Description
During the ten-year period of 1992-2001, weather-related disasters worldwide killed over 622,000 and affected over 2 billion people (WMO, 2004). In 2004, Japan saw a record year for typhoons,with ten tropical cyclones making landfall, two within ten days.Tokage was the most powerful typhoon to hit Japan in 16 years.
W o r l d M e t e o r o l o g i c a l O r g a n i z a t i o n
GUIDELINES ON INTEGRATING
SEVERE WEATHER WARNINGS INTO
DISASTER RISK MANAGEMENT
WMO/TD-No. 1292 PWS-13
The designations employed and the presentation of material in this
publication do not imply the expression of any opinion whatsoever on
the part of the Secretariat of the World Meteorological Organization
concerning the legal status of any country, territory, city or area, or of
its authorities, or concerning the delimitation of its frontiers or bound-
aries.
It should be noted that this Report is not an official WMO
Publication and has not been subjected to the Organization’s standard
editorial procedures. The views expressed by individuals or group of
experts and published in a WMO Technical Document, do not neces-
sarily have the endorsement of the Organization.
© 2005, World Meteorological Organization
WMO/TD No. 1292
NOTE
Co-authors: Jim Davidson and M.C. Wong
(Contributions by: C.Y. Lam, M.C. Wong, C. Alex, S. Wass, and C. Dupuy)
Edited by: Haleh Kootval
Cover design by: F. Decollogny
Figure of Risk Management vs Crisis Management
Courtesy of: Donald Wilhite (2000)
Page
CHAPTER 1. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
CHAPTER 2. EXPLANATION OF KEY CONCEPTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
CHAPTER 3. RISK MANAGEMENT PHILOSOPHY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
CHAPTER 4. NMSs’ RESPONSIBILITIES WITHIN A RISK MANAGEMENT FRAMEWORK . . . . . . . . . 7
4.1 CORPORATE CULTURE: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
4.2 CAPACITY BUILDING: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
4.3 DEVELOPING PARTNERSHIPS: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
4.4 PUBLIC EDUCATION & AWARENESS: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
4.5 ESTABLISHING A KNOWLEDGE BASE: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
4.5.1 Risk Databases: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
4.5.2 Risk Assessments: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
4.5.3 Cost-Benefit Analyses: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
4.5.4 Research Activities: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
4.6 DISASTER RISK MANAGEMENT EXPERTISE: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
4.7 FORMULATING A RISK MANAGEMENT ACTION PLAN: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
4.7.1 Step 1: Communicate & Consult with Stakeholders: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
4.7.2 Step 2: Establish the Context: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
4.7.3 Step 3: Identify & Quantify the Risks: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
4.7.4 Step 4: Analyse and Evaluate the Risks: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
4.7.5 Step 5: Treat the Avoidable and Manageable Risks: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4.7.6 Step 6: Monitor and Review the Risk Management Action Plan: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4.8 SECURING FUNDING: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
CHAPTER 5. EFFECTIVE EARLY WARNINGS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
5.1 STAKEHOLDER INVOLVEMENT: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
5.2 WARNING PRESENTATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
5.3 WARNING COMMUNICATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
5.4 WARNING RESPONSE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
5.5 MONITORING AND REVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
CHAPTER 6. SUCCESSFUL INITIATIVES OF NMSs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
6.1 EXAMPLES OF SUCCESSFUL RISK MANAGEMENT INITIATIVES: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
6.1.1 Examples from Hong Kong, China . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
6.1.2 The French Vigilance System After the Heat Wave of August 2003 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
6.1.3 Examples from United States of America:
All Hazards Emergency Message Collection System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
6.1.4 Examples from Australia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
6.2 EXAMPLES OF PROJECTS FUNDED FROM INNOVATIVE SOURCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
6.2.1 World Bank Loan to the Algerian Government to Improve
Flood Warnings, Infrastructure Improvements, and Enhance
the Relationship with Disaster Response Agencies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
6.2.2 Grant from the European Development Fund to
the Caribbean Forum of Africa, Caribbean and
Pacific States to Reduce Vulnerability to Floods and Hurricanes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
TABLE OF CONTENTS
6.2.3 Central America Flash Flood Warning System
Funded by the United States Agency for International
Development’s Office of Foreign Disaster Assistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
6.2.4 Examples from Hong Kong, China . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
6.2.5 Examples from France: Building a Comprehensive
Mainland Hydro-meteorological Radar Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
6.2.6 Examples from Australia: Studies on Climate Change
and Tropical Cyclone Vulnerability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
CHAPTER 7. REFERENCES, FURTHER READING AND USEFUL WEB SITES . . . . . . . . . . . . . . . . . . . . . . 24
I. REFERENCES AND USEFUL READINGS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
II. USEFUL WEBSITES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
APPENDIX A UNDERSTANDING PROBABILITIES OF OCCURRENCE AND
CALCULATION OF BENEFIT COST RATIO (BCR): AN EXAMPLE . . . . . . . . . . . . . . . . . . . 26
Contents iv
During the ten-year period of 1992-2001, weather-related
disasters worldwide killed over 622,000 and affected over 2
billion people (WMO, 2004). In 2004, Japan saw a record year
for typhoons, with ten tropical cyclones making landfall, two
within ten days. Tokage was the most powerful typhoon to hit
Japan in 16 years. The year’s number of tropical cyclones in
Japan surpassed the previous record of six set in 1990 and left
the largest number of people dead (some 220) and injured
since 1983 (MeteoWorld, 2004). A related disaster was the
corresponding lack of rainfall in the same season in China
resulting in widespread drought resulting in the migration of
birds carrying the avian flu virus to populated areas.
In the same year, United States was severely affected by
four hurricanes making landfall in Florida, with five other
hurricanes of different intensities making landfall elsewhere
along the US coastline (National Hurricane Centre/Tropical
Prediction Centre, 2004). The last time with such frequent
tropical cyclone activity was in 1916 when eight hurricanes
impacted the US coastline. Total damage in 2004 as a result
of tropical cyclone landfalls could approach 50 billion US
dollars (Colorado State University News, 2004). Charley
brought death and destruction to the Caribbean and Florida.
Damage to property in Cuba was in excess of 1 billion US
dollars. Preliminary estimates of Charley’s damage in Florida
ranged from 13 to 15 billion US dollars, making it the second
costliest hurricane in the US history. Then came Ivan, the
most powerful hurricane to hit the Caribbean in ten years.
Following a trail of havoc across Grenada, Jamaica and
Alabama, Ivan left in its wake more than 100 deaths and prop-
erty damage estimated at 12 million US dollars. Jeanne,
weaker but no less deadly, swept along the north coast of
Haiti on 16 September, killing more than 2,000 people and
destroying the country’s economy.
The South Pacific islands also had their share of weather-
related disasters. In the austral summer of 2002-2003, a
powerful tropical cyclone Zoe impacted a number of remote
island communities in the Solomon Islands and caused
significant economic loss. This was followed in 2003-2004 by
another intense tropical cyclone Heta that left a trail of utter
devastation through the small island nation of Niue. The
economic loss at Niue was so great that the continuing viabil-
ity of the country came into question, with survival possible
only through international aid. In early 2005, the Cook
Islands in the Southwest Pacific were affected by four intense
tropical cyclones (Meena, Nancy, Olaf and Percy) causing
significant damage being reported.
On the morning of Sunday, 26 December 2004 a magni-
tude 9.3 earthquake, the world's most severe in 40 years,
ripped apart the seafloor off the coast of northwest Sumatra.
It unleashed a devastating tsunami that swept thousands of
kilometres across the Indian Ocean, taking the lives of some
300,000 people in countries as far apart as Indonesia,
Thailand, the Maldives, Sri Lanka and Somalia, making it the
deadliest in recorded history. Those hardest hit were the
people living in low-lying coastal areas. Beyond the loss of
human lives, the tsunami also destroyed livelihoods, trau-
matized whole populations and severely damaged habitats.
The scale of the disaster shocked the international commu-
nity and the United Nations (UN) made a "Flash Appeal"
immediately after the event to raise funds to support relief
and recovery operations.While attention was understandably
drawn towards this tsunami and its aftermath in south Asia,
the impact caused by the perennial threat of global severe
weather should not be under-estimated.
The latest event at the time of writing these guidelines
was category 5 hurricane Katrina, which struck the Gulf coast
of the United States in August 2005 causing many deaths and
estimated economic losses in excess of US$100 billion, the
largest loss on record resulting from a single weather-related
event. Between 1998 and 2002, 45 weather-related disasters
in the United States caused damage with a total cost of
approximately US$200 billion.
The dramatic impact of natural disasters and the subse-
quent response activities often attract much international
interest.Attention is being increasingly focused upon natural
disasters inflicting tremendous economic losses (in addition
to human suffering and casualties) and the efforts expended
on the mitigation and reduction of such disasters. Disaster
mitigation is now a recognized international priority. WMO
cooperates within many other organizations and interna-
tional programmes, particularly the International Strategy
for Disaster Reduction (ISDR), in its efforts to improve
natural disaster prevention and mitigation. The UN Secretary
General, Mr Kofi Annan, recommended in his report to the
General Assembly on 21 March 2005 the establishment of a
global early warning system for all natural hazards.
Increasingly it is recognized that disasters are linked. The
impacts of many types of natural disasters do not happen in
isolation, but are sequential (Rogers 2005). For example, a
drought in one region followed by famine can be exacerbated
by dust storms transporting locusts spawned by heavy rains
in another location. Similarly, the diversion of moisture from
one region to another causes flooding in one region and
drought in another. The recognition of such cause and effect
on a global and regional scale is leading to the creation of
early warning systems that can accommodate multiple
hazards and cross-boundary impacts. At the same time,
governments are becoming aware that a paradigm shift from
crisis management to risk management is necessary if the
finite resources available are spent in the most efficient way
to assist the populations at risk to prevent or mitigate disas-
ters.
The Public Weather Services Programme was given the
mandate by the WMO Commission for Basic Systems to
produce disaster risk management guidelines for National
Meteorological Services and National Meteorological and
Hydrological Services (hereafter NMSs). These guidelines on
"the application of risk management principles in the provi-
Chapter 1
INTRODUCTION
Guidelines on integrating severe weather warnings into disaster risk management
sion of severe weather warnings, and the use of this approach
in efforts to secure extra funding from innovative sources" are
prepared with the central purpose of recommending a risk
management action plan suitable for adoption by NMSs.
They are not meant to be an exhaustive treatise on disaster
risk management. There is already a wealth of literature on
this subject and relevant documents are referenced here. The
intention is to promote the role of NMSs in disaster risk
management, to encourage their more proactive involvement,
and where appropriate lead in developing and implementing
a risk management action plan. Effective risk management is
often the difference between a natural hazard and an impact
disaster.
Risks can be characterized as being a complex combina-
tion of unavoidable, acceptable and treatable circumstances.
Having identified and evaluated risks, a decision must be
made as to what level of risk is acceptable, and at what level a
risk is to be treated. Increasingly, this is being discussed in
terms of consequences of the event and cost effectiveness of
the treatment. These guidelines relate to the various risks that
are considered treatable or manageable in the context of the
provision of severe weather warnings. They are prepared with
due reference to the outcomes of various Disaster reduction
conferences, workshops and seminars held around the world
in recent years, including:
(i) 1st World Conference on Natural Disaster Reduction
held in Yokohama, Japan in 1994;
(ii) 1st International Conference on Early Warning –
Potsdam, Germany – 1998;
(iii) Risk Management and Assessments of Natural Hazards
Forum – Washington, USA – 2001;
(iv) 2nd International Conference on Early Warning – Bonn,
Germany – 2003;
(v) Workshop on Early Warning Systems – Shanghai, China
– 2003;
(vi) 2nd World Conference on Natural Disaster Reduction,
Kobe, Japan 2005.
It is noted that the 3rd International Early Warning
Conference will be held in Bonn, Germany in March 2006.
2
Issuing weather forecasts and warnings is a core business
for NMSs. Most of the time, NMSs focus on the applica-
tion of science. In fact, one of the great achievements of
meteorological science and technology during the 20th
century has been our increased ability, through improved
warning and forecasting systems, to provide much more
reliable weather information for effective protection of
life and property from natural hazards. However, in an
information age with higher public expectation, more
intense media coverage as well as much closer connection
with the global community, NMSs can no longer afford to
play a passive role in disaster reduction and mitigation
effort. It is just as critical for NMSs to be actively
involved in disaster management programmes to ensure
that the users get the full benefit of reliable forecasts and
warnings.
It is important therefore not just how we forecast a
natural hazard but how we plan for and minimize the
impact of the hazard. In the context of the current guide-
lines, a natural hazard is a weather or flood-related
situation with potential to inflict loss or damage to the
community or environment. A natural disaster is a cata-
strophic event caused by a natural hazard that severely
disrupts the fabric of a community and usually requires
the intervention of government to return the community
to normality. While hazards may induce a crisis, they do
not necessarily lead to disasters. Though most natural
hazards may be inevitable, natural disasters are not
totally unavoidable. A disaster will depend on the char-
acteristics, probability and intensity of the hazard, as well
as the susceptibility of the exposed community based on
physical, social, economic and environmental conditions.
In some instances, natural disasters cannot be
prevented from occurring. However, their overall impact
can be significantly reduced through disaster mitigation.
Disaster mitigation is the process of managing the "risks"
associated with potential natural disasters so that loss
may be minimized or even eliminated. This includes the
disaster response which are actions taken in anticipa-
tion of, during and immediately after, a natural disaster
to ensure that its effects are minimized, and that people
affected are given immediate relief and support. A
systematic approach should be taken to manage the risk
of natural disasters. This process of disaster risk
management should consider the likely effects of natural
hazards and the measures by which they can be mini-
mized. A disaster risk management system should
incorporate response actions that are appropriate to the
social and economic conditions of the community under
threat. Because of this, disaster reduction has become
increasingly associated with practices that define efforts
to achieve sustainable developments.
The concept of disaster risk is used to describe the
likelihood of harmful consequences arising from the
interaction of natural hazards and the community. Two
elements are essential in the formulation of disaster risk:
the probability of occurrence of a hazard, and the vulner-
ability of the community to that hazard.
Risk = Hazard Probability x Vulnerability
A closer look at (a) the nature of hazards and (b) the
notions of vulnerability allows for a better and more
comprehensive understanding of the challenges posed
by disaster mitigation:
(i) Nature of hazard - By seeking to understand hazards of
the past, monitoring of the present, and prediction of
the future, a community or public authority is poised to
minimize the risk of a disaster. The NMSs play a key role
in this aspect of risk management of weather-related
natural disasters.
(ii) Notions of Vulnerability - The prevailing conditions
within any group of people in a society can determine
the extent of their susceptibility or resilience to loss or
damage from a natural hazard. The community vulner-
ability is the susceptibility and resilience of the
community and environment to natural hazards.
Different population segments can be exposed to greater
relative risks because of their socio-economic conditions
of vulnerability. Reducing disaster vulnerability requires
increasing knowledge about the likelihood, conse-
quences, imminence and presence of natural hazards,
and empowering individuals, communities and public
authorities with that knowledge to lower the risk before
severe weather events, and to respond effectively imme-
diately afterwards. Increasing this knowledge depends
on focusing science and technology investment to
improve disaster mitigation and resilience by identifying
and meeting needs and closing knowledge gaps wher-
ever possible.
As the decade progresses, a broader global aware-
ness of the social and economic consequences of natural
disasters has developed highlighting the increasing
importance of engaging a much broader community in
hazards awareness and risk management practices. The
importance given to socio-economic vulnerability as a
rapidly increasing factor of risk in most of today’s soci-
eties underlined the need to encourage the participation
of wide spectrum of stakeholders in hazard and risk
reduction activities. Stakeholders are those people or
organizations who may affect, be affected by, or perceive
themselves to be affected by, a decision or activity.
Stakeholders can include governments, delivery agencies,
private sector and the public. In developing a disaster
risk management system, no single agency can provide a
fully comprehensive solution. It is essential that agencies
Chapter 2
EXPLANATION OF KEY CONCEPTS
Guidelines on integrating severe weather warnings into disaster risk management
work together and with stakeholders to narrow knowl-
edge gaps and to develop disaster risk management plans
using a coordinated approach.
4
The devastating effects of natural disasters can quite
commonly be associated with inadequacies in a community’s
development strategy. Poorly planned development turns
recurring natural hazards into disasters. Housing a dense
population on a flood plain or permitting poor building
codes in cyclone-prone and earthquake-prone areas aggra-
vate not only the vulnerability of the exposed communities
but also increase the losses due to natural hazards. Wilhite
(2005) has taken the commonly accepted cycle of disaster
management and redefined it in terms of crisis management
and risk management. Crisis management emphasizes post
disaster impact assessment, response, recovery and recon-
struction; whereas risk management emphasizes protection
through mitigation, preparedness, prediction and early warn-
ing. Traditionally, the focus of disaster management has
almost been exclusively on actions taken immediately before,
during and shortly after a disaster. While many governments
nowadays still employ reactive crisis management, rather
than proactive risk management, approaches to disasters, the
tide appears to be turning.
There is no doubt that the role of relief assistance during
a crisis will remain important and need to be enhanced at all
levels. However, the question must be asked: "Can we afford
to value lives and properties only after they have been lost in
a disaster?" Much greater attention will need to be given to
preventive strategies that can contribute to saving lives and
protecting assets before they are lost. In fact, a paradigm shift
is occurring with a move away from reactive response and
recovery to a much more proactive and holistic concern with
preparedness and prevention. There is now an increased
emphasis placed on risk management rather than crisis
management, and an acceptance that natural disasters, devel-
opment problems and environmental issues are inextricably
linked. Instead of diverting resources away from ongoing
projects through budget re-allocation in order to finance
recovery and re-construction efforts, proactive mechanisms
are sought to reduce the economic costs and impacts of
hazards, improve the countries’ response capacity, decrease
vulnerability and enhance communities’ resilience to disas-
ters.
Responses to warnings on natural disasters/hazards
generally involve making decisions based on calculated risks
and uncertainty. Although safety assurance of life and prop-
erty is a common ideal underlying all warnings, we have to
accept that risks can never be totally eliminated. Thus risk
management generally involves the challenging task of mini-
mizing threats to life, property and the general environment,
yet at the same time without creating unnecessary disruption
and chaos to the communities likely to be affected by natural
disasters/hazards.
With this new philosophy of risk management, coun-
tries will need to develop policies that focus on the
importance of taking appropriate preparedness and preven-
tion measures. Through such measures, the overall impact of
natural hazards can be significantly reduced:
(i) improved scientific understanding of the range of
natural hazards that can occur and their potential to
occur at specific locations (hazard/risk mapping on a
firm scientific basis);
(ii) improved understanding of the cost implications of the
above hazards;
(iii) better design and construction of buildings, bridges and
other structures (including flood plain management) to
reduce the risk of damage during a severe weather or
flood event;
(iv) increased community awareness through public educa-
tion of the impacts of natural hazards and the measures
individuals can take to reduce damage and protect life;
(v) improved technology and modelling effort to extend the
lead time for warnings of potential natural hazards; and
(vi) integrated risk management approach, treated as a total
package instead of separate components.
Among the measures mentioned, NMSs would be
involved to a greater or lesser extent. But in all aspects of risk
management, effective communication with stakeholders,
especially the public, is vital.A community at risk to a natural
hazard will not be prepared to meet it if the general public is
not well informed. Communication of mitigation and
preparedness measures can transform a community’s readi-
ness to face an extreme event. Effective communication is
especially critical immediately before, during, and after a
hazard event to ensure the public has adequate warnings and
complete information to maximize public safety. Ultimately,
the messages must also be understood and acted upon by
those communities at the greatest risk, including the disad-
vantaged people.
As such, severe weather warning systems must be viewed
as an integral element of long-term strategies in the sustain-
able development for a safer and more adaptable community.
Such systems thus constitute an essential and highly cost-
effective component of national strategies and must be
properly embedded and integrated in national risk manage-
ment programmes. Building on this foundation, there is
inevitably a need to ensure that severe weather warnings of
natural hazards becomes an integral part of government
policy in every disaster-prone country, and it also follows
that NMSs need to be recognized as a major component of
the corresponding infrastructure in support of risk manage-
ment and disaster mitigation.
With continuing scientific and technological develop-
ments, there is a growing expectation that forecasters will be
able to warn at longer lead-times, and with more emphasis or
greater accuracy regarding the ‘where’,‘when’ and magnitude
of an impending natural hazard. While the advances and
successful applications of weather warning systems are most
encouraging, the common pitfall is to over-emphasize on the
Chapter 3
RISK MANAGEMENT PHILOSOPHY
technical details and under-state the practical implications for
the communities exposed to the hazard. In the communica-
tion process, not only should the communities be adequately
informed, but also they should be sufficiently impressed that
preparedness actions are taken before and during the antici-
pated hazardous event. Warning messages must convey to an
often-sceptical public that they are vulnerable, that the danger
is real, and that specific measures can be taken to protect
themselves and their property. Messages must also consider
the target audience, making provisions to reach high-risk
areas and to overcome language barriers in diverse popula-
tions.
In the spirit of an integrated risk management approach,
the multi-discipline and multi-sector nature of the warning
process must not be overlooked. Although based on scientific
and technology, warnings must be tailored to serve users'
needs, with due consideration given to their situations and
available resources. While severe weather warning capabili-
ties must continue to be strengthened at the global level, it is
just as important that enough emphasis should be given to
developing capacities that are relevant, and responsive to, the
needs of local communities. Fundamental to the process is
the existence of a Risk Management Action Plan, which is
the subject of discussion in the next chapter.
Guidelines on integrating severe weather warnings into disaster risk management 6
Disaster risk management consists of three broad activities:
mitigation, preparedness, and early warning. Mitigation
includes risk reduction; preparation includes risk assessment,
risk analysis, hazard identification, research and develop-
ment, public education and outreach; whereas early warning
consists of hazard forecasts, communication and dissemina-
tion. NMSs contribute to risk management in three ways:
The first is to provide early warning of weather, water and
climate hazards for operational decisions; the second is to
support risk and impact assessments to determine who and
what is at risk and why; and the third is to improve forecasts
and analyses to help reduce or remove risks (Rogers 2005).
The following sections provide a framework for NMSs to
consider in developing and implementing a Risk
Management Action Plan in partnership with other agencies
and organizations involved in risk management. This frame-
work focuses on the role of the NMS in developing and
executing the Risk Management Action Plan with regard to
severe weather warnings, but acknowledges the ultimate Risk
Management framework should also address non-weather
hazards and draw expertise from many other partners. Some
NMSs may have implemented some or most of these
elements in a Risk Management Action Plan; where that is
true, the framework may prompt these countries to increase
or enhance their efforts.
4.1 CORPORATE CULTURE
For meteorological hazards, the NMS is the sole warning
authority for its citizens. NMSs are the experts on the mete-
orological aspects of severe weather hazards, and are
therefore a critical player in the development of the country’s
risk management plans.
The NMS must be a credible authority for information
on severe weather warnings and has a reputation for accuracy,
reliability, and timeliness. It is also increasingly recognized
that NMSs need to develop a corporate culture of being
caring and people-centred, in addition to the more tradi-
tional culture of being professional and science-centred.
Developing working relationships with partners such as
emergency managers and the media (see section 4.3) and
involving stakeholders in the development and review of the
warning system (see section 5.1) is essential.
