forest management can increase the risk of fires – deforestation can increase the risk of land-slides – deforestation can increase the risk of avalanches
• Humans often interfere with natural systems without understanding the consequences
32
Risk assessment and
Risk management
33
Risk Management - Components
Several consecutive and complimentary actions • Risk assessment
• What-if scenarios for mitigation/avoidance
• Cost-benefit analyses - opportunity costs
• Risk communication to stakeholders
• Risk-informed decision-making procedures
• Implementation of measures
• Monitoring
34
Risk Assessment
• to quantify the magnitude of the risk
• to describe the areas at risk
• to inform the public about the risk
• to provide a basis for avoidance or mitigation
35
Mapping of risk areas
• traditionally people ‚knew‘ risk areas, based on past experience and knowledge traded-on by their ancestors, e.g. areas exposed to avalanche or flooding risks
• modern mapping techniques based on earth-observation (e.g. satellite imagery) in conjunction with geographical information systems (GIS) allow to superimpose maps for different properties, e.g. risk types, items to be protected, etc.
36
Mapping of risk zones
Seismic risks in the continental USA
37
Mapping of risk zones
Flooding risk by River Thames in Greater London
38
Mapping of risk zones
Complex vulnerability map of a village in Moçambique
39
Multiple Natural Hazards in Iceland
Sea ice Land slides Avalanches
Earth quakes Volcanic eruptions Glacial river surges
40
Frequencies of occurrence and magnitude
• Statistical analysis of historical records – e.g. weather records, evidence of flooding events
• Regular time series – e.g. river/sea level measurements
• Expert opinion
• Anecdotical evidence / traditional knowledge – e.g. narrative descriptions of events and impacts
• Synthetic Monte Carlo-simulations
41
Extreme events
• Often risks with high Losses, but very low Probability are ignored, as are risks with high Probability, but very low Losses
• They can also pose a perception issue
Probability
Loss
unlikely events
neglible loss
42
Scenario analyses
• Sets of Features, Events and Processes (FEPs)
• Sets of boundary and initial conditions
• Determination of ranges of parameter values
• Deterministic analyses - worst/best cases
• Estimating the probability of parameter values and events
• Development of probability density functions (PDFs)
• Probabilistic modelling with parameter variations according to the PDFs
43
Deriving PDFs • Statistical analysis of past events - ex post facto evaluation • Sampling of expert opinion • Anecdotical evidence • Synthetic Monte Carlo-simulations
44
Confounded by anthropogenic feedback
• Risk management often relies on time series of past events to construct predictive models for future events
• These time series by force reflect a past world, e.g. with respect to the structure of catchment areas
• Catchment areas are changed by humans (e.g. deforestation) and thus are the drainage patterns
• Hence the predictive models e.g. for flooding are based on inadequate databases and are often bound to fail
45
Predictions and forecasts
• Certain natural processes are slow and nearly linear at a human timescale - e.g. basin subsidence
• Fast natural processes are often random from a human point of view as not the whole history and triggering events are observable - e.g. earthquakes
• Some natural processes can be followed over a few days and their development can be forecast with some accuracy - e.g. tropical storms, flooding
• Predictions and forecast require conceptual models of the processes and numerical realisation of these models
46
System Models • Conceptual model represent the system, its components and
their interactions • Numerical models facilitate a quantitative analysis
47
Probabilistic risk assessment
• Formal numerical modelling of system behaviour and resulting consequences by using probability density functions to vary boundary conditions and system variables
• Aggregates different risks of differing probability
• Not all boundary conditions and system variables need to be PDFs
• Frequently used (and required by regulators) in plant safety assessment or for the safety assessment of radioactive waste disposal
48
Consequence analysis What-if-Scenarios to assess
• Severity • Assessment of affected areas, people • Morbidity, mortality • Statistic vs. deterministic effects • Estimation of damage caused • Estimation of further consequences - chains of
consequences
• Acceptability
• (Ir)reversibility
49
Example: Assessment of flooding zones
• What-if scenarios
• Estimation of affected areas, people
• Estimation of damage caused
• Estimation of further consequences - chain of consequences
50
Summarising risk assessment
Australian Journal of Emergency Management16
Levels of acceptability are, however, ‘builtin’ to such things as urban planning designconstraints and the Building Code, wherecriteria are based on ‘design’ levels. Forexample, under the earthquake loadingcode AS1170.4–1993 Earthquake Loads(Standards Australia, 1993), the ‘designlevel of earthquake shaking’ is one in whichthere is an estimated 10% probability of theground motions being exceeded in a 50year period i.e. the acceptability criterionis set at a 10% chance of exceedence overthe nominal lifetime of a ‘typical’ building.
