A Guide To GIS Applications
A Guide To GIS Applications
A Guide To GIS Applications
Version 1.0
This is version 1.0 of this guide, issued on November 30th 2005. Revised versions (as they
are published) will be available on the Emergency Planning College website
www.epcollege.gov.uk
The author
This guide has been authored by Dr Robert MacFarlane, Visiting Fellow at the Emergency
Planning College and Director of the Centre for Environmental and Spatial Analysis (CESA)
at Northumbria University.
Acknowledgements
A large number of people in a wide range of agencies supported the writing of this
document, supplying material for case studies, providing illustrations and discussing and
helping to formulate the ideas. There are too many to mention by name, and
acknowledgements of source are given where relevant in the text, but sincere thanks go to
all who supported the project.
Contents
Page No
1.0
Glossary of Abbreviations
Introduction
1.1
1.2
1.3
6
6
8
2.0
11
3.0
15
4.0
17
4.1
4.2
17
19
5.0
6.0
7.0
20
5.1
5.2
5.3
20
20
24
Introduction
Data, Information and Communication
Models of Decision Making
GIS: an overview
28
6.1
6.2
6.2.1
6.2.2
6.2.3
6.2.4
6.2.5
6.2.6
6.2.7
Introduction
The Key Functions of a GIS
Data Integration
Data Analysis (i): Querying
Data Analysis (ii): Spatial Analysis
Data Modelling
Data Mining
Terrain Analysis
Information Outputs and Cartographic Standards
28
32
32
37
38
45
46
47
48
53
7.1
7.1.1
7.1.2
7.1.3
7.1.4
7.2
7.2.1
7.3
7.3.1
7.4
7.4.1
7.4.2
7.4.3
7.4.4
7.4.5
7.4.6
53
54
55
56
58
60
60
62
64
67
67
67
70
72
74
75
7.5
7.5.1
7.5.2
7.5.3
7.5.4
8.0
9.0
10.0
11.0
77
79
80
80
80
82
8.1
8.2
8.3
8.4
8.4.1
8.4.2
8.4.3
8.4.4
8.4.5
8.5
82
83
84
85
88
90
92
98
99
100
Overview
Hardware
Software
Data
Data Quality Issues
The Significance of Metadata
Security, Confidentiality and Access to Data and Information
Copyright Issues and Licensing
Spatial Data Storage
Staffing and Training
101
9.1
9.2
101
105
Introduction
Issues arising from GIS Applications in Multi-Agency Operations
109
10.1
10.2
10.3
10.3.1
10.3.2
10.3.3
10.3.4
10.3.5
10.6
10.7
109
111
111
114
115
116
117
118
118
119
Introduction
Obstacles to Interoperability
Working towards Interoperable Systems
Web services
Metadata standards
Spatial Frameworks
Semantic Interoperability
Spatial Data Infrastructures
Commercial Off The Shelf (COTS) Software
Political Will
Future developments
120
11.1
11.2
11.2.1
11.2.2
11.3
11.4
120
120
120
120
123
123
Location-Based Information
Data Sources
Geodemographics
Remotely Sensed Data
Mobile Technologies
Concluding Comment
Appendix 1
Glossary of Terms
125
Appendix 2
126
Glossary of Abbreviations
5 Cs
3D
Three Dimensional
AGI
C2
CCA
CCTV
CEFAS
CD
COMAH
COP
COTS
DEFRA
Compact Disk
Control Of Major Accident Hazards
Common Operational Picture
Commercial Off The Shelf Software
Department for Environment, Food and Rural Affairs
DEM
DLM
DNF
DPA
EOC
EPC
FEMA
FOI
Freedom of Information
FoIA
FMD
GI
Geographical Information
GIS
GI/S
GML
GP
General Practitioner
GSM
HCl
Hydrogen Chloride
HTTP
IEM
IGGI
IRMP
JACC
KM
Knowledge Management
LAN
LRF
MAFF
NSCWIP
NYC
OS
OSAPR
OSS
OSLO
PC
Ordnance Survey
Ordnance Survey Address Point Reference
Open Source Software
Ordnance Survey Liaison Officer
Personal Computer
PDA
PSI
QA
Quality Assurance
RAM
RRF
SDI
SESMIC
SOP
SVG
SWIM
TOID
TOpographic IDentifier
UPRN
VMDS
WAN
WTC
XML
Section One
Introduction
Summary
This guide is intended to establish authoritative guidance on the application of
GIS in civil protection, to assist users in the specification, acquisition and
maintenance of a GIS and to stimulate debate in the user community about
the future development and application of GIS and related technologies.
The primary audience is anticipated to be staff in Category One responders
identified in the Civil Contingencies Act 2004, most notably in Local Authority
Emergency Planning Units. However, it is also suited to a much wider
audience as it assumes no significant prior knowledge of either GIS or civil
protection. The structure and style of the guide is such that it can be worked
through from beginning to end, dipped in and out of as required or used as a
reference source. It is, very deliberately, a wide ranging document that is not
restricted to technical issues, and the coverage of data, information and
decision making and interoperability issues are very significant.
1.1
The aim of this document is to provide authoritative guidance on the application of GIS to
managers and end-users operating in the joint, multi-agency civil protection environment in
order to:
1.
maximise the potential benefits of GIS to the process of planning for and
managing emergencies and disasters, thereby enhancing national resilience
to such events;
2.
3.
4.
This is a wide-ranging document that takes the perspective that GIS is a tool to generate
information from a wide range of different datasets. In common with any tool, effective use is
dependent upon the quality of what might be termed the raw material, in this case data, the
skills and insight of those that use it and the wider organisational context within which it is
employed. All of these issues are covered in this guide.
1.2
Those involved in preparing for, responding to, and recovering from emergencies have a
need for information. However, that need is more precise, for information that is relevant,
appropriate, accurate, timely and delivered in a form that is appreciable under their
circumstances. However, this need, or demand, for information is often only partially met as,
and when, it is most needed.
The quality of the response is only as effective as the reliability of the information
which is available (Neil Macintosh, local authority Chief Executive, speaking about
the Lockerbie disaster, 1988).
Although there are a range of issues relating to the nature and transfer of information (see
Section 5), Figure 1 illustrates the fact that demand for information, most acutely during an
emergency, accelerates at a rate far above that of supply. This leads to what may be termed
a demand-provision gap. In most cases this is not because the information, or at least the
data from which the information could be generated, does not exist, but because it is not
accessible at the point and time of need.
Box 1: GI and GIS
In some text books GIS is disaggregated, and this can be helpful:
Geographical the spatial key or location of features is central to data handling, analysis
and reporting, which sets GIS apart from other data base management systems.
Information without data and information GIS can have no role to play and good quality
data are critical if the results of analysis are to be reliable.
Systems at a basic level they are computer-based systems, but it is important to remember
that GIS are rarely personal technology, so an understanding of how organisations manage
data and use information is critical to understanding and achieving effective use of GIS.
More recently Geographical Information (GI) as a term has become more widely used in its
own right. GI handling has become much more tightly embedded into a wider range of
technologies than ten years ago and GIS as a term is being precisely defined as desktop
systems with a powerful range of functionality. GI handling technologies including, for
example, addressing software which is used by call centre operators who ask for postcode
and house number only, and indeed such technologies are instrumental in the increase in
both amount and quality of GI that is available for application and analysis in a GIS. The
term GIScience has also become widespread in recent years, and is defined as the set of
scientific principles that should govern the use and analysis of GI in GIS (see Longley et al.,
2005 in Appendix 2).
This is of course a generic issue, and one that is far wider than GIS alone, but the need for
information is the key driver for the development and implementation of GIS in Integrated
Emergency Management. The specific value of GIS is that many of the issues that need to
be considered in preparing for, responding to and recovering from emergencies are explicitly
geographical: roads, rivers, floodplains, industrial hazards, towns and cities are all
geographically distributed in a way that is of clear relevance to emergency planning and
management. In short, where things are matters a great deal if something may, or does, go
wrong there. GIS is a tool that enables us to account for geography, and geography is critical
in understanding, planning for and communicating hazards, risks and vulnerabilities.
Response &
Demand for
Information
Availability of
Information
Time
The guide also explicitly places GIS applications within the principles of Integrated
Emergency Management and the framework of the Civil Contingencies Act 2004. One of the
underpinning considerations in IEM, which is elaborated in detail in section three, are the 5
Cs:
x
Command the ability to effectively direct operations at levels from the strategic,
through tactical to operational;
Control the ability to ensure that directions are implemented in line with the
command instructions;
Co-operation the ability for individuals and organisations to work effectively and
efficiently together in pursuit of common objectives;
Communication the ability to derive and pass information between individuals and
organisations in such a way that:
o
The media are supplied with appropriate, suitable and sufficient information to
meet their requirements;
The public, affected businesses and other individuals and agencies are
warned and informed about the developing situation and any actions that they
may be advised to take.
8
It will be demonstrated in this guide that data, information and GIS have a critical role in the
effective discharge of these functions in preparing for, responding to and recovering from
emergencies. None of these functions, of course, take place for the first time in an
emergency situation and the Civil Contingencies Act and the associated regulations and
guidance focus to a large degree on preparing for emergencies.
Consider the findings of two reports of 2004:
The FBIs information systems were woefully inadequate. The FBI lacked the ability
to know what it knew: there was no effective mechanism for capturing or sharing its
institutional knowledge.
The 9/11 Commission Report, July 2004
We should never forget how important apparently dry looking systems can be and
we should never undervalue the people who administer them. The consequence
when these systems go wrong can be devastating.
Sir Michael Bichard, Press Conference on the release of the Bichard Inquiry Report,
June 2004
All of the processes of IEM are information hungry and much of the required information is
Geographical Information (GI). It is for this reason that GIS represents a significant tool to
decision makers at all levels in an IEM context, not only because GIS supports the effective
management of existing data, but also because analytical and modelling tools support the
generation of new information, and permit the integration of data from multiple sources. In an
information management context this is termed adding value or leveraging information; in
an IEM context it supports evidence-based decision making and the development and
maintenance of a Common Operational Picture, the cornerstone of a co-ordinated approach.
Staff and students were evacuated safely, using map-based instructions and a
geographical awareness of the layout of the building;
Local fire crews were dispatched using address and street-based routing systems;
Additional fire crews were dispatched, as risk assessments permitted, from other parts of
London;
Aerial thermal imagery of hot spots in the fire were digitally relayed from helicopter to
fire crews on the ground;
The search, by commercial firms using GIS, for appropriate and suitably located office
space to house displaced staff during the reconstruction work;
The decision to target publicity in the international press on the basis of coverage of the
fire: many overseas newspapers gave the impression that the University had all been
destroyed so students might look elsewhere for degree courses and where that coverage
had been highest, so the positive publicity also had to be highest;
The restructuring programme in the aftermath of the fire achieved positive gains in the
use of space, in part facilitated through space-planning software.
This example illustrates the geographical dimensions of a single emergency and this is
typical of emergencies in general. As there is a clear geographical dimension to
emergencies, GIS as a set of tools that enable planners and responders to account for this
dimension is of critical importance. This guide elaborates this point and sets out in detail how
to capture Geographical Information and develop GIS for effective and integrated emergency
management.
10
Section Two
Emergencies and Disasters
Summary
Emergencies and disasters usually have a very clear geography to them:
they happen in places or in areas, affecting other places or areas and the
severity of the impacts depends upon the spread of the impacts in relation to
the distribution of vulnerable communities, individuals, facilities, resources,
infrastructure and environments. Before moving on to GIS it is important to
consider the geographical dimensions of emergencies and Integrated
Emergency Management.
2.1
Emergencies can, by their very nature, be extremely diverse. Some of the key variables are:
x
x
x
x
x
x
x
As a consequence of this, there are widely varying requirements of planners and responders
at different levels of command, and within and between multiple agencies. For instance, a
rapid onset emergency such as a serious fire and chemical release demands rapid and
decisive action in a timeframe that does not necessarily allow for a highly detailed analysis of
potential consequences and the implications of different response scenarios. In contrast, a
creeping crisis or slow onset emergency, especially where there is prior warning of key
characteristics such as magnitude, severity, location and timescale, may permit a detailed
analysis of the various options for possible prevention, mitigation and response. Indeed the
case for detailed problem analysis and assessment of response options makes very sound
business sense. For instance, the School of Veterinary Medicine at Penn University in the
US reports that an outbreak of avian influenza in 1997 took several months and cost the
State of Pennsylvania $3.5 million to control, and this was before the University had
developed a functioning GIS for animal disease control. In 2001, when the GIS was
operational, researchers were able to identify the infected poultry flocks, identify surrounding
flocks which were at risk by virtue of their location and plan for the transport and disposal of
infected carcasses to minimise risk of further infection. The outbreak was controlled in less
than a month and at a cost of $400,000.
From a geographical perspective, different kinds of emergencies have different
characteristics, illustrated in figure 2 and table 2.
11
Type of Emergency
Example
A. Single Location
Fixed site
Industrial plant, school, airport, train station, sports stadium or city centre
Corridor
Unpredictable
B. Multiple locations
Multiple locations
C. Wide area
Large area
Whole area
D. Outside Area
External emergency
Causal Factors
Fixed
Mobile
Restricted
B
D
Spread of
Consequences
Diffused
12
Region
Examples / Characteristics
The cause and direct consequences are focused on a small area. The sudden slump of a
coal spoil heap onto a school in Aberfan, Wales in 1966 is an example of this. However,
the human consequences of such a disaster can radiate through social networks over a
wide area and be very long lasting.
An incident such as the nuclear reactor fire at Chernobyl, Ukraine in 1986 is one where the
causes of the incident are site-specific, but where the direct consequences (radiation
fallout) were international in scale.
The 1953 floods on the East Coast of England were caused by a storm surge combining
with high spring tides. Both of these were multiple, widespread causal factors and the
consequences were spread from the Humber to the Thames with over 300 deaths.
The South Asian Tsunami of 26th December 2004 was initially caused by a sub-sea
earthquake off the coast of Sumatra but the tsunami, itself the cause of the death and
destruction of property, was both fast moving and international in scale.
Health emergencies can vary from the highly localised (e.g. Legionnaires Disease
outbreak from a single cooling system, affecting a local community), through to outbreaks
of animal (e.g. Foot and Mouth) or human disease (e.g. SARS) which have the potential to
spread between countries and continents.
13
Figure 3: the relative severity of a gust of wind of 75mph across the UK as a whole
(Courtesy of the Met Office)
By the same token the impact of given level of flooding in a densely populated urban area
outweighs the impacts of a flood of equal magnitude in a sparsely populated rural area; this
is effectively common sense, but we still need tools to identify hazards, assess risks and
measure degrees of magnitude of impact, all of which have an explicitly geographical
dimension.
So, different kinds of emergencies need to be prepared for and responded to in different
ways and recovery from different types and levels of emergencies clearly poses different
types and scales of requirements. The UK model of Integrated Emergency Management is
intended to establish a framework to prepare for and have the capabilities to respond to and
recover from such a range of potential emergencies.
14
Section Three
Integrated Emergency Management
Summary
This section provides an overview of the processes of Integrated Emergency
Management (IEM) established in the Civil Contingencies Act 2004 and
associated guidance. IEM is based around six processes - Anticipate,
Assess, Prevent, Prepare, Respond and Recover each of which is
elaborated here.