Many NMSs are now emphasizing a user-centred
approach to improving their products and services, in addi-
tion to the traditional science-oriented approach. Some
examples of a user-centred approach are:
(i) plain-language descriptions that are clearly understood
by the public in weather bulletins and outreach material
about the hazards and warnings;
(ii) making observations, forecasts, and warnings available
in real-time on the Internet;
(iii) frequent updates of the latest information, even if noth-
ing is new; and
(iv) distributing purposely designed educational and
outreach material to educate stakeholders about severe
weather events, the associated risks, and actions to take
in advance of, during, and after the event.
4.2 CAPACITY BUILDING
In support of an effective severe weather warning system, it
is vital that NMSs secure the necessary financial and person-
nel resources for:
(i) ongoing maintenance and improvement of the national
meteorological observation infrastructure;
(ii) pure and applied research – both meteorological and in
other disciplines associated with risk management (most
likely through partnerships and collaboration);
(iii) ongoing training for NMS staff, partners, and stake-
holders;
(iv) public education and awareness; and
(v) development of technical, operational, and dissemina-
tion capabilities.
NMSs should ensure that:
(i) the capacity building of their severe weather warning
specialists is high on the agenda;
(ii) the operational procedures are sound; and
(iii) the communication mechanisms and systems are robust.
Training should not be limited to NMS staff, but must
also include partner agencies as well as communities at risk.
For example, in the United States, the National Weather
Service (NWS) works with the Federal Emergency
Management Agency (FEMA) to develop and teach emer-
gency managers how to use NWS product and services, using
resident and distance-learning components:
(i) Hazard Weather and Flood Preparedness;
(ii) Warning Coordination;
(iii) Partnerships for Creating and Maintaining Spotter
Groups; and
(iv) Hurricane Planning.
NMSs also need to allocate sufficient resources to
develop a knowledge base for the effective provision of severe
weather warnings. Examples of such initiatives are:
(i) sponsor or conduct applied research regarding the severe
weather hazards of their country;
(ii) maintain a historical database of past severe weather
events;
(iii) support the production of hazard risk assessments; and
(iv) play an active role in the development of a risk manage-
ment plan for their country for regional and local
applications.
NMSs and other agencies involved in risk management
planning should establish collaborations and methods for
Chapter 4
NMSs’ RESPONSIBILITIES WITHIN
A RISK MANAGEMENT FRAMEWORK
Guidelines on integrating severe weather warnings into disaster risk management
effective exchange of information among relevant hazard
databases to facilitate monitoring, assessment, and predic-
tion.
4.3 DEVELOPING PARTNERSHIPS
The design and operation of severe weather warning systems,
and on a larger scale, the risk management plan, must be
based on a commitment to cooperation and information
exchange and the concept of partnership in the overall public
interest. The benefits of such partnerships include:
(i) drawing expertise from a wide range of disciplines, such
as social science, community planning, engineering, etc.;
(ii) accomplishing tasks that cannot be managed by a single
agency or organization;
(iii) demonstrating to government budget planners a
commitment to work together towards a common goal
and making better use of scarce financial resources;
(iv) leveraging resources for research, awareness, prepared-
ness, etc.;
(v) sharing costs, knowledge, and lessons learned;
(vi) ensuring a consistent message (the warning bulletins and
other outreach material) from multiple credible sources;
and
(vii)yielding wider distribution of the message through
multiple outlets and receiving feedbacks from a whole
range of users.
To identify and evaluate the weather information needs
of the users, NMSs need to build relationships and work in
partnership with users in both the public and private sectors.
NMS partners include:
(i) other government agencies with missions involving the
protection of life and property, such as the NHS where
that is a separate agency from the NMS, national,
regional or local emergency management agencies, first
responders, infrastructure managers (dams, highway
departments, bridges);
(ii) the media;
(iii) non-government organizations;
(iv) emergency relief organizations, such as the Red Cross
and Red Crescent Society;
(v) academic institutions and schools;
(vi) trained volunteers associated with the NMS, such as
cooperative observers, storm spotters, and amateur radio
operators;
(vii)meteorological societies and other professional associa-
tions in risk management disciplines;
(viii) private sector weather companies, and
(ix) utility services, telecommunication operators and other
operation-critical or weather-sensitive businesses.
A typical partnership would involve disaster, warning,
and risk management experts from government, business,
academia, non-government organizations such as the Red
Cross and Red Crescent Society, as well as emergency
management officials to agree on warning standards, proce-
dures, and systems. An example of this is the Partnership for
Public Warning in the United States. More discussion and
examples of partnerships can be found in Section 5.1,
Stakeholder Involvement, of this document, as well as the
WMO document PWS-8, Guide on Improving Public
Understanding of and Response to Warnings.
4.4 PUBLIC EDUCATION & AWARENESS
Public education and awareness programmes need to focus
on the NMS’s severe weather warning systems, the hazards,
the vulnerability and appropriate responses. The goal of
effective public information materials and programmes
should be to ensure all those who are potentially at risk fully
understand the nature of the threat and the appropriate
action to take at various stages as the threat develops and
disaster strikes (Zillman, 2004).
Working with partners can enhance the effectiveness of
education and outreach efforts and spread the cost of devel-
opment and printing among several groups. In addition to
printed material, the same information should also be made
available electronically via the Internet, and in alternative
languages for different ethnic groups or in formats that are
accessible to people with disabilities.
Educational and outreach material should highlight
scientific capability, hazard awareness, risk, and prepared-
ness. The effectiveness of brochures and educational material
can be increased if they follow a similar format, featuring
sections such as:
(i) a description of what causes the weather hazard;
(ii) a map highlighting areas of the country vulnerable to the
hazard;
(iii) historical information about major events of this type of
weather hazard in the country;
(iv) definitions of terminology associated with the hazard;
(v) how to anticipate and monitor the hazard in one’s own
situation;
(vi) measures that individuals and communities can take in
preparation for the occurrence of hazard;
(vii)recommended actions to take before, during, and after
the event;
(viii) how to stay informed during the event;
(ix) contacts for more information; and
(x) contingency measures in the event of a disaster (for the
family, the community, the business, etc.).
Even the selection of graphics and photographs is signif-
icant. Rather than images of destroyed structures, social
scientists recommend including pictures of people taking the
proper actions in preparation for and during the event, such
as people boarding up windows, heading to a storm shelter,
shopping for items in their emergency supplies kit, etc.
NMSs should support hazard awareness education in
schools by contributing to the development of educational
material for teachers and students. In educating the students,
the messages are also brought home to their families and
neighbourhood as well.
The WMO document PWS-8, Guide on Improving Public
Understanding of and Response to Warnings, includes more
discussion and examples of public education and awareness
initiatives.
8
4.5 ESTABLISHING A KNOWLEDGE BASE
One must have relevant knowledge and information in order
to properly understand the hazard and vulnerability. At the
same time, the value of local knowledge, community
"memory", and relevant experience during past events should
not be overlooked. Furthermore, knowledge and informa-
tion is central to almost every aspect and every stage of
natural disaster risk management. It is essential for assessing
risk and potential vulnerability in the earliest stages of
community planning for construction of new facilities (such
as dams, bridges and population centres) or for individuals
planning to move to new locations (such as beaches, flood
plains and mountain sides). NMSs therefore need to allocate
or secure sufficient resources for the development of an
adequate knowledge base for the effective provision of severe
weather warnings. The knowledge base shall consist of (i)
risk databases, (ii) risk assessments, (iii) cost-benefit analy-
ses, and (iv) research activities. These will be discussed in
turn in the following sections.
4.5.1 Risk Databases
Information that is most vital in the early stages of sound
risk management consists of the full suite of climatological
data and products, including model studies of extreme
events, that will enable the potential natural hazard to be
accurately characterized (e.g. through hazard maps) and the
necessary decisions on siting, construction, protection and
precaution to be taken on a fully informed basis.
Information is essential when natural hazards threaten
and when communities prepare to withstand the potential
onset of disaster; and it is often even more essential in the
critical post-disaster recovery phase when affected commu-
nities are shattered and confused, when fear of the
unexpected is greatly heightened, and when relief authorities
need to know everything that is going on to enable them to
manage the complex mix of issues involved in restoring
essential facilities and in meeting the physical or social needs
of devastated communities.
Effective risk management of and preparedness for
natural hazards require free and unlimited access to relevant
databases to facilitate monitoring, assessment, and predic-
tion. It is recommended that all agencies responsible for these
databases develop good collaborating links for the exchange
of information included in the databases.
Communities and civic authorities need to know the
nature, severity and likely return periods of all kinds of
potential hazards. One of most important sources of infor-
mation is still the careful analysis of records of what has
happened in the past. In this regard, meteorological data-
bases (including warnings and climate databases) contain
information relating to natural hazards of a meteorological
origin. On the other hand, disaster databases in general
contain information largely driven by economic and financial
considerations such as insured or un-insured losses, but often
with very little emphasis on information relating to the
weather and climate conditions at the time of the disaster
event. There is a growing recognition that in order to effec-
tively assess the vulnerability and risks involved, analysis has
to be done based on information drawn from both types of
databases. At present, there is no easy means of integrating
or cross-referencing such data and information for complet-
ing an adequate risk analysis. More research on the effective
application of these data and information is therefore
required.
4.5.2 Risk Assessments
Risk assessment requires a "by the people, for the people"
mindset. Risk is the outcome of the interaction between a
hazard phenomenon and the elements at risk within the
community (e.g. the people, buildings and infrastructure)
that are vulnerable to such an impact.
Each hazard likely to impact upon a community can be
systematically analyzed in this fashion. The vast majority of
information, relationships and processes involved in under-
standing risk are spatial in nature. Graphical Information
Systems (GIS) are especially useful for this purpose.
In risk assessment, one has to consider the probabilities
of hazardous events affecting the community and the conse-
quent harm to the community. Probability is a concept and
skill that most people have problems with understanding, as
many cannot handle statistical concepts or effectively factor
probabilities into their decision-making. More research is
needed to help describe how low probability/high conse-
quence events (such as an impact by a very intense cyclone)
affect the community’s mitigation desires and actions.
4.5.3 Cost-Benefit Analyses
General Principles
We will never be in a position to reduce hazard-induced
losses to zero. The measurement of how much reduction is
achieved is usually referred to as the Loss Avoided (La).
Against this, there will be an attributable cost in providing
warnings and taking preventive actions, called Cost of
Prevention (Cp). The ratio, ?La/?Cp, gives a simple but
effective Benefit Cost Ratio (BCR). Clearly we need to aim
for as high a value as possible. The BCR can be used to:
(i) support investment proposals to improve disaster
prevention and mitigation, and
(ii) prioritise work programmes and financial investment.
Understanding of the Benefit Cost Ratio (La/Cp) allows
us to make effective use of probability forecasts.
Calculation of BCR
In practice it is difficult, even impossible, to get an accurate
measure of La. However, this should not deter attempts to do
so with varying levels of sophistication. In a rudimentary
approach, a subjective value can be derived by making esti-
mates of:
(i) number of people or organisations receiving a warning;
(ii) percentage of recipients taking action;
9 Chapter 4 — NMSs’ responsibilities within a risk management framework
Guidelines on integrating severe weather warnings into disaster risk management
(iii) effectiveness (average cost saving) of these individual
actions;
(iv) frequency of the phenomena and accuracy of forecast-
ing.
In addition, in cost-benefit analyses, a nominal monetary
value to human life saved and a surrogate value for public or
political goodwill are sometimes assigned. This valuation is
merely a matter of convenience and does not mean a judge-
ment on the intrinsic value of lives. Any other common
measure of value would, in principle, do equally well.
Cost of prevention, Cp, is usually easier to measure or
more accurate to estimate. A decision has to be made as to
whether to use full or marginal costs. Marginal costs are those
attributable to taking mitigating actions such as extra
manpower and resources (for example from emergency
authorities) to reduce losses and save lives.
It follows that there needs to be detailed discussions with
other stakeholders in order to derive Cpand La. The strive for
higher BCRis more likely to attract extra funding for capac-
ity building leading to improved effectiveness of severe
weather warnings. Often it will be found that efforts to
improve response to warnings, rather than a higher accuracy
of forecasts, offer the most economical way of improving the
BCR. This again highlights the need to work closely with
professional partners in designing a risk management plan
and in bidding for funds.
Probabilities and Consequences
In trying to forecast a severe weather event, we are dealing
with extreme events and thus working within the tail-end
regime of a statistical distribution. The main issue is the soci-
etal impact caused by the weather, i.e. amount of damage,
disruption, economic loss or number of deaths. These factors
usually increase exponentially with an increase in the sever-
ity of the phenomena.
From a risk management viewpoint, it will be more valu-
able to deal with the probability of various thresholds being
exceeded. Unfortunately our professional partners (and the
public) are generally not well versed in the use of probability
forecasts as a decision making tool.
Correct and accurate use of probability forecasts means
that, given a large sample, on average an event will occur at the
same frequency as the forecast probability. For example,
events assigned with a 50% probability will occur on half of
the number of occasions they are predicted. Similarly, those
assigned with a 10% probability should occur 1 in 10 times,
and those with 90% probability on 9 out of 10 occasions. As
such, the full BCR is only realised over a period of time.
Therefore our partners must appreciate that they need
management strategies that are different from those dealing
with deterministic forecasts. In particular, the response must
be commensurate with the risk involved.
As part of the partnership approach it is necessary to
have a common understanding and use of thresholds. An
example illustrating the use of probabilities of a severe
weather event affecting the community in calculating BCR
and the determination of threshold for decision making is
described in Appendix A.
4.5.4 Research Activities
Every NMS, according to the available resources in qualified
workforce and technical facilities, will currently perform
some research activities in the fields of meteorology, clima-
tology, hydrology, oceanography, or even seismology. An
important part of these efforts is the understanding and
prediction of hazards. Storing and updating the results of
such research activities is a fundamental requirement for any
scientific approach for risk management planning of hazards.
An aspect, which probably deserves more attention, is
the benefits that may be derived from international collabo-
ration of research activities, through networking of
institutions or individual scientists. A primary motivating
factor is that the understanding of physical processes in the
atmosphere often requires full scale field experiments to
gather data and information; and such efforts are usually out
of the reach of individual scientists, a single NMS or even
national consortia of research institutions in many countries.
A good example is the progress made in the understanding
of Atlantic storms after FASTEX (Fronts and Atlantic Storm-
Track Experiment) (Browning et al, 1999). This has lead to
the development of the decade long WMO research project
THORPEX (The Observing System Research and
Predictability Experiment), which is addressing the use of
ensemble forecasts and targeted observations to improve
early warning of hazard weather (Shapiro and Thorpe 2004).
International cooperation among meteorologists and
other scientists in different fields is also an effective way of
securing funding and support from international institutions.
For example, the five-year cycle of the R&D programmes of
the European Union, which various European meteorologi-
cal services contribute to and get funding for, contains
components that directly address the hazard risk manage-
ment issues (e.g. GMES (Global Monitoring for Environment
and Security)).
Another related structure, different but related to the
European Union and is more of a networking mechanism, is
the European Cooperation in the field of Scientific and
Technical Research (COST) organisation. It is well known
that the origin of the European Centre for Medium-range
Weather Forecasting (ECMWF) is to be found in a COST
action decided 30 years ago. More recently, actions in the
domains of nowcasting, urban pollution, radar techniques
and waves modelling also provide the international commu-
nity with valuable results and deliverables (exchange formats,
data bases, tested algorithms, etc.) (further details can be
found inhttp://www.cordis.lu/cost/home.html). The fact that
the COST approach is "bottom up" (i.e. actions initially
proposed by individual scientists) and multi-disciplinary is a
great advantage; at the same time, there are also effective
coordination and controls ("top down") by national repre-
sentatives that look after users’ interests and ensure
productive outcome of the actions. Eventually, application
programmes would be derived from such actions, currently
funded by the European Commission of the European
Union. In this sense, participation in COST can be an effi-
cient leverage for obtaining much-needed funds for
meteorological institutes to increase their knowledge base.
10
As already mentioned, a mere "natural sciences"
approach is not sufficient. Risk management requires a
combined multi-disciplinary approach because of the socio-
economic, political and in some respects, psychological,
dimensions involved in the formulation of policies and strate-
gies. A workshop on early warnings held in Shanghai in 2003
noted (Glantz, 2003): "With regard to geological or hydro-
meteorological hazards, the physical processes are either well
understood or are under close scrutiny by scientific
researchers. With regard to socio-economic and political
processes, there is a heightened need to understand their roles
in the conversion of a potential hazard into an actual disas-
ter."
NMSs will increase their capacity to fulfil their role in a
risk management plan when the results of such research
activities in other specialized fields are integrated, to a greater
or lesser extent, into their own knowledge base. Participation
in multi-disciplinary seminars or colloquia dealing with
natural hazards is necessary, but not sufficient. In building a
real knowledge base, the mere exchange of results among
meteorologists, hydrologists, sociologists, psychologists,
media specialists is a good start but will probably still fall
some way short of providing an overall approach and under-
standing of hazards in the context of attendant societal or
economical problems. A recent prominent report on Risk
Management in the USA notes: "In developing the nation’s
all-hazards approach to disaster vulnerability reduction, no
single federal agency can provide a fully comprehensive solu-
tion. Agencies must work together to narrow programme
gaps through a coordinated science and applications research
agenda to pursue common solutions to common problems"
(Executive Office of the President of the USA, 2003).
A conspicuous illustration of this may be derived from
the lessons learnt from the deadly heat wave that hit Western
Europe, especially France, during the first fortnight of August
2003. A comprehensive evaluation, from the risk manage-
ment point of view, is to be found in Section 6.1.2. It has been
publicly acknowledged that in such an occurrence, the collab-
oration and synergy between key stakeholders in the fields of
meteorology, public health, and civil security have been very
insufficient.As far as the research implications are concerned,
it is clear that at the time there were available neither a
comprehensive understanding of all the factors that resulted
in such a catastrophe, nor in the aftermath adequate and
detailed indicators that could have been of precious use for
the authorities and the public (the cited Shanghai workshop
noted: "The selection of indicators is very important, because
monitoring will centre on them. The wrong indicators can
lead to wasted time, effort and resources").
The predictions for the onset, duration and end of the
heat wave by the relevant meteorological services were mostly
correct and timely (as allowed by the state of the art)
(Grazzini et al., 2003), but still quite general and insufficiently
echoed by the media and relevant institutions. Taking the
case of Météo France, its warning action (press releases
mentioning the people at risk) was actually more efficient
and reflected the user-oriented approach of Météo France.
Yet only a few years ago, a bio-meteorological conference
(Conseil Supérieur de la Météorologie, 2002) had correctly
pointed out the deadly potential of such heat waves and
roughly described its consequences. Nevertheless, it is now
understood that the setting up of combined indicators for
such events requires definitions and testing that can be
achieved only through multi-disciplinary research into
biological, demographical and sociological behaviours affect-
ing the most vulnerable populations, as well as much needed
precise documentation and understanding of the meteoro-
logical circumstances, not only minimum and maximum
temperatures but also wind, humidity and air quality.
In France, these conclusions have led to formalized
agreements between Météo France and two institutions in
charge of public health. By the end of 2003 (see also Chapter
6), a first agreement was quickly set out to provide some new
indicators, albeit on a provisional basis, with the opera-
tionally oriented "Institut de Veille Sanitaire" (Institute for
Sanitary Watch). Another collaboration which is much more
far-reaching was established in July 2004 with the medical
research institute INSERM. According to the press release, it
is due to enlarge and enhance the combined research on heat
(and cold) waves and their impact on public health, with an
aim to document and study in details the bio-meteorological
mechanisms at much finer scales and over a longer period of
30 years.
In building up a comprehensive research basis for risk
management planning, NMSs are advised to actively enter
(according to their resources and/or those that could be
provided by interested sponsors) into joint research agree-
ments with other scientific institutions. This would normally
imply the constitution of mixed research teams or the
exchange of scientific missions; or to host post-doctorate
researchers coming from different scientific horizons. In the
medium to long term, the results of such research will prove
to be of great use for the prevention and mitigation of
complex hazards. The same cited American report
(Executive Office of the President of the USA , 2003) lists as
follows the research and development areas to be considered:
"- fundamental and applied research on geological,
meteorological, epidemiological and fire hazards devel-
opments;
- application of remote sensing technologies, software
models, infrastructure models, organizational and social
behaviour models, emergency medical techniques; and
- many other science disciplines applicable to all facets of
disasters and disaster management".
4.6 DISASTER RISK MANAGEMENT EXPERTISE
NMSs should give due consideration to gaining access (full-
time or occasional) to risk management expertise to
complement its core operations. As an example, the
Australian Bureau of Meteorology has over the past few years
established a cross-cutting disaster mitigation programme
with the recruitment of a disaster management specialist.
This programme supports the Bureau’s severe weather warn-
ing services and works in partnership with various
stakeholders throughout the community to ensure the appli-
cation of risk management principles in the communication
of severe weather warning information and related risk
messages.
11 Chapter 4 — NMSs’ responsibilities within a risk management framework
4.7 FORMULATING A RISK MANAGEMENT
ACTION PLAN
Risk management provides structured systems for identify-
ing and analyzing potential risks, as well as devising and
implementing responses appropriate to their impact.
According to the Secretariat of the ISDR (International
Strategy for Disaster Reduction, 2004), disaster risk manage-
ment focuses on three key areas:
(i) assessment of the risk factors present;
(ii) tools and practices to reduce the risks; and
(iii) institutional mechanisms to support both risk assess-
ment and risk management.
It thus spans a wide range of methods and activities –
assessment and analysis, public information, community
participation, warning systems, policy and regulation, project
impact assessment, education programmes, conservation
practices and political processes. When advantage is taken of
the mutual benefits achieved through adoption of a system-
atic integrated approach to the various activities, it is
appropriately referred to as "integrated disaster risk manage-
ment".
Risk management consists of a systematic process of
assessing and dealing with risks. This process involves
consideration of the context, followed by identification,
analysis, evaluation, and treatment of risks. It is a repetitive
cyclic process that also requires monitoring and review, and
should include communication and consultation with stake-
holders all along. The process of disaster risk management is
shown schematically in Fig. 1. This process can be used in
formulating the Risk Management Action Plan. The steps to
follow are:
(i) Communicate and consult with stakeholders throughout
the process;
(ii) Establish the Context;
(iii) Identify and Quantify the risks;
(iv) Analyse and Evaluate the risks;
(v) Treat the avoidable and manageable risks; and
(vi) Monitor and Review the Risk Management Plan.
Each of these steps will be discussed in detail in the
following sections.
4.7.1 Step 1: Communicate & Consult with
Stakeholders
Communication and consultation are important considera-
tions at all stages of the process to ensure that stakeholders are
involved. This is the first step in the disaster risk manage-
ment process. Consultation is a two-way process that enables
disaster risk managers to be aware of perceptions and to
accept input to the process from stakeholders. Stakeholders
will have perceptions of risks that can be useful in the risk
assessment stages. In developing risk responses, it is also
desirable to consider communication strategies for both
internal and external stakeholders. Consultation ultimately
aims at developing partnerships. Communication must be
effective to ensure that those agencies and/or individuals
responsible for implementing risk treatment measures are
given sufficient information about the measures and the
reasons for the adopted strategies.