Not all acceptability criteria can be
those in which deep soft sediments aremost likely to maximise earthquake im-pact. Conversely they are the areas that areat least risk from landslide impact and, tosome degree, from severe wind impact.Additionally, the impact on the Cairnscommunity of cyclone hazard with a 150-year return period is likely to be moresevere than the impact of the shakingassociated with a 150-year return periodearthquake; but the maximum credibleearthquake event may have a greaterpotential for catastrophe, than the maxi-mum credible cyclone.
ings are typically taken to mean short-termwarnings such as those issued by theBureau of Meteorology for the hazards thatcan literally be seen coming, such ascyclones, floods and severe storms, theymay also embrace the longer-term esti-mates of the ‘hazardousness’ of areas suchas those contained in the earthquakehazard (acceleration coefficient) maps thataccompany AS1170.4-1993 or by hazardmaps specifically prepared for a city. Theycan both be significantly enhanced throughthe scenario analysis process.
Mitigation strategies and response options.
Figure 4: Number of buildings affected by different storm tide scenariosin Cairns with heights above highest astronomical t ide (HAT).
No
of b
uild
ings
(100
0s)
12
14
HAT 0.5 1.0 1.51.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
On property
< 1m overfloor
> 1m overfloor
10
8
6
4
2
0
Metres
is the practice to compare thelevel of risk found during theassessment process with pre-viously established risk criteria,so that it can be judged whetherthe risk is ‘acceptable’ or not. Atfirst glance this may seem to besomething of a ‘chicken-and-egg’ process—if you do notknow what the level of riskposed by earthquake is inCairns, for example, how can yourealistically determine whatlevel of risk is acceptable?
expressed as categorically as thisbecause they deal with humannature and the political ‘outrage’dimension of risk management.They also vary considerably overtime—the threshold of accep-tance is typically much lowerimmediately after a hazard im-pact than it was immediatelybefore the impact, hence theneed for a strong feedbackmechanism between establish-ing acceptability and the formu-lation of risk mitigation and responsestrategies.
Perhaps the risk ‘formula’ could be betterexpressed as:
Risk (Total) = (Hazard x Vulnerabilityx Elements at Risk)Acceptability
to reflect the strong modifying influenceof acceptability. Clearly, a key element indetermining limits of acceptability restswith effective risk communication andpublic policy development.
The ‘acceptability’ factor is central to theprocess of risk prioritisation which is thefirst step in the allocation of resources torisk mitigation, especially if considered ina multi-hazard context. We are beginningto address the complex issue of comparingthe risks posed by hazards with greatlydifferent impact potential. In Cairns, forexample, there is a strong spatial corre-lation between the areas that are most atrisk from major inundation hazards (riverflooding, storm tide and tsunami) and
Risk mitigation strategiesWhilst the Cities Project is concernedprimarily with risk identification, analysisand assessment, it does have some linkagewith elements of the risk mitigation pro-cess.
Monitoring and surveillance: One ofthe principal sources of historical hazardevent information and hazard phenom-enon knowledge is the extensive networkof monitoring stations and remote sensingresources that have been established acrossAustralia. For example, the Bureau ofMeteorology maintains some 45 weatherradar sites, 246 automatic weather stationsand 3,029 stream gauging stations, whilstAGSO has access to more than 150 seis-mographs across the country. The Bureaualso takes data from the Japanese Geo-stationary Meteorological Satellite 26 timesa day in addition to data taken from thepolar orbiting US NOAA. Whilst weathermonitoring of Cairns is comprehensive and
Figure 5: Cities Project understanding of the risk management process
elements atrisk and theirvulnerability
probabilityand process
models
history ofevents and their
consequences
hazardphenomena and
environment
monitoringand
surveillance
levels ofcommunityacceptance
scenarioanalysis
warnings andforecasts
Risk mitigationstrategies and
response options
safe, sustainableand prosperous
communities
Risk assessments are made sothat strategies may be developedthat ultimately will lead to theelimination, reduction, transferor acceptance of the risks, and toensure that the community isprepared to cope with a hazardimpact. Whilst the developmentand implementation of thesestrategies lie essentially outsidethe remit of the Cities Project, ourexperience in working withemergency managers and others
to date suggests that amongst the mosteffective strategies are:• well maintained and appropriate infor-
mation that is fundamental to riskassessment
• risk-based planning of settlement, dev-elopment and key facilities (such ashospitals)
• protection plans for key facilities andlifelines
• cost-effective engineered defences suchas levees and retrofit programs
• appropriate and enforced building andplanning codes
• emergency management plans, resourcesand training based on risk assessments
• wide-spread and ongoing communityawareness programs based on risk his-tory, scenario analysis and an effectiverisk communication capability.These components of the Cities Project’s
understanding of the risk managementprocess are illustrated in Figure 5.
has reasonable historical depth,seismic monitoring coverage hasuntil recently been relativelypoor, with only the larger (andless frequent) events being meas-ured by distant instruments.
Warnings and forecasts. Aneffective warning and forecastingsystem, combined with a highlevel of community awarenessand risk appreciation, is clearlyone of the most potent mech-anisms by which to achieve riskmitigation. Whilst these warn-
51
Summary
• Hazards and risk are defined by what humans would like to see protected
• Thus natural hazards are a societal problem
• Most natural processes cannot be stopped or even diverted
• Risk management has to balance cost and benefits
„Nature to be commanded, must be obeyed“