3.1
UK doctrine for IEM identifies six processes. Note that these are processes and activities as
distinct from phases. They are:
1. Anticipate: knowing what might happen is important in being able to frame and scale
an appropriate response. Emergencies arise from either hazards (non-malicious)
which may be natural (e.g. severe weather) or human (e.g. industrial accidents) or
threats (malicious and deliberate) and their very nature is that they are more or less
unpredictable in detail. However, horizon scanning and effective anticipation of
hazards and threats is essential.
2. Assess: appreciating the spread, severity and consequences of anticipated hazards
and threats needs to be set within a risk assessment framework. Risk registers are
developed and maintained at the local, regional and national level and it is important
that they reflect the changing nature of hazards and threats and the nature of the
population, environment and national security context in their makeup.
3. Prevent: it is intrinsically preferable to prevent an emergency than have to deal with
its consequences. If an area that has suffered repeated flooding is assessed to be at
high risk of flooding on an annual basis and bankside engineering and floodwater
storage works have the potential to significantly reduce that risk, this is likely to result
over time in both financial savings and reduced potential for loss of life and damage
to quality of life.
4. Prepare: not all hazards and threats are foreseen and not all of those that are can be
prevented. It is therefore critical to have structures, processes and resources in place
to deal with emergencies and mitigate their effects. Central to this is emergency
planning which falls into development (creating, implementing, reviewing and
maintaining) and exercising and training processes.
5. Respond: emergencies are almost always responded to at the operational level by
one or more of the blue light emergency services. In the event of an incident that
requires a co-ordinated multi-agency response, a specialized (e.g. CBRN) response
or the rapid establishment of a higher level of command, procedures are established
to escalate that response in a way that is appropriate. This response will draw heavily
on established procedures, frameworks and resources that have been the subject of
training and exercising prior to a real incident.
6. Recover: although the involvement of the emergency services may be relatively
limited in time, the process of recovering from an emergency can take months or
years and there are effects, perhaps most notably those of personal loss and trauma,
that extend over decades. There are medical, site clearance, decontamination,
reconstruction, risk assessment, counseling and many other dimensions to recovery,
15
some of which will overlap with the emergency response phase, others of which
succeed it over varying timescales.
These processes are underpinned by eight guiding principles for emergency response,
which are:
1. DIRECTION clarity of purpose is delivered through a strategic aim and supporting
objectives that are agreed and understood by all involved to prioritise and focus the
response.
2. INTEGRATION effective co-ordination exists between and within organisations and
tiers of response as well as timely access to appropriate guidance and support.
3. SUBSIDIARITY co-ordination occurs at the lowest appropriate level; local
responders are the building blocks of response on any scale.
4. PREPAREDNESS all individuals and organisations that might have to respond to
emergencies are prepared and clear about roles and responsibilities.
5. CONTINUITY response to emergencies is grounded in the existing functions of
organisations and familiar ways of working, though delivered at a greater tempo, on a
larger scale and in more testing circumstances. Sustainability is a key issue.
6. COMMUNICATION good two-way communication is critical to effective response.
Reliable information is passed correctly and without delay between those who need
to know, including the public.
7. CO-OPERATION there is positive engagement based on mutual trust and
understanding to facilitate information sharing and deliver effective solutions to
issues as they arise.
8. ANTICIPATION there is ongoing risk identification, analysis and mitigation so that
potential direct and indirect developments are anticipated and managed flexibly.
Although these relate to IEM as a whole, these principles equally underpin the development
and application of GIS within the context of IEM. The following presentation and discussion
of GIS will identify a number of key themes, around anticipation, direction, preparation,
integration, leadership, communication, continuity and co-operation the principles are the
same.
16
Section Four
The Civil Contingencies Act
Summary
This section provides an overview of the Civil Contingencies Act 2004,
focusing specifically on issues around information sharing and co-ordinated
working.
4.1
The Civil Contingencies Act (2004), referred hereafter as the Act, together with
accompanying regulations and non-legislative measures, delivers a single framework for UK
Civil Protection to meet the new challenges of the 21st Century. It is a wide-ranging piece of
legislation and only the key elements are summarised here. For the act itself, the
accompanying regulations, issues in relation to the devolved administrations and guidance
consult www.ukresilience.info
Key to the Act is a definition of what constitutes an emergency:
x
x
x
It is important to note that the focus is on consequences rather than causes, so the Act
applies equally to events or situations that originate outside of the UK as it does for those
within UK boundaries.
Part 1 of the Act focuses on preparations by local responders for localised emergencies and
Part 2 sets out the means to establish emergency powers for very serious emergencies
which affect a larger geographical area. Part 1 of the Act divides local responders into two
categories (1 and 2) and different duties apply to each. Category one responders include:
Local Authorities
x
Local Authorities
Government agencies
x
x
x
Environment Agency
Scottish Environment
Protection Agency
Maritime and
Coastguard Agency
Emergency Services
x
x
x
x
x
Police Forces
British Transport Police
Police Service of
Northern Ireland
Fire Authorities
Ambulance Services
NHS Bodies
x
x
x
x
x
x
x
x
17
x
x
Put in place arrangements to make information available to the public about civil
protection matters and maintain arrangements to warn, inform and advise the public
in the event of an emergency;
Co-operation and information sharing;
Provide advice and assistance to businesses and voluntary organisations about BCM
(local authorities only).
It should be very clear that information and the effective and efficient flow of information is
pivotal to almost all of these duties.
Category two organisations include:
Utilities
x
x
x
x
Electricity
Gas
Water and Sewerage
Public communications
providers (landlines and
mobiles)
Transport
x
x
x
x
x
x
x
Government
Network Rail
Train Operating Companies
(Passenger and Freight)
Transport for London
London Underground
Airports
Harbours and Ports
Highways Agency
Health
x
Common Health
Services Agency
(Scotland)
These Category 2 organisations are placed under the lesser duties of co-operating with
Category 1 organisations and sharing relevant information.
The Act also establishes the means through which the activities of these responders are to
be co-ordinated at local and regional levels. The key groups that have responsibility and
authority to drive forward co-operation and information sharing in preparing for and
responding to emergencies are as follows:
Local Resilience Forums (LRFs)
The LRF is a strategic co-ordinating group which matches, in the anticipation, prevention
and planning phases, the strategic co-ordination group that is usually established by the
police during the response and recovery phases of a major incident. It is a senior group, with
a primary focus on co-operation and co-ordination. Outside of London, LRF areas equate
with those of Police Force Areas.
Regional Resilience Teams (RRTs)
RRTs (and the National Assembly for Wales) are established to ensure effective two-way
communication between local responders and central government. They ensure that
planning is co-ordinated and that local responders have the support that they need to meet
their responsibilities.
Regional Resilience Forums (RRFs)
The RRF is a mechanism for ensuring multi-agency co-operation at the regional level. It is a
body for facilitating and supporting rather than directing co-operation and it does not have
any statutory powers.
18
4.2
Information Sharing
Under the Act local responders have a duty to share information. This information will take
may forms, for instance describing capabilities, resources, processes, contact details for key
personnel. Only some of these will be spatial data and information, but these are critical in
IEM.
Information is shared between Category one and two responders as they work
together to perform their duties under the Act. Information sharing is a crucial
element of civil protection work, underpinning all forms of co-operation (Section 3.1).
The process of sharing information is crucial to the duty [for example] sound risk
assessment relies on obtaining accurate information about the nature of the hazard,
the probability of a hazardous event occurring, and the potential effects and impact
on the community if it does (Section 3.3).
Information sharing is necessary so that Category one and two responders are able
to make the right judgements. If Category one and two responders have access to all
the information they need, they can make the right decisions about how to plan and
what to plan for. If they do not have access to all information, their planning will be
weakened (Section 3.4)
Emergency Preparedness (2005)1
In sharing information the Act states that the initial presumption should be that all
information should be shared, although these are some exceptions to this. It is important that
these are set out clearly as uncertainty about roles, rights and responsibilities in this regard
is well known to be corrosive of attempts to foster information sharing for co-operative
working. Sections 5, 8 and 9 elaborate on these issues, with 8.4.3 providing some detailed
information on security, confidentiality and access to data and information.
www.ukresilience.info
19
Section Five
Data, Information and Decision Making
Summary
GIS are computer-based tools for supporting decision making. For the use of
GIS to be effective it has to be informed by an appreciation of issues around
data availability and quality and the way in which information is used in
decision-making, especially under the specific conditions of emergency
situations. This section provides a relatively brief overview of issues which
are critical to the efficient and effective application of GIS in different kinds of
emergencies and at different levels of incident command. Fundamentally it is
about informed decision making in an emergency context.
5.1
Introduction
Data and information are different things. Data are results, observations, counts,
measurements, locations, attributes and other basic descriptive characteristics of things and
places in a fairly raw format. Information is more processed, ordered, summarised,
selective and user-friendly with the intention of assisting correct interpretation. Typically
data are high-volume (a lot of records) whereas information is more tightly geared to the
requirements of a specific end-use. One of the key strengths of tools such as spreadsheets,
databases and GIS is their ability to transform, if appropriately used, data into information
that can be appreciated and acted on more readily. However, it is important to recognise that
data are almost universally imperfect, therefore the decisions that are based on them may
be misguided, and even when data and information are strong, decisions may still be
misguided. Evidence is also widely used as a term and it is defined here as something that is
created from information, through further sorting, selection, distillation or triangulation with
other sources. In this respect it is similar to the term intelligence; although specifically
associated with the work of the security and intelligence services, the term is also widely
used in contexts such as local government and regional development in a way that broadly
equates with information and evidence.
Data do not just exist, they are created. They are usually created with specific purposes in
mind, and for this reason they may be sub-optimal when evaluated for an unforeseen
purpose. As emergency management is to a large degree the process of dealing with
unforeseen incidents, this is especially pertinent in this context. Data created for the
purposes of asset management, public health, community safety or education may be
neither structured, appropriately detailed or attributed for the purposes of emergency
managers, but the reality is that we have to work with the available data, while also ensuring
that future data are more suited in quality, content, coverage and availability.
20
Relevant
Timely
Accurate
incident commanders and strategic issues that may be critical at the latter level could simply
confuse those at lower levels of command. GIS and allied technology, including mobile data
bearers such as the Airwave digital radio system, provide a framework for disseminating
information in such a way that it is appropriately targeted to the requirements of different
groups, and can support not just incident management and command information but also
the back-flow of situational information for tactical management and higher strategic
decisions. Too much information can be a bad thing, and the emphasis must be on fitness
for purpose, a theme that will come up again.
However, information does not just float between those who have it and those that need it, to
be used the same way irrespective of recipient or context. A whole series of processes are
involved in handling and communicating information, including:
x
x
x
x
x
Providing
Receiving
Summarising
Checking
Collating
x
x
x
x
x
Relaying
Capturing
Retrieving
Prioritising
Logging
x
x
x
x
x
Distributing
Responding
Editing
Filtering
Recording
These are not of course specific to GIS, and the primary point is that the construction,
transfer and application of information is far from simple.
Originator
Data
Onward
Communication
Processing
Analysis
Information
Information
Portrayal:
Message Sent
Interference
Message
Transmission
Interference
Decoding and
Interpretation:
Message Received
Recipient
Processed Amounts
KM
Knowledge
Information
GIS
Time Needed
KM however is premised on the condition that much information, and the knowledge based
on that information, is held (and often not very well organised) at the organisational as well
as the personal level. To be semantically precise much of what is defined as KM is in fact
focused on data, information and evidence, but as a set of organising principles for
governing efficiency and equity of access to such resources, it is indisputably a good thing.
Data
23
well as human systems to work seamlessly together to provide information as, when and
where required).
5.3
IEM is to a large degree about decisions, which are in turn about making choices. Effective
IEM is about making the right choices. Making the right choices is about (a) approach, (b)
information and (c) ability and authority to pursue the determined course of action.
There is of course an almost impossible variety of decisions, but some of the key types are:
x
x
x
x
x
x
x
x
x
It is clear that decisions relating to emergencies are of the more challenging type i.e. they
are complex, contingent, relate to multiple objectives that are defined by a range of groups,
they are commonly unstructured at the outset of an incident and a range of levels of
Command and Control are involved.
Decision-making has been studied extensively and from a variety of different perspectives,
including business management, informatics, sociology, economics and psychology. Early
work established what is known as the rational model of decision making. In this model
people are presented with a problem, they gather the relevant data to address it, analyse the
data as appropriate to generate supporting information, evaluate the different options and
then make what is the optimal decision under the circumstances. There are of course a
number of assumptions underpinning this, namely that:
1. the problem is clearly defined;
2. the required data to understand the dimensions of the problem are available;
3. the tools to generate information on the problem are available and correctly applied;
4. the different options are accurately identified;
5. what constitutes the optimal decision is understood and agreed upon.
Subsequent work has studied the reality of decision-making in different contexts, and while
the rational model remains attractive as the basis for making informed and considered
decisions under ideal conditions, the specific circumstances surrounding an emergency are
less than ideal and the assumptions outlined above may be invalid for the following reasons:
1. the problems may be multiple, developing rapidly and contingent on factors that are
not yet fully appreciated;
2. the data requirements are not yet fully appreciated for the reasons above, and even
core datasets may not be available to incident controllers due to inadequate
preparation;
3. the tools to translate data into information may not be available or the available staff
may not be sufficiently trained to make correct and effective use of them;
4. identifying options for tackling the incident is far more complex under pressure and
where the contingencies are poorly understood;
24
uncertainty
complexity
time pressure
a dynamic event that is innately unpredictable
information and communication problems (overload, paucity or ambiguity)
heightened levels of stress for participants, coupled with potential personal danger.
Decision-making styles under these conditions vary between individuals and organisations,
but a general distinction between an essentially intuitive approach and an analytical
approach has been identified. The intuitive or naturalistic approach is based on what feels
right, based on previous experience, training and personal assessment of the observed
circumstances and it supports rapid decision-making. The analytical approach is inherently
slower, more intellectually and resource demanding, but it is permits a fuller consideration of
the evolving situation, the resources available to address it and the risks associated with
different paths of action. Different styles of decision-making fit different processes of IEM
and different roles and responsibilities. A major emphasis in IEM is on preparing adequately
for a range of incidents that vary in their severity, location, contingencies, interdependencies,
consequences and time-space overlap with other incidents. Preparation cannot reduce the
complexity, time pressure or dynamic nature of emergencies, but it can permit controllers
and responders to better appreciate what is happening, what might happen and be able to
coordinate and communicate information flows that support effective decision making by all
parties, at all levels of control and response. Figure 7 updates Figure 1 to illustrate how
information needs to keep closer pace with the demand for information as both the intuitive
and the analytical styles require information, although the latter is much more informationhungry.
Response &
Demand for
Information
Availability of
Information
rr
Cu
itu
tS
n
e
n
io
at
Time
Figure 7: accelerating the provision of information to keep closer pace with demand
The AAPPRR processes of IEM do not all operate under the stress of emergency conditions
and the emphasis in Anticipating, Assessing, Preventing, Preparing and Recovering must be
on an analytical approach that is based upon data and information of the required breadth
and depth and an appropriate level of quality. Indeed a critical element of preparation
focuses on information as a resource that underpins effective response. In responding to an
emergency different requirements are observed at different levels of incident control, and
these are summarised in Table 3.
The picture is one where there are different kinds of requirements at different levels, but
these requirements are more likely to be adequately met if preparation focuses on the core
questions of:
x
x
x
Imagination, anticipation and analysis are key to this, not just in horizon scanning or in
contingency planning, but also in identifying what it is you need to know. At present
information is dangerously underrated by many organisations in relation to emergencies, and
this is something that has the potential to severely undermine their ability to conduct
effective IEM.