4.7.2 Step 2: Establish the Context
The problem must first be defined by determining the nature
and scope of the risk management project. This includes
defining the community involved and the types of natural
hazards to be addressed. A project management structure
must also be established. A simple SWOT (strengths and
weaknesses, opportunities and threats) analysis can be an
excellent means in the initial consideration of the context of
risk assessment.
4.7.3 Step 3: Identify & Quantify the Risks
In the world of natural hazards, there are many different ways
of defining and quantifying risks depending on the intended
purposes (e.g. for deciding what protective measures to put
in place or what insurance premiums to charge). However,
most definitions would include some combinations of:
(i) the nature of the hazard, including its severity;
(ii) the probability of its occurrence;
(iii) the exposure and vulnerability of human and natural
systems; and
(iv) the numbers of people and the costs of the facilities at
risk.
Risk identification is achieved by:
(i) identifying and describing the natural hazards – the
sources of risk;
(ii) identifying and describing the community and its envi-
ronment – the elements at risk;
(iii) determining the vulnerability – the balance between
susceptibility (the level to which a particular natural
hazard will effect a community) and resilience (the abil-
ity of a community to recover from the impact of a
natural hazard); and
(iv) describing the risk.
The process of quantifying and mapping risks
thus involves an integrated approach to the use of physi-
cal/geographical/environmental information in conjunction
with both social and economic data, usually through inte-
grated assessment models of various kinds.
4.7.4 Step 4: Analyse and Evaluate the Risks
Risk is analysed by determining the likelihood of a natural
hazard occurring and the consequences of that hazard event.
This is done in both qualitative and quantitative terms. The
analysis considers community vulnerability and existing risk
management measures with all assumptions clearly stated.
The relationship between likelihood and consequences then
enables the level of risk to be determined. The level is not an
absolute level, but reflects a multi-faceted set of criteria that
enables societal judgments on the risk to be made.
Guidelines on integrating severe weather warnings into disaster risk management 12
13 Chapter 4 — NMSs’ responsibilities within a risk management framework
Fig. 1. The Disaster Risk Management Process
Establish Context
Identify Risk ...
What? (event)
How? (source)
Why? (impact)
Analyse Risks
yes
Consequences
Estimate Level of Risk
Evaluate Risk
(priorities)
Is risk acceptable?
Treat Risks
Stakeholder
Dialogue -
Communication /
Consultation
Monitoring /
Review
No
Likelihood
Risk is evaluated by comparing the risk evaluation crite-
ria with the level of risk. This establishes the priority for the
treatment of each risk and/or the acceptability of the resid-
ual risk. Acceptable risk is sufficiently low risk at a level that
the society is comfortable with. A particular risk may be
accepted when the cost of treatment is considered excessive
compared to the benefit of the treatment. The process is
achieved by consultation with all stakeholders and is subject
to review and modification if required.
4.7.5 Step 5: Treat the Avoidable and Manageable Risks
Risk treatments are designed to reduce any or all of the
vulnerability of elements at risk, the likelihood of risk occur-
ring and the consequences of the event. The process involves
identifying and evaluating options, selecting the most appro-
priate treatment(s), and planning and implementing the
treatment programme(s).
The treatment of identified risks will almost certainly lie
in technical innovation, personnel skills development and
the ability to do both pure and applied research. There is a
range of actions that can be taken to reduce the risk associ-
ated with natural hazards. These measures fall into four main
categories:
(i) adoption of adaptive strategies such as temporary or
permanent migration away from the areas of high risks;
(ii) education and awareness programmes for key stake-
holders;
(iii) implementation of a works programme aimed at forti-
fying the infrastructure; and
(iv) adoption of diversified responses, such as a combina-
tion of the above measures along with land use planning,
refined warning systems, and insurance incentives.
Formal reporting is an important phase of the risk
management process. A Risk Management Action Plan is a
helpful formal way to report on designated or major under-
takings. It should summarize the results of the risk
management process, action strategies and implementation
framework. In particular, it should describe the risk response
measures to be implemented to reduce and control risks. For
major risks, risk action schedules should be prepared to
assign individual responsibilities and time frames and to
identify those who are responsible for follow-up. The Plan
should also include provisions for implementation and ongo-
ing reporting.
The most important task in risk management is imple-
menting the Risk Management Action Plan and allocating
management resources. This should be followed by monitor-
ing the effectiveness of these measures over time (Step 6).
Planning for implementation requires particular attention to
be given to resources needed, management responsibilities
and timing of tasks.
4.7.6 Step 6: Monitor and Review the Risk
Management Action Plan
Risk is not static. It is therefore necessary to continually moni-
tor the status of the risk being managed and the interaction
of risk, community and the environment; and to review the
risk management processes in place. Continual monitoring
enables the process to dynamically adapt to changes in risk
as well as changes in stakeholders’ needs. The frequency of
monitoring and the responsibility for conducting the review
should be specified in the Risk Management Action Plan.
4.8 SECURING FUNDING
In planning and implementing the risk management
programme, resources for ongoing training activities, public
education as well as the development of both technical and
operational capabilities are essential. Severe weather warnings
are effective only to the extent that policy makers at national
levels of authority have the will to make a sustained commit-
ment of resources that establish and operate the required risk
management mechanisms.
To obtain and maintain such resources, a collaborative
approach to funding requests is more likely to be successful.
If the NMS is able to join forces with partners, it not only adds
influence and weight to the scope and substance of the
proposed project, it also enables the politicians to consider
the merit of the project in the context of the complete risk
management picture, and not simply confining it to the
narrow and more technical role of the NMS.
Presenting a well considered risk management plan is
also a very significant step in soliciting funding from both
national and international sources, such as the World Bank or
the European Union.Weather and climate hazards have never
been bounded by national boundaries; and in an increas-
ingly globalized community with active information
exchange and population movement, the impact of any severe
weather events would be even more keenly felt. As such,
NMSs need not be too inward looking in the search of possi-
ble funding options; sponsors with international interest may
just as likely be prepared to contribute. A joint funding bid
with key stakeholders or other NMSs in the region would
inevitably broaden the horizon and hence enhance the like-
lihood of success. Sometimes, it may even lead to a larger
funding allocation than a NMS working alone would other-
wise have achieved.
Apart from international sponsorships and traditional
sources such as the NMS budget, other appropriate funding
alternatives can also be explored in connection with different
components in the overall risk management plans.
Collaboration with universities or research organizations is
one such option through the utilization of research grants.
Often, the academia can also offer the necessary expertise
across a multitude of disciplines and provide a dedicated
environment for research studies. This kind of collaboration
is particularly rewarding for NMSs that do not have in-house
capacity to undertake basic research and technology devel-
opment. NMSs can also consider the possibility of securing
funding through joint venture with other community or
infrastructure projects, particularly those that are weather-
sensitive and may require meteorological support for
planning and operation; e.g. the setting-up of observation
networks along highways or bridges, consultancy role in
Guidelines on integrating severe weather warnings into disaster risk management 14
15 Chapter 4 — NMSs’ responsibilities within a risk management framework
urbanization or conservation projects, forecasting and warn-
ings services for recreational or major sporting events, etc.
As an illustration of what can be achieved, success stories
of funding acquired from some innovative sources are
presented in more details in Section 6.2.
Natural disasters can lead to extensive property damage,
disruptions on social activities and loss of life. As noted in
the Introduction, during the period 1992-2001, natural disas-
ters worldwide have killed over 622,000 and affected over 2
billion people. While the figures have been high, it should be
pointed out that they would have been even higher without
pre-disaster efforts, particularly early warning systems, which
contribute significantly to an effective risk management
strategy.
The primary objective of a warning system is to
empower individuals and communities to respond appro-
priately to a threat in order to reduce the risk of death, injury,
property loss and damage. Warnings need to get the message
across and stimulate those at risk to take action.
Effective inclusion of the severe weather warning system
in a risk management plan relies on NMSs to appreciate the
needs of a multi-cultural, economically stratified and often
mobile community, and the community understanding the
hazard, its vulnerability and the most suited protective action
to take.
Greater focus towards disaster mitigation also means
(Gunasekera, 2004):
(i) further increasing the emphasis on extending the lead
time of warnings;
(ii) improving the accuracy of warnings at varying lead
times;
(iii) satisfying greater demand for probabilistic forecasts;
(iv) better communication and dissemination of warnings;
(v) using new technologies to alert the public;
(vi) better targeting of the warning services to relevant and
specific users (right information to right people at right
time at the right place); and
(vii)ensuring that warning messages are understood and the
appropriate action taken in response.
5.1 STAKEHOLDER INVOLVEMENT
Stakeholders need to be consulted as partners in the design
and refinement of severe weather warning systems, and on
the larger scale, the risk management plan. Stakeholders
include the public, other national government agencies,
emergency management agencies, local authorities, non-
government organizations, the media, social scientists,
national and regional infrastructure authorities, academia,
etc.
Involving stakeholders in developing and enhancing the
end-to-end severe weather warning system has many bene-
fits, such as:
(i) improved presentation, structure, and wording of the
warnings themselves;
(ii) more effective communication of the risks and actions to
take in response to severe weather;
(iii) better understanding of how, and how often, stakehold-
ers want to receive warnings; and
(iv) increased sense of ownership, and therefore, credibility
in the warning system.
5.2 WARNING PRESENTATION
Effective warnings are short, concise, understandable, and
actionable, answering the questions of "what?", "where?",
"when?", "why?", and "how to respond?". The use of plain
language in simple, short sentences or phrases enhances the
user’s understanding of the warning. In addition, the most
important information in the warning should be presented
first, followed by supporting information. Effective warnings
should also include detailed information about the threat
with recognizable or localized geographical references.
Warnings should be presented in several different
formats – text, graphics, colour-coded categories, audio –
and should include specific actions for people to take to
respond to the event. The various formats also make it easier
for people with disabilities to receive and act on the warnings.
All formats, however, must present the information
accurately and consistently.
The suggestions above are based on social scientist
recommendations to structure the message to encourage the
public to feel personally affected. For a more thorough
discussion of warning presentation, see WMO document
PWS-8, Guide on Improving Public Understanding of and
Response to Warnings.
5.3 WARNING COMMUNICATION
Dissemination is delivery of the warning messages; but
communication is accomplished only after the information is
received and understood. So the foundation of warning
communication builds on the format and wording of the
warnings, dissemination methods, education and prepared-
ness of stakeholders, and their understanding of the risks
they face. Communication is also significantly enhanced
when consistent warning information is received from multi-
ple credible sources.
NMSs should ensure their severe weather warnings are
communicated using a variety of formats (text, graphics,
voice) and disseminated via as wide a range of media as is
available (press, radio and television, e-mail, cell phone,
Internet, etc.). Media broadcasts from the weather office
and/or radio and television interviews with one or more
authoritative figures can be effective in triggering response
from people. These authoritative figures may be from the
NMS (such as the NWS director or the manager of a local
Chapter 5
EFFECTIVE EARLY WARNINGS
17 Chapter 5 — Effective early warnings
weather office) or a community leader (such as the governor
or emergency manager).
The NMS should identify and designate the appropriate
source of warnings and information from within the NMS
structure for different types of risks occurring at different
locations. For local hazards, the community is more likely to
trust information from someone with a local knowledge of
the area, rather than someone from a distant office who may
not be as sensitive to the local needs.
Effective communication about risks and warnings
requires knowledge about the recipients. In most countries,
the public is very diverse, with different backgrounds, expe-
riences, perceptions, circumstances and priorities. Any
attempts to communicate with the public must reflect this
diversity.
5.4 WARNING RESPONSE
Public education and awareness (section 4.4), stakeholder
involvement (section 5.1), warning presentation (section
5.2), and warning communication (section 5.3) all contribute
to an appropriate response to the warning.
The warning message by itself does not stimulate an
immediate response from individuals. Individuals receiving
the warning will first assess their own personal sense of risk.
The additional information required before they take action
depends on the content and clarity of the initial warning and
the credibility of the issuing organization. The potential
for individuals to respond appropriately is dramatically
increased if they are provided with information to assist them
in defining their personal risk and highlighting what life- or
property-saving actions to take.
Two excellent references on warning systems and public
response are Mileti and Sorenson (1990) and Mileti (1999).
Their work is summarized below.
Successful warning programmes strive to ensure that
every person or organization at risk:
(i) receives the warning;
(ii) understands the information presented;
(iii) believes the information;
(iv) personalizes the risk;
(v) makes correct decisions; and
(vi) responds in a timely manner.
An individual’s perception of risk is enhanced if:
(i) warning messages before and during a particular event
are issued and updated frequently;
(ii) warnings are delivered by multiple credible sources;
(iii) warning messages are consistent;
(iv) the basis for the warning is clear; and
(v) suggested response actions are included.
Preparedness prior to a hazardous event is critical to an
effective response to the warning. Individuals, families,
schools, businesses, communities and public facilities should
have an advance plan in place, so they know what to do and
where to go when a warning is received. The NMS and its risk
management partners have a key role to play in educating
their constituents in how to prepare a plan of action.
For a more thorough discussion of warning response, see
WMO document PWS-8, Guide on Improving Public
Understanding of and Response to Warnings.
5.5 MONITORING AND REVIEW
Verification and assessment of the warning services after a
severe weather event are essential to measure performance,
identify and correct deficiencies, and capture best practices,
which can be shared with other parts of the NMS or with
other risk management partners. In addition to quantitative
measurement, objective assessment is also valuable.
Interviews or surveys conducted with partners and stake-
holders can yield significant insight into how products and
services are received, interpreted, and what actions have been
taken as a result of the warning. This feedback can then be
used to make adjustments for similar future warning events.
Publishing verification scores and post-event assess-
ments can add to the credibility of the NMS and, for
stakeholders and partners, reinforce the perception of the
NMS as being user-oriented and dedicated to the cause.
6.1 EXAMPLES OF SUCCESSFUL RISK
MANAGEMENT INITIATIVES
The following sections describe initiatives by several NMSs
using risk management principles to improve their warning
systems. These examples provide practical applications of
some of the principles summarized in these guidelines, and
may trigger other NMSs to implement, expand upon, or
create their own applications within their organization or
country.
6.1.1 EXAMPLES FROM HONG KONG, CHINA
Hong Kong is a metropolis with over 6 million people and
frequented by tropical cyclones and heavy rain. Tropical
cyclones could bring gales or even hurricane force winds as
well as torrential rain to the city. It is not uncommon to regis-
ter more than 100 mm of rain a day in tropical cyclone or
monsoon trough situations, causing landslips and flooding.
The following examples illustrate how the Hong Kong
Observatory, the NMS in Hong Kong, China, mitigates the
impacts of these and other natural hazards.
Tropical Cyclone Warning Signal System as a Triggering
Mechanism for Protective Actions Against
Hazardous Weather
On average, six tropical cyclones affect Hong Kong each
year. To mitigate the impact of tropical cyclones, Hong Kong
operates a graded Tropical Cyclone Warning Signal System to
warn the public of the threat of winds associated with a
tropical cyclone. The System consists of five numbers repre-
senting increasing levels of wind strength. The Tropical
Cyclone Signal No. 1 is issued whenever a tropical cyclone is
within 800 km of Hong Kong and may affect the territory
later. Signals No. 3 and No. 8 warn the public of strong and
gale/storm force winds in the city respectively. Signal No. 9
signifies increasing gale or storm force winds, while No. 10
warns of hurricane force winds.
In view of the stringent building codes in Hong Kong,
home is generally considered the safest place for people to
take refuge from a tropical cyclone. When Signal No. 8 is
issued, the Hong Kong Observatory advises the public to stay
home or return home. Practically all activities in the city
move towards shutdown. All schools, government offices,
banks, the stock market, and courts are closed. Most public
transport services will start to cease operation. When Signal
No. 9 goes up, even the underground train stops. With Signal
No. 10, the city grinds to a complete halt to prepare for the
onslaught of a full-fledged typhoon.
However, the issuance of Signal No. 8 itself has the poten-
tial of causing chaos as millions of people try to go home all
at once. To facilitate an orderly shutdown, a special
announcement to the public (the Pre-8 Announcement)
about the impending No. 8 is issued two hours before the
actual issuance of the Signal. This enables transport opera-
tors to take measures to cope with the surge in demand for
public transport and to allow the public to go home in a safe
and orderly manner before the cyclone hits.
The tropical cyclone warning system has been in use for
many years and the public is already familiar with it. Together
with the well-coordinated response actions taken by relief
agencies, the system has proved very effective in reducing the
loss of life and property due to tropical cyclones.
Linking Rainstorm Signals to School Operation
Hong Kong is affected by severe local rainstorms, typi-
cally between April and September. The rainstorms can
develop explosively with very intense rainfall. The instanta-
neous rainfall rates can exceed 300 mm/hr. Heavy rain,
together with landslip and flooding, will lead to chaos, partic-
ularly during rush hours, in a densely populated city if the
situation is not properly managed.
Since 1992, the Hong Kong Observatory operates a
colour-coded rainstorm warning system which consists of 3
levels, namely: Amber, Red, and Black. The Amber signal is
issued to give alert on potential heavy rain which may develop
into Red/Black rainstorms. The Red and Black signals are
issued to warn the public about the occurrence of heavy rain
(50 and 70 mm/hr respectively) which can be hazardous and
may result in major disruptions.
By the time the Red signal situation is reached, road
conditions are considered unsuitable for students to
commute between home and school. The issuance of the Red
signal will trigger a series of response measures in relation to
school operation. Prior instructions are given to school
administrators, school-bus operators and parents on the
actions to take. Students are advised to stay home if they have
not left for school. For those on the way to or already at
school, the schools will be open and have sufficient staff to
take care of them until conditions are safe for them to return
home. Classes already in session will not be affected by the
issuance of the signals and students will only be released
when the threat recedes. Forecasters will liaise closely with the
education authority when heavy rain is expected and the Red
signal is imminent. Such close coordination has been in
operation for over ten years and there is practically no death
nor injury of students as a result of such inclement weather
conditions.
Chapter 6
SUCCESSFUL INITIATIVES OF NMSs
Very Hot and Cold Weather Warnings and Temporary Relief
Centres for the Needy
In the subtropical climate of Hong Kong, extreme
temperatures are rare, seldom exceeding 36°C and never
falling below 0°C. Nonetheless, social consequences and
health impacts can still be felt whenever temperatures rise
above or drop below certain "comfort" levels during heat
waves or prolonged cold spells that may affect Hong Kong
several times each year. The elderly and patients with chronic
illnesses are particularly at risk. Hypothermia and heat stroke
could lead to death for people who are over-exposed to the
elements or engaged in outdoor activities.
To alert the public of such risks, the Hong Kong
Observatory operates the Very Hot and Cold Weather
Warnings. Taking into account the combined effects of wind
and humidity, the reference urban temperature criteria for
operating the warnings are usually around 33°C and above
for the Very Hot Weather Warning, or around 12°C and below
for the Cold Weather Warning. The warnings are broadcast
on radio and TV, with advisories on actions to take. Upon the
issuance of such warnings, temporary relief shelters operated
by the Home Affairs Department, where accommodation is
provided with air conditioning in very hot weather and blan-
kets as well as hot food in cold weather, will be opened to help
the needy through difficult times.
6.1.2 The French Vigilance System After the Heat Wave
of August 2003
Following the exceptional heat wave that struck several coun-
tries in Western Europe, particularly France, during the first
fortnight of August 2003, a thorough revision of the risk
management plan for such situations was made.
Apart from various socio-economic effects, the over-
whelming impact was the increase of mortality among elderly
people, either living in pensioners institutions that were not
equipped with adequate resources to deal with such unusual
weather conditions, or in isolation in the big cities, suffering
from over-heating and deadly dehydration and were detected
too late by neighbours, medical or special rescue services.
During that fortnight of August 2003, the number of deaths
attributable to the heat wave was numbered by the [French]
National Statistics Institute to be almost 15,000 people.
One of the lessons learned (Lalande, 2003) was that the
weather warnings, though scientifically correct and timely,
were not assigned enough weight in the risk management
plan and other decision making processes, and were not given
sufficient exposure by the media coverage at the time.
In France, a combined risk management tool, named the
"Vigilance System", is jointly managed by the Civil Security
Authorities and Météo France, and directed towards local
authorities, the media, and the public at large (through two
Internet sites). Details of this system are described in a WMO
publication (WMO, 2003). The Vigilance System was initially
built for short-range warnings (with 24-hour lead time,
updated twice a day) and a limited set of phenomena (heavy
winds and rains, snow falls/avalanches and thunderstorms).
Other hazards have also been considered, including cold or
heat waves, but their characteristics in lead-time and societal
impacts were thought at the time to be too different from
those of other considered hazards, and as such the related
information communication was left to another kind of
procedure in the form of press releases.
This conclusion was clearly reversed after the event in
2003, and the partners in the Vigilance System decided to
add new indicators and adopt new procedures to take account
of this kind of menace. The basic view was that the potential
of the Vigilance System, already a corner stone in the "risk
culture" of the French people, would be enhanced rather than
blurred with the introduction of new concepts and proce-
dures.
Accordingly, Météo France in collaboration with
researchers from the "Institut de Veille Sanitaire" (Institute for
Sanitary Watch) expedited a study so that relevant indicators
were soon integrated into the popular four-color mechanisms
of the Vigilance System. The new procedures take into
account heat or cold waves forecast over three days of the
onset, and the persistence of maximum and minimum
temperatures above relevant thresholds in different cities
studied where the statistics of "excess number of deaths", or
a proxy of it, were correlated with the observed meteorolog-
ical conditions. They were tested during the summer of 2004;
and are now part of the overall Risk Management Plan, with
a comprehensive set of downstream actions organized by
Civil Security authorities, rescue services and health institu-
tions under the aegis of the "Prefects" (governors).
6.1.3 Examples from United States of America: All
Hazards Emergency Message Collection System
In the United States, the National Oceanic and Atmospheric
Administration (NOAA) is assigned responsibility by the
Department of Homeland Security’s (DHS) Federal
Response Plan to "provide public dissemination of critical pre
and post event information on the all hazards NOAA Weather
Radio (NWR) system, the NOAA Weather Wire Service, and
the Emergency Managers Weather Information Network
(EMWIN)"(Federal Response Plan, 2003). In 2002, in
support of the nation’s homeland security efforts, NOAA’s
National Weather Service (NWS) submitted a Fiscal Year 2004
Federal Budget funding initiative to streamline the creation,
authentication, and collection of all types of non-weather
emergency messages in a quick and secure fashion for subse-
quent alert, warning, and notification purposes. NWS
received specific funding to develop the All Hazards
Emergency Message Collection System (HazCollect) in its
Fiscal Year 2004 (FY04) budget.
HazCollect will improve NWS’s support to DHS’s
Federal Response Plan by integrating the automated collec-
tion and acceptance of non-weather related emergency
messages into the NWS Information Technology (IT) infra-
structure. NWS will implement a new, centralized point of
collection for non-weather related emergency messages
broadcast over NWS dissemination systems. A critical
component of NOAA’s Homeland Security Initiative,
HazCollect will reduce the time to disseminate non-weather
emergency messages from 7 to 2 minutes.
19 Chapter 6 — Successful initiatives NMSS
HazCollect will replace the existing system, which is a
manual, error-prone process. Currently, the non-weather
related emergency message is received at the local Weather
Forecast Office (WFO) by telephone or fax from local and/or
state government agencies. The author of the information is
manually authenticated and authorized by the WFO by vali-
dating a verbal password and/or checking the author or
agency name against an approved source and scope list. The
message is evaluated for reasonableness, manually typed by
WFO staff into their workstation, and then disseminated
using the existing NWS dissemination systems, including
NOAAPORT, NOAA Weather Wire Service, Family of
Services, Emergency Managers Weather Information
Network, and NOAA Weather Radio.
The current system has grave shortfalls. The manual
security and message composition process is time-consum-
ing and error-prone. Even in the best circumstances during
emergencies, when time is most critical and seconds can be
the difference between life and death, manual intervention by
the WFO staff adds minutes to the interval between the time
the emergency manager desires to warn the public and the
time the public actually receives the message. Additionally,
the manual text entry process is prone to typographical and
grammatical errors when transcribing the emergency
manager’s input and composing a message for transmission.
HazCollect will meet emergency managers’ requirement
for a fast, reliable way to inject non-weather related emer-
gency messages into the United States’ Emergency Alert
System (EAS), which is operated on commercial and public
radio and television stations nationwide; until now, no single
technical solution has been federally mandated or locally
selected to do this. Each state is currently free to choose any
system it prefers. When implemented, HazCollect will be
able to import emergency managers’ critical non-weather
emergency messages into the NWS dissemination infra-
structure and ensure efficient distribution to other national
systems and to EAS for re-broadcast. NWR is the NWS’ entry
point into EAS, and DHS and the Federal Emergency
Management Agency (FEMA) recognize NWR as a critical
component of the U.S. Warning Dissemination
Infrastructure.
NWS expects to deploy HazCollect in the summer and
fall of 2005. HazCollect will provide an information tech-
nology interface between state and local computer systems
and the NWS communications and dissemination infra-
structure utilizing FEMA’s Disaster Management
Interoperability Services. Through agreements with local
and state governments and at the request of government offi-
cials, NOAA’s NWS accepts emergency messages of
non-weather related nature, such as chemical spills/releases,
abducted child alerts, and radiological events, and
informs/warns the public of these events through the various
existing NWS dissemination systems.
6.1.4 Examples from Australia
Standard Emergency Warning Signal
In 1999, Australia’s lead disaster management authorities
reached agreement on the need for a Standard Emergency
Warning Signal (SEWS) to be used in assisting the delivery of
public warnings and messages for major emergency events.
The SEWS is a wailing siren sound used in northern Australia
for many years to attract attention to cyclone warnings where
a destructive impact was expected. The SEWS is intended for
use as an alert signal to be played on public media to draw
listeners’ attention to an emergency warning immediately
following.
It is considered vital that the status and effectiveness of
the SEWS are preserved as a "disaster mitigation" mecha-
nism by ensuring that it is used for serious events only. As a
general rule, the following 4 factors should be present:
(i) potential for loss of life and/or a major threat to a signif-
icant number of properties or the environment; usually
the threat/impact would be the lead item in local news
bulletins;
(ii) a significant number of people need to be warned;
(iii) impact is expected within 12 hours – or is occurring at
the time; and
(iv) one or more phenomena are classified as destructive.
When the SEWS is to be used, the following advice to the
media is included in the warning header: "Transmitters serv-
ing the area (defined) are requested to use the Standard
Emergency Warning Signal before broadcasting this
message". The media is requested to isolate the use of the
SEWS to the threatened defined area to reduce the risk of
unnecessarily alarming others. In regard to severe weather
and flood warnings, the use of the SEWS is authorized by the
State/Territory Director of the Australian Bureau of
Meteorology. Below are a number of severe weather and
flood examples where the use of the SEWS could be autho-
rized:
(i) wind gusts > 125 km/h;
(ii) storm tide > 0.5 m above (effectively) high water mark;
(iii) large hail > 4 cm in diameter (i.e. > golf-ball size);
(iv) tornado(s);
(v) major flood in a river or creek system;
(vi) intense rainfall leading to flash floods and/or landslides;
and
(vii)major urban and rural fires.
The information above is largely extracted from a multi-
authored booklet on the SEWS, reprinted in Queensland in
2004, and available online at:http://www.disaster.qld.gov.au/disasters/warning.asp
Disaster Mitigation Initiative After the Sydney to Hobart
Yacht Race 1998
Of the 115 yachts that set sail on Boxing Day 1998 in the
Sydney to Hobart Yacht Race, only 44 reached their destina-
tion. The destruction caused by a storm encountered by the
fleet triggered a massive search and rescue operation involv-
ing numerous personnel from the defence forces and disaster
management authorities. Even so, it resulted in the abandon-
ment of several yachts and the death of 6 people. It was the
20 Guidelines on integrating severe weather warnings into disaster risk management
most disastrous event in the 54-year history of the yachting
classic.
While the issuance of warnings on weather and sea
conditions independently assessed at the subsequent
Coroner’s enquiry were considered to be "excellent forecast
by world standards", the investigating Coroner did include
one recommendation concerning the terminology used in
weather forecasts and warnings to be specifically provided for
yacht-racing fleets. Since many yacht crews did not appreci-
ate how forecasts and warnings are to be interpreted despite
the various Bureau publications handed out to them prior to
the race, the recommendation called for the inclusion in fore-
casts and warnings of additional information on maximum
wind gusts and maximum wave heights likely to be encoun-
tered by yacht-racing fleets.
Consequently, the Australian Bureau of Meteorology
started to include the following safety preamble in the head-
ers of all marine forecasts and warnings: "Please be aware –
Wind gusts can be a further 40 percent stronger than the
averages given here, and maximum waves may be up to twice
the height". On the basis of surveys, this "disaster mitigation"
initiative has proven popular with mariners and the practice
continues today.
The information above is extracted mainly from the
Bureau’s "Preliminary Report on Meteorological Aspects of
the 1998 Sydney to Hobart Yacht Race" (Bureau of
Meteorology, 1999).
6.2 EXAMPLES OF PROJECTS FUNDED FROM
INNOVATIVE SOURCES
The following sections describe initiatives by several NMSs
to improve their warning system with funding and coopera-
tion from partners or organizations outside the usual funding
mechanisms. These examples may inspire other NMSs to
find creative and innovative sources for funding improve-
ments within their organization or country.
6.2.1 World Bank Loan to the Algerian Government to
Improve Flood Warnings, Infrastructure
Improvements, and Enhance the Relationship
with Disaster Response Agencies
The Algerian meteorological service, Office National de la
Météorologie (ONM), has been successful in securing fund-
ing through their national government for a World Bank
loan. This followed a devastating flood in 2001 in which
more than 700 people drowned and 24,000 were left home-
less. The funding will pay for, among other developments, a
meteorological component which will help enhance ONM's
capability to forecast such events and to make recommenda-
tions for infrastructure improvements - both technical and
personnel. This meteorological component is part of a
broader project, which will also help to improve the liaison
with other disaster response agencies to improve future
response to severe weather warnings.
6.2.2 Grant from the European Development Fund to
the Caribbean Forum of Africa, Caribbean and
Pacific States to Reduce Vulnerability to Floods
and Hurricanes
The Caribbean Forum of Africa, Caribbean and Pacific States
(CARIFORUM) has been awarded a grant from the
European Development Fund through the Caribbean
Regional Indicative Programme (CRIP). This grant is to fund
a project to reduce the vulnerability of the Caribbean region
in relation to adverse weather - primarily floods and hurri-
canes. Four new weather radars will be added to the existing
network of five. The composite images and data will be made
available to the participating meteorological services.
6.2.3 Central America Flash Flood Warning System
Funded by the United States Agency for
International Development’s Office of Foreign
Disaster Assistance
Following the catastrophic flooding of Hurricane Mitch in
1998 in Central America, the United States Agency for
International Development (USAID) provided funding for
the re-construction of damaged infrastructure. The National
Oceanic and Atmospheric Administration (NOAA) National
Weather Service (NWS) provided technology transfer, train-
ing, and technical assistance to the meteorological and
hydrologic services in the countries hardest hit (Honduras,
Nicaragua, El Salvador, and Guatemala). The USAID/Office
of Foreign Disaster Assistance (OFDA) also initiated a project
in 2000 to have NWS develop and implement a Central
America Flash Flood Guidance (CAFFG) system. This proto-
type system was developed with the Hydrologic Research
Centre, a public-benefit non-profit research, technology
transfer, and training corporation in San Diego, California,
United States. The purpose of this system is to provide oper-
ational meteorological and hydrological services in the seven
Central American countries (Belize, Costa Rica, El Salvador,
Guatemala, Honduras, Nicaragua, and Panama) with guid-
ance to provide flash flood warnings for small river basins.
This system became operational in August 2004.
The flash flood warning system uses real-time
Geostationary Operational Environmental Satellite (GOES)
imagery and in situ rain gauge data captured by a Digital
Direct Readout Ground Station (DRGS) downlink located in
Costa Rica. Hourly rainfall estimates are calculated on a 4-
km grid. Flash Flood guidance, which is the rainfall required
to produce flash flooding, is calculated daily for river basins
from 100 sq km to 300 sq km. A distributed hydrologic
model is run hourly to simulate flows and soil moisture for
the region. Graphical and text rainfall, soil moisture, and
flash flood guidance products are created and posted to the
Internet for access by the NMSs and disaster preparedness
response agencies in the seven Central American countries.
The next steps in this project are to evaluate the effectiveness
of this new approach and to continue with the training initia-
tives to ensure sustainability of the technology.
21 Chapter 6 — Successful initiatives NMSS
6.2.4 Examples from Hong Kong, China
The basic source of funding for meteorological service warn-
ing projects in Hong Kong is from the Hong Kong Special
Administrative Region Government (HKSARG). A few
recent examples in which projects are funded by other
sources are given below:
(i) Partnership with Higher Education Institutions on
Research Projects
The Hong Kong Observatory recently collaborated with
the Hong Kong Polytechnic University in the development of
real-time estimation of the integrated precipitable water
vapour (PWV) content in the atmosphere based on Global
Positioning Satellite (GPS) measurements to support the
operation of the rainstorm warning system in Hong Kong.
The project attracted funding support from the Hong Kong
Research Grants Council. The project started in September
2000 and the first phase was successfully completed in
December 2003.
Funding support has also been obtained from the
University Grants Council for partnership with the Hong
Kong University to jointly develop automatic identification
and tracking of tropical cyclone centres using radar imagery
in support of the Tropical Cyclone Warning Service. The
project has yielded fruitful results.
(ii) Cooperation with Engineering Departments and
Emergency Response Agencies to Promote Public
Awareness of Natural Disasters
Improvements in the drainage and slope safety systems,
as well as stringent building codes, have in recent years
successfully reduced the loss of life and property due to severe
weather. However, the awareness of potential hazards among
the general public has fallen.
To counter this decline in awareness, a large-scale public
education programme on natural disaster preparedness and
prevention was organized jointly by the Hong Kong
Observatory, engineering departments and other emergency
response agencies in Hong Kong. The programme aims to
promote public awareness of weather-related disasters and to
raise the community’s preparedness to deal with natural
disasters in Hong Kong.
The year-long "Safer Living – Reducing Natural
Disasters" public education programme commenced in
March 2005, prior to the rain and tropical cyclone season.
Activities included a tropical cyclone naming contest, exhi-
bitions, public seminars, open days, emergency operation
drills, carnivals, feature stories on TV and newspapers, etc.
Resources of the participating organizations were pooled
together to support this programme, without which such a
large-scale endeavour would not be possible.
6.2.5 Examples from France: Building a
Comprehensive Mainland Hydro-meteorological
Radar Network
One of the most prominent and recurring hazards in France
is flooding. Heavy tolls in lives and property were recorded,
for example in Nîmes (1987) and Vaison-la-Romaine (1992).
More recently, severe disasters (fortunately with a lower loss
in lives) repeatedly struck the south of France in 2000 (the
Aude département), 2002 (the Gard département) and 2003
(the Rhone Delta).
Recognizing the importance of a focused and integrated
approach to this kind of hydro-meteorological risk, Météo
France and the relevant ministry in charge of hydrology (now
the Ministère de l’Ecologie et du Développement Durable, or
the Ministry of Ecology) have established two five-year
programmes to install a comprehensive radar network,
thereby enhancing the existing network assembled by Météo
France over the years.
The first programme, completed in 2001, set up five new
radars in the most vulnerable part of France, the so-called
"Mediterranean arch". This resulted in almost all of the radar
coverage gaps in the region being filled and increased the
total number of radars to 18. The cost of this programme was
co-financed and shared by the two institutions involved, with
marginal extra funding coming from the European Union
(because of the international relevance of the coverage by
some of the radars).
As a follow-up of this first fruitful partnership, a new
programme (called PANTHERE) was launched for 2001-
2006 with new installations totalling 12.2 million Euros
(almost US$15 million). The main part of the programme
will be another five new radars, completing the coverage both
in the southern and northern parts of the country. The
programme also includes components for research and
development in new technologies.
For implementation, the project is undertaken by a team
of Météo France scientists and managers, under the control
of a bi-partite board of the two main partners. Regional
branches of the Ministry of Ecology and of Météo France
are actively required to help the core team in gathering local
users’ requirements, soliciting additional funding, looking
for adequate locations for the radars; assisting in tendering
processes, and supervising the building of necessary infra-
structure. This mechanism is very effective: bulk tendering
for the equipment lowers the unit price; centralized manage-
ment by a highly qualified and experienced team ensures
good coordination and progress; while the local initiatives
generate interest from new partners. It has been so success-
ful that funds collected from local authorities, the European
Union, and a Belgian Ministry now account for one-third of
the total funding. As a result, a sixth radar can also be accom-
modated within the time frame of the present PANTHERE
programme, which is on schedule and due to be completed
in 2006.
6.2.6 Examples from Australia: Studies on Climate
Change and Tropical Cyclone Vulnerability
The Queensland east coast in Australia is threatened by one
or two tropical cyclones most seasons. Tropical cyclones typi-
cally form in the Coral Sea and cause on landfall severe
winds, storm surge, and extreme wave conditions in the
impact zone. In 1999, an opportunity arose to further study
the risks posed by these phenomena when the Queensland
State Government provided A$1M funding under the
"Greenhouse Special Treasury Initiative" through the
22 Guidelines on integrating severe weather warnings into disaster risk management
23 Chapter 6 — Successful initiatives NMSS
Department of Natural Resources and Mines. The Australian
Bureau of Meteorology in partnership with the Queensland
Department of Emergency Services was successful in bidding
for project funding (totalling A$265K). The study was titled
"Queensland Climate Change and Community Vulnerability
to Tropical Cyclones", with coastal engineers from the
Queensland Environmental Protection Agency subsequently
joining the project team.
When the study was partially completed, the Project
Management Team led by the Australian Bureau of
Meteorology assessed that there were insufficient funds to
complete the study. On this occasion, a successful bid for
supplementary funding (totalling A$80K) was made under
the Federal/State Natural "Disasters Risk Management
Studies Program". More recently, another successful bid
(totalling A$105K) was made under the same program to
undertake a similar study in the Gulf of Carpentaria where
the coast is also cyclone prone. Without the funding provided
under the "Greenhouse Special Treasury Initiative" and the
"Disaster Risk Management Studies Program", a "disaster
mitigation" study of this magnitude and complexity would
not have been possible.
Incidentally, the project has already won three presti-
gious awards – an Engineering Excellence Award in 2002 and
both Queensland and Australian Safer Communities Awards
in 2004. Both awards were shared with the project consul-
tants/contractors – Systems Engineering Australia Pty Ltd
and James Cook University (Marine Modelling Unit and the
Cyclone Testing Station).
I. REFERENCES AND USEFUL READINGS
Browning, K.A., Chalon, J-P, and Thorpe, A.J., editors, 1999.
Fronts and Atlantic Storm-Track Experiment, Quarterly
Journal of the Royal Meteorological Society, Vol. 125,
part C, No. 561, October 1999.
Bureau of Meteorology, 1999. Preliminary Report on
Meteorological Aspects of the 1998 Sydney to Hobart
Yacht Race, February 1999. (Available on the Internet:
www.bom.gov.au/inside/services_policy/marine/sydney
_hobart/contents.shtml)
Colorado State University News, 2004: "Hurricane season
review", Colorado State University. (Available on the
Internet:http://comment.colostate.edu/index.asp?page=display_
article&article_id=64147507).
Conseil Supérieur de la Météorologie, 2002. Colloquium on
"Climate and Health", March 2002.
Counter Disaster and Rescue Services, 2000. Disaster Risk
Management Guide: A How-to Manual for Local
Government, Queensland Department of Emergency
Services, January 2000.
Executive Office of the President of the USA, 2003. Reducing
Disaster Vulnerability Through Science and Technology -
National Science and Technology Council, Committee
on the Environment and Natural Resources, An interim
report of the Subcommittee on Disaster Reduction, July
2003. (Available on the Internet:http://www.ostp.gov/NSTC/html/SDR_Report_Reduci
ngDisasterVulnerability2003.pdf)
Federal Response Plan 2003, 9230.1-PL, January 2003.
[United States]
Lalande, Dr Françoise (under the direction of), 2003, Mission
d’expertise et d’évaluation du système de santé pendant
la canicule 2003 (Expert mission for the evaluation of the
health system during the hot spell of 2003), published by
la Documentation Française. (Available on the Internet:http://ladocumentationfrançaise.fr/BRP/034000558/000
0.pdf.)
Glantz, Michel H., 2003. Usable Science 8: Early Warning
Systems: Do’s and Don’t’s - Report of the workshop held
in Shanghai, China; October 16, 2003. (Available on the
Internet:http://www.esig.ucar.edu/warning/report.pdf).
Grazzini, F., L. Ferranti, F. Lalaurette and F.Vitard., 2003. The
exceptional warm anomalies of summer 2003. ECMWF
Newsletter no. 99- autumn/winter 2003. (Available on
the Internet:http://www.ecmwf.int).
Gunasekera, Don, 2004. Natural Disaster Mitigation: Role and
Value of Warnings, Outlook 2004 Speaker Papers,
Disaster Management Workshop session, Canberra,
Australia.
MeteoWorld, 2004: "2004: the year of the tropical cyclone",
World Meteorological Organization. (Available on the
Internet:http://www.wmo.int/meteoworld/archive/en/dec2004/t
ropicalcyclone.htm)
Mileti, Dennis .S, 1999. Disasters by Design - A Reassessment
of Natural Hazards in the United States, National
Academy of Sciences, Joseph Henry Press, 2101
Constitution Avenue, N.W., Washington, D.C., 20418,
USA.
Mileti, Dennis S. and John H. Sorenson, 1990.
Communication of Emergency Public Warnings – A Social
Science Perspective and State-of-the-Art Assessment,
Oak Ridge National Laboratory, ORNL-6609, Oak
Ridge, Tennessee, USA. (Available on the Internet:http://emc.ornl.gov/EMC/PDF/CommunicationFinal.p
df).
National Hurricane Centre/Tropical Prediction Centre, 2004:
"2004 Atlantic Hurricane Season", National Weather
Service, NOAA. (Available on the Internet:http://www.nhc.noaa.gov/2004atlan.shtml)
Proceedings of the Forum on Risk Management and
Assessments of Natural Hazards, 5-6 February 2001,
Washington, D.C., sponsored by the [United States]
Office of the Federal Coordinator for Meteorological
Services and Supporting Research, 2001. (Available on
the Internet:http://www.ofcm.noaa.gov/risk/proceedings/pdf/riskpr
oceedingsentire.pdf).
Rogers, D.P., 2005: Turning Crisis Management into Risk
Management – The Role of Weather Forecasting. In
proceedings of the seminar on "Safer Living – Reducing
Natural Disasters", Hong Kong, China, 2005.
Shapiro, M. and A. Thorpe, THORPEX International Science
Plan Version 3. WMO/TD No. 1246, WWRP/THOR-
PEX No. 2, 51pp.
Sorenson, John H., 2000. Hazard Warning Systems: Review
of 20 Years of Progress, Natural Hazards Review, Vol. 1,
No. 2, 2000.
World Meteorological Organization (WMO), 2002. Guide on
Improving Public Understanding of and Response to
Warnings, PWS-8, WMO/TD No. 1139, Geneva,
Switzerland.
WMO, 2003. Guidelines on Cross-Border Exchange of
Warnings, PWS-9, WMO/TD No. 1179, Geneva,
Switzerland.
WMO, 2004. Natural Disaster Prevention and Mitigation: Role
and Contribution of the WMO and NMHSs, WMO
discussion paper, 13 September, 2004.
Wilhite, Donald, 2005: Moving from Crisis Response to
Drought Risk Management: Shifting the Pradigm.
National Drought Mitigation Centre, International
Chapter 7
REFERENCES, FURTHER READING
AND USEFUL WEB SITES
Drought Information Centre, University of Nebraska,
Lincoln, Nebraska, USA.
Zillman, John W., 1999. The National Meteorological Service.
World Meteorological Organization Bulletin,Vol. 48, No.
2, pp. 129-159, 1999.
Zillman, John W., 2004. Social and Economic Impacts of
Weather-Related Disaster Management, Symposium on
Planning and Preparedness for Weather-Related
Disasters, 29-30 March 2004, Hong Kong, China.
II. USEFUL WEBSITES
Billion Dollar U.S. Weather Disasters, 1980 - 2003http://www.ncdc.noaa.gov/oa/reports/billionz.html
Central America Flash Flood Guidance (CAFFG) system
(available in English or Spanish):http://www.hrc-web.org/Whats_New/
Early Warning Systemshttp://www.esig.ucar.edu/warning/
HAZUS.org
www.hazus.org
NOAA Economic Statisticshttp://www.publicaffairs.noaa.gov/pdf/economic-statis-
tics2004.pdf
Partnership for Public Warninghttp://www.partnershipforpublicwarning.org
[United States] Federal Response Plan, 9230.1-PL, January
2003. Available online at:http://www.fema.gov/pdf/rrr/frp/frp2003.pdf
NOTE: This document will be absorbed into a more
comprehensive plan named the National Response Plan
[United States] Natural Hazards Centrehttp://www.colorado.edu/hazards/
25 Chapter 7 — References, further reading and useful web sites
The ratio La/Cp can provide a decision threshold for action.
Suppose that, on each occasion, loss avoided (La) is $10,000 and the cost of protection (Cp) is $4,000; then Cp/La = 0.4.
Consider the financial implications over 10 events for 5 different forecast probabilities. If action is taken on each of the 10 events,
then the total cost of protection (?Cp) remains the same at $40,000, but the total loss avoided (?La) increases:
So there is a break-even point (i.e. BCR = 1) when probability is 0.4 or P ? Cp/La. This becomes the decision point for
taking action. Emergency authorities can then put in place a whole range of options and strategies depending on the likely
impact of an event. These will start with relatively low-cost preparatory actions such as ensuring equipment and manpower
are available if needed and raising the state of awareness amongst responders. These may be initiated even with a low proba-
bility of occurrence because the BCR will still be high. Similarly, when confidence (probability) is high, more expensive
strategies can be brought into play.
Forecast
Probability
P
0.1
0.2
0.3
0.4
0.5
Number of
occasions
loss avoided
1
2
3
4
5
Total Loss
Avoided
?La ($)
10,000
20,000
30,000
40,000
50,000
Total cost of
Protection
?Cp ($)
40,000
40,000
40,000
40,000
40,000
BCR over
10 events
0.25
0.50
0.75
1.00
1.25
Appendix A
UNDERSTANDING PROBABILITIES OF OCCURRENCE AND
CALCULATION OF BENEFIT COST RATIO (BCR): AN EXAMPLE
doc_941774093.pdf
During the ten-year period of 1992-2001, weather-related disasters worldwide killed over 622,000 and affected over 2 billion people (WMO, 2004). In 2004, Japan saw a record year for typhoons,with ten tropical cyclones making landfall, two within ten days.Tokage was the most powerful typhoon to hit Japan in 16 years.