26
Level
Information outputs are fed upwards to maintain an accurate and relevant Common Operational Picture.
Decision-making may be characterised by an intuitive approach, based on problem-recognition from previous experience, training and exercising.
Goals at Bronze level may be single or fewer in number, but should be specific, clear and explicit.
Bronze commanders work within a functional and/or geographical area of responsibility to implement the tactical plan as defined by silver command. The
Bronze commander must have a clear understanding of the tactical plan and have access to information that is critical to its execution on the ground,
including the activities of other agencies that may be pertinent to their own goals and actions.
Information outputs are task-specific to Bronze level, concerned with maintaining situational awareness at Silver and Bronze and the upward transfer of
changing situational information that is of relevance to Gold.
Information requirements are more specific than those of Gold, focusing on hazards, vulnerabilities, risks and resources that shape the translation of
policy and strategy into practice.
Decision-making needs to identify and evaluate options which necessitates an analytical approach, although pressure or rapidly changing circumstances
may force an intuitive approach.
Goals at Silver level may be multiple and relatively general, although they should be clear and explicit.
Silver determines and directs the tactics of incident management within the strategy, parameters and with the resources defined at Gold level, which may
include resources from multiple agencies.
Information outputs are varied, including tasking, situational awareness to Silver and meeting public, media and political interests.
Information requirements are broad and relatively unpredictable, but Gold commanders should avoid the long stick intervention at lower levels of
command that too much and too detailed information may encourage.
Decision-making should be analytical, in-depth and broadly-referenced, making use of specialist resources, and being able to develop and maintain a
Common Operational Picture, identify and assess options and evaluate progress.
Gold determines policy, overall strategy, resource deployment and the parameters within which lower levels of command will operate. This will include
resources from multiple agencies if a single agency is identified as overall commander. This is most likely to be Police during a response, and local
authority during recovery. A multi-agency Strategic Coordinating Group (SCG) may be established at Gold level if required and is intended to
complement individual agencies strategic management structures and procedures, not replace them. The SCG also has a communication and
coordination function where an incident crosses boundaries and/or involves Government Offices, Lead Government Departments or Devolved
Administrations.
Gold has overall command and responsibility for an incident. In relatively minor incidents gold command may not be formally established, but is just
nominally identified.
Characteristics
27
Military terminology reverses Operational and Tactical in the hierarchy, so that Operational equates with Silver and Tactical with Bronze.
Table 3: characteristics of civilian levels of civilian3 command and control during an incident response
Bronze
Gold
Nomenclature
Silver
Strategic
Tactical
Operational
Section Six
GIS: an overview
Summary
This section provides an overview of GIS data models, tools and techniques.
Related concepts such as scale, the integration of data using a spatial key
and effective communication with maps are covered. The emphasis in this
section is on identifying and illustrating key functionalities of GIS but it is not
intended to be a detailed training manual of how to implement such functions.
6.1
Introduction
As set out in the introduction, geography matters to IEM: hazards are spatially distributed,
and generally very uneven in that distribution, vulnerable facilities are distributed and
clustered in space, and resources may be sub-optimally located to deal with anticipated and
actual emergencies. GIS are computer-based systems for the integration, storage, querying,
analysis, modelling, reporting and mapping of geographically-referenced data.
The most commonly used conceptual model for a GIS is the overlay model (Figure 8). At its
core, GIS stores layers of data that share a common mode of geographical referencing, for
instance the GB Ordnance Survey Grid or Latitude and Longitude. Because the layers of
data have a common geography they can be superimposed upon one another. These layers
are typically thematic, so one layer may be a terrain model, another may be hydrology,
another 100 year flood plains, another all residential properties, another roads and so on. By
holding each theme as a separate layer they can be analysed separately, and also in
combination with each other.
Figure 8: GIS depends upon layers of data being accurately georeferenced relative
to each other and an externally accepted standard framework such as the
National Grid of Great Britain (Image Courtesy of Bristol City Council)
28
For instance, epidemiologists might be interested, as the first stage of an analysis, in the
geographical distribution of patients with a specific disease. This might show a level of
clustering that requires further investigation, and the second stage might be to superimpose
other thematic layers which may have a causal association with the distribution of the
disease.
The classic example of this is
the work, 150 years ago, of
Dr John Snow who identified
cholera as being a waterborne illness. In September
1854, during the last great
cholera epidemic in Great
Britain, 500 people living in
the Broad Street area of
London died of the disease
within a ten-day period.
Dr Snow established that
cholera was a water-borne
illness and he arrived at this
conclusion by plotting the
location of each of the
deaths on a street map and
then adding (as a separate
thematic layer) the location
of public water pumps. Dr
Snow ordered the removal of
the handle of a pump that
was in the centre of the
cluster of deaths, and thus
ended the cholera epidemic.
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
http://www.ph.ucla.edu/epi/snow/mapsbroadstreet.html
29
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
Figure 11: exactly the area in the centre of Newcastle upon Tyne, illustrated using
1:250,000 scale raster map and 1:1,250 Vector LandLineTM data
Figure 11 illustrates a relatively small area of the centre of Newcastle upon Tyne, less than
1km from East to West. The image on the left is of 1:250,000 scale mapping. The image on
the right is vastly more detailed, illustrating with much greater accuracy the location of
buildings, football stadia, shopping centres, hospitals and roads. In the 1:250,000 map the
Hospital (the collection of buildings upper right in the LandLineTM map) has been generalised
to a small white square and the letter H. Roads in contrast, have been magnified to many
times their true size and consequently they have had to be relocated to fit them all in.
Nominal scale describes the scale at which a given mapping product was intended to be
used. Clearly the road atlas style map with a nominal scale of 1:250,000 can be displayed at
a scale of 1:2,500, but it yields very little of value and may in fact be misleading, yet the
LandLineTM data with a nominal scale of 1:1,250 would simply degenerate into an
indistinguishable mass of lines at around 1:10,000, yet it retains a great deal of value at
larger scales.
However, GIS can take us a long way past mapping and the visual analysis of spatial
features. Three key attributes of any record within a GIS database are:
x
What is it? This would be the defining and secondary attributes of the object or area.
A defining attribute might be whether the object is a residential building, an industrial
site or a nature reserve. Secondary attributes might then respectively be whether the
building is a private dwelling or a care home, whether the industrial site is a COMAH
site or whether the nature reserve is designated for its geological, flora or fauna
interest.
30
When do the records relate to? The temporal dimension of records can be highly
variable. For instance, some records have a start and an end date/time, some will
have a single date (e.g. date built or designated) and others may be cyclical or
periodic, such as opening or operating times.
Where are the objects on the ground? It has been estimated that approximately 85%
of corporate datasets are geographically referenced in one way or another. For many
organisations this is integral to their operations (for instance, in many utilities asset
management), whereas in others it is less direct or it may be incidental (for instance
GP practices where patient records are referenced to their home address). However,
even where the geographical location of individual records is not critical to the data
originator, when used in another context (for instance during the targeted evacuation
of those with specific health conditions) it becomes very significant.
The geography of records is thus significant in two ways:
o
31
Work Orders
Firearms
licenses
Census Area
Specific Site
Hazards
Planning
Zones
Occupants
Records
Street Trees
Land Use
Ownership
Nature
Reserves
Emergency
Dispatches
Parcel /
Area
Address
Service Bills
Flood Plains
Property
Rights
Fire Hydrants
Mastermap
TOID
Ownership
Landmarks
Inventory
Environmental
Sampling
X,Y
Coordinate
Occupancy
Key Holder
Incident
Locations
Addresspoint
OSAPR
Building
Street
Segment
Emergency
Calls
Road Class
& Speed
Responsible
Authority
Specific Site
Hazards
Traffic Flow
Restrictions
UPRN
6.2
This section outlines each of the key functions of a GIS, with examples. Further examples
appear in Section 7, which illustrates GIS applications in the different processes of IEM.
6.2.1
Data Integration
Data that relate to defined geographical places or features (points, lines or areas in vector
format data or grid-cells or pixels in raster format data see Box 5) can be integrated within
GIS, irrespective of their origin. This is of course a sweeping statement and a great deal
depends on the format of those data, but as a general statement this is true; where records
32
are geographically referenced to an accepted format they can be imported into GIS. It does
not always have to be the case, however, that individual datasets have to have a spatial
identifier themselves to be integrated into GIS. Figure 13 illustrates the way in which some
datasets with no explicit spatial location (e.g. table in top left) can be integrated, on the basis
of a common field, with others that have either a relatively inaccurate (e.g. postcode) or a
highly accurate (e.g. Easting and Northing grid reference) location. This joining of tables is
extremely significant in the context of a data creation and integration exercise.
ID
Parcel
Checked
111-000-101
Yes
111-000-102
Yes
111-000-103
No
111-000-104
Yes
111-000-105
Yes
Address
Postcode
Easting
Northing
1 High Street
AZ1 1QT
869123
253476
3 High Street
AZ1 1QT
869133
253478
5 High Street
AZ1 1QT
869143
253480
7 High Street
AZ1 1QV
869152
253482
9 High Street
AZ1 1QV
869164
253483
ID
Parcel
Type
Address
Postcode
111-000-101
Residential
1 High Street
AZ1 1QT
111-000-102
Residential
3 High Street
AZ1 1QT
111-000-103
Retail
5 High Street
AZ1 1QT
111-000-104
Residential
7 High Street
AZ1 1QV
111-000-105
Retail
9 High Street
AZ1 1QV
Postcode
Geodemographic
AZ1 1QT
C15
AZ1 1QT
C17
AZ1 1QT
C17
AZ1 1QV
D21
AZ1 1QV
C15
x
x
x
x
x
x
Health Authorities
Primary Care Trusts
Emergency Services
Ministry of Defence
Private Sector Companies
Regional Development Agencies
Data and information sharing is critical to the effective application of GIS to IEM. GIS
depends on data, and data that are fit for use. Sharing data is therefore pivotal to the longterm development of GIS in this field and much else besides (see sections 9 and 10).
The concept of a Common Operational Picture (COP) is significant. For agencies to work
effectively together as part of a multi-agency response to emergencies, each agency has got
to be appraised of the bigger picture. This is significant as (a) the perception of a specific .
33
Other communal
establishments associated
with high density
occupancy
Care Centres
Vulnerable Individuals
(e.g. physical & mental
health or mobility)
Deprivation data
Census data
National Parks
Cultural & Historical Sites
Country Parks
Recreation Sites
Service Units
Industrial Units
Hotels
Woodland
Farmland
Financial Services
Nature Reserves
Retail Units
Transportation (incl.
routes, depots, ports,
bridges, etc)
Groundwater
Centres of Employment
Hospitals
Geology
Vertical Structures
Telecommunications
Emergency Services
resource sites, service
boundaries and control
rooms
Terrain
Government Agency
Boundaries
Resources
Infrastructure
Natural Environment
Built Environment
and Economy
Community and
Demographics
34
Table 4: examples of thematic data sources relevant to emergency planning and management, illustrating the need to develop multi-agency
databases (Note: the categories are not of course entirely mutually exclusive, and are used indicatively here)
Transport-related hazards
Fire hazards
Slope instability
Pipelines
REPIR sites
COMAH sites
Storm tides
Hazards
incident will differ, both in control and on the ground, between agencies, and (b) because the
resources of those individual agencies are themselves part of the bigger picture.
Box 5: Raster and Vector data
Vector data are those which are stored as points, lines and areas (below right). Raster data
are those which are stored in a regular gird, or matrix of cells (below left in which U = urban,
F = forest, A = Arable and W = water).
urban
forest
arable
water
Bridge ID
BG0001
BG0002
BG0003
Built
1901
1959
2001
Type
Arch
Suspension
Beam
Lanes
1
4
2
Traffic Control
Lights
Toll Booths
None
Traffic Flow
(veh/hour)
50
25,000
5,000
Maintained By
County Council
Bridges Inc
County Council
The raster data format is now most commonly used for remotely sensed data, that is satellite
images or aerial photographs. Figure 15 illustrates a false colour (red shows vegetation, light
blue or blue/grey shows sealed surfaces such as buildings and roads and very dark blue is
water, with the North Sea visible in the upper right) SPOT satellite image. These data are
measured solar radiation that is being given off by the earths surface, and the data are
collected in raster format, with one value for each cell (or pixel) of 20m x 20m resolution. The
pixel-based nature of the data is clear when you zoom into a small area, also illustrated in
Figure 15.
The raster format is also commonly used in modelling data, where the grid is used to
identify, for example, the predicted value of an atmospherically pollutant. Figure 27
(radioactivity atmospheric dispersal modelling by the Met Office) clearly illustrates the grid
based nature of such model output.
35
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
Figure 16: OS StrategiTM data illustrating geographically precise nature of vector data
The spatial resolution (pixel size) of aerial photographs is much small than most satellite
data (1m x 1m or rather less is typical) although the resolution of satellite data varies
dramatically between sensors (the equipment mounted on the satellite which gathers the
data) as weather satellites have very different applications to modern systems such as
IKONOS (see section 11) which can produce data that are comparable with aerial
photographs.
36
In the New York City Emergency Operations Center (NYC EOC), in common with many in
the UK, including the Lothian and Borders Police Joint Agencies Control Centre (JACC) in
Edinburgh, a series of agencies with responsibility for emergency response are brought
together and a GIS capacity has been established. In the JACC information from the GIS
can be displayed on large wall displays. However, in the NYC EOC representatives from
each agency sit at a designated PC which provides access to an Intranet GIS. This GIS is
constantly maintained by EOC staff during an incident so that all agencies have access to
precisely the same information about the incident, its implications, the response and
associated issues. Although neither the EOC nor the JACC have access to real time data
from all relevant agencies (but see section 7.4.6 on Automatic Vehicle Location Systems),
the benefits from all agencies facing the same information, the Common Operational Picture
(COP), is widely accepted.
There are technical issues to be addressed to achieve a COP, but more complex are the
cultural and organisational issues which hamper effective inter-agency communication and
working at all levels of operations and command.
6.2.2
Data analysis is the process of generating added value from a dataset or a number of
datasets, and beginning to get to grips with details and causes. For instance, analysing the
relationship between hoax 999 calls, the demographics of an area and the distribution of
prolific youth offenders is an approach that can target school-based educational work by fire
prevention officers. Also taking risk as its focus, an analysis of the spatial distribution of
accidents over time, in relation to the spatial distribution of police, fire and ambulance
stations and travel time zones around those locations, is a technique to analyse the
efficiency and effectiveness of service provision and plan for the optimal re-location of
emergency service locations. See the case studies in section 7 for further details.
Data can be queried in two main ways in a GIS:
a) On the basis of location: records can be selected on the basis of where they are, and
this is unique to GIS. For example, Figure 17 illustrates a map of total hazard score
for an area in the Eastern United States. Overlain on the hazard map are the
footprints of all buildings in that area. Selecting those buildings that fall within the
highest hazard score zones and then accessing a table of their attributes is a simple
operation in GIS. As set out in Box 4 these records could then be integrated with a
range of other attributes such as occupancy details, telephone numbers and details
of any vulnerable individuals or groups located in them. This is returned to in section
7.
37
Spatial analysis is concerned with the patterns and associations that exist within and
between layers of spatial data, patterns and associations that might go unnoticed unless an
explicitly geographical perspective is taken. This is an extensive field and reference to a
specialist text book is recommended for further details (see Appendix 2), but the key
operations are set out here.