W o r l d M e t e o r o l o g i c a l O r g a n i z a t i o n
GUIDELINES ON INTEGRATING
SEVERE WEATHER WARNINGS INTO
DISASTER RISK MANAGEMENT
WMO/TD-No. 1292 PWS-13
The designations employed and the presentation of material in this
publication do not imply the expression of any opinion whatsoever on
the part of the Secretariat of the World Meteorological Organization
concerning the legal status of any country, territory, city or area, or of
its authorities, or concerning the delimitation of its frontiers or bound-
aries.
It should be noted that this Report is not an official WMO
Publication and has not been subjected to the Organization’s standard
editorial procedures. The views expressed by individuals or group of
experts and published in a WMO Technical Document, do not neces-
sarily have the endorsement of the Organization.
© 2005, World Meteorological Organization
WMO/TD No. 1292
NOTE
Co-authors: Jim Davidson and M.C. Wong
(Contributions by: C.Y. Lam, M.C. Wong, C. Alex, S. Wass, and C. Dupuy)
Edited by: Haleh Kootval
Cover design by: F. Decollogny
Figure of Risk Management vs Crisis Management
Courtesy of: Donald Wilhite (2000)
Page
CHAPTER 1. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
CHAPTER 2. EXPLANATION OF KEY CONCEPTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
CHAPTER 3. RISK MANAGEMENT PHILOSOPHY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
CHAPTER 4. NMSs’ RESPONSIBILITIES WITHIN A RISK MANAGEMENT FRAMEWORK . . . . . . . . . 7
4.1 CORPORATE CULTURE: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
4.2 CAPACITY BUILDING: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
4.3 DEVELOPING PARTNERSHIPS: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
4.4 PUBLIC EDUCATION & AWARENESS: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
4.5 ESTABLISHING A KNOWLEDGE BASE: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
4.5.1 Risk Databases: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
4.5.2 Risk Assessments: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
4.5.3 Cost-Benefit Analyses: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
4.5.4 Research Activities: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
4.6 DISASTER RISK MANAGEMENT EXPERTISE: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
4.7 FORMULATING A RISK MANAGEMENT ACTION PLAN: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
4.7.1 Step 1: Communicate & Consult with Stakeholders: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
4.7.2 Step 2: Establish the Context: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
4.7.3 Step 3: Identify & Quantify the Risks: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
4.7.4 Step 4: Analyse and Evaluate the Risks: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
4.7.5 Step 5: Treat the Avoidable and Manageable Risks: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4.7.6 Step 6: Monitor and Review the Risk Management Action Plan: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4.8 SECURING FUNDING: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
CHAPTER 5. EFFECTIVE EARLY WARNINGS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
5.1 STAKEHOLDER INVOLVEMENT: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
5.2 WARNING PRESENTATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
5.3 WARNING COMMUNICATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
5.4 WARNING RESPONSE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
5.5 MONITORING AND REVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
CHAPTER 6. SUCCESSFUL INITIATIVES OF NMSs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
6.1 EXAMPLES OF SUCCESSFUL RISK MANAGEMENT INITIATIVES: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
6.1.1 Examples from Hong Kong, China . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
6.1.2 The French Vigilance System After the Heat Wave of August 2003 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
6.1.3 Examples from United States of America:
All Hazards Emergency Message Collection System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
6.1.4 Examples from Australia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
6.2 EXAMPLES OF PROJECTS FUNDED FROM INNOVATIVE SOURCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
6.2.1 World Bank Loan to the Algerian Government to Improve
Flood Warnings, Infrastructure Improvements, and Enhance
the Relationship with Disaster Response Agencies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
6.2.2 Grant from the European Development Fund to
the Caribbean Forum of Africa, Caribbean and
Pacific States to Reduce Vulnerability to Floods and Hurricanes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
TABLE OF CONTENTS
6.2.3 Central America Flash Flood Warning System
Funded by the United States Agency for International
Development’s Office of Foreign Disaster Assistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
6.2.4 Examples from Hong Kong, China . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
6.2.5 Examples from France: Building a Comprehensive
Mainland Hydro-meteorological Radar Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
6.2.6 Examples from Australia: Studies on Climate Change
and Tropical Cyclone Vulnerability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
CHAPTER 7. REFERENCES, FURTHER READING AND USEFUL WEB SITES . . . . . . . . . . . . . . . . . . . . . . 24
I. REFERENCES AND USEFUL READINGS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
II. USEFUL WEBSITES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
APPENDIX A UNDERSTANDING PROBABILITIES OF OCCURRENCE AND
CALCULATION OF BENEFIT COST RATIO (BCR): AN EXAMPLE . . . . . . . . . . . . . . . . . . . 26
Contents iv
During the ten-year period of 1992-2001, weather-related
disasters worldwide killed over 622,000 and affected over 2
billion people (WMO, 2004). In 2004, Japan saw a record year
for typhoons, with ten tropical cyclones making landfall, two
within ten days. Tokage was the most powerful typhoon to hit
Japan in 16 years. The year’s number of tropical cyclones in
Japan surpassed the previous record of six set in 1990 and left
the largest number of people dead (some 220) and injured
since 1983 (MeteoWorld, 2004). A related disaster was the
corresponding lack of rainfall in the same season in China
resulting in widespread drought resulting in the migration of
birds carrying the avian flu virus to populated areas.
In the same year, United States was severely affected by
four hurricanes making landfall in Florida, with five other
hurricanes of different intensities making landfall elsewhere
along the US coastline (National Hurricane Centre/Tropical
Prediction Centre, 2004). The last time with such frequent
tropical cyclone activity was in 1916 when eight hurricanes
impacted the US coastline. Total damage in 2004 as a result
of tropical cyclone landfalls could approach 50 billion US
dollars (Colorado State University News, 2004). Charley
brought death and destruction to the Caribbean and Florida.
Damage to property in Cuba was in excess of 1 billion US
dollars. Preliminary estimates of Charley’s damage in Florida
ranged from 13 to 15 billion US dollars, making it the second
costliest hurricane in the US history. Then came Ivan, the
most powerful hurricane to hit the Caribbean in ten years.
Following a trail of havoc across Grenada, Jamaica and
Alabama, Ivan left in its wake more than 100 deaths and prop-
erty damage estimated at 12 million US dollars. Jeanne,
weaker but no less deadly, swept along the north coast of
Haiti on 16 September, killing more than 2,000 people and
destroying the country’s economy.
The South Pacific islands also had their share of weather-
related disasters. In the austral summer of 2002-2003, a
powerful tropical cyclone Zoe impacted a number of remote
island communities in the Solomon Islands and caused
significant economic loss. This was followed in 2003-2004 by
another intense tropical cyclone Heta that left a trail of utter
devastation through the small island nation of Niue. The
economic loss at Niue was so great that the continuing viabil-
ity of the country came into question, with survival possible
only through international aid. In early 2005, the Cook
Islands in the Southwest Pacific were affected by four intense
tropical cyclones (Meena, Nancy, Olaf and Percy) causing
significant damage being reported.
On the morning of Sunday, 26 December 2004 a magni-
tude 9.3 earthquake, the world's most severe in 40 years,
ripped apart the seafloor off the coast of northwest Sumatra.
It unleashed a devastating tsunami that swept thousands of
kilometres across the Indian Ocean, taking the lives of some
300,000 people in countries as far apart as Indonesia,
Thailand, the Maldives, Sri Lanka and Somalia, making it the
deadliest in recorded history. Those hardest hit were the
people living in low-lying coastal areas. Beyond the loss of
human lives, the tsunami also destroyed livelihoods, trau-
matized whole populations and severely damaged habitats.
The scale of the disaster shocked the international commu-
nity and the United Nations (UN) made a "Flash Appeal"
immediately after the event to raise funds to support relief
and recovery operations.While attention was understandably
drawn towards this tsunami and its aftermath in south Asia,
the impact caused by the perennial threat of global severe
weather should not be under-estimated.
The latest event at the time of writing these guidelines
was category 5 hurricane Katrina, which struck the Gulf coast
of the United States in August 2005 causing many deaths and
estimated economic losses in excess of US$100 billion, the
largest loss on record resulting from a single weather-related
event. Between 1998 and 2002, 45 weather-related disasters
in the United States caused damage with a total cost of
approximately US$200 billion.
The dramatic impact of natural disasters and the subse-
quent response activities often attract much international
interest.Attention is being increasingly focused upon natural
disasters inflicting tremendous economic losses (in addition
to human suffering and casualties) and the efforts expended
on the mitigation and reduction of such disasters. Disaster
mitigation is now a recognized international priority. WMO
cooperates within many other organizations and interna-
tional programmes, particularly the International Strategy
for Disaster Reduction (ISDR), in its efforts to improve
natural disaster prevention and mitigation. The UN Secretary
General, Mr Kofi Annan, recommended in his report to the
General Assembly on 21 March 2005 the establishment of a
global early warning system for all natural hazards.
Increasingly it is recognized that disasters are linked. The
impacts of many types of natural disasters do not happen in
isolation, but are sequential (Rogers 2005). For example, a
drought in one region followed by famine can be exacerbated
by dust storms transporting locusts spawned by heavy rains
in another location. Similarly, the diversion of moisture from
one region to another causes flooding in one region and
drought in another. The recognition of such cause and effect
on a global and regional scale is leading to the creation of
early warning systems that can accommodate multiple
hazards and cross-boundary impacts. At the same time,
governments are becoming aware that a paradigm shift from
crisis management to risk management is necessary if the
finite resources available are spent in the most efficient way
to assist the populations at risk to prevent or mitigate disas-
ters.
The Public Weather Services Programme was given the
mandate by the WMO Commission for Basic Systems to
produce disaster risk management guidelines for National
Meteorological Services and National Meteorological and
Hydrological Services (hereafter NMSs). These guidelines on
"the application of risk management principles in the provi-
Chapter 1
INTRODUCTION
Guidelines on integrating severe weather warnings into disaster risk management
sion of severe weather warnings, and the use of this approach
in efforts to secure extra funding from innovative sources" are
prepared with the central purpose of recommending a risk
management action plan suitable for adoption by NMSs.
They are not meant to be an exhaustive treatise on disaster
risk management. There is already a wealth of literature on
this subject and relevant documents are referenced here. The
intention is to promote the role of NMSs in disaster risk
management, to encourage their more proactive involvement,
and where appropriate lead in developing and implementing
a risk management action plan. Effective risk management is
often the difference between a natural hazard and an impact
disaster.
Risks can be characterized as being a complex combina-
tion of unavoidable, acceptable and treatable circumstances.
Having identified and evaluated risks, a decision must be
made as to what level of risk is acceptable, and at what level a
risk is to be treated. Increasingly, this is being discussed in
terms of consequences of the event and cost effectiveness of
the treatment. These guidelines relate to the various risks that
are considered treatable or manageable in the context of the
provision of severe weather warnings. They are prepared with
due reference to the outcomes of various Disaster reduction
conferences, workshops and seminars held around the world
in recent years, including:
(i) 1st World Conference on Natural Disaster Reduction
held in Yokohama, Japan in 1994;
(ii) 1st International Conference on Early Warning –
Potsdam, Germany – 1998;
(iii) Risk Management and Assessments of Natural Hazards
Forum – Washington, USA – 2001;
(iv) 2nd International Conference on Early Warning – Bonn,
Germany – 2003;
(v) Workshop on Early Warning Systems – Shanghai, China
– 2003;
(vi) 2nd World Conference on Natural Disaster Reduction,
Kobe, Japan 2005.
It is noted that the 3rd International Early Warning
Conference will be held in Bonn, Germany in March 2006.
2
Issuing weather forecasts and warnings is a core business
for NMSs. Most of the time, NMSs focus on the applica-
tion of science. In fact, one of the great achievements of
meteorological science and technology during the 20th
century has been our increased ability, through improved
warning and forecasting systems, to provide much more
reliable weather information for effective protection of
life and property from natural hazards. However, in an
information age with higher public expectation, more
intense media coverage as well as much closer connection
with the global community, NMSs can no longer afford to
play a passive role in disaster reduction and mitigation
effort. It is just as critical for NMSs to be actively
involved in disaster management programmes to ensure
that the users get the full benefit of reliable forecasts and
warnings.
It is important therefore not just how we forecast a
natural hazard but how we plan for and minimize the
impact of the hazard. In the context of the current guide-
lines, a natural hazard is a weather or flood-related
situation with potential to inflict loss or damage to the
community or environment. A natural disaster is a cata-
strophic event caused by a natural hazard that severely
disrupts the fabric of a community and usually requires
the intervention of government to return the community
to normality. While hazards may induce a crisis, they do
not necessarily lead to disasters. Though most natural
hazards may be inevitable, natural disasters are not
totally unavoidable. A disaster will depend on the char-
acteristics, probability and intensity of the hazard, as well
as the susceptibility of the exposed community based on
physical, social, economic and environmental conditions.
In some instances, natural disasters cannot be
prevented from occurring. However, their overall impact
can be significantly reduced through disaster mitigation.
Disaster mitigation is the process of managing the "risks"
associated with potential natural disasters so that loss
may be minimized or even eliminated. This includes the
disaster response which are actions taken in anticipa-
tion of, during and immediately after, a natural disaster
to ensure that its effects are minimized, and that people
affected are given immediate relief and support. A
systematic approach should be taken to manage the risk
of natural disasters. This process of disaster risk
management should consider the likely effects of natural
hazards and the measures by which they can be mini-
mized. A disaster risk management system should
incorporate response actions that are appropriate to the
social and economic conditions of the community under
threat. Because of this, disaster reduction has become
increasingly associated with practices that define efforts
to achieve sustainable developments.
The concept of disaster risk is used to describe the
likelihood of harmful consequences arising from the
interaction of natural hazards and the community. Two
elements are essential in the formulation of disaster risk:
the probability of occurrence of a hazard, and the vulner-
ability of the community to that hazard.
Risk = Hazard Probability x Vulnerability
A closer look at (a) the nature of hazards and (b) the
notions of vulnerability allows for a better and more
comprehensive understanding of the challenges posed
by disaster mitigation:
(i) Nature of hazard - By seeking to understand hazards of
the past, monitoring of the present, and prediction of
the future, a community or public authority is poised to
minimize the risk of a disaster. The NMSs play a key role
in this aspect of risk management of weather-related
natural disasters.
(ii) Notions of Vulnerability - The prevailing conditions
within any group of people in a society can determine
the extent of their susceptibility or resilience to loss or
damage from a natural hazard. The community vulner-
ability is the susceptibility and resilience of the
community and environment to natural hazards.
Different population segments can be exposed to greater
relative risks because of their socio-economic conditions
of vulnerability. Reducing disaster vulnerability requires
increasing knowledge about the likelihood, conse-
quences, imminence and presence of natural hazards,
and empowering individuals, communities and public
authorities with that knowledge to lower the risk before
severe weather events, and to respond effectively imme-
diately afterwards. Increasing this knowledge depends
on focusing science and technology investment to
improve disaster mitigation and resilience by identifying
and meeting needs and closing knowledge gaps wher-
ever possible.
As the decade progresses, a broader global aware-
ness of the social and economic consequences of natural
disasters has developed highlighting the increasing
importance of engaging a much broader community in
hazards awareness and risk management practices. The
importance given to socio-economic vulnerability as a
rapidly increasing factor of risk in most of today’s soci-
eties underlined the need to encourage the participation
of wide spectrum of stakeholders in hazard and risk
reduction activities. Stakeholders are those people or
organizations who may affect, be affected by, or perceive
themselves to be affected by, a decision or activity.
Stakeholders can include governments, delivery agencies,
private sector and the public. In developing a disaster
risk management system, no single agency can provide a
fully comprehensive solution. It is essential that agencies
Chapter 2
EXPLANATION OF KEY CONCEPTS
Guidelines on integrating severe weather warnings into disaster risk management
work together and with stakeholders to narrow knowl-
edge gaps and to develop disaster risk management plans
using a coordinated approach.
4
The devastating effects of natural disasters can quite
commonly be associated with inadequacies in a community’s
development strategy. Poorly planned development turns
recurring natural hazards into disasters. Housing a dense
population on a flood plain or permitting poor building
codes in cyclone-prone and earthquake-prone areas aggra-
vate not only the vulnerability of the exposed communities
but also increase the losses due to natural hazards. Wilhite
(2005) has taken the commonly accepted cycle of disaster
management and redefined it in terms of crisis management
and risk management. Crisis management emphasizes post
disaster impact assessment, response, recovery and recon-
struction; whereas risk management emphasizes protection
through mitigation, preparedness, prediction and early warn-
ing. Traditionally, the focus of disaster management has
almost been exclusively on actions taken immediately before,
during and shortly after a disaster. While many governments
nowadays still employ reactive crisis management, rather
than proactive risk management, approaches to disasters, the
tide appears to be turning.
There is no doubt that the role of relief assistance during
a crisis will remain important and need to be enhanced at all
levels. However, the question must be asked: "Can we afford
to value lives and properties only after they have been lost in
a disaster?" Much greater attention will need to be given to
preventive strategies that can contribute to saving lives and
protecting assets before they are lost. In fact, a paradigm shift
is occurring with a move away from reactive response and
recovery to a much more proactive and holistic concern with
preparedness and prevention. There is now an increased
emphasis placed on risk management rather than crisis
management, and an acceptance that natural disasters, devel-
opment problems and environmental issues are inextricably
linked. Instead of diverting resources away from ongoing
projects through budget re-allocation in order to finance
recovery and re-construction efforts, proactive mechanisms
are sought to reduce the economic costs and impacts of
hazards, improve the countries’ response capacity, decrease
vulnerability and enhance communities’ resilience to disas-
ters.
Responses to warnings on natural disasters/hazards
generally involve making decisions based on calculated risks
and uncertainty. Although safety assurance of life and prop-
erty is a common ideal underlying all warnings, we have to
accept that risks can never be totally eliminated. Thus risk
management generally involves the challenging task of mini-
mizing threats to life, property and the general environment,
yet at the same time without creating unnecessary disruption
and chaos to the communities likely to be affected by natural
disasters/hazards.
With this new philosophy of risk management, coun-
tries will need to develop policies that focus on the
importance of taking appropriate preparedness and preven-
tion measures. Through such measures, the overall impact of
natural hazards can be significantly reduced:
(i) improved scientific understanding of the range of
natural hazards that can occur and their potential to
occur at specific locations (hazard/risk mapping on a
firm scientific basis);
(ii) improved understanding of the cost implications of the
above hazards;
(iii) better design and construction of buildings, bridges and
other structures (including flood plain management) to
reduce the risk of damage during a severe weather or
flood event;
(iv) increased community awareness through public educa-
tion of the impacts of natural hazards and the measures
individuals can take to reduce damage and protect life;
(v) improved technology and modelling effort to extend the
lead time for warnings of potential natural hazards; and
(vi) integrated risk management approach, treated as a total
package instead of separate components.
Among the measures mentioned, NMSs would be
involved to a greater or lesser extent. But in all aspects of risk
management, effective communication with stakeholders,
especially the public, is vital.A community at risk to a natural
hazard will not be prepared to meet it if the general public is
not well informed. Communication of mitigation and
preparedness measures can transform a community’s readi-
ness to face an extreme event. Effective communication is
especially critical immediately before, during, and after a
hazard event to ensure the public has adequate warnings and
complete information to maximize public safety. Ultimately,
the messages must also be understood and acted upon by
those communities at the greatest risk, including the disad-
vantaged people.
As such, severe weather warning systems must be viewed
as an integral element of long-term strategies in the sustain-
able development for a safer and more adaptable community.
Such systems thus constitute an essential and highly cost-
effective component of national strategies and must be
properly embedded and integrated in national risk manage-
ment programmes. Building on this foundation, there is
inevitably a need to ensure that severe weather warnings of
natural hazards becomes an integral part of government
policy in every disaster-prone country, and it also follows
that NMSs need to be recognized as a major component of
the corresponding infrastructure in support of risk manage-
ment and disaster mitigation.
With continuing scientific and technological develop-
ments, there is a growing expectation that forecasters will be
able to warn at longer lead-times, and with more emphasis or
greater accuracy regarding the ‘where’,‘when’ and magnitude
of an impending natural hazard. While the advances and
successful applications of weather warning systems are most
encouraging, the common pitfall is to over-emphasize on the
Chapter 3
RISK MANAGEMENT PHILOSOPHY
technical details and under-state the practical implications for
the communities exposed to the hazard. In the communica-
tion process, not only should the communities be adequately
informed, but also they should be sufficiently impressed that
preparedness actions are taken before and during the antici-
pated hazardous event. Warning messages must convey to an
often-sceptical public that they are vulnerable, that the danger
is real, and that specific measures can be taken to protect
themselves and their property. Messages must also consider
the target audience, making provisions to reach high-risk
areas and to overcome language barriers in diverse popula-
tions.
In the spirit of an integrated risk management approach,
the multi-discipline and multi-sector nature of the warning
process must not be overlooked. Although based on scientific
and technology, warnings must be tailored to serve users'
needs, with due consideration given to their situations and
available resources. While severe weather warning capabili-
ties must continue to be strengthened at the global level, it is
just as important that enough emphasis should be given to
developing capacities that are relevant, and responsive to, the
needs of local communities. Fundamental to the process is
the existence of a Risk Management Action Plan, which is
the subject of discussion in the next chapter.
Guidelines on integrating severe weather warnings into disaster risk management 6
Disaster risk management consists of three broad activities:
mitigation, preparedness, and early warning. Mitigation
includes risk reduction; preparation includes risk assessment,
risk analysis, hazard identification, research and develop-
ment, public education and outreach; whereas early warning
consists of hazard forecasts, communication and dissemina-
tion. NMSs contribute to risk management in three ways:
The first is to provide early warning of weather, water and
climate hazards for operational decisions; the second is to
support risk and impact assessments to determine who and
what is at risk and why; and the third is to improve forecasts
and analyses to help reduce or remove risks (Rogers 2005).
The following sections provide a framework for NMSs to
consider in developing and implementing a Risk
Management Action Plan in partnership with other agencies
and organizations involved in risk management. This frame-
work focuses on the role of the NMS in developing and
executing the Risk Management Action Plan with regard to
severe weather warnings, but acknowledges the ultimate Risk
Management framework should also address non-weather
hazards and draw expertise from many other partners. Some
NMSs may have implemented some or most of these
elements in a Risk Management Action Plan; where that is
true, the framework may prompt these countries to increase
or enhance their efforts.
4.1 CORPORATE CULTURE
For meteorological hazards, the NMS is the sole warning
authority for its citizens. NMSs are the experts on the mete-
orological aspects of severe weather hazards, and are
therefore a critical player in the development of the country’s
risk management plans.
The NMS must be a credible authority for information
on severe weather warnings and has a reputation for accuracy,
reliability, and timeliness. It is also increasingly recognized
that NMSs need to develop a corporate culture of being
caring and people-centred, in addition to the more tradi-
tional culture of being professional and science-centred.
Developing working relationships with partners such as
emergency managers and the media (see section 4.3) and
involving stakeholders in the development and review of the
warning system (see section 5.1) is essential.
Many NMSs are now emphasizing a user-centred
approach to improving their products and services, in addi-
tion to the traditional science-oriented approach. Some
examples of a user-centred approach are:
(i) plain-language descriptions that are clearly understood
by the public in weather bulletins and outreach material
about the hazards and warnings;
(ii) making observations, forecasts, and warnings available
in real-time on the Internet;
(iii) frequent updates of the latest information, even if noth-
ing is new; and
(iv) distributing purposely designed educational and
outreach material to educate stakeholders about severe
weather events, the associated risks, and actions to take
in advance of, during, and after the event.
4.2 CAPACITY BUILDING
In support of an effective severe weather warning system, it
is vital that NMSs secure the necessary financial and person-
nel resources for:
(i) ongoing maintenance and improvement of the national
meteorological observation infrastructure;
(ii) pure and applied research – both meteorological and in
other disciplines associated with risk management (most
likely through partnerships and collaboration);
(iii) ongoing training for NMS staff, partners, and stake-
holders;
(iv) public education and awareness; and
(v) development of technical, operational, and dissemina-
tion capabilities.
NMSs should ensure that:
(i) the capacity building of their severe weather warning
specialists is high on the agenda;
(ii) the operational procedures are sound; and
(iii) the communication mechanisms and systems are robust.
Training should not be limited to NMS staff, but must
also include partner agencies as well as communities at risk.
For example, in the United States, the National Weather
Service (NWS) works with the Federal Emergency
Management Agency (FEMA) to develop and teach emer-
gency managers how to use NWS product and services, using
resident and distance-learning components:
(i) Hazard Weather and Flood Preparedness;
(ii) Warning Coordination;
(iii) Partnerships for Creating and Maintaining Spotter
Groups; and
(iv) Hurricane Planning.
NMSs also need to allocate sufficient resources to
develop a knowledge base for the effective provision of severe
weather warnings. Examples of such initiatives are:
(i) sponsor or conduct applied research regarding the severe
weather hazards of their country;
(ii) maintain a historical database of past severe weather
events;
(iii) support the production of hazard risk assessments; and
(iv) play an active role in the development of a risk manage-
ment plan for their country for regional and local
applications.
NMSs and other agencies involved in risk management
planning should establish collaborations and methods for
Chapter 4
NMSs’ RESPONSIBILITIES WITHIN
A RISK MANAGEMENT FRAMEWORK
Guidelines on integrating severe weather warnings into disaster risk management
effective exchange of information among relevant hazard
databases to facilitate monitoring, assessment, and predic-
tion.
4.3 DEVELOPING PARTNERSHIPS
The design and operation of severe weather warning systems,
and on a larger scale, the risk management plan, must be
based on a commitment to cooperation and information
exchange and the concept of partnership in the overall public
interest. The benefits of such partnerships include:
(i) drawing expertise from a wide range of disciplines, such
as social science, community planning, engineering, etc.;
(ii) accomplishing tasks that cannot be managed by a single
agency or organization;
(iii) demonstrating to government budget planners a
commitment to work together towards a common goal
and making better use of scarce financial resources;
(iv) leveraging resources for research, awareness, prepared-
ness, etc.;
(v) sharing costs, knowledge, and lessons learned;
(vi) ensuring a consistent message (the warning bulletins and
other outreach material) from multiple credible sources;
and
(vii)yielding wider distribution of the message through
multiple outlets and receiving feedbacks from a whole
range of users.
To identify and evaluate the weather information needs
of the users, NMSs need to build relationships and work in
partnership with users in both the public and private sectors.
NMS partners include:
(i) other government agencies with missions involving the
protection of life and property, such as the NHS where
that is a separate agency from the NMS, national,
regional or local emergency management agencies, first
responders, infrastructure managers (dams, highway
departments, bridges);
(ii) the media;
(iii) non-government organizations;
(iv) emergency relief organizations, such as the Red Cross
and Red Crescent Society;
(v) academic institutions and schools;
(vi) trained volunteers associated with the NMS, such as
cooperative observers, storm spotters, and amateur radio
operators;
(vii)meteorological societies and other professional associa-
tions in risk management disciplines;
(viii) private sector weather companies, and
(ix) utility services, telecommunication operators and other
operation-critical or weather-sensitive businesses.
A typical partnership would involve disaster, warning,
and risk management experts from government, business,
academia, non-government organizations such as the Red
Cross and Red Crescent Society, as well as emergency
management officials to agree on warning standards, proce-
dures, and systems. An example of this is the Partnership for
Public Warning in the United States. More discussion and
examples of partnerships can be found in Section 5.1,
Stakeholder Involvement, of this document, as well as the
WMO document PWS-8, Guide on Improving Public
Understanding of and Response to Warnings.
4.4 PUBLIC EDUCATION & AWARENESS
Public education and awareness programmes need to focus
on the NMS’s severe weather warning systems, the hazards,
the vulnerability and appropriate responses. The goal of
effective public information materials and programmes
should be to ensure all those who are potentially at risk fully
understand the nature of the threat and the appropriate
action to take at various stages as the threat develops and
disaster strikes (Zillman, 2004).
Working with partners can enhance the effectiveness of
education and outreach efforts and spread the cost of devel-
opment and printing among several groups. In addition to
printed material, the same information should also be made
available electronically via the Internet, and in alternative
languages for different ethnic groups or in formats that are
accessible to people with disabilities.
Educational and outreach material should highlight
scientific capability, hazard awareness, risk, and prepared-
ness. The effectiveness of brochures and educational material
can be increased if they follow a similar format, featuring
sections such as:
(i) a description of what causes the weather hazard;
(ii) a map highlighting areas of the country vulnerable to the
hazard;
(iii) historical information about major events of this type of
weather hazard in the country;
(iv) definitions of terminology associated with the hazard;
(v) how to anticipate and monitor the hazard in one’s own
situation;
(vi) measures that individuals and communities can take in
preparation for the occurrence of hazard;
(vii)recommended actions to take before, during, and after
the event;
(viii) how to stay informed during the event;
(ix) contacts for more information; and
(x) contingency measures in the event of a disaster (for the
family, the community, the business, etc.).
Even the selection of graphics and photographs is signif-
icant. Rather than images of destroyed structures, social
scientists recommend including pictures of people taking the
proper actions in preparation for and during the event, such
as people boarding up windows, heading to a storm shelter,
shopping for items in their emergency supplies kit, etc.
NMSs should support hazard awareness education in
schools by contributing to the development of educational
material for teachers and students. In educating the students,
the messages are also brought home to their families and
neighbourhood as well.
The WMO document PWS-8, Guide on Improving Public
Understanding of and Response to Warnings, includes more
discussion and examples of public education and awareness
initiatives.
8
4.5 ESTABLISHING A KNOWLEDGE BASE
One must have relevant knowledge and information in order
to properly understand the hazard and vulnerability. At the
same time, the value of local knowledge, community
"memory", and relevant experience during past events should
not be overlooked. Furthermore, knowledge and informa-
tion is central to almost every aspect and every stage of
natural disaster risk management. It is essential for assessing
risk and potential vulnerability in the earliest stages of
community planning for construction of new facilities (such
as dams, bridges and population centres) or for individuals
planning to move to new locations (such as beaches, flood
plains and mountain sides). NMSs therefore need to allocate
or secure sufficient resources for the development of an
adequate knowledge base for the effective provision of severe
weather warnings. The knowledge base shall consist of (i)
risk databases, (ii) risk assessments, (iii) cost-benefit analy-
ses, and (iv) research activities. These will be discussed in
turn in the following sections.
4.5.1 Risk Databases
Information that is most vital in the early stages of sound
risk management consists of the full suite of climatological
data and products, including model studies of extreme
events, that will enable the potential natural hazard to be
accurately characterized (e.g. through hazard maps) and the
necessary decisions on siting, construction, protection and
precaution to be taken on a fully informed basis.
Information is essential when natural hazards threaten
and when communities prepare to withstand the potential
onset of disaster; and it is often even more essential in the
critical post-disaster recovery phase when affected commu-
nities are shattered and confused, when fear of the
unexpected is greatly heightened, and when relief authorities
need to know everything that is going on to enable them to
manage the complex mix of issues involved in restoring
essential facilities and in meeting the physical or social needs
of devastated communities.
Effective risk management of and preparedness for
natural hazards require free and unlimited access to relevant
databases to facilitate monitoring, assessment, and predic-
tion. It is recommended that all agencies responsible for these
databases develop good collaborating links for the exchange
of information included in the databases.
Communities and civic authorities need to know the
nature, severity and likely return periods of all kinds of
potential hazards. One of most important sources of infor-
mation is still the careful analysis of records of what has
happened in the past. In this regard, meteorological data-
bases (including warnings and climate databases) contain
information relating to natural hazards of a meteorological
origin. On the other hand, disaster databases in general
contain information largely driven by economic and financial
considerations such as insured or un-insured losses, but often
with very little emphasis on information relating to the
weather and climate conditions at the time of the disaster
event. There is a growing recognition that in order to effec-
tively assess the vulnerability and risks involved, analysis has
to be done based on information drawn from both types of
databases. At present, there is no easy means of integrating
or cross-referencing such data and information for complet-
ing an adequate risk analysis. More research on the effective
application of these data and information is therefore
required.
4.5.2 Risk Assessments
Risk assessment requires a "by the people, for the people"
mindset. Risk is the outcome of the interaction between a
hazard phenomenon and the elements at risk within the
community (e.g. the people, buildings and infrastructure)
that are vulnerable to such an impact.
Each hazard likely to impact upon a community can be
systematically analyzed in this fashion. The vast majority of
information, relationships and processes involved in under-
standing risk are spatial in nature. Graphical Information
Systems (GIS) are especially useful for this purpose.
In risk assessment, one has to consider the probabilities
of hazardous events affecting the community and the conse-
quent harm to the community. Probability is a concept and
skill that most people have problems with understanding, as
many cannot handle statistical concepts or effectively factor
probabilities into their decision-making. More research is
needed to help describe how low probability/high conse-
quence events (such as an impact by a very intense cyclone)
affect the community’s mitigation desires and actions.
4.5.3 Cost-Benefit Analyses
General Principles
We will never be in a position to reduce hazard-induced
losses to zero. The measurement of how much reduction is
achieved is usually referred to as the Loss Avoided (La).
Against this, there will be an attributable cost in providing
warnings and taking preventive actions, called Cost of
Prevention (Cp). The ratio, ?La/?Cp, gives a simple but
effective Benefit Cost Ratio (BCR). Clearly we need to aim
for as high a value as possible. The BCR can be used to:
(i) support investment proposals to improve disaster
prevention and mitigation, and
(ii) prioritise work programmes and financial investment.
Understanding of the Benefit Cost Ratio (La/Cp) allows
us to make effective use of probability forecasts.
Calculation of BCR
In practice it is difficult, even impossible, to get an accurate
measure of La. However, this should not deter attempts to do
so with varying levels of sophistication. In a rudimentary
approach, a subjective value can be derived by making esti-
mates of:
(i) number of people or organisations receiving a warning;
(ii) percentage of recipients taking action;
9 Chapter 4 — NMSs’ responsibilities within a risk management framework
Guidelines on integrating severe weather warnings into disaster risk management
(iii) effectiveness (average cost saving) of these individual
actions;
(iv) frequency of the phenomena and accuracy of forecast-
ing.
In addition, in cost-benefit analyses, a nominal monetary
value to human life saved and a surrogate value for public or
political goodwill are sometimes assigned. This valuation is
merely a matter of convenience and does not mean a judge-
ment on the intrinsic value of lives. Any other common
measure of value would, in principle, do equally well.
Cost of prevention, Cp, is usually easier to measure or
more accurate to estimate. A decision has to be made as to
whether to use full or marginal costs. Marginal costs are those
attributable to taking mitigating actions such as extra
manpower and resources (for example from emergency
authorities) to reduce losses and save lives.
It follows that there needs to be detailed discussions with
other stakeholders in order to derive Cpand La. The strive for
higher BCRis more likely to attract extra funding for capac-
ity building leading to improved effectiveness of severe
weather warnings. Often it will be found that efforts to
improve response to warnings, rather than a higher accuracy
of forecasts, offer the most economical way of improving the
BCR. This again highlights the need to work closely with
professional partners in designing a risk management plan
and in bidding for funds.
Probabilities and Consequences
In trying to forecast a severe weather event, we are dealing
with extreme events and thus working within the tail-end
regime of a statistical distribution. The main issue is the soci-
etal impact caused by the weather, i.e. amount of damage,
disruption, economic loss or number of deaths. These factors
usually increase exponentially with an increase in the sever-
ity of the phenomena.
From a risk management viewpoint, it will be more valu-
able to deal with the probability of various thresholds being
exceeded. Unfortunately our professional partners (and the
public) are generally not well versed in the use of probability
forecasts as a decision making tool.
Correct and accurate use of probability forecasts means
that, given a large sample, on average an event will occur at the
same frequency as the forecast probability. For example,
events assigned with a 50% probability will occur on half of
the number of occasions they are predicted. Similarly, those
assigned with a 10% probability should occur 1 in 10 times,
and those with 90% probability on 9 out of 10 occasions. As
such, the full BCR is only realised over a period of time.
Therefore our partners must appreciate that they need
management strategies that are different from those dealing
with deterministic forecasts. In particular, the response must
be commensurate with the risk involved.
As part of the partnership approach it is necessary to
have a common understanding and use of thresholds. An
example illustrating the use of probabilities of a severe
weather event affecting the community in calculating BCR
and the determination of threshold for decision making is
described in Appendix A.
4.5.4 Research Activities
Every NMS, according to the available resources in qualified
workforce and technical facilities, will currently perform
some research activities in the fields of meteorology, clima-
tology, hydrology, oceanography, or even seismology. An
important part of these efforts is the understanding and
prediction of hazards. Storing and updating the results of
such research activities is a fundamental requirement for any
scientific approach for risk management planning of hazards.
An aspect, which probably deserves more attention, is
the benefits that may be derived from international collabo-
ration of research activities, through networking of
institutions or individual scientists. A primary motivating
factor is that the understanding of physical processes in the
atmosphere often requires full scale field experiments to
gather data and information; and such efforts are usually out
of the reach of individual scientists, a single NMS or even
national consortia of research institutions in many countries.
A good example is the progress made in the understanding
of Atlantic storms after FASTEX (Fronts and Atlantic Storm-
Track Experiment) (Browning et al, 1999). This has lead to
the development of the decade long WMO research project
THORPEX (The Observing System Research and
Predictability Experiment), which is addressing the use of
ensemble forecasts and targeted observations to improve
early warning of hazard weather (Shapiro and Thorpe 2004).
International cooperation among meteorologists and
other scientists in different fields is also an effective way of
securing funding and support from international institutions.
For example, the five-year cycle of the R&D programmes of
the European Union, which various European meteorologi-
cal services contribute to and get funding for, contains
components that directly address the hazard risk manage-
ment issues (e.g. GMES (Global Monitoring for Environment
and Security)).
Another related structure, different but related to the
European Union and is more of a networking mechanism, is
the European Cooperation in the field of Scientific and
Technical Research (COST) organisation. It is well known
that the origin of the European Centre for Medium-range
Weather Forecasting (ECMWF) is to be found in a COST
action decided 30 years ago. More recently, actions in the
domains of nowcasting, urban pollution, radar techniques
and waves modelling also provide the international commu-
nity with valuable results and deliverables (exchange formats,
data bases, tested algorithms, etc.) (further details can be
found inhttp://www.cordis.lu/cost/home.html). The fact that
the COST approach is "bottom up" (i.e. actions initially
proposed by individual scientists) and multi-disciplinary is a
great advantage; at the same time, there are also effective
coordination and controls ("top down") by national repre-
sentatives that look after users’ interests and ensure
productive outcome of the actions. Eventually, application
programmes would be derived from such actions, currently
funded by the European Commission of the European
Union. In this sense, participation in COST can be an effi-
cient leverage for obtaining much-needed funds for
meteorological institutes to increase their knowledge base.
10
As already mentioned, a mere "natural sciences"
approach is not sufficient. Risk management requires a
combined multi-disciplinary approach because of the socio-
economic, political and in some respects, psychological,
dimensions involved in the formulation of policies and strate-
gies. A workshop on early warnings held in Shanghai in 2003
noted (Glantz, 2003): "With regard to geological or hydro-
meteorological hazards, the physical processes are either well
understood or are under close scrutiny by scientific
researchers. With regard to socio-economic and political
processes, there is a heightened need to understand their roles
in the conversion of a potential hazard into an actual disas-
ter."
NMSs will increase their capacity to fulfil their role in a
risk management plan when the results of such research
activities in other specialized fields are integrated, to a greater
or lesser extent, into their own knowledge base. Participation
in multi-disciplinary seminars or colloquia dealing with
natural hazards is necessary, but not sufficient. In building a
real knowledge base, the mere exchange of results among
meteorologists, hydrologists, sociologists, psychologists,
media specialists is a good start but will probably still fall
some way short of providing an overall approach and under-
standing of hazards in the context of attendant societal or
economical problems. A recent prominent report on Risk
Management in the USA notes: "In developing the nation’s
all-hazards approach to disaster vulnerability reduction, no
single federal agency can provide a fully comprehensive solu-
tion. Agencies must work together to narrow programme
gaps through a coordinated science and applications research
agenda to pursue common solutions to common problems"
(Executive Office of the President of the USA, 2003).
A conspicuous illustration of this may be derived from
the lessons learnt from the deadly heat wave that hit Western
Europe, especially France, during the first fortnight of August
2003. A comprehensive evaluation, from the risk manage-
ment point of view, is to be found in Section 6.1.2. It has been
publicly acknowledged that in such an occurrence, the collab-
oration and synergy between key stakeholders in the fields of
meteorology, public health, and civil security have been very
insufficient.As far as the research implications are concerned,
it is clear that at the time there were available neither a
comprehensive understanding of all the factors that resulted
in such a catastrophe, nor in the aftermath adequate and
detailed indicators that could have been of precious use for
the authorities and the public (the cited Shanghai workshop
noted: "The selection of indicators is very important, because
monitoring will centre on them. The wrong indicators can
lead to wasted time, effort and resources").
The predictions for the onset, duration and end of the
heat wave by the relevant meteorological services were mostly
correct and timely (as allowed by the state of the art)
(Grazzini et al., 2003), but still quite general and insufficiently
echoed by the media and relevant institutions. Taking the
case of Météo France, its warning action (press releases
mentioning the people at risk) was actually more efficient
and reflected the user-oriented approach of Météo France.
Yet only a few years ago, a bio-meteorological conference
(Conseil Supérieur de la Météorologie, 2002) had correctly
pointed out the deadly potential of such heat waves and
roughly described its consequences. Nevertheless, it is now
understood that the setting up of combined indicators for
such events requires definitions and testing that can be
achieved only through multi-disciplinary research into
biological, demographical and sociological behaviours affect-
ing the most vulnerable populations, as well as much needed
precise documentation and understanding of the meteoro-
logical circumstances, not only minimum and maximum
temperatures but also wind, humidity and air quality.
In France, these conclusions have led to formalized
agreements between Météo France and two institutions in
charge of public health. By the end of 2003 (see also Chapter
6), a first agreement was quickly set out to provide some new
indicators, albeit on a provisional basis, with the opera-
tionally oriented "Institut de Veille Sanitaire" (Institute for
Sanitary Watch). Another collaboration which is much more
far-reaching was established in July 2004 with the medical
research institute INSERM. According to the press release, it
is due to enlarge and enhance the combined research on heat
(and cold) waves and their impact on public health, with an
aim to document and study in details the bio-meteorological
mechanisms at much finer scales and over a longer period of
30 years.
In building up a comprehensive research basis for risk
management planning, NMSs are advised to actively enter
(according to their resources and/or those that could be
provided by interested sponsors) into joint research agree-
ments with other scientific institutions. This would normally
imply the constitution of mixed research teams or the
exchange of scientific missions; or to host post-doctorate
researchers coming from different scientific horizons. In the
medium to long term, the results of such research will prove
to be of great use for the prevention and mitigation of
complex hazards. The same cited American report
(Executive Office of the President of the USA , 2003) lists as
follows the research and development areas to be considered:
"- fundamental and applied research on geological,
meteorological, epidemiological and fire hazards devel-
opments;
- application of remote sensing technologies, software
models, infrastructure models, organizational and social
behaviour models, emergency medical techniques; and
- many other science disciplines applicable to all facets of
disasters and disaster management".
4.6 DISASTER RISK MANAGEMENT EXPERTISE
NMSs should give due consideration to gaining access (full-
time or occasional) to risk management expertise to
complement its core operations. As an example, the
Australian Bureau of Meteorology has over the past few years
established a cross-cutting disaster mitigation programme
with the recruitment of a disaster management specialist.
This programme supports the Bureau’s severe weather warn-
ing services and works in partnership with various
stakeholders throughout the community to ensure the appli-
cation of risk management principles in the communication
of severe weather warning information and related risk
messages.
11 Chapter 4 — NMSs’ responsibilities within a risk management framework
4.7 FORMULATING A RISK MANAGEMENT
ACTION PLAN
Risk management provides structured systems for identify-
ing and analyzing potential risks, as well as devising and
implementing responses appropriate to their impact.
According to the Secretariat of the ISDR (International
Strategy for Disaster Reduction, 2004), disaster risk manage-
ment focuses on three key areas:
(i) assessment of the risk factors present;
(ii) tools and practices to reduce the risks; and
(iii) institutional mechanisms to support both risk assess-
ment and risk management.
It thus spans a wide range of methods and activities –
assessment and analysis, public information, community
participation, warning systems, policy and regulation, project
impact assessment, education programmes, conservation
practices and political processes. When advantage is taken of
the mutual benefits achieved through adoption of a system-
atic integrated approach to the various activities, it is
appropriately referred to as "integrated disaster risk manage-
ment".
Risk management consists of a systematic process of
assessing and dealing with risks. This process involves
consideration of the context, followed by identification,
analysis, evaluation, and treatment of risks. It is a repetitive
cyclic process that also requires monitoring and review, and
should include communication and consultation with stake-
holders all along. The process of disaster risk management is
shown schematically in Fig. 1. This process can be used in
formulating the Risk Management Action Plan. The steps to
follow are:
(i) Communicate and consult with stakeholders throughout
the process;
(ii) Establish the Context;
(iii) Identify and Quantify the risks;
(iv) Analyse and Evaluate the risks;
(v) Treat the avoidable and manageable risks; and
(vi) Monitor and Review the Risk Management Plan.
Each of these steps will be discussed in detail in the
following sections.
4.7.1 Step 1: Communicate & Consult with
Stakeholders
Communication and consultation are important considera-
tions at all stages of the process to ensure that stakeholders are
involved. This is the first step in the disaster risk manage-
ment process. Consultation is a two-way process that enables
disaster risk managers to be aware of perceptions and to
accept input to the process from stakeholders. Stakeholders
will have perceptions of risks that can be useful in the risk
assessment stages. In developing risk responses, it is also
desirable to consider communication strategies for both
internal and external stakeholders. Consultation ultimately
aims at developing partnerships. Communication must be
effective to ensure that those agencies and/or individuals
responsible for implementing risk treatment measures are
given sufficient information about the measures and the
reasons for the adopted strategies.