(a)
Overlay analysis
As the name suggests this is the process of superimposing one or more thematic layers
upon another. At its simplest level this can be a visual process, to see how the distribution of
one set of features relates to another. GIS also supports the combination and subsequent
analysis of layers of datasets. For instance, points can be appended with the unique
identifier of each polygon that they fall within. From this the number of points per polygon
can be calculated, and if each polygon is associated with a population figure, the rate per
1000 can also be calculated.
Polygon layers can also be combined, and a series of different operations are possible,
including spatial union (see Fig 18) and a cookie-cutter approach (see Fig 19).
38
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
39
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
Figure 19: Using one thematic layer as a cookie cutter to clip out only those features of
another layer which fall within its spatial extent
(b)
Boolean analysis
As has been identified, much of the power of GIS derives from its ability to combine spatial
and attribute queries. In a conventional database management system queries are usually
structured to identify records that meet defined criteria. A simple query might be to select all
records where Type = B or where Quality is > 27. More complex queries use what are
termed Boolean operators such as AND, OR and EXCLUDING. Simple set theory can be
used to illustrate the point.
A series of sets can be identified in the Venn
diagram illustrated right:
x
x
x
x
x
A combined approach can also be applied to identify records that both meet certain attribute
criteria as well as defined spatial criteria (e.g. identify all the schools with grounds large
enough to land a heavy helicopter and sufficient indoor space to act as a major casualty
clearing station that are also within 5km of the crash site but not within the probable
contaminated area). These kinds of queries could be answered without GIS, but this would
take time and resources that incident responders are unlikely to have. Time and resources
need to be allocated to develop GIS in advance of an incident to support a more efficient,
better informed and more effective response when it matters.
40
(c)
Buffering
Buffering is the process of creating a new
polygon feature around an existing feature,
or set of features. Buffers can be generated
around points, lines or areas and for areas
they may be either within the polygon,
outside the polygon, or both. Once created,
buffers or a defined extent (e.g. 200m or
4km) can be used in any of the other
operations described here, for example in
selecting features that fall within the
buffered polygon such as key holders within
an affected area.
Neighbourhood analysis
The spatial relationship between datasets can be analysed through overlay analysis as
identified above, but neighbourhood analysis permits a more interactive approach to define
features by their spatial relationship to other features. For instance, in Figure 22 all
properties falling within 200m of a river have been selected. Alternative approaches may
define all features that fall entirely, or partly within other features, or be within given
distances of other features.
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
Surface modelling
In many applications that deal with point data, there is an emphasis on understanding the
density or level of concentration of points over space. In many cases simple visual analysis
will be unreliable, as many points may be so closely clustered as to be indistinguishable, or
may be even be superimposed. Under such circumstances using analytical techniques to
convert the point layer into a grid-based (raster) layer (see Box 3) that illustrates density of
points per unit area gives a much more effective picture of the actual distribution, and
clustering, of the dataset.
Figures 23 and 24 illustrate this with an example of child pedestrian road traffic accidents
(RTAs) in a UK city. Each point in Figure 23 represents a single RTA involving a child
pedestrian over a ten year period. The overall pattern is relatively clear and the areas of
41
particular concentration can be relatively well discerned. Which areas are the worst, the real
blackspots, is not however very clear.
Figure 24: point locations of Child Pedestrian RTAs converted into a density surface
42
(f)
Spatial Statistics
In the example of child pedestrian RTAs above, spatial statistics could be used to identify
both the most highly statistically significant clusters of incidents, and also test for the
strength of association with other variables such as demographic profiles and census data.
This is a complex area and reference to a specialist textbook is recommended for further
details (see Appendix 2).
(g)
Network analysis
Network analysis focuses on the movement of objects (usually, but not always vehicles)
along an interlinked network of, routes or pathways (usually but not always, roads). The
analysis includes a number of parameters, including:
x
Connectivity: network analysis can only work for a network of roads that has been
correctly linked together. For instance, even the smallest error in data creation that
leaves one section un-linked to another will lead to the software assuming they are not
meant to be linked computers do not compensate for human failures. The network
also needs to be structured in such a way that features such as overpasses, bridges
and underpasses are correctly interpreted by the software so that drivers are not
erroneously instructed to leap off a bridge onto the road below!
Speed: travel time along a network is calculated as a simple function of distance and
speed. Therefore speed must be included as an attribute of each section of the road
network. Usually it is the speed limit that is recorded, although this can lead to
dangerously optimistic results for many applications bear in mind that fire engines
responding to a 999 call in urban areas usually travel at an average speed of less than
20mph. For most of the roads in an urban area the speed limit is 30 or 40mph, and
even then average actual speeds take no account of variations over the course of a
day, such as rush hour. Many data re-sellers enhance road networks to include
attributes such as likely travel speed at different times of the day.
Direction: most road networks are two-way but some streets are one way systems and
these need to be represented as such in a road network that is intended to support
network analysis, for obvious reasons.
Different software packages operate in slightly different ways in respect of how they report
results of analysis, but typically areas that can be reached within a given time (e.g. <5
minutes) can be identified and isochrones (contour type lines of equal travel time) can be
identified.
43
44
Although GIS is a set of tools which can be applied individually, the greatest leverage of
information is often achieved by using a range of such tools within an overall project to
analyse different dimensions of a given problem.
6.2.4
Data Modelling
Modelling is the process of generating additional elements from existing datasets. For
example a model of atmospheric pollution dispersion from a point source takes a number of
known or estimated facts, such as the location of release, the characteristics of the
compound, wind strength and direction and terrain characteristics to predict the likely
dispersion of the compound over space and time.
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
45
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
Figure 27: the modelled the longer term spread of Caesium-137 following the hypothetical
incident established in Figure 26 (Courtesy of the Met Office)
This ability to run different scenarios and assess their likely implications is pivotal to activities
associated with preparing for emergencies, but it is also significant in an emergency context
during which the implications of a shift in wind direction could be estimated and worked
through.
Models are based on assumptions and data that may be flawed, so their outputs should be
treated with a degree of caution, but if the data and the models are robust then the output is
of potentially great value to emergency managers. See the case studies in section 7 for
further details.
6.2.5
Data Mining
Although data mining is a highly advanced set of techniques for identifying anomalies in
extensive digital datasets, at a basic level it is a process of applying a set of queries to a
dataset, except it is done automatically and with specific purposes in mind. For example, an
increasing number of Police Forces in the UK use Automatic Number Plate Recognition
(ANPR), and once read each number plate can be compared with a set of records. These
records may, for example, be whether the car is associated with a wanted person or whether
it is untaxed. Boolean queries that include a spatial element can also be associated with this,
for example to narrow the query to be whether a set of number plates associated with
wanted people appear in a given area.
46
6.2.6
Terrain Analysis
Terrain analysis is a specific set of techniques which analyse the surface of the earth. The
purposes of such analysis might be to identify slope angle or aspect (both of which are
relevant to slope stability and hydrological modelling) or carry out what is termed
intervisibility analysis.
Terrain analyses are typically carried out on a Digital Elevation Model (DEM), which is a
raster data layer in which each individual cell is allocated a value which is the average
altitude of the land represented by that pixel (Figure 28).
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
Figure 28: a Digital Elevation Model of the Northumberland Coast and Hills
Intervisibility analysis can calculate whether one point on a DEM is visible from another and
also carry out more sophisticated viewshed analysis, heavily used in environmental
assessments of new developments such as windfarms, which can calculate the cumulative
visibility of a feature from the surrounding landscape. Precisely the same technique is also
widely used in communications planning, where analogue radios in particular depend upon
line of sight between transmitter and receiver, and mobile phone companies use this
technique for service assessment and planning.
47
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
Figure 30: Layering infrastructure data and draping aerial photography over a DEM
(Image Courtesy of Bristol City Council and Cities Revealed Aerial Photography
The GeoInformation Group5)
6.2.7
Maps are the primary output from GIS. However, GIS also support a range of other forms of
output, including:
5
www.CitiesRevealed.com
48
x
x
x
x
The emphasis is on graphical communication, and the adage that a picture is worth a
thousand words is worth bearing in mind. Visual images can interest and engage people
and also help them to understand relatively complex issues in terms that are accessible to
them. Some of the advantages of using images, including maps are that they:
x
x
x
x
x
x
x
map design (including animations and 3D graphics) is a complex field where a huge
range of decisions about the look, feel and accessibility of the map have been
automated. Making a map that has the desired effect, which should be a rapid and
49
Know your audience: just as a comic will fall flat if s/he misjudges the audience, a
map that is appreciated and understood amongst a technical group, may be
impenetrable, even misleading, to a more generalist group. Figures 32 to 34 illustrate
some basic map design issues, but this is the tip of a sizeable iceberg and reference
to Appendix 2 for further material is recommended.
Figure 32
Figure 33
Figure 32 is a simple map but one where the
key message of population density is rapidly
appreciable using an intuitive low to high,
light to dark colour scheme.
Figure 33 uses the same basic scheme, but
there are fewer categories which means that
some of the detail is lost. This may be an
advantage in making impressionistic
interpretation easier, but a disadvantage if
important distinctions are lost.
Figure 34
Figure 34 is a map of exactly the same dataset, but it is not at all intuitive: if a reader has to
pay very close and repeated attention to a key then much of the impact of the map has
immediately been lost. By failing to use a scaled approach, in favour of a categorical
classification, the map fails to be readily accessible.
IEM demands that agencies cooperate in preparing for, responding to and recovering from
emergencies, which means that data and information must be able to pass between
agencies in a way that they can effectively support joint-working. Section 10 provides an
overview of the issues around interoperability which is a multi-dimensional pre-requisite to
be addressed over the coming years to realise and manage the level of data and information
mobility that is required. One dimension of this is what has been termed soft interoperability
or the ability of organisations to work together towards a common goal. Another dimension is
that of semantics, or what data and information mean, as these may vary widely between
organisations. Differences in terminology, map design and expectations of these have the
potential to cause serious inefficiencies, or worse, on decision making.
50
In the UK there are no agreed principles or details for map design in an emergency planning
context; it is a case of locally or internal agency-agreed (or often not-agreed) colours,
shadesets, and symbols being unilaterally adopted. As one of the most basic principles of
map design is that the basic message should be understandable with only limited reference
to the legend, this clearly cannot be the case where expectations from within one agency
may not match with those from others.
Figure 35: Pre-defined symbolisation for areas within a desktop GIS (ESRI, 2004)
Figure 35 illustrates some symbols for colouring areas that are available within a widely used
desktop GIS, but these are part of a commercial product rather than a common position
within the UK. Two instances where common symbolisation has been agreed are in UK
nature conservation and the military. Templates for colouring land cover and vegetation
classifications are promoted by government agencies such as Natural England and accepted
by professionals in the field; although many of these colours are not intuitive or realistic they
have been widely adopted and are appreciated by relevant staff. The military imperatives
(and indeed the consequences of failure) to arrive at a common symbolisation for mapping
are self-evident and these have now been implemented at a NATO level to ensure
communication between international forces, although the antecedents of common graphical
symbolisation in a military context goes back two millennia to the Roman army.
Work in progress in the United States by the Federal Geographic Data Committee Homeland
Security Working Group is developing a common symbol set for GIS applications in
emergency management (see Figure 36). Although there is currently no equivalent in the UK
it is essential that common symbolisation is developed and that it is done through
consultation and reference to both international experience and standards and also local
circumstances. This is a critical issue where agencies work together and it becomes
especially acute in wide area emergencies and cross-border and well as inter-agency
operations.
51
Figure 36: proposed emergency map symbols for the United States
(Source: http://www.fgdc.gov/HSWG/index.html)
52
Section Seven
GIS Applications in Integrated Emergency Management
Summary
The basic capabilities of GIS have been identified in the previous section.
This section develops and illustrates these capabilities in the context of the
IEM model, broken down by processes. The emphasis throughout is on the
capabilities of GIS as a tool.
7.1
Although the anticipation and assessment processes are separate, they have a common
focus on hazards, threats, vulnerabilities, risks and interdependencies. GIS applications
primarily fall within the assessment process, in which identified hazards and threats are
analysed in their spatial context, which supports vulnerability and risk assessments. For
instance, Figure 37 illustrates different aspects of the demography of an urban area. The
four maps illustrate (a) the total population (which says relatively little about daytime
population or vulnerability), (b) the population over 65 years of age (which, all other things
being equal, are more likely than other groups to be at home Monday to Friday 9 to 5), (c)
the proportion of the population that is linguistically isolated (so may be unable to appreciate
aural warnings given in English) and (d) the proportion of people with mobility impairments
(who would need assistance in an evacuation scenario). Such information is extremely
valuable in the process of anticipating and assessing risks from various scenarios.
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
Figure 37: profiling an urban area for risk and vulnerability assessment
53
7.1.1
Assessing hazards is not just restricted to emergency planners; it is something that the
public and private sectors alike need to consider, and public-facing systems such as the
Environment Agencys Floodline system (see Figure 41) enables individual businesses and
householders to identify the hazards and assess the risks to which they may be exposed.
One example of a private sector initiative to assess risk comes from the insurer Norwich
Union. Their business interests make it essential that they charge insurance rates which are
proportionate to risk, and losses from flooding represent a major business concern for the
insurance industry as a whole. However, flood modelling for the UK has previously been
hampered by terrain data that are insufficiently detailed to discriminate between properties.
As a consequence of this, many homes and businesses which are at low risk of flooding may
previously have been refused cover or charged excessive premiums. Norwich Union
commissioned a digital map to help pinpoint and understand exactly which areas of the
country are at risk from river flooding.
Box 6: spatial interdependencies
IEM has to be concerned with interdependencies. One event may be caused by others, and
the severity of consequences may depend upon a series of other factors such as nature and
time of incident and proximity to vulnerable facilities. The nature of financial, industrial,
social, transportation, communications and other systems is that failures or emergencies in
one part of intermeshed systems has widespread implications elsewhere, and in ways that
may not have been precisely foreseen. The seriousness of the fuel protests of 2000 was not
in proportion to the numbers taking direct action against fuel supplies. In the same year the
Food and Mouth Disease epidemic caused serious financial loss for many farm businesses,
including those that were not directly affected. However, what was the effective closure of
public access to much of the UK countryside also had very serious implications for rural
tourism and leisure businesses these are interdependencies.
Many of these interdependencies are spatial in nature. For instance, in July 2001 in the US a
train carrying chemicals and paper products derailed in a tunnel in central Baltimore, caught
fire and, in the ensuing five days, caused a series of infrastructure failures and public safety
problems. The train leaked several thousand gallons of hydrochloric acid into the tunnel, and
the fire caused a water main to burst. More than 70 million gallons of water spread over the
downtown area, flooding buildings and streets and leaving businesses without water. The
fire also burned through fibre-optic cables, causing widespread telecommunication
problems. The fire and burst water main damaged power cables and left 1,200 Baltimore
buildings without electricity6. This is a very clear example of interdependencies that are
physical, geographical and information-related (the loss of fibre-optic cables).
GIS can help, in both planning and response, to analyse and visualise such potential
interdependencies that might not be identified with a solely non-spatial view.