4.7.2 Step 2: Establish the Context
The problem must first be defined by determining the nature
and scope of the risk management project. This includes
defining the community involved and the types of natural
hazards to be addressed. A project management structure
must also be established. A simple SWOT (strengths and
weaknesses, opportunities and threats) analysis can be an
excellent means in the initial consideration of the context of
risk assessment.
4.7.3 Step 3: Identify & Quantify the Risks
In the world of natural hazards, there are many different ways
of defining and quantifying risks depending on the intended
purposes (e.g. for deciding what protective measures to put
in place or what insurance premiums to charge). However,
most definitions would include some combinations of:
(i) the nature of the hazard, including its severity;
(ii) the probability of its occurrence;
(iii) the exposure and vulnerability of human and natural
systems; and
(iv) the numbers of people and the costs of the facilities at
risk.
Risk identification is achieved by:
(i) identifying and describing the natural hazards – the
sources of risk;
(ii) identifying and describing the community and its envi-
ronment – the elements at risk;
(iii) determining the vulnerability – the balance between
susceptibility (the level to which a particular natural
hazard will effect a community) and resilience (the abil-
ity of a community to recover from the impact of a
natural hazard); and
(iv) describing the risk.
The process of quantifying and mapping risks
thus involves an integrated approach to the use of physi-
cal/geographical/environmental information in conjunction
with both social and economic data, usually through inte-
grated assessment models of various kinds.
4.7.4 Step 4: Analyse and Evaluate the Risks
Risk is analysed by determining the likelihood of a natural
hazard occurring and the consequences of that hazard event.
This is done in both qualitative and quantitative terms. The
analysis considers community vulnerability and existing risk
management measures with all assumptions clearly stated.
The relationship between likelihood and consequences then
enables the level of risk to be determined. The level is not an
absolute level, but reflects a multi-faceted set of criteria that
enables societal judgments on the risk to be made.
Guidelines on integrating severe weather warnings into disaster risk management 12
13 Chapter 4 — NMSs’ responsibilities within a risk management framework
Fig. 1. The Disaster Risk Management Process
Establish Context
Identify Risk ...
What? (event)
How? (source)
Why? (impact)
Analyse Risks
yes
Consequences
Estimate Level of Risk
Evaluate Risk
(priorities)
Is risk acceptable?
Treat Risks
Stakeholder
Dialogue -
Communication /
Consultation
Monitoring /
Review
No
Likelihood
Risk is evaluated by comparing the risk evaluation crite-
ria with the level of risk. This establishes the priority for the
treatment of each risk and/or the acceptability of the resid-
ual risk. Acceptable risk is sufficiently low risk at a level that
the society is comfortable with. A particular risk may be
accepted when the cost of treatment is considered excessive
compared to the benefit of the treatment. The process is
achieved by consultation with all stakeholders and is subject
to review and modification if required.
4.7.5 Step 5: Treat the Avoidable and Manageable Risks
Risk treatments are designed to reduce any or all of the
vulnerability of elements at risk, the likelihood of risk occur-
ring and the consequences of the event. The process involves
identifying and evaluating options, selecting the most appro-
priate treatment(s), and planning and implementing the
treatment programme(s).
The treatment of identified risks will almost certainly lie
in technical innovation, personnel skills development and
the ability to do both pure and applied research. There is a
range of actions that can be taken to reduce the risk associ-
ated with natural hazards. These measures fall into four main
categories:
(i) adoption of adaptive strategies such as temporary or
permanent migration away from the areas of high risks;
(ii) education and awareness programmes for key stake-
holders;
(iii) implementation of a works programme aimed at forti-
fying the infrastructure; and
(iv) adoption of diversified responses, such as a combina-
tion of the above measures along with land use planning,
refined warning systems, and insurance incentives.
Formal reporting is an important phase of the risk
management process. A Risk Management Action Plan is a
helpful formal way to report on designated or major under-
takings. It should summarize the results of the risk
management process, action strategies and implementation
framework. In particular, it should describe the risk response
measures to be implemented to reduce and control risks. For
major risks, risk action schedules should be prepared to
assign individual responsibilities and time frames and to
identify those who are responsible for follow-up. The Plan
should also include provisions for implementation and ongo-
ing reporting.
The most important task in risk management is imple-
menting the Risk Management Action Plan and allocating
management resources. This should be followed by monitor-
ing the effectiveness of these measures over time (Step 6).
Planning for implementation requires particular attention to
be given to resources needed, management responsibilities
and timing of tasks.
4.7.6 Step 6: Monitor and Review the Risk
Management Action Plan
Risk is not static. It is therefore necessary to continually moni-
tor the status of the risk being managed and the interaction
of risk, community and the environment; and to review the
risk management processes in place. Continual monitoring
enables the process to dynamically adapt to changes in risk
as well as changes in stakeholders’ needs. The frequency of
monitoring and the responsibility for conducting the review
should be specified in the Risk Management Action Plan.
4.8 SECURING FUNDING
In planning and implementing the risk management
programme, resources for ongoing training activities, public
education as well as the development of both technical and
operational capabilities are essential. Severe weather warnings
are effective only to the extent that policy makers at national
levels of authority have the will to make a sustained commit-
ment of resources that establish and operate the required risk
management mechanisms.
To obtain and maintain such resources, a collaborative
approach to funding requests is more likely to be successful.
If the NMS is able to join forces with partners, it not only adds
influence and weight to the scope and substance of the
proposed project, it also enables the politicians to consider
the merit of the project in the context of the complete risk
management picture, and not simply confining it to the
narrow and more technical role of the NMS.
Presenting a well considered risk management plan is
also a very significant step in soliciting funding from both
national and international sources, such as the World Bank or
the European Union.Weather and climate hazards have never
been bounded by national boundaries; and in an increas-
ingly globalized community with active information
exchange and population movement, the impact of any severe
weather events would be even more keenly felt. As such,
NMSs need not be too inward looking in the search of possi-
ble funding options; sponsors with international interest may
just as likely be prepared to contribute. A joint funding bid
with key stakeholders or other NMSs in the region would
inevitably broaden the horizon and hence enhance the like-
lihood of success. Sometimes, it may even lead to a larger
funding allocation than a NMS working alone would other-
wise have achieved.
Apart from international sponsorships and traditional
sources such as the NMS budget, other appropriate funding
alternatives can also be explored in connection with different
components in the overall risk management plans.
Collaboration with universities or research organizations is
one such option through the utilization of research grants.
Often, the academia can also offer the necessary expertise
across a multitude of disciplines and provide a dedicated
environment for research studies. This kind of collaboration
is particularly rewarding for NMSs that do not have in-house
capacity to undertake basic research and technology devel-
opment. NMSs can also consider the possibility of securing
funding through joint venture with other community or
infrastructure projects, particularly those that are weather-
sensitive and may require meteorological support for
planning and operation; e.g. the setting-up of observation
networks along highways or bridges, consultancy role in
Guidelines on integrating severe weather warnings into disaster risk management 14
15 Chapter 4 — NMSs’ responsibilities within a risk management framework
urbanization or conservation projects, forecasting and warn-
ings services for recreational or major sporting events, etc.
As an illustration of what can be achieved, success stories
of funding acquired from some innovative sources are
presented in more details in Section 6.2.
Natural disasters can lead to extensive property damage,
disruptions on social activities and loss of life. As noted in
the Introduction, during the period 1992-2001, natural disas-
ters worldwide have killed over 622,000 and affected over 2
billion people. While the figures have been high, it should be
pointed out that they would have been even higher without
pre-disaster efforts, particularly early warning systems, which
contribute significantly to an effective risk management
strategy.
The primary objective of a warning system is to
empower individuals and communities to respond appro-
priately to a threat in order to reduce the risk of death, injury,
property loss and damage. Warnings need to get the message
across and stimulate those at risk to take action.
Effective inclusion of the severe weather warning system
in a risk management plan relies on NMSs to appreciate the
needs of a multi-cultural, economically stratified and often
mobile community, and the community understanding the
hazard, its vulnerability and the most suited protective action
to take.
Greater focus towards disaster mitigation also means
(Gunasekera, 2004):
(i) further increasing the emphasis on extending the lead
time of warnings;
(ii) improving the accuracy of warnings at varying lead
times;
(iii) satisfying greater demand for probabilistic forecasts;
(iv) better communication and dissemination of warnings;
(v) using new technologies to alert the public;
(vi) better targeting of the warning services to relevant and
specific users (right information to right people at right
time at the right place); and
(vii)ensuring that warning messages are understood and the
appropriate action taken in response.
5.1 STAKEHOLDER INVOLVEMENT
Stakeholders need to be consulted as partners in the design
and refinement of severe weather warning systems, and on
the larger scale, the risk management plan. Stakeholders
include the public, other national government agencies,
emergency management agencies, local authorities, non-
government organizations, the media, social scientists,
national and regional infrastructure authorities, academia,
etc.
Involving stakeholders in developing and enhancing the
end-to-end severe weather warning system has many bene-
fits, such as:
(i) improved presentation, structure, and wording of the
warnings themselves;
(ii) more effective communication of the risks and actions to
take in response to severe weather;
(iii) better understanding of how, and how often, stakehold-
ers want to receive warnings; and
(iv) increased sense of ownership, and therefore, credibility
in the warning system.
5.2 WARNING PRESENTATION
Effective warnings are short, concise, understandable, and
actionable, answering the questions of "what?", "where?",
"when?", "why?", and "how to respond?". The use of plain
language in simple, short sentences or phrases enhances the
user’s understanding of the warning. In addition, the most
important information in the warning should be presented
first, followed by supporting information. Effective warnings
should also include detailed information about the threat
with recognizable or localized geographical references.
Warnings should be presented in several different
formats – text, graphics, colour-coded categories, audio –
and should include specific actions for people to take to
respond to the event. The various formats also make it easier
for people with disabilities to receive and act on the warnings.
All formats, however, must present the information
accurately and consistently.
The suggestions above are based on social scientist
recommendations to structure the message to encourage the
public to feel personally affected. For a more thorough
discussion of warning presentation, see WMO document
PWS-8, Guide on Improving Public Understanding of and
Response to Warnings.
5.3 WARNING COMMUNICATION
Dissemination is delivery of the warning messages; but
communication is accomplished only after the information is
received and understood. So the foundation of warning
communication builds on the format and wording of the
warnings, dissemination methods, education and prepared-
ness of stakeholders, and their understanding of the risks
they face. Communication is also significantly enhanced
when consistent warning information is received from multi-
ple credible sources.
NMSs should ensure their severe weather warnings are
communicated using a variety of formats (text, graphics,
voice) and disseminated via as wide a range of media as is
available (press, radio and television, e-mail, cell phone,
Internet, etc.). Media broadcasts from the weather office
and/or radio and television interviews with one or more
authoritative figures can be effective in triggering response
from people. These authoritative figures may be from the
NMS (such as the NWS director or the manager of a local
Chapter 5
EFFECTIVE EARLY WARNINGS
17 Chapter 5 — Effective early warnings
weather office) or a community leader (such as the governor
or emergency manager).
The NMS should identify and designate the appropriate
source of warnings and information from within the NMS
structure for different types of risks occurring at different
locations. For local hazards, the community is more likely to
trust information from someone with a local knowledge of
the area, rather than someone from a distant office who may
not be as sensitive to the local needs.
Effective communication about risks and warnings
requires knowledge about the recipients. In most countries,
the public is very diverse, with different backgrounds, expe-
riences, perceptions, circumstances and priorities. Any
attempts to communicate with the public must reflect this
diversity.
5.4 WARNING RESPONSE
Public education and awareness (section 4.4), stakeholder
involvement (section 5.1), warning presentation (section
5.2), and warning communication (section 5.3) all contribute
to an appropriate response to the warning.
The warning message by itself does not stimulate an
immediate response from individuals. Individuals receiving
the warning will first assess their own personal sense of risk.
The additional information required before they take action
depends on the content and clarity of the initial warning and
the credibility of the issuing organization. The potential
for individuals to respond appropriately is dramatically
increased if they are provided with information to assist them
in defining their personal risk and highlighting what life- or
property-saving actions to take.
Two excellent references on warning systems and public
response are Mileti and Sorenson (1990) and Mileti (1999).
Their work is summarized below.
Successful warning programmes strive to ensure that
every person or organization at risk:
(i) receives the warning;
(ii) understands the information presented;
(iii) believes the information;
(iv) personalizes the risk;
(v) makes correct decisions; and
(vi) responds in a timely manner.
An individual’s perception of risk is enhanced if:
(i) warning messages before and during a particular event
are issued and updated frequently;
(ii) warnings are delivered by multiple credible sources;
(iii) warning messages are consistent;
(iv) the basis for the warning is clear; and
(v) suggested response actions are included.
Preparedness prior to a hazardous event is critical to an
effective response to the warning. Individuals, families,
schools, businesses, communities and public facilities should
have an advance plan in place, so they know what to do and
where to go when a warning is received. The NMS and its risk
management partners have a key role to play in educating
their constituents in how to prepare a plan of action.
For a more thorough discussion of warning response, see
WMO document PWS-8, Guide on Improving Public
Understanding of and Response to Warnings.
5.5 MONITORING AND REVIEW
Verification and assessment of the warning services after a
severe weather event are essential to measure performance,
identify and correct deficiencies, and capture best practices,
which can be shared with other parts of the NMS or with
other risk management partners. In addition to quantitative
measurement, objective assessment is also valuable.
Interviews or surveys conducted with partners and stake-
holders can yield significant insight into how products and
services are received, interpreted, and what actions have been
taken as a result of the warning. This feedback can then be
used to make adjustments for similar future warning events.
Publishing verification scores and post-event assess-
ments can add to the credibility of the NMS and, for
stakeholders and partners, reinforce the perception of the
NMS as being user-oriented and dedicated to the cause.
6.1 EXAMPLES OF SUCCESSFUL RISK
MANAGEMENT INITIATIVES
The following sections describe initiatives by several NMSs
using risk management principles to improve their warning
systems. These examples provide practical applications of
some of the principles summarized in these guidelines, and
may trigger other NMSs to implement, expand upon, or
create their own applications within their organization or
country.
6.1.1 EXAMPLES FROM HONG KONG, CHINA
Hong Kong is a metropolis with over 6 million people and
frequented by tropical cyclones and heavy rain. Tropical
cyclones could bring gales or even hurricane force winds as
well as torrential rain to the city. It is not uncommon to regis-
ter more than 100 mm of rain a day in tropical cyclone or
monsoon trough situations, causing landslips and flooding.
The following examples illustrate how the Hong Kong
Observatory, the NMS in Hong Kong, China, mitigates the
impacts of these and other natural hazards.
Tropical Cyclone Warning Signal System as a Triggering
Mechanism for Protective Actions Against
Hazardous Weather
On average, six tropical cyclones affect Hong Kong each
year. To mitigate the impact of tropical cyclones, Hong Kong
operates a graded Tropical Cyclone Warning Signal System to
warn the public of the threat of winds associated with a
tropical cyclone. The System consists of five numbers repre-
senting increasing levels of wind strength. The Tropical
Cyclone Signal No. 1 is issued whenever a tropical cyclone is
within 800 km of Hong Kong and may affect the territory
later. Signals No. 3 and No. 8 warn the public of strong and
gale/storm force winds in the city respectively. Signal No. 9
signifies increasing gale or storm force winds, while No. 10
warns of hurricane force winds.
In view of the stringent building codes in Hong Kong,
home is generally considered the safest place for people to
take refuge from a tropical cyclone. When Signal No. 8 is
issued, the Hong Kong Observatory advises the public to stay
home or return home. Practically all activities in the city
move towards shutdown. All schools, government offices,
banks, the stock market, and courts are closed. Most public
transport services will start to cease operation. When Signal
No. 9 goes up, even the underground train stops. With Signal
No. 10, the city grinds to a complete halt to prepare for the
onslaught of a full-fledged typhoon.
However, the issuance of Signal No. 8 itself has the poten-
tial of causing chaos as millions of people try to go home all
at once. To facilitate an orderly shutdown, a special
announcement to the public (the Pre-8 Announcement)
about the impending No. 8 is issued two hours before the
actual issuance of the Signal. This enables transport opera-
tors to take measures to cope with the surge in demand for
public transport and to allow the public to go home in a safe
and orderly manner before the cyclone hits.
The tropical cyclone warning system has been in use for
many years and the public is already familiar with it. Together
with the well-coordinated response actions taken by relief
agencies, the system has proved very effective in reducing the
loss of life and property due to tropical cyclones.
Linking Rainstorm Signals to School Operation
Hong Kong is affected by severe local rainstorms, typi-
cally between April and September. The rainstorms can
develop explosively with very intense rainfall. The instanta-
neous rainfall rates can exceed 300 mm/hr. Heavy rain,
together with landslip and flooding, will lead to chaos, partic-
ularly during rush hours, in a densely populated city if the
situation is not properly managed.
Since 1992, the Hong Kong Observatory operates a
colour-coded rainstorm warning system which consists of 3
levels, namely: Amber, Red, and Black. The Amber signal is
issued to give alert on potential heavy rain which may develop
into Red/Black rainstorms. The Red and Black signals are
issued to warn the public about the occurrence of heavy rain
(50 and 70 mm/hr respectively) which can be hazardous and
may result in major disruptions.
By the time the Red signal situation is reached, road
conditions are considered unsuitable for students to
commute between home and school. The issuance of the Red
signal will trigger a series of response measures in relation to
school operation. Prior instructions are given to school
administrators, school-bus operators and parents on the
actions to take. Students are advised to stay home if they have
not left for school. For those on the way to or already at
school, the schools will be open and have sufficient staff to
take care of them until conditions are safe for them to return
home. Classes already in session will not be affected by the
issuance of the signals and students will only be released
when the threat recedes. Forecasters will liaise closely with the
education authority when heavy rain is expected and the Red
signal is imminent. Such close coordination has been in
operation for over ten years and there is practically no death
nor injury of students as a result of such inclement weather
conditions.
Chapter 6
SUCCESSFUL INITIATIVES OF NMSs
Very Hot and Cold Weather Warnings and Temporary Relief
Centres for the Needy
In the subtropical climate of Hong Kong, extreme
temperatures are rare, seldom exceeding 36°C and never
falling below 0°C. Nonetheless, social consequences and
health impacts can still be felt whenever temperatures rise
above or drop below certain "comfort" levels during heat
waves or prolonged cold spells that may affect Hong Kong
several times each year. The elderly and patients with chronic
illnesses are particularly at risk. Hypothermia and heat stroke
could lead to death for people who are over-exposed to the
elements or engaged in outdoor activities.
To alert the public of such risks, the Hong Kong
Observatory operates the Very Hot and Cold Weather
Warnings. Taking into account the combined effects of wind
and humidity, the reference urban temperature criteria for
operating the warnings are usually around 33°C and above
for the Very Hot Weather Warning, or around 12°C and below
for the Cold Weather Warning. The warnings are broadcast
on radio and TV, with advisories on actions to take. Upon the
issuance of such warnings, temporary relief shelters operated
by the Home Affairs Department, where accommodation is
provided with air conditioning in very hot weather and blan-
kets as well as hot food in cold weather, will be opened to help
the needy through difficult times.
6.1.2 The French Vigilance System After the Heat Wave
of August 2003
Following the exceptional heat wave that struck several coun-
tries in Western Europe, particularly France, during the first
fortnight of August 2003, a thorough revision of the risk
management plan for such situations was made.
Apart from various socio-economic effects, the over-
whelming impact was the increase of mortality among elderly
people, either living in pensioners institutions that were not
equipped with adequate resources to deal with such unusual
weather conditions, or in isolation in the big cities, suffering
from over-heating and deadly dehydration and were detected
too late by neighbours, medical or special rescue services.
During that fortnight of August 2003, the number of deaths
attributable to the heat wave was numbered by the [French]
National Statistics Institute to be almost 15,000 people.
One of the lessons learned (Lalande, 2003) was that the
weather warnings, though scientifically correct and timely,
were not assigned enough weight in the risk management
plan and other decision making processes, and were not given
sufficient exposure by the media coverage at the time.
In France, a combined risk management tool, named the
"Vigilance System", is jointly managed by the Civil Security
Authorities and Météo France, and directed towards local
authorities, the media, and the public at large (through two
Internet sites). Details of this system are described in a WMO
publication (WMO, 2003). The Vigilance System was initially
built for short-range warnings (with 24-hour lead time,
updated twice a day) and a limited set of phenomena (heavy
winds and rains, snow falls/avalanches and thunderstorms).
Other hazards have also been considered, including cold or
heat waves, but their characteristics in lead-time and societal
impacts were thought at the time to be too different from
those of other considered hazards, and as such the related
information communication was left to another kind of
procedure in the form of press releases.
This conclusion was clearly reversed after the event in
2003, and the partners in the Vigilance System decided to
add new indicators and adopt new procedures to take account
of this kind of menace. The basic view was that the potential
of the Vigilance System, already a corner stone in the "risk
culture" of the French people, would be enhanced rather than
blurred with the introduction of new concepts and proce-
dures.
Accordingly, Météo France in collaboration with
researchers from the "Institut de Veille Sanitaire" (Institute for
Sanitary Watch) expedited a study so that relevant indicators
were soon integrated into the popular four-color mechanisms
of the Vigilance System. The new procedures take into
account heat or cold waves forecast over three days of the
onset, and the persistence of maximum and minimum
temperatures above relevant thresholds in different cities
studied where the statistics of "excess number of deaths", or
a proxy of it, were correlated with the observed meteorolog-
ical conditions. They were tested during the summer of 2004;
and are now part of the overall Risk Management Plan, with
a comprehensive set of downstream actions organized by
Civil Security authorities, rescue services and health institu-
tions under the aegis of the "Prefects" (governors).
6.1.3 Examples from United States of America: All
Hazards Emergency Message Collection System
In the United States, the National Oceanic and Atmospheric
Administration (NOAA) is assigned responsibility by the
Department of Homeland Security’s (DHS) Federal
Response Plan to "provide public dissemination of critical pre
and post event information on the all hazards NOAA Weather
Radio (NWR) system, the NOAA Weather Wire Service, and
the Emergency Managers Weather Information Network
(EMWIN)"(Federal Response Plan, 2003). In 2002, in
support of the nation’s homeland security efforts, NOAA’s
National Weather Service (NWS) submitted a Fiscal Year 2004
Federal Budget funding initiative to streamline the creation,
authentication, and collection of all types of non-weather
emergency messages in a quick and secure fashion for subse-
quent alert, warning, and notification purposes. NWS
received specific funding to develop the All Hazards
Emergency Message Collection System (HazCollect) in its
Fiscal Year 2004 (FY04) budget.
HazCollect will improve NWS’s support to DHS’s
Federal Response Plan by integrating the automated collec-
tion and acceptance of non-weather related emergency
messages into the NWS Information Technology (IT) infra-
structure. NWS will implement a new, centralized point of
collection for non-weather related emergency messages
broadcast over NWS dissemination systems. A critical
component of NOAA’s Homeland Security Initiative,
HazCollect will reduce the time to disseminate non-weather
emergency messages from 7 to 2 minutes.
19 Chapter 6 — Successful initiatives NMSS
HazCollect will replace the existing system, which is a
manual, error-prone process. Currently, the non-weather
related emergency message is received at the local Weather
Forecast Office (WFO) by telephone or fax from local and/or
state government agencies. The author of the information is
manually authenticated and authorized by the WFO by vali-
dating a verbal password and/or checking the author or
agency name against an approved source and scope list. The
message is evaluated for reasonableness, manually typed by
WFO staff into their workstation, and then disseminated
using the existing NWS dissemination systems, including
NOAAPORT, NOAA Weather Wire Service, Family of
Services, Emergency Managers Weather Information
Network, and NOAA Weather Radio.