Commercially available terrain data that Norwich Union were using to model flood risk prior
to this project had an error estimate of +/- 5 metres. Clearly, this is the difference between
no flooding and complete inundation of a 2 story building. Data were collected by an airborne
radar system which was used to produce a Digital Elevation Model, which shows the height
of the ground above sea level. This was then combined with a flood software model to
produce information on where floods are likely to occur and how far they might extend. The
Peerenboorn, J.P., Fisher, R.E., Rinaldi, S.M. and Kelly, T.K. (2002). Studying the chain reaction,
Electric Perspectives, January/February 2002.
54
new terrain data has an error estimate of +/- 0.5 metres, or 10% of the previous level,
illustrated in Figure 38.
Figure 38: an area near Shrewsbury mapped using the commercially available 5m contours
(left) and the new 1m contour data (right) derived from airborne survey
(courtesy of Norwich Union)
By investing in consistently accurate information on the height of land surrounding rivers
Norwich Union have the best available information about the risk of flood for insurance
purposes, which enables them to better understand whether a customer has been a victim of
a one-off occurrence or is at risk from potentially frequent flooding.
7.1.2
To develop contingency plans, prepare emergency responses and identify areas and groups
at risk of flooding this case study is intended to illustrate the way in which the implications of
different severities of flooding can be identified, prepared for and publicised. In this scenario
extremely high tides and rising temperatures have combined with forecast high rain,
following a period of heavy snowfall to threaten an area with severe flooding.
This illustrates the study area
for this scenario (the coastal
area within the box). Two
rivers run, over a relatively
short distance, from the coast
up to high ground, thereby
establishing a clear hazard
from rapid snow melt.
Much of the coastal area is
relatively low lying and
therefore at risk from coastal
flooding.
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
55
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
Such scenarios are extremely helpful in contingency planning, and the private sector, most
notably insurers, have invested a great deal in this so that premiums reflect the spatial
distribution of risk, balanced against the probability of different levels of flooding occurring.
7.1.3
Over the past decade on-line mapping applications have been developed in internet,
extranet and intranet sites for a wide range of agencies and across a diverse set of
application areas. They have proven to be effective as one way of communicating
information within organisations, between organisations and to the general public. Many online, or web-based mapping applications portray spatial information that is relatively static, or
is changeable only over relatively long periods of time. For example, electoral
representation, landscape character and the location of bus stops are all on-line in many UK
Local Authorities publicly accessible web-mapping applications.
Such applications can also deal with more dynamic situations and one heavily used example
of this is the Automobile Associations Travel Watch Service, which provides an up-to-date
view of roadworks and delays on the road network in both map and textual form (Figure 39).
56
The Highways Agency and the RAC also offer similar services, and the Highways Agency
have extended this to include details of scheduled future works on the major routes7.
Figure 39: the AAs Travel Watch Internet Mapping service, through which individual delays
can be interrogated to give further details (www.theaa.com)
In recent years a growing number of authorities have started to develop web-mapping
applications in areas of direct relevance to IEM. For example, the Centre for Environment
Fisheries and Aquaculture Science (CEFAS), website provides a mapping application that
permits access to multiple layers of marine environmental data that are of clear relevance to
coastal emergency planning, for instance in respect of assessing the risk of pollution
incidents8 (Figure 40).
http://www.highways.gov.uk/
http://www.cefas.co.uk/
9
http://www.nscwip.info/
8
57
epidemic of 2001, the DEFRA website that provided public access to maps of affected areas
was receiving up to 600,000 hits a day.
The Environment
Agency website10
provides a mapping
service for members of
the general public to
identify both current
flood warnings and
also the likely extent of
different severities of
flooding.
While this is
unwelcome news for
those who live or work
in the identified areas,
it is a necessary
element of preparing
for flooding
emergencies.
Surrey County Council, working in partnership with the Surrey Emergency Services Major
Incident Committee (SESMIC), has pioneered the use of web-based GIS as an element of
the Surrey Alert system. Surrey Alert is intended to provide spatially related information for
use in planning, training and responding to incidents. Designed to be used in conjunction
with existing emergency planning procedures, Surrey Alert is intended to be the definitive
source of spatial information during an incident, so that all responders can trust that the
information they are using is the most up to date, and accurate.
There are four main elements to Surrey Alert:
i)
10
http://www.environment-agency.gov.uk/
58
x
x
x
x
x
x
Figure 42 illustrates the publicly accessible flood-warning application for the whole of the
County and Figure 43 illustrates much more detailed, large-scale mapping, for a much
smaller area. The intention is to develop the GIS to become the basis of a Community Risk
Register as required under the Civil Contingencies Act.
It should be noted that web-mapping applications are not the same thing, in respect of
functionality, as desktop GIS. Specifically, the analytical capabilities of such systems (this
relates to currently available Commercial Off The Shelf Software - COTS - systems) are
weak and their main use is restricted to displaying and querying spatial information. Many
organisations have recognised that this lightweight characteristic of such applications is in
fact an advantage in certain contexts, and cost savings have been realised by replacing
desktop GIS with web applications for casual users (see Section 9 and Box 11). The
organisation benefits from reduced costs and the user benefits from a system that is usually
less complex to use and where there is no risk of over-writing or otherwise corrupting core
datasets, as access is usually read-only. A further gain is that versions of datasets can be
controlled more tightly in web-applications and users access to current (or approved)
datasets can be more easily facilitated and managed.
59
Preventing Emergencies
Case study: Integrated Risk Management Planning in Surrey Fire and Rescue
Service
GIS was adopted by Surrey Fire & Rescue Service to analyse the implications for service
delivery of changing crewing patterns at their fire stations. The question was whether
savings could be made without increasing the risk to people, properties and businesses
within the fire service area. Underlying this was the need to identify fire stations that could
change the role of their personnel from being dedicated to crewing a fire engine, to being
able to carry out more flexible roles. To achieve this way of working, it would need to shown
that effective operational cover could still be provided, even if the change in working took
place. The ability of GIS to plot isochrones, lines of equal travel time, and then to validate
60
these models with data of real times taken to reach certain geographical areas proved an
objective means of proving this risk management approach.
Figure 45: historical incident data illustrated as a hotspot map where red shows high
operational activity, blue shows low operational activity and fire engine icons show the
locations of fire stations
61
The data used for the IRMP process was already available in the organisation within an
operational database, which held details of all incidents attended by the organisation. A
degree of data cleansing was required, but as it was used for performance indicator returns
data quality was almost all fit for purpose.
The use of GIS within the IRMP process has been successful, and different crewing patterns
at certain fire stations within the organisation have now been implemented, leading to
efficiency gains across the service. Building on this experience SFRS plan to extends the
use of GIS initially for enhancing existing risk profiling within the county through integrating
various data sets including census data and incident data. IRMP has also strengthened
partnership and inter-agency working, particularly with the police and community safety
partnerships, to direct activities such as home fire risk assessments and smoke alarm fitting
(integral to emergency prevention), and also crime and disorder reduction initiatives in the
county.
Although this case study relates to Fire and Rescue, the principles of data analysis for risk
assessment and service assessment are highly transferable, and have been adopted by
other emergency services. In the ambulance service in particular, it is well established that
response time is crucial to the survival of trauma victims, and analysis of frequently occurring
traumas such as those associated with road traffic collisions has been influential in guiding
the spatial distribution of emergency services, at particular times of day, week and year. The
principle is the same: data analysis informs risk assessment which drives service planning to
meet defined criteria and standards.
7.3
Emergencies will happen and preparing for them is critical. The nature of emergencies is
that there are a range of unknowns involved (location, cause, magnitude, severity of
consequences, etc) and so the emphasis must be on building capabilities that are suitable,
sufficient and also flexible. The emphasis is therefore on planning for a flexible response.
Planning is core to preparing for emergencies, and there are three basic types of plan: (a)
generic (including generic capability or procedure), (b) hazard or contingency-specific and
(c) site or location-specific. These are detailed below.
a) Generic Plans: these plans define how an organisation should respond to any
emergency. They are not specific in respect of location, nature and severity, but they
are premised on the understanding that responding to many emergencies will require
the same initial steps to be taken in terms of staff alerting, resource deployment,
media relations and warning and informing the public. The role of GIS in generic
plans is relatively limited, but generic plans should offer guidance on how GIS should
be deployed in the event of an emergency.
b) Hazard or Contingency-Specific Plans: many events such as festivals, sporting
occasions or conferences give rise to a set of conditions that require specific plans to
be set in place. The combination of large numbers of people, a high level of
concentration, overloaded services and security considerations make it very likely
that any incident would be resource-demanding and complex to respond to. Although
planning is, by its very nature, based on assessments and assumptions, GIS has a
range of applications in risk assessment, service analysis and response if an
emergency such as a bomb threat, building collapse or disease outbreak was to
occur during the event.
62
An inner and outer cordon, the placement of which would be determined by the
nature of the incident and the local environment
Forward control point
Incident control point
Reception point for utility company staff
Establishment of internal traffic routes for emergency services
Casualty clearing station(s)
Ambulance loading point
Survivor assembly point
Helicopter landing sites
Rendezvous point(s) for responding personnel
Vehicle and equipment marshalling area
Body holding area
Media liaison point.
The precise, on-the-ground delineation of these zones and the location of facilities such as
decontamination, casualty clearing and access points will be determined by the arrangement
of factors including the spread of any damage or contamination, wind direction, terrain, road
networks, ground conditions, rivers and the availability of suitable buildings, all of which can
be held, interrogated and analysed within GIS. Achieving this capability does, however,
require that GIS databases are prepared and in place to help frame and direct a response at
short notice. At a basic level, access to digital (or indeed hard copy) backdrop mapping will
permit a rapid appraisal of an emergency scene and context, for instance to identify potential
helicopter landing sites. However, the advantages of an integrated GIS database with
63
multiple thematic layers must be acknowledged; in the context of helicopter landing sites, not
all overhead power lines are marked and only in the MoD over-printed 1:50,000 OS maps
are they all present and associated with height above ground level figures. This is just one
example of adding value through data combination.
Figure 46: Schematic organisation for incident scene management at the tactical level
(Source: Emergency Preparedness www.ukresilience.info )
GIS applications in preparing for emergencies are not restricted to developing a multiagency database for integrated emergency management, contingency planning and tabletop exercises, for they also include the ability to analyse the distribution of resources relative
to identified hazards, risks and vulnerabilities with a view to optimising response, something
that is at the core of the IRMP project referred to in 7.2.1.
7.3.1
The Met Office has developed a Severe Weather Impacts Model (SWIM) that is based on
multiple layers of spatial data and a knowledge-base that is distilled from extensive research
into the impacts of severe weather incidents. Severe weather is defined as events that lie at
the boundaries of what is normally experienced. These include extremes of temperature
(both hot and cold), precipitation and wind.
The Met Office is seeking to support emergency planners and responders in both the public
and private sectors in their ability to anticipate, plan for and respond to severe weather
events. It should be noted that although this model will predict impacts from flooding there is
no significant overlap with the Environmental Agencys work as the latter focuses on fluvial
(rivers) flooding while SWIM focuses on pluvial (directly from rainfall) flooding.
SWIM itself is comprised of a number of elements, which enable the prediction of impacts
from:
x Strong winds
x Dense fog
x Heavy snow
x Ice accretion
x Heavy rain
Scenarios can be defined to run within defined parameters (type and severity of weather
conditions, geographical area, and model outputs) and there are a number of information
outputs, primarily:
64
x
x
x
x
Figure 47 illustrates the system calculating the number of electricity company customers
who are predicted to experience a loss of power in the Midlands. The darker the colour, the
higher the number of those who will be affected.
Figure 47: predicted levels of power loss consequent upon severe winds
(Courtesy of the Met Office)
The results are also made available in tabular form, and can be aggregated to a range of
geographical units including postcode areas and local authorities (Figure 48).
Figure 48: tabular reporting of impacts from SWIM (Courtesy of the Met Office)
65
In addition to aggregated impacts information such as this, SWIM also supports more highly
disaggregated predictions for infrastructure damage. For example, Figure 49 illustrates three
layers of information:
1. forecast maximum gust speed
2. forecast direction of maximum gust
3. the location of specific high voltage electricity transmission lines.
Figure 49: predicted maximum gust speed and direction, overlain with vulnerable
electricity transmission lines (Courtesy of the Met Office)
The combination of these three variables, when combined within the model with a range of
additional characteristics of the transmission lines such as age, type and construction,
enables a more precise estimation of damage, and where it is likely to occur. Supported by
this information Distribution Network Operators can undertake planning to mitigate the
impacts of infrastructure damage, which is typically carried out in a very tight timeframe for
decisions.
This is by no means a comprehensive overview of SWIM, but given that average losses per
annum from severe weather range from 1.2 to 1.5 Billion in the UK, the significance of the
model to assess risks, evaluate scenarios, warn relevant authorities of predicted impacts
and support mitigation of those impacts is very clear. SWIM is an excellent example of the
knowledge-based analytical applications of GIS that individual planners and responders are
unlikely to undertake internally, but where the ability to assimilate such GI into decision
making is critical.
66
7.4
Responding to Emergencies
7.4.1
Introduction
If data are available to a GIS, staff to operate the system are present and trained,
expectations of the system have been appropriately managed and the organisational
structures to support inter-agency working are in place, then GIS can enhance the efficiency
and effectiveness of an integrated emergency response. Some of the key elements of this
contribution might include:
x
x
x
x
x
x
x
x
x
x
Scenario: an anonymous caller claims to a newspaper that ground up radioactive waste from
a hospital (Caesium-137) has been deposited in a certain reservoir that supplies a major
city. No further details are available and it has to be assumed, if this is genuine, that
contaminated water has now entered the downstream rivers and public water supply system.
The overriding objective in the short term is to prevent contaminated water from reaching
homes and businesses and, where this has already happened, inform those affected of the
actions they should take. In the longer term, contaminated water bodies and rivers must be
identified for appropriate controls to be implemented. From a nature conservation
perspective such as event would have limited implications unless the dose was massive, so
it would be allowed to flush through, but fishing, water sports and other activities would
need to be controlled.
67
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
Step 2: Identify all streams flowing from this reservoir, any potable water abstraction
points and they supply and associated sewerage plants.
Here data on water abstraction
points and sewage treatment
facilities that are associated
with the reservoir and the
outflow stream are called up
and mapped.
Abstraction and sewerage
facilities that are affected
would be closed down.
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
68
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
Step 5: Establish a cordon around the affected water bodies and streams.
With regard to animal health,
protection of the human food
chain and the control of
recreational use of the
contaminated stream until the
threat and its potential impacts
can be verified, a 500m buffer
is drawn to identify an
exclusion zone for livestock
and humans. This would also
be applied to the reservoir
itself.
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
69
7.4.3
Scenario: a fire at a COMAH site (hazardous chemical production site) has led to the
explosion of a fuel tank and the release into the atmosphere of an unknown quantity of
hydrogen chloride.
Step 1: task response crews
Vehicle Mounted Data
Systems (see 7.4.5) are used
to access site hazard
information by response crews
and relay situational
information back to incident
control.
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
70
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
71
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
In addition to these stages there would clearly be a host of other, much more detailed tasks
that GIS could support, perhaps involving very large scale mapping, including site plans or
Ordnance Survey MastermapTM, for instance.
7.4.4
Neighbourhood analysis: identifying features that meet defined spatial and attribute
queries can be used in risk analysis. For instance, in the case of Foot and Mouth
Disease (FMD) atmospheric transmission was a primary concern. If livestock on a
farm was confirmed as being infected then a query to identify all livestock farms
within a 10km radius would be a necessary first step in controlling the potential
domino effect of successive infections. In reality, during the 2001 FMD outbreak in
the UK, MAFFs GIS staff were supported by the Met Office who applied plume
modelling techniques to define likely airborne spread of the virus. All livestock farms
within that plume could then be identified through a spatial query.