The current system has grave shortfalls. The manual
security and message composition process is time-consum-
ing and error-prone. Even in the best circumstances during
emergencies, when time is most critical and seconds can be
the difference between life and death, manual intervention by
the WFO staff adds minutes to the interval between the time
the emergency manager desires to warn the public and the
time the public actually receives the message. Additionally,
the manual text entry process is prone to typographical and
grammatical errors when transcribing the emergency
manager’s input and composing a message for transmission.
HazCollect will meet emergency managers’ requirement
for a fast, reliable way to inject non-weather related emer-
gency messages into the United States’ Emergency Alert
System (EAS), which is operated on commercial and public
radio and television stations nationwide; until now, no single
technical solution has been federally mandated or locally
selected to do this. Each state is currently free to choose any
system it prefers. When implemented, HazCollect will be
able to import emergency managers’ critical non-weather
emergency messages into the NWS dissemination infra-
structure and ensure efficient distribution to other national
systems and to EAS for re-broadcast. NWR is the NWS’ entry
point into EAS, and DHS and the Federal Emergency
Management Agency (FEMA) recognize NWR as a critical
component of the U.S. Warning Dissemination
Infrastructure.
NWS expects to deploy HazCollect in the summer and
fall of 2005. HazCollect will provide an information tech-
nology interface between state and local computer systems
and the NWS communications and dissemination infra-
structure utilizing FEMA’s Disaster Management
Interoperability Services. Through agreements with local
and state governments and at the request of government offi-
cials, NOAA’s NWS accepts emergency messages of
non-weather related nature, such as chemical spills/releases,
abducted child alerts, and radiological events, and
informs/warns the public of these events through the various
existing NWS dissemination systems.
6.1.4 Examples from Australia
Standard Emergency Warning Signal
In 1999, Australia’s lead disaster management authorities
reached agreement on the need for a Standard Emergency
Warning Signal (SEWS) to be used in assisting the delivery of
public warnings and messages for major emergency events.
The SEWS is a wailing siren sound used in northern Australia
for many years to attract attention to cyclone warnings where
a destructive impact was expected. The SEWS is intended for
use as an alert signal to be played on public media to draw
listeners’ attention to an emergency warning immediately
following.
It is considered vital that the status and effectiveness of
the SEWS are preserved as a "disaster mitigation" mecha-
nism by ensuring that it is used for serious events only. As a
general rule, the following 4 factors should be present:
(i) potential for loss of life and/or a major threat to a signif-
icant number of properties or the environment; usually
the threat/impact would be the lead item in local news
bulletins;
(ii) a significant number of people need to be warned;
(iii) impact is expected within 12 hours – or is occurring at
the time; and
(iv) one or more phenomena are classified as destructive.
When the SEWS is to be used, the following advice to the
media is included in the warning header: "Transmitters serv-
ing the area (defined) are requested to use the Standard
Emergency Warning Signal before broadcasting this
message". The media is requested to isolate the use of the
SEWS to the threatened defined area to reduce the risk of
unnecessarily alarming others. In regard to severe weather
and flood warnings, the use of the SEWS is authorized by the
State/Territory Director of the Australian Bureau of
Meteorology. Below are a number of severe weather and
flood examples where the use of the SEWS could be autho-
rized:
(i) wind gusts > 125 km/h;
(ii) storm tide > 0.5 m above (effectively) high water mark;
(iii) large hail > 4 cm in diameter (i.e. > golf-ball size);
(iv) tornado(s);
(v) major flood in a river or creek system;
(vi) intense rainfall leading to flash floods and/or landslides;
and
(vii)major urban and rural fires.
The information above is largely extracted from a multi-
authored booklet on the SEWS, reprinted in Queensland in
2004, and available online at:http://www.disaster.qld.gov.au/disasters/warning.asp
Disaster Mitigation Initiative After the Sydney to Hobart
Yacht Race 1998
Of the 115 yachts that set sail on Boxing Day 1998 in the
Sydney to Hobart Yacht Race, only 44 reached their destina-
tion. The destruction caused by a storm encountered by the
fleet triggered a massive search and rescue operation involv-
ing numerous personnel from the defence forces and disaster
management authorities. Even so, it resulted in the abandon-
ment of several yachts and the death of 6 people. It was the
20 Guidelines on integrating severe weather warnings into disaster risk management
most disastrous event in the 54-year history of the yachting
classic.
While the issuance of warnings on weather and sea
conditions independently assessed at the subsequent
Coroner’s enquiry were considered to be "excellent forecast
by world standards", the investigating Coroner did include
one recommendation concerning the terminology used in
weather forecasts and warnings to be specifically provided for
yacht-racing fleets. Since many yacht crews did not appreci-
ate how forecasts and warnings are to be interpreted despite
the various Bureau publications handed out to them prior to
the race, the recommendation called for the inclusion in fore-
casts and warnings of additional information on maximum
wind gusts and maximum wave heights likely to be encoun-
tered by yacht-racing fleets.
Consequently, the Australian Bureau of Meteorology
started to include the following safety preamble in the head-
ers of all marine forecasts and warnings: "Please be aware –
Wind gusts can be a further 40 percent stronger than the
averages given here, and maximum waves may be up to twice
the height". On the basis of surveys, this "disaster mitigation"
initiative has proven popular with mariners and the practice
continues today.
The information above is extracted mainly from the
Bureau’s "Preliminary Report on Meteorological Aspects of
the 1998 Sydney to Hobart Yacht Race" (Bureau of
Meteorology, 1999).
6.2 EXAMPLES OF PROJECTS FUNDED FROM
INNOVATIVE SOURCES
The following sections describe initiatives by several NMSs
to improve their warning system with funding and coopera-
tion from partners or organizations outside the usual funding
mechanisms. These examples may inspire other NMSs to
find creative and innovative sources for funding improve-
ments within their organization or country.
6.2.1 World Bank Loan to the Algerian Government to
Improve Flood Warnings, Infrastructure
Improvements, and Enhance the Relationship
with Disaster Response Agencies
The Algerian meteorological service, Office National de la
Météorologie (ONM), has been successful in securing fund-
ing through their national government for a World Bank
loan. This followed a devastating flood in 2001 in which
more than 700 people drowned and 24,000 were left home-
less. The funding will pay for, among other developments, a
meteorological component which will help enhance ONM's
capability to forecast such events and to make recommenda-
tions for infrastructure improvements - both technical and
personnel. This meteorological component is part of a
broader project, which will also help to improve the liaison
with other disaster response agencies to improve future
response to severe weather warnings.
6.2.2 Grant from the European Development Fund to
the Caribbean Forum of Africa, Caribbean and
Pacific States to Reduce Vulnerability to Floods
and Hurricanes
The Caribbean Forum of Africa, Caribbean and Pacific States
(CARIFORUM) has been awarded a grant from the
European Development Fund through the Caribbean
Regional Indicative Programme (CRIP). This grant is to fund
a project to reduce the vulnerability of the Caribbean region
in relation to adverse weather - primarily floods and hurri-
canes. Four new weather radars will be added to the existing
network of five. The composite images and data will be made
available to the participating meteorological services.
6.2.3 Central America Flash Flood Warning System
Funded by the United States Agency for
International Development’s Office of Foreign
Disaster Assistance
Following the catastrophic flooding of Hurricane Mitch in
1998 in Central America, the United States Agency for
International Development (USAID) provided funding for
the re-construction of damaged infrastructure. The National
Oceanic and Atmospheric Administration (NOAA) National
Weather Service (NWS) provided technology transfer, train-
ing, and technical assistance to the meteorological and
hydrologic services in the countries hardest hit (Honduras,
Nicaragua, El Salvador, and Guatemala). The USAID/Office
of Foreign Disaster Assistance (OFDA) also initiated a project
in 2000 to have NWS develop and implement a Central
America Flash Flood Guidance (CAFFG) system. This proto-
type system was developed with the Hydrologic Research
Centre, a public-benefit non-profit research, technology
transfer, and training corporation in San Diego, California,
United States. The purpose of this system is to provide oper-
ational meteorological and hydrological services in the seven
Central American countries (Belize, Costa Rica, El Salvador,
Guatemala, Honduras, Nicaragua, and Panama) with guid-
ance to provide flash flood warnings for small river basins.
This system became operational in August 2004.
The flash flood warning system uses real-time
Geostationary Operational Environmental Satellite (GOES)
imagery and in situ rain gauge data captured by a Digital
Direct Readout Ground Station (DRGS) downlink located in
Costa Rica. Hourly rainfall estimates are calculated on a 4-
km grid. Flash Flood guidance, which is the rainfall required
to produce flash flooding, is calculated daily for river basins
from 100 sq km to 300 sq km. A distributed hydrologic
model is run hourly to simulate flows and soil moisture for
the region. Graphical and text rainfall, soil moisture, and
flash flood guidance products are created and posted to the
Internet for access by the NMSs and disaster preparedness
response agencies in the seven Central American countries.
The next steps in this project are to evaluate the effectiveness
of this new approach and to continue with the training initia-
tives to ensure sustainability of the technology.
21 Chapter 6 — Successful initiatives NMSS
6.2.4 Examples from Hong Kong, China
The basic source of funding for meteorological service warn-
ing projects in Hong Kong is from the Hong Kong Special
Administrative Region Government (HKSARG). A few
recent examples in which projects are funded by other
sources are given below:
(i) Partnership with Higher Education Institutions on
Research Projects
The Hong Kong Observatory recently collaborated with
the Hong Kong Polytechnic University in the development of
real-time estimation of the integrated precipitable water
vapour (PWV) content in the atmosphere based on Global
Positioning Satellite (GPS) measurements to support the
operation of the rainstorm warning system in Hong Kong.
The project attracted funding support from the Hong Kong
Research Grants Council. The project started in September
2000 and the first phase was successfully completed in
December 2003.
Funding support has also been obtained from the
University Grants Council for partnership with the Hong
Kong University to jointly develop automatic identification
and tracking of tropical cyclone centres using radar imagery
in support of the Tropical Cyclone Warning Service. The
project has yielded fruitful results.
(ii) Cooperation with Engineering Departments and
Emergency Response Agencies to Promote Public
Awareness of Natural Disasters
Improvements in the drainage and slope safety systems,
as well as stringent building codes, have in recent years
successfully reduced the loss of life and property due to severe
weather. However, the awareness of potential hazards among
the general public has fallen.
To counter this decline in awareness, a large-scale public
education programme on natural disaster preparedness and
prevention was organized jointly by the Hong Kong
Observatory, engineering departments and other emergency
response agencies in Hong Kong. The programme aims to
promote public awareness of weather-related disasters and to
raise the community’s preparedness to deal with natural
disasters in Hong Kong.
The year-long "Safer Living – Reducing Natural
Disasters" public education programme commenced in
March 2005, prior to the rain and tropical cyclone season.
Activities included a tropical cyclone naming contest, exhi-
bitions, public seminars, open days, emergency operation
drills, carnivals, feature stories on TV and newspapers, etc.
Resources of the participating organizations were pooled
together to support this programme, without which such a
large-scale endeavour would not be possible.
6.2.5 Examples from France: Building a
Comprehensive Mainland Hydro-meteorological
Radar Network
One of the most prominent and recurring hazards in France
is flooding. Heavy tolls in lives and property were recorded,
for example in Nîmes (1987) and Vaison-la-Romaine (1992).
More recently, severe disasters (fortunately with a lower loss
in lives) repeatedly struck the south of France in 2000 (the
Aude département), 2002 (the Gard département) and 2003
(the Rhone Delta).
Recognizing the importance of a focused and integrated
approach to this kind of hydro-meteorological risk, Météo
France and the relevant ministry in charge of hydrology (now
the Ministère de l’Ecologie et du Développement Durable, or
the Ministry of Ecology) have established two five-year
programmes to install a comprehensive radar network,
thereby enhancing the existing network assembled by Météo
France over the years.
The first programme, completed in 2001, set up five new
radars in the most vulnerable part of France, the so-called
"Mediterranean arch". This resulted in almost all of the radar
coverage gaps in the region being filled and increased the
total number of radars to 18. The cost of this programme was
co-financed and shared by the two institutions involved, with
marginal extra funding coming from the European Union
(because of the international relevance of the coverage by
some of the radars).
As a follow-up of this first fruitful partnership, a new
programme (called PANTHERE) was launched for 2001-
2006 with new installations totalling 12.2 million Euros
(almost US$15 million). The main part of the programme
will be another five new radars, completing the coverage both
in the southern and northern parts of the country. The
programme also includes components for research and
development in new technologies.
For implementation, the project is undertaken by a team
of Météo France scientists and managers, under the control
of a bi-partite board of the two main partners. Regional
branches of the Ministry of Ecology and of Météo France
are actively required to help the core team in gathering local
users’ requirements, soliciting additional funding, looking
for adequate locations for the radars; assisting in tendering
processes, and supervising the building of necessary infra-
structure. This mechanism is very effective: bulk tendering
for the equipment lowers the unit price; centralized manage-
ment by a highly qualified and experienced team ensures
good coordination and progress; while the local initiatives
generate interest from new partners. It has been so success-
ful that funds collected from local authorities, the European
Union, and a Belgian Ministry now account for one-third of
the total funding. As a result, a sixth radar can also be accom-
modated within the time frame of the present PANTHERE
programme, which is on schedule and due to be completed
in 2006.
6.2.6 Examples from Australia: Studies on Climate
Change and Tropical Cyclone Vulnerability
The Queensland east coast in Australia is threatened by one
or two tropical cyclones most seasons. Tropical cyclones typi-
cally form in the Coral Sea and cause on landfall severe
winds, storm surge, and extreme wave conditions in the
impact zone. In 1999, an opportunity arose to further study
the risks posed by these phenomena when the Queensland
State Government provided A$1M funding under the
"Greenhouse Special Treasury Initiative" through the
22 Guidelines on integrating severe weather warnings into disaster risk management
23 Chapter 6 — Successful initiatives NMSS
Department of Natural Resources and Mines. The Australian
Bureau of Meteorology in partnership with the Queensland
Department of Emergency Services was successful in bidding
for project funding (totalling A$265K). The study was titled
"Queensland Climate Change and Community Vulnerability
to Tropical Cyclones", with coastal engineers from the
Queensland Environmental Protection Agency subsequently
joining the project team.
When the study was partially completed, the Project
Management Team led by the Australian Bureau of
Meteorology assessed that there were insufficient funds to
complete the study. On this occasion, a successful bid for
supplementary funding (totalling A$80K) was made under
the Federal/State Natural "Disasters Risk Management
Studies Program". More recently, another successful bid
(totalling A$105K) was made under the same program to
undertake a similar study in the Gulf of Carpentaria where
the coast is also cyclone prone. Without the funding provided
under the "Greenhouse Special Treasury Initiative" and the
"Disaster Risk Management Studies Program", a "disaster
mitigation" study of this magnitude and complexity would
not have been possible.
Incidentally, the project has already won three presti-
gious awards – an Engineering Excellence Award in 2002 and
both Queensland and Australian Safer Communities Awards
in 2004. Both awards were shared with the project consul-
tants/contractors – Systems Engineering Australia Pty Ltd
and James Cook University (Marine Modelling Unit and the
Cyclone Testing Station).
I. REFERENCES AND USEFUL READINGS
Browning, K.A., Chalon, J-P, and Thorpe, A.J., editors, 1999.
Fronts and Atlantic Storm-Track Experiment, Quarterly
Journal of the Royal Meteorological Society, Vol. 125,
part C, No. 561, October 1999.
Bureau of Meteorology, 1999. Preliminary Report on
Meteorological Aspects of the 1998 Sydney to Hobart
Yacht Race, February 1999. (Available on the Internet:
www.bom.gov.au/inside/services_policy/marine/sydney
_hobart/contents.shtml)
Colorado State University News, 2004: "Hurricane season
review", Colorado State University. (Available on the
Internet:http://comment.colostate.edu/index.asp?page=display_
article&article_id=64147507).
Conseil Supérieur de la Météorologie, 2002. Colloquium on
"Climate and Health", March 2002.
Counter Disaster and Rescue Services, 2000. Disaster Risk
Management Guide: A How-to Manual for Local
Government, Queensland Department of Emergency
Services, January 2000.
Executive Office of the President of the USA, 2003. Reducing
Disaster Vulnerability Through Science and Technology -
National Science and Technology Council, Committee
on the Environment and Natural Resources, An interim
report of the Subcommittee on Disaster Reduction, July
2003. (Available on the Internet:http://www.ostp.gov/NSTC/html/SDR_Report_Reduci
ngDisasterVulnerability2003.pdf)
Federal Response Plan 2003, 9230.1-PL, January 2003.
[United States]
Lalande, Dr Françoise (under the direction of), 2003, Mission
d’expertise et d’évaluation du système de santé pendant
la canicule 2003 (Expert mission for the evaluation of the
health system during the hot spell of 2003), published by
la Documentation Française. (Available on the Internet:http://ladocumentationfrançaise.fr/BRP/034000558/000
0.pdf.)
Glantz, Michel H., 2003. Usable Science 8: Early Warning
Systems: Do’s and Don’t’s - Report of the workshop held
in Shanghai, China; October 16, 2003. (Available on the
Internet:http://www.esig.ucar.edu/warning/report.pdf).
Grazzini, F., L. Ferranti, F. Lalaurette and F.Vitard., 2003. The
exceptional warm anomalies of summer 2003. ECMWF
Newsletter no. 99- autumn/winter 2003. (Available on
the Internet:http://www.ecmwf.int).
Gunasekera, Don, 2004. Natural Disaster Mitigation: Role and
Value of Warnings, Outlook 2004 Speaker Papers,
Disaster Management Workshop session, Canberra,
Australia.
MeteoWorld, 2004: "2004: the year of the tropical cyclone",
World Meteorological Organization. (Available on the
Internet:http://www.wmo.int/meteoworld/archive/en/dec2004/t
ropicalcyclone.htm)
Mileti, Dennis .S, 1999. Disasters by Design - A Reassessment
of Natural Hazards in the United States, National
Academy of Sciences, Joseph Henry Press, 2101
Constitution Avenue, N.W., Washington, D.C., 20418,
USA.
Mileti, Dennis S. and John H. Sorenson, 1990.
Communication of Emergency Public Warnings – A Social
Science Perspective and State-of-the-Art Assessment,
Oak Ridge National Laboratory, ORNL-6609, Oak
Ridge, Tennessee, USA. (Available on the Internet:http://emc.ornl.gov/EMC/PDF/CommunicationFinal.p
df).
National Hurricane Centre/Tropical Prediction Centre, 2004:
"2004 Atlantic Hurricane Season", National Weather
Service, NOAA. (Available on the Internet:http://www.nhc.noaa.gov/2004atlan.shtml)
Proceedings of the Forum on Risk Management and
Assessments of Natural Hazards, 5-6 February 2001,
Washington, D.C., sponsored by the [United States]
Office of the Federal Coordinator for Meteorological
Services and Supporting Research, 2001. (Available on
the Internet:http://www.ofcm.noaa.gov/risk/proceedings/pdf/riskpr
oceedingsentire.pdf).
Rogers, D.P., 2005: Turning Crisis Management into Risk
Management – The Role of Weather Forecasting. In
proceedings of the seminar on "Safer Living – Reducing
Natural Disasters", Hong Kong, China, 2005.
Shapiro, M. and A. Thorpe, THORPEX International Science
Plan Version 3. WMO/TD No. 1246, WWRP/THOR-
PEX No. 2, 51pp.
Sorenson, John H., 2000. Hazard Warning Systems: Review
of 20 Years of Progress, Natural Hazards Review, Vol. 1,
No. 2, 2000.
World Meteorological Organization (WMO), 2002. Guide on
Improving Public Understanding of and Response to
Warnings, PWS-8, WMO/TD No. 1139, Geneva,
Switzerland.
WMO, 2003. Guidelines on Cross-Border Exchange of
Warnings, PWS-9, WMO/TD No. 1179, Geneva,
Switzerland.
WMO, 2004. Natural Disaster Prevention and Mitigation: Role
and Contribution of the WMO and NMHSs, WMO
discussion paper, 13 September, 2004.
Wilhite, Donald, 2005: Moving from Crisis Response to
Drought Risk Management: Shifting the Pradigm.
National Drought Mitigation Centre, International
Chapter 7
REFERENCES, FURTHER READING
AND USEFUL WEB SITES
Drought Information Centre, University of Nebraska,
Lincoln, Nebraska, USA.
Zillman, John W., 1999. The National Meteorological Service.
World Meteorological Organization Bulletin,Vol. 48, No.
2, pp. 129-159, 1999.
Zillman, John W., 2004. Social and Economic Impacts of
Weather-Related Disaster Management, Symposium on
Planning and Preparedness for Weather-Related
Disasters, 29-30 March 2004, Hong Kong, China.
II. USEFUL WEBSITES
Billion Dollar U.S. Weather Disasters, 1980 - 2003http://www.ncdc.noaa.gov/oa/reports/billionz.html
Central America Flash Flood Guidance (CAFFG) system
(available in English or Spanish):http://www.hrc-web.org/Whats_New/
Early Warning Systemshttp://www.esig.ucar.edu/warning/
HAZUS.org
www.hazus.org
NOAA Economic Statisticshttp://www.publicaffairs.noaa.gov/pdf/economic-statis-
tics2004.pdf
Partnership for Public Warninghttp://www.partnershipforpublicwarning.org
[United States] Federal Response Plan, 9230.1-PL, January
2003. Available online at:http://www.fema.gov/pdf/rrr/frp/frp2003.pdf
NOTE: This document will be absorbed into a more
comprehensive plan named the National Response Plan
[United States] Natural Hazards Centrehttp://www.colorado.edu/hazards/
25 Chapter 7 — References, further reading and useful web sites
The ratio La/Cp can provide a decision threshold for action.
Suppose that, on each occasion, loss avoided (La) is $10,000 and the cost of protection (Cp) is $4,000; then Cp/La = 0.4.
Consider the financial implications over 10 events for 5 different forecast probabilities. If action is taken on each of the 10 events,
then the total cost of protection (?Cp) remains the same at $40,000, but the total loss avoided (?La) increases:
So there is a break-even point (i.e. BCR = 1) when probability is 0.4 or P ? Cp/La. This becomes the decision point for
taking action. Emergency authorities can then put in place a whole range of options and strategies depending on the likely
impact of an event. These will start with relatively low-cost preparatory actions such as ensuring equipment and manpower
are available if needed and raising the state of awareness amongst responders. These may be initiated even with a low proba-
bility of occurrence because the BCR will still be high. Similarly, when confidence (probability) is high, more expensive
strategies can be brought into play.
Forecast
Probability
P
0.1
0.2
0.3
0.4
0.5
Number of
occasions
loss avoided
1
2
3
4
5
Total Loss
Avoided
?La ($)
10,000
20,000
30,000
40,000
50,000
Total cost of
Protection
?Cp ($)
40,000
40,000
40,000
40,000
40,000
BCR over
10 events
0.25
0.50
0.75
1.00
1.25
Appendix A
UNDERSTANDING PROBABILITIES OF OCCURRENCE AND
CALCULATION OF BENEFIT COST RATIO (BCR): AN EXAMPLE
doc_941774093.pdf