72
Figure 50: the spread of FMD during 2001 (Source: DEFRA, Central Science Labs 11)
11
Buffering: this basic GIS tool is extremely helpful in defining potential areas of risk,
especially at the early stage of an emergency when little detailed information is
available. For instance, until plume models which account for a wide range of
atmospheric, topographic, environmental and subject-specific information (e.g.
survival rates of virus outside of the human/animal body) are available to emergency
managers, a simple buffer around infected premises or individuals is a sensible initial
step for disease control measures to be applied.
Network analysis: many animal disease outbreaks are managed through controlled
and carefully targeted livestock slaughtering. The disposal of carcasses is
problematic on many levels, and site selection itself is highly suited to GIS, but
transporting infected carcasses to disposal points requires careful planning, most
critically to avoid non-infected areas. This kind of problem is highly suited to route
http://footandmouth.csl.gov.uk/secure/fmdstatistics/spat.cfm
73
planning through network analysis, for example <identify route from Farm A to
Disposal site avoiding roads that travel within 5 miles of Farms Q to T>.
In addition to analytical functions of GIS, the ability to visualise datasets in map form is in
itself highly significant. For instance, during the 2001 FMD outbreak, small scale maps were
used at the national level for monitoring spread of the disease and for briefing purposes,
livestock populations and their movements were used by epidemiologists working to
understand and then manage the spread of disease, and mid-scale maps were used by the
army to identify potential burial sites. Large-scale maps of farm buildings were used for
briefing and recording the work of cleaning and disinfection teams12. Later in the outbreak,
when the outbreak was coming under control and the GIS element of the response had
developed significantly, the Ministry for Agriculture, Fisheries and Food (MAFF, now
DEFRA) intranet mapping facility was used across the Ministry for a range of applications,
including the issuing of movement licenses and responding to queries relating to restricted
areas.
7.4.5
Mobile GIS
GIS have been around, in one form or another, for approximately 40 years, although it was
the early 1980s before systems first became commercially available. For the vast majority of
that time the software was limited to essentially immoveable computers (mainframes,
workstations and desktop PCs). With the advent of laptop and notebook computers that had
comparable processing power to desktop machines, GIS could be mobilised for use out of
the office. Although this was most significant when laptops could be combined with GPS
units and other data logging devices, for most GIS users laptops only promised a change of
scenery rather than a dramatic shift in working processes.
It was with the advent of ruggedised
computers (which were typically far lower on
processing power than mainstream laptops),
quite closely followed by the emergence of
Personal Digital Assistants (PDAs) that
software developers began to offer mapping
software and lightweight GIS for field
applications.
this network and offering you some service-related information. By extension, mobile phone
users can request information from service providers on the nearest facility such as a
restaurant. Most commonly at the moment, location is calculated using the cell from which
the call was made. However, an increasing number of mobile phones now contain a GPS
receiver which permits such queries to be executed with much greater precision; the
technological boundaries between GPS, in-car navigation systems, PDAs, mobile phones
and digital radios are breaking down. Just as users can request information based on their
location, unsolicited information, potentially including emergency messages, can be sent to
all such devices within a given area.
As a final example of this, some years ago Cambridgeshire Police control room took a 999
call that was an automated emergency message giving a location at which there had been
an accident. The location was given in latitude and longitude. The corresponding OS grid
reference was worked out and a patrol car was dispatched to find an American car had
driven off the road. The airbag had inflated, triggering a call from an embedded phone that
was associated with a GPS. Being an American phone it was set to dial 911, but this was redirected to 999. The driver, a US serviceman, was recovered from the vehicle and survived
the accident.
The Fire Service has been developing Vehicle Mounted Data Systems (VMDS) over several
years. A range of models exist, but they have a common focus on permitting fire and rescue
personnel to access risk information relating to locations they have been tasked to attend.
The initial stimulus to develop VMDS was the death, in 1993, of two firefighters in
Herefordshire who were killed when a suspended ceiling collapsed on them; had the incident
commander known that the building was constructed in this way, the firefighters would most
likely not have been tasked to enter the building. As a result of this and following HSE
reviews of the incident, a programme of work to develop mobile Decision Support Systems
(DSS) to enable crews and commanders to access site-based risk information was initiated
by the Chief and Assistant Chief Fire Officers Association (CACFOA).
Durham Constabulary are developing the Airwave digital radio system as a data bearer to
enhance their Command and Control (C2) systems. Although Airwave has the potential to
support two-way data flows of many different types between control and field units, their
75
initial focus in developing the system has been to support an Automatic Vehicle Location
System (AVLS). A GPS transmitter in police vehicles is allied with the Airwave radio, and the
location of that unit is passed back to control every second, although more frequent updates
are triggered when the unit is travelling above 50mph or when the blue lights are activated
(Airwave coverage within Durham is in excess of 97%). These track data are archived and
each point on the route can be associated with a range of attributes including speed,
direction, whether the blue lights are on and, in the case of Armed Response Vehicles,
whether the box containing the weapons has been opened. This is of clear relevance of
post-incident de-briefing and retaining an audit trail for possible legal proceedings. In a
similar vein, local authorities are using the same technology with resources such as road
gritters where the recorded attributes for each point include time, level of grit remaining in
the hopper and rate of dispersion onto the road surface.
Although the flow of data from Airwave is mediated through the Co-ordinator system, the
data are mapped in real time onto the forces standard GIS software, Blue 8. This permits
staff to see the distribution of mobile units as they are at any given time, and each location is
associated with its call-sign, supporting ease of understanding and communication. Although
officers in mobile units do not have access to mapping in their vehicles, the ability of control
room staff to integrate a range of geographical datasets and to overlay these with the
locations of their mobile units has considerable potential for effective and efficient working.
For instance, if an area can be identified on the control officers screen which shows an area
that is known to be contaminated following a chemical incident, units in that area can be
almost instantly advised of risks and how they should respond for their own safety and in
implementing the required response.
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
Figure 53: Illustration of AVLS operating in Durham Constabulary on a medium scale map
(left) and a large scale map, illustrating the positional accuracy of the system (right). The
green diamonds illustrate current location of police vehicles and their callsign.
Control staff are also able to access emergency plans and procedures through the GIS and
a range of triggers for SoPs are embedded into the system, and updated as required. This
enables control staff to advise officers on appropriate actions if called to premises that are
known to be hazardous and then initiate a wider, multi-agency response as appropriate.
Trials to extend the same approach to individual officers with hand-held Airwave sets are
underway at the time of writing.
Durham Constabulary are in the process of implementing a digital C2 system which will
support tasking of resources on the basis of location (which unit is nearest), attributes (is it
the right kind of resource, the right number of officers and with the right level of equipment)
and availability (have the officers indicated that they are available for tasking). It is
acknowledged that implementing such systems, and placing them at the core of C2, will
require significant changes to business processes, but Durham Constabulary are working to
76
effect these changes to realise the effectiveness and efficiency gains that they promise. The
Force is also liaising with other emergency services and council emergency planning staff to
identify the most appropriate way to use the emerging framework and systems to support
multi-agency response to serious and wide-area emergencies.
7.5
The transition from emergency response to recovery is not sharply bounded and there is
strong evidence that a recovery which is initiated as early as possible will be more effective
and efficient than one which is effectively a separate operation in the aftermath of an
emergency response. Although it is not a neat distinction, recovery can be divided into short
and longer term processes.
Short-term recovery is focused on the restoration of what might be termed life-support
systems, notably the provision of water, food, medical assistance and shelter. It is at this
stage that overlap with emergency response is the clearest. Following closely behind these
most basic provisions are the restoration of electricity, sanitation and transport and
communications. Ensuring that resources are allocated effectively, equitably and efficiently,
within an overall plan, can be significantly promoted through the application of the GIS tools
that have been set out previously.
Longer term recovery can take months through to years and beyond, and many of the
requirements are not for physical reconstruction but for more human forms of support to
individuals and communities that are recovering from trauma and loss. In respect of physical
reconstruction, it is important to remember that many of the earliest public sector
applications of GIS were in land-use planning, an innately spatial activity, and the hazard
and risk landscape may have shifted through and after the emergency in such a way that
planning for re-development (and not necessarily a simple restoration) is required.
Many emergencies will be scrutinised very closely for issues of liability, and there may be
criminal investigations in some cases. Similarly, where there have been fatalities, extremely
rigorous standards for the recovery of human remains have to be observed. In such cases
the accuracy of recording and auditing is paramount and two cases where GI technologies
have been used in this way come from the US (Box 9).
Box 9: GI technologies in recovery operations
GI technologies such as GPS and GIS can offer those managing a recovery operation an
efficient and effective set of tools to record what has been found and where and then, if
required, literally track them as they are recovered. Two examples of this are (a) wreckage
recovery from the Space Shuttle Columbia in 2003, and (b) the clearance of the Ground
Zero site in New York following September 11th 2001.
Recovery of Space Shuttle Columbia
The explosion over Texas of the Space Shuttle Columbia, on 1st February 2003, created a
debris field over a wide area due to the high altitude at which the Shuttle disintegrated.
Three main agencies (the National Aeronautical and Space Administration, the
Environmental Protection Agency and the Federal Bureau of Investigation) had separate
reasons for needing to rapidly and accurately map the debris:
(i)
(ii)
concerns that the debris may be toxic and/or explosive and as such,
represented a public health risk
the requirement to rapidly recover the human remains of the astronauts
77
(iii)
(iv)
the need to accurately record the spread of debris for a forensic engineering
investigation
the fact that heavy rain threatened to obscure debris in mud and leaf litter if the
task was not accomplished rapidly.
To facilitate this mapping and recovery operation large numbers of individuals were given
GPS receivers for logging the location of debris. However, in retrospect, the fact that there
was insufficient training given meant that many units were set to different map projections
and co-ordinate systems, so many hours had to be spent converting field records prior to
data integration and mapping. This is a clear case of a failure to recognise the need for a
common standard and make a limited investment to converge on this at the start of a project,
causing disproportionate additional work further downstream. In many cases the use of
different co-ordinate systems was not down to personal decisions, but reflected the fact that
many US government agencies utilise different map projections and coordinate systems. In
the UK it is unlikely that there would be this level of divergence in spatial referencing,
although the map projection settings in GPS need to be consistent for data from different
receivers to overlay without additional transformation and reference to technical manuals
must be made prior to any data collection exercise.
The speed of map production was further hampered by the extremely slow printing of what
were physically large and complex maps this is a clear bottleneck in the use of GI in
emergency operations. Together with the need for achieving standardisation for a common
effort, based on awareness and training, these were the key lessons in relation to the use of
GI, from the Columbia recovery operation13.
Recovery Operations at Ground Zero
Aspects of the GIS operation in the aftermath of September 11th 2001 have been quite
widely documented and Greene (2002) in Appendix 2 brings together most of this in a single
source. It is well known that the emergency response to this operation was in many respects
extremely short, given the magnitude and severity of its impacts, and from the earliest days
the operation as a whole was dominated by the recovery effort. A specific characteristic of
this emergency was that the NYC EOC, which was located in the WTC complex, was
evacuated and subsequently destroyed when the building collapsed shortly after the attacks
themselves. It was then necessary to establish a temporary GIS operation which was
achieved in a short space of time through the combined efforts of core staff from the City
Authorities, the Office for Emergency Management and a wide range of other partners
ranging from Universities and Colleges, the National Park Service and the military, with
private sector support to ensure hardware and software was in place very rapidly.
A wide range of datasets were required, many of which were not readily available, some of
which had to be captured for the first time (see Box 10 on remotely sensed data sources in
this operation) and others of which where there were serious quality and compatibility
issues. During the recovery operation an extremely wide range of GIS techniques were
applied, including three-dimensional modelling of both the surface rubble pile and the many
sub-surface floors of the WTC complex, all of which were instrumental in the search and
recovery operations and the subsequent site clearance operations.
The site clearance operation itself was of course a massive undertaking, and the
requirement for an audit trail itself made use of GI technologies. For each lorry load of
rubble, which were taken outside of New York for sorting and disposal, GPS was used to
map the path of the lorries. This enabled records to be maintained in the event of
investigators requiring access to specific pieces of debris, the lorry routes were recorded in
13
Brown, S.C., Crum, S. and Foote, V.S. (2003). GIS and GPS Emergency Response Lessons from
the Space Shuttle Columbia Disaster, Journal of Extension, 41(4).
78
the event of any concerns around debris contamination and the process was managed as
efficiently as possible, with safeguards in place to prevent illegal sale or use of WTC debris.
It is also important to bear in mind that it is not just the public sector that is involved, and
private sector applications of GIS may include the verification and resolution of insurance
claims, where the interests of the individual are for rapid settlement and the private sector
company is ensuring that potentially fraudulent claims can be reliably identified.
In information terms placing a comprehensive picture of what has happened at the centre of
the recovery plan and implementation is critical. A number of key themes are identified:
x
Damage assessment and resource allocation: establishing the context for recovery
operations in the aftermath of an emergency is critical, and recovery operations
themselves need to be planned.
Revised risk assessment: the emergency may in itself indicate that the initial risk
assessment was flawed, and/or the new situation has introduced new hazards and
altered the risk profile.
Public Facing Systems: the requirement to keep the public, media and other
stakeholders informed does not end when the emergency services leave a scene
and the recovery operation gets fully underway.
7.5.1
Figure 54: QuickBird Satellite Imagery illustrating the Impact of the December 26th 2004
Tsunami in Indonesia Digital Globe14
In the immediate aftermath of the massive earthquake of 8th October 2005, centred in
Pakistan administered Kashmir, high resolution imagery was used for initial damage
assessments. However, in the absence of available corresponding pre-disaster imagery and
the low lateral displacement of buildings (which tended to collapse vertically) it was difficult
14
www.digitalglobe.com
79
to assess the appraisal of urban damage; on the other hand landslips were in many cases
very obvious and this assisted route status assessment in several cases (Personal
Communication with Mapaction http://www.mapaction.org/ ).
At a more analytical level, resource allocation decisions can be rationally supported through
the GIS tool box which can, for instance, help to define priority areas of greatest need and
then analyse the distribution of resources and their movement to the areas of greatest need.
7.5.2
In many emergencies risk of re-occurrence is not affected by the event itself. Although there
is an underlying trend towards increased incidence of flooding over time, the statistical
probability of an event is not changed by the fact that one such event may have just
happened. What can happen however is that new hazards are created, or exacerbated and
risk assessments must be open to this possibility.
7.5.3
It is often said that attention is paid when potential problems become actual problems and
until that time their significance goes unrecognised. A fundamental theme in this guide has
been that adequately resourced, data-rich, appropriately operated and managed systems
have the potential to effectively and efficiently support all aspects of emergency
management, but realising that potential requires adequate preparation i.e. do not leave it
until it is needed, because at that point it is too late. However, learning from the experience
80
of emergencies is critical, however well prepared the staff, systems and processes were.
The emphasis should be on ensuring that the information infrastructure itself is resilient in
respect of secondary and backup facilities, something that was notable by its absence on
September 11th in New York, when the NYC Emergency Operations Centre was lost in the
attacks and building collapses. Other lessons from 9/11 are covered in Section 9.2.
81
Section Eight
Acquiring and Implementing a GIS
Summary
This section provides an overview of the key issues involved in the
acquisition, effective implementation and efficient maintenance of a GIS.
Issues around data, hardware, software and staffing are covered, and in
addition the significance of metadata, quality, data security and confidentially
issues are introduced.
8.1
Overview
The potential gains of GIS in the context of IEM has been established, but how can these
gains be realised? For many users their organisations may already have invested in GIS,
and for others this may have been extended to the emergency planning function. If this is the
case then the following section should be reviewed to see if it addresses any shortcomings
that you may have identified. If you have no access to GIS capability, then it is intended to
serve as an overview of how, and where, to get started.
Magnitude of
Costs and Benefits
Clearly there are costs associated with GIS and hardware, software, data and training are
the main headings. Figure 56 illustrates the fact that costs tend to increase sharply at the
start of a project, while benefits (whether they are measured in financial or in more broadly
referenced terms) accrue more slowly.
Return on Investment
Benefits
Costs
Time
Figure 56: costs and benefits are apparent over different timescales with GIS projects
It should be borne in mind that costs will fall, and from the point they meet accelerating
benefits there is a return on investment, but there are costs such as licenses, hardware
renewal, data cleansing and data purchasing that are recurrent over various timescales. The
rate of benefits growth will slow over time, but the level of benefit will be maintained over
time as effectiveness and efficiency gains are realised.
Figure 57 provides a general overview of how an organisation should approach the
acquisition of a GIS, and it has a great deal in common with any procurement process,
82
certainly in the IT field. The figure is drawn from Longley et al., 2005 (see Appendix 2) and
reference to chapter 17 Managing GIS should be made for an elaboration of each stage. In
general terms, however, the process is self-explanatory.
Analysis of
Requirements
Specification of
Requirements
Evaluation of
Alternatives
1. Definition of
Objectives
Implementation
of System
6. Final Design
8. Shortlisting
2. Requirements
Analysis of Users
11. Implementation
Plan
7. Request for
Proposal
9. Benchmark
Testing
3. Preliminary
Design
12. Contract
10. Evaluation
4. Cost-benefit
Analysis
5. Pilot Study
14. Implementation
Figure 57: General model of the GIS Acquisition Process (Longley et al., 2005)
Although this section starts with hardware, software and data, the critical point is that the
emphasis in your approach must not be on equipment, but rather on embedding GIS as a
bundle of capabilities into the business processes of IEM.
8.2
Hardware
GIS software will run on most modern desktop PCs and dedicated workstations such as
those required 10 years ago are no longer necessary. If buying a new PC then as fast a
processor and as much memory (RAM) as possible is recommended. A substantial hard disk
drive is appropriate, especially if the machine is to be used standalone (i.e. not usually
connected to a network, with the consequent requirement that data are stored on the PC
itself), such as a laptop for field use. A second hard drive for backup and additional capacity
is also recommended.
Remember that the resilience of this element of the IEM structure is critical so the ability to
back up data (which will include data created during a project or incident, in addition to
basemap and pre-existing thematic data layers) must be considered. A CD writer for PCs
should be included in the specification and more portable devices such as USB Memory
Sticks that plug into USB ports are valuable and can rapidly transfer data, maps, documents
and other files between machines if there is no network facility available (servers and
networks are briefly discussed in Box 11). It should also be noted that wireless
communications, for instance through the Bluetooth system, can easily support PC to PC
communications and the establishment of a limited network, rapidly, easily and at low cost in
almost any operating environment.
83
A PC without a printer is of limited use. Although other forms of display, such as LCD
projectors and interactive display boards (Figure 58) are becoming increasingly widely and
effectively used, the demand for paper maps will always remain high. An A3 colour printer
that can also print at A4 and work at high speed in draft setting should be regarded as the
minimum specification. Above this size printers become increasingly cumbersome, although
A1 or A0 maps are ideal for wall display and briefing purposes, and maps of A1/A2 size are
well suited to annotation and planning work by groups.
8.3
Software
There is a wide variety of GIS software products on the market, although it is dominated by a
small number. It is not the place of this document to review individual products. For a
catalogue of government-listed suppliers consult the Office for Government Commerce
(http://online.ogcbuyingsolutions.gov.uk/).
84
8.4
Data
Data are key to GIS and data capture, assembly and integration are typically the most costly,
and time-consuming elements of any GIS project. A wide range of different types of data can
be integrated within GIS, including:
Points
Lines
Areas
Grids
Images
15
Orthorectification is the process of adjusting a vertical aerial photograph to account for the nature
of the terrain and distance away from the vertical below the camera. This enables aerial
photographs to be integrated within GIS and overlain with other layers of data.
85
Box 10: the use of remotely sensed data in the aftermath of the 9/11 Trade Center
attacks
In the aftermath of the terrorist attacks on, and subsequent collapse of ,the World Trade
Center Twin Towers on 11th September 2001 there was a critical need for information about
the site. The base mapping available for the area had become suddenly out of date and
there was a need for new sorts of information, including hazards on Ground Zero. Some of
these demands for data were met by using remote sensing of two types: (i) terrain analysis,
and (ii) surface temperatures.
(i) Terrain analysis: in the days following the
collapse of the Twin Towers there were
concerns that the rubble pile was very
unstable and that recovery operations were
at risk of a further collapse. To assess this
risk, a technique known as LIDAR was
applied to the site. LIght Detection And
Ranging uses an aircraft-mounted device
to gather height data at a high level of
accuracy that can be used to create
detailed terrain models such as the one
illustrated here of the Ground Zero site. In
addition to helping visualise the new
landscape of the site, this technique
confirmed that the rubble pile was not
undergoing any significant subsidence.
(ii)
These examples are not included here to suggest that such datasets would be a normal
part of an emergency response and longer-term recovery, but to make the point that a wide
range of data sources can be integrated within GIS for purposes that may be unforeseen
prior to an emergency. However, it should be borne in mind that, under Military Aid to the
Civil Community (MACC) arrangements a range of sensors not normally applied outside of a
military context may be available to responders. No details of these capabilities are given
here, but responders should be aware that such resources may be brought forward in an
emergency context and an approach under MACC that defines a particular need may initiate
this.
86
Backdrop and framework mapping (see Glossary) are usually commercially supplied and the
Ordnance Survey is the leading UK provider. Many thematic datasets are referenced to OS
products (see Figure 61) such as AddresspointTM , BoundarylineTM or MastermapTM which
facilitates sharing between departments and agencies as the spatial framework is common
and there may be a common attribute or identifier which can be used to integrate datasets.
StrategiTM
MasterMapTM Imagery
MasterMapTM
Crown Copyright. All rights reserved, Cabinet Office, License No 100038675, 2005
Figure 61: examples of Ordnance Survey Digital Map Products relating to Southampton
(Courtesy of the Ordnance Survey)
87
Additional relevant datasets from the OS include a Points of Interest dataset of over three
and a half million entries which can help to fix a location and they include details of features,
business and leisure sites and the OS Master MapTM Integrated Transport Network, which
has road routing information such as one way streets and bridge heights which are critical
for detailed route planning and service analysis. In addition to the OS there is a range of
private sector spatial data suppliers, most of which specialise in added-value products such
as historical maps, flood defences, coal mining affected areas, explosives sites and
groundwater vulnerability data.
8.4.1
Quality as a term is often used in a rather loose fashion, so a wider view of the relevant
issues is taken at the outset. The primary measures of data usefulness are:
(a) availability: data may exist somewhere in a form that is highly suited to a required
application, but if it is not available at the point and time of need it is effectively
useless.
(b) utility: data may be available, but it may just not be suited or relevant to the intended
application. For instance, if data on the distribution of infants of 12 months or less is
required and the only available data are for the population as a whole, you are clearly
looking at two different things and cannot safely use one to support decisions relating
to the other.
(c) quality: if the data are available and relevant to the application but for one or more
reasons (see below) are of a standard that is unacceptable to the required
application, it may even be worse than useless as the evidence that decisions are
based on may be misleading or otherwise erroneous (remember the term Garbage In
Garbage Out).
Having access to data does not in itself guarantee an effective and efficient GIS application.
Data quality as well as availability is a key issue to be considered. As previously discussed,
data and information are rarely perfect. However, how good data and information are can
only be judged in a specific context. There is no single, universally applicable, quality
standard for spatial data. Rather the concept of fitness for use is adopted. For instance, a
GIS that is used to establish a situational picture at a regional or national level does not need
to adhere to the same standards as a VMDS which has to direct fire crews to a specific entry
point and then locate hydrants and specific pipelines and silos on site. For the latter
application both detail and a high level of spatial accuracy are required.
Fitness for use is not a concept that has any relevance without context and it is the users of
information who ultimately determine this. In Figure 62 the whole cycle of generating
information from data must start with a specification of information requirements, and the
parameters of timescale, accuracy and quantity which will determine, in the technical sphere,
precise judgements about fitness for use. Figure 62 defines two broad realms, one with a
focus on end users and ensuring as far as possible that information is appropriate and valid
for the intended applications, which in turn rests on the other (technical) realm within which
the application requirements are translated into technical requirements and the required
quality (of data, process and information) can be determined.
88
Specification of
Information
Requirements
Data
Collection
Information
Dissemination
Data
Integration
Data
Analysis
Ensuring Data and
Analytical Integrity
and System Resilience
Ensuring Appropriateness
and Validity of Information
for End Users
Operational Level
Strategic Level
Type
Sources
Internal
External
Breadth of Sources
Frequency of use
High
Low
Precision
Highly precise
Resolution
Highly detailed
Time span
Spatial accuracy: in short, things need to appear on the map in the correct place. If
roads are not aligned with buildings and elements of the road network are not
completely joined up then information based on these data may be seriously flawed.
Attribute accuracy: even if buildings are located in precisely the right place, if they are
wrongly attributed or confused with other buildings this could have serious
repercussions for IEM, from anticipating and assessing through to responding and
recovering.
89
Temporal relevance: if data were accurate ten years ago it is unlikely that they will
remain as accurate today. Maintaining spatial databases is essential to ensure
quality remains appropriately high.
Metadata extent: metadata is information about data and it is important that they are
accessible to users and would-be-users of data. Without metadata, users cannot
make informed decisions about suitability, appropriateness and fitness for use. This
is elaborated in the next section.
8.4.2
If data are known or suspected to be flawed in some way, it is critical that anyone using, or
looking to use that data is aware of its shortcomings. Decisions made in the knowledge of
uncertain information are likely to be different to those where the information is (perhaps
wrongly) understood to be of high quality in all regards. As with other forms of data, it is
crucial that metadata are kept up to date. Metadata are often divided into Discovery Level
Metadata and Application Metadata16.
Discovery Level Metadata (DLM) is intended to support searches for data. On the premise
that you cannot hope to find what you are looking for unless details of existing / available
data are published in one form or another, DLM is effectively a summary of what the dataset
refers to, a description of how it was created and information to help people access and use
it. To support wide access, DLM should be accessible over the internet, or other networks
and searching of one level or another should be supported. Sufficient information to enable a
potential user to judge whether this meets their requirements should be made available. This
may be a sub-set of a full set of metadata records. As an example of this, in the UK the
Association for Geographical Information (AGI) operates a metadata search engine, the
Data Locator, which can be accessed at http://www.gigateway.co.uk/ (see Figure 63).
Application metadata is divided in some references to exploration and exploitation metadata, but
as both relate to using the data rather than finding it in the first place they are treated as one in this
overview. See the metadata section of Global Spatial Data Initiative SDI Cookbook
http://www.gsdi.org/gsdicookbookindex.asp
90
Title
Abstract
Date
Frequency of update
Topic category
West bounding coordinate
East bounding coordinate
North bounding coordinate
South bounding coordinate
x
x
x
x
x
x
x
x
Extent
Spatial reference system
Data format
Additional information source
Supplier + web address
Date of update of metadata
Dataset reference date
Dataset language
This is intended to flag up the significance of metadata and provide an overview of the
subject, and the IGGI Guide: Principles of Good Metadata Management provides an
excellent introduction to this subject18.
17
The UK GEMINI Discovery Metadata Standard is a defined element set for describing geo-spatial,
discovery level metadata within the United Kingdom ( www.govtalk.gov.uk ).
18
www.iggi.gov.uk/publications/index.htm
91
8.4.3
Data and information can be highly sensitive, either for commercial, personal confidentiality
or national security reasons. It can also be (potentially) commercially valuable, and the
European Directive on Public Sector information (PSI) has opened the door for public
agencies to realise that value. Freedom of Information (FoI) is also pertinent, and can be
used by members of the public to gain access to public-sector information. The context for
data and information sharing is relatively complex.
In this section the terms data and information are used extensively. The Civil
Contingencies Act refers to information sharing which also embraces data sharing. The
position taken in this guide is that information is created through the processing of data (see
section 5). As such they are closely related and in some contexts they are used
interchangeably. In some cases the terms geographical information or spatial data are
used as overarching terms for both. At the risk of being clumsy, this section refers to the
sharing of data and information for the sake of completeness.
As stated above, this is a complex area, and this section is limited to an overview of the main
issues, covering the following:
a)
b)
c)
d)
e)
(a)
Principles of e-Government
The Civil Contingencies Act and Sensitive Information
Data and Information Sharing Agreements and Protocols
Freedom of Information
Public Sector Information
Principles of e-Government
It is the last of these that is most pertinent here, and one of the principles that flows out of
this can be paraphrased as collect it once, get it right and use it many times. The re-use of
information, often in contexts for which it was not originally envisaged, is key to egovernment. Although there are restrictions on the legitimate use of information (see below),
and issues around the fitness for use of data in unforeseen applications and settings, this
principle is intended to drive wider use of information for evidence-based practice and this is
strongly applicable in an IEM context.
(b)
Chapter three of Emergency Preparedness19 deals with information sharing, and a familiarity
with its contents is strongly recommended.
As set out earlier, the initial presumption of the Civil Contingencies Act is that all information
that is required to meet obligations defined under the Act should be shared by category one
and two responders, and it also establishes that local responders have a duty to share
information. However, it is recognised that many agencies and personnel have reservations
19
available at www.ukresilience.info
92
about the sharing of information, so this section provides an overview of the roles, rights,
responsibilities and obligations in regard of IEM and the Act.
As with all data and information sharing issues, the first point to consider is whether vires
(legal powers) exist for the activity in question, including any data sharing which is a
necessary part of that activity. Section 2 of the Local Government Act 2000 provides local
authorities with the power to do anything (unless otherwise barred in law from doing so) to
promote or improve the economic, social or environmental well-being of their area. This
power is likely to provide the lawful basis for such systems, but issues of Human Rights,
confidentiality and compliance with the provisions of the Data Protection Act 1998 will still
need to be considered.
Examples of the types of data and information that need to be collated for the processes of
IEM have been set out in table four. It is very clear that this is a wide range of data that will
be held by an almost equally wide range of authorities, agencies and private sector
organisations. For various reasons these data may be regarded as sensitive and there may
be degrees of unwillingness of share them. Although the initial presumption is that
information should be shared, certain types of information may be subject to controls if its
release would be counterproductive or otherwise damaging. Emergency Preparedness
(Section 3.17) defines four different types of sensitive information which may not be shared:
i)
ii)
iii)
iv)
Where data are judged to be sensitive for any of reasons (i) to (iv) then they should not
usually be shared. However, these considerations should not reasonably apply to the vast
majority of the datasets relating to hazards, community and demographics, built environment
and economy, natural environment, infrastructure and resources. Where these sensitivity
considerations do not reasonably apply then sharing of data and information as it is required
for IEM can, and should, take place.
The Department for Constitutional Affairs observes that there is no need to always rely on
express powers (often referred to as information gateways) to enable data sharing. Data
sharing is usually incidental to the exercise of a function; it is a tool that helps achieve a
designated activity. If the powers are there to carry out a function or activity, then it is likely
that a power to share data can be implied (as long as the data sharing is clearly for the
purposes of achieving that function)20. The CCA defines the roles and responsibilities of
20
http://www.dca.gov.uk/foi/sharing/faqs.htm
93
Category One and Two responders and the power to share data and information for the
purposes of IEM flows from this.
The underlined section above is significant: the Civil Contingencies Act does not entitle
agencies to share, or seek to share, data without a clear rationale for why individual datasets
are required for IEM. As data sharing takes place to permit category one and two responders
to fulfil their responsibilities under the legislation, the data they are seeking to share must be
directly relevant to those responsibilities.
Perhaps the most significant piece of legislation in this regard is the Data Protection Act, and
it is also the legislation that causes the most anxiety amongst would-be data sharers. The
key aspects and implications of the DPA are set out below (with the relevant sections of
Emergency Preparedness in brackets):
a) The DPA is concerned with the sharing of personal data.
b) Personal data are those which identify individuals. Such datasets are held by the
public and private sectors alike, and the legislation exists to protect the rights of
individuals to privacy and, where data sharing is required, to operate in a way that is
both lawful and fair.
c) A tight definition of personal data is one that includes name, date of birth, address
and/or other attributes or characteristics that allow an individual to be identified.
However, the definition also includes potentially identifiable records in a dataset. An
example of this might be a postcode where just one person is resident under these
(albeit unusual) conditions the individual concerned can be identified.
d) The DPA does not prohibit the sharing of personal data (including potentially
identifiable records). Category one responders in public authorities must ensure that
any data sharing they undertake which involves the processing of personal data has
a lawful basis, whether explicit or implied. An example of the need to share personal
information in an emergency context would be the requirement to know the location
of kidney dialysis patients during a water outage; if social services cannot provide
information that identifies individuals to the emergency managers then the care of
these vulnerable individuals cannot be assured.
e) The first data protection principle requires that personal data are processed fairly and
lawfully. In short, in order for personal data to be processed fairly, individuals must be
informed of the purposes for which their personal data are to be processed including
details of to whom it may be disclosed and given any further information that in the
circumstances would be necessary in order that the processing would be fair. This is,
of course, assuming that the information can be lawfully shared in the first place (see
section 3.28b).
f)
The second principle states that personal data should not be further processed in any
manner incompatible with the original purpose for which it was obtained. In the view
of the Department for Constitutional Affairs, the requirement of compatibility does not
have to mean "identical to" and provided the further processing is for a purpose that
is not contradictory to the original purpose or purposes, it will be consistent with the
second principle. Compliance advice can be found on the Information
Commissioner's website at www.dataprotection.gov.uk
g) Agencies seeking to share personal data should apply a necessity test; public
authorities must only collect and use the information that is legally permissible and
strictly necessary in order to fulfil their obligations in protecting the public in the case
of emergency. So, datasets which are not required to carry out activities in relation to
IEM, thereby executing responsibilities under the Civil Contingencies Act, should not
be shared.
h) In relation to (f) a dataset can be rendered non-personal, through anonymisation or
aggregation. For instance, if the only requirement for a risk assessment was the
94
number of pupils with physical disabilities in any given school then aggregated data
rather than full records should be shared. It should be noted that even records that
are aggregated in this way can be potentially identifiable, but if a necessity test
identified this level of aggregation as the minimum requirement, then this passes the
test. As another example, providing the name, age, sex, address and treatment
details of kidney dialysis patients for risk analysis is clearly not consistent with the
original purpose for which these data were collected. However, if the details of the
individual and/or their location are generalised in a way that retains their validity for
IEM this should be acceptable to both sets of interests. The basic issue is that where
there is a lawful basis for the sharing of personal data, only the minimum data that
needs to be shared should be.
Full Record Personal Data
Name:
Mr Peter Small
Age:
36
Address:
39 Main Street,
Midtown,
BB27 8XY
Telephone: 0111982
Condition: Kidney Failure, Weekly
Dialysis
Reference: BB7645HG981Q
Consultant: Dr C.Monk
Depersonalised Record
Condition: Kidney Failure, Weekly Dialysis
Sex:
Male
Postcode: BB27 8XY
j)
95
It is important that potential restrictions on the sharing of data and information are
considered and any conditions carefully adhered to, but the support of the Act for information
sharing as a cornerstone of collaborative working must be borne in mind and Category one
and two responders must be active in this regard.
(c)
One concrete step that can be taken to achieve transparency of aims, objectives and
processes is the development of data and information sharing agreements or protocols. Note
that Chapter three of Emergency Preparedness (sections 3.7 to 3.11) indicates that
formalisation of requests for information is not a preferred option, recommending that
existing and/or informal information sharing arrangements are pursued in the first instance.
This principle, that data and information sharing should not be over-bureaucratised, is
echoed here, but it is recommended that some form of written agreement (for instance a
Memorandum of Agreement) that sets out the purpose and specific requirements, quality
and technical standards and update frequency of data is established.
It is critical to note that a formal protocol or other form of agreement, however well received
by partner agencies, does not alter the legal requirements under the legislation outlined
above. It does, however, formalise what will be done and what roles and responsibilities flow
from this. Some key issues around data and information sharing agreements and protocols
are:
x
The responsibility for maintaining and updating the data should rest, and be
understood to rest, with a single agency (the data originator).
Protocols for sharing, providing and maintaining data must be at the organisational
rather than inter-personal level, and clearly-defined roles and responsibilities should
be established and developed between those organisations and relevant competent
staff.
The establishment of any agreement will not make unlawful processing lawful. The
reasons for such arrangements are to clear up uncertainty about what is and what is
not permissible, ensure openness and transparency and to ensure consistency of
practice.
Protocols make sure that everyone is clear about their data exchange responsibilities
and liabilities. They promote trust between partner organisations and the public.
Agencies do not necessarily need a separate protocol for each area of work it may
suffice to have one protocol that deals with all circumstances where data and
information are exchanged.
Section 3.41 of Emergency Preparedness proposes that Category one responders should,
wherever possible, seek to channel requests for information through as small a number of
routes as possible. So, if economies of scale in sourcing information can be realised (for
example in working through the LRF or RRF or even at national level. In addition to the
efficiency dimension, this can help to ensure consistency of process and data and
information across wider areas.
96
(d)
Freedom of Information
Under the Freedom of Information Act (FoIA) 2000, which came into force on 1st January
2005, anybody may request information from a public authority which has functions in
England, Wales and/or Northern Ireland21. The Act confers two statutory rights on applicants:
x
x
To be told whether or not the public authority holds that information; and if so,
To have that information communicated to them.
The Freedom of Information (Scotland) Act 2002 confers equivalent rights in Scotland22.
There are exceptions to the freedom of information and these are broadly the same as those
defined in the Civil Contingencies Act (i.e. considerations in relation to national security,
public safety, commercial sensitivities and personal information).
The FoIA is concerned with giving people a general right of access to information held by or
on behalf of public authorities, promoting a culture of openness and accountability across the
public sector. This should lead to a better understanding about how public authorities carry
out their duties, why they make the decisions they do and how they spend public money.
The development of public-facing systems that proactively make such information available
is of course consistent with the principles of the FoIA, but information released under the
terms of the FoIA must itself be consistent with other relevant legislation, including the Data
Protection Act.
(e)
Regulations on the re-use of Public Sector Information (PSI Regulations) came into force on
1 July 2005 SI 2005 no. 1515). They implement a European Directive on the re-use of
Public Sector Information (PSI) which was adopted by the European Parliament and Council
in 2003, and the regulations became binding on 1st July 2005. The directive recognises that
the public sector is the largest producer of information in Europe and that this is a resource
with considerable economic potential, which can be realised by re-use. In a sense, the PSI
regulations takes FOI a stage further in that it covers how the information is used once it has
been received. Re-use of PSI is different from FOI in that FOI is about access, on request,
to information whereas the PSI regulations relate to the dissemination of information to
others. It therefore has obvious copyright implications. This can be illustrated by the fact that
supplying information under FOI does not automatically confer a right of re-use. This means
that if a company that had received information under an FOI request would need to obtain
permission if it wanted to publish the information. Dissemination often takes place within a
market environment and PSI can be sold onto users with a range of intended purposes. For
further details consult the Office for Public Sector Information23.
In summary:
x
It is possible that Category one and two responders who are seeking to access data and
information for IEM purposes will be treated as customers in the reuse of information
21
http://www.foi.gov.uk/index.htm
22
http://www.itspublicknowledge.info/
http://www.opsi.gov.uk/
23
97
market and expected to pay. However, there are specific conditions relating to IEM which
mean that such requests should be exempted. These are that:
a) the Civil Contingencies Act is enabling legislation that supports and facilitates data
and information sharing;
b) the data and information are not, under the terms of the Act, being sought for
commercial exploitation.
As requests for data and information sharing are to fulfill legislative obligations and because
there should be no intention to pass on this data or information to any third party for
commercial gain, no charges for re-use of PSI should be levied. This is because such
activities do not constitute re-use under the PSI Regulations as the activities described form
part of the public task of the organisations concerned.
8.4.4
Copyright issues are different to those surrounding data protection, but they are both
concerned with proper and legitimate use of data and information. To summarise quite a
complex field, many spatial datasets, especially those which are purchased or otherwise
licensed for use in a given context, will be subject to conditions that prohibit or constrain their
use in other applications or by other organisations. Copyright and licence conditions will
always need to be consulted before granting approval to use data in a new application, or
context, or by a larger number of users. In the specific case of the Ordnance Survey, all
agencies should have a defined OS Liaison Officer (OSLO) who can advise on such issues.
Finally, most data suppliers require a licensing statement to appear on maps using, or
derived from their data (see, for example, OS maps in this document).
98
8.4.5
24
see www.knowledgemanagement.org.uk
99
8.5
People and their skills and experience are a critical part of realising a GIS that will
adequately meet an organisations needs. Public sector staff, including those in emergency
planning roles, that have an appreciation and experience of GIS are becoming more
common as the technology pervades ever more sectors. However, skilled GIS professionals
remain in relatively short supply. In terms of resilience, a system that is designed, developed
and used by a single, highly skilled member of staff, but remains a mystery to all others in
that organisation, is of little use if the operator is on leave and out of contact when it is
needed. Conversely, if an organisation undertakes to bring a larger number of staff up to
speed on the basic operation of a system, it is well known that regular use to retain and
develop that knowledge is required, and higher-level requirements may be out of their reach.
These are two extremes and appropriate training and continuing professional development
are the key to developing and diffusing competency in core skills throughout a body of staff.
100
Section Nine
Embedding GIS in, and across, organisations
Summary
GIS are most commonly implemented within single organisations, but the
reality of Integrated Emergency Management is that agencies need to work
closely and co-operatively together. This means that spatial data and
information must be able to move both within and between organisations, and
this section provides an overview of structures, processes and standards to
achieve this, along with a discussion of functional, organisational and
technical issues to be addressed to achieve this.
9.1
Introduction
One of the key benefits of using GIS that has been identified in a series of studies is the
ability to integrate data from seemingly disparate sources, in pursuit of a common objective.
This is accepted, yet realising this level of data integration can be problematic. GIS are most
commonly developed and used within organisations. The history of GIS is that most public
sector applications were specific to individual departments, for instance planning or facility
management. There is no doubt that designing, implementing, embedding and using GIS is
simpler within smaller organisations that have a unified and focused purpose for that
application. More recently, the emphasis has been on corporate or enterprise GIS where
the system is rolled out over the whole of an organisation25. The gains that can potentially be
realised from a corporate-level approach include:
x
x
x
x
x
x
x
However, designing and implementing a pan-organisational system and then ensuring its
effective uptake and application is far more complex than the process within a smaller and
more tightly focused unit. Different departments may be resistant to changes that the new
system requires and the potentially significant not invented here syndrome is selfexplanatory. These problems are compounded when a system is intended to tie together the
work of a number of different organisations. Competing conceptions of the costs and
benefits, data models, technical standards and organisational structures pose severe
challenges to data and systems integration for multi-agency working. However, the benefits
25
Enterprise GIS does not necessarily imply that all potential users have a fully-fledged GI System on
their desktop. Intranet mapping systems or customised, stripped down versions of the full software
installations has been widely and successfully implemented in many organisations, with financial
savings for the organisation and a simpler system to stay on top of for relatively infrequent users
(see Box 11).
101
of getting this right are considerable, broadly reflecting those observed above, but with the
enhanced benefits of a wider coverage and participation.
Box 11: GIS Architecture Overview
GIS can be configured in different ways, depending on a range of factors including the
requirements of users, the availability of resources and finance and the physical distribution
of users, including access to mobile devices and wireless connectivity. This box gives a brief
overview of the primary configurations.
1. Standalone PCs with desktop GIS software and data on local hard drives
In this configuration individual
desktop PCs have GIS software on
them and all the data that are
required are stored locally, on the
PCs hard drive or associated
peripherals. This is not incompatible
with the PCs being on a network or
the internet, but this connectivity is
not used to access or serve data from
other locations.
2. Networked PCs with desktop GIS software and data on centralised / workgroup
fileserver
In this approach PCs have desktop
GIS software installed on them, but
the data are centralised to one
degree (e.g. an entire organisation) or
another (e.g. a single workgroup) on
a fileserver. This server would contain
both read-only (e.g. OS backdrop
mapping) and read-write (e.g.
thematic layers for editing and
analysis) layers. In this configuration
DBMS
users can be differentiated with
different levels of access, or no
access, to some datasets if required.
3. PCs operating as Thin Clients, with Desktop GIS software accessed through an
application fileserver and data on centralised / workgroup fileserver
In many larger organisations desktop
PCs are used as thin clients and do
not have GIS software installed on
them locally. Instead, the application
is accessed through software such as
Citrix, which gives PC users access
to the software, but from a central
application server. This approach
can have financial advantages as the
number of concurrent users can be
limited. Typically, data are also
served from a remote location in this
configuration.
Application
Server
DBMS
102
4. PCs and Mobile Devices access mapping application and data through web browsers
from remote application and data servers
One of the constraints of the previous
configurations is that it they tend to
be specific to individual organisations,
with limited scope to offer the
application or datasets to third
parties. The development of webmapping applications, sometimes
termed Server GIS, enables remote
users (including mobile devices
accessing through wireless
connections) to access an application
Application
Server
through a standard web-browser,
which in turn accesses data from a
fileserver. The users interface with
the system is entirely through the
DBMS
browser, so the scope to interact
directly with the data (which has
issues for system security) can be
very tightly controlled.
Although these are presented as separate and discrete configurations, within an
organisation they may be combined for different types of users and application settings. For
example, some GIS projects have a high level of sensitivity especially in respect of data that
are held, so these may operate only on standalone PCs. Another variant might be of
configuration (2), where external network links to other organisations or the internet
complement data that are held on a local fileserver. Also, option (4) does not preclude
individual PCs, which may be used to access a web-mapping application over the internet,
also being configured with local GIS software and data (1) or as appears in (2).
Achieving evidence-based multi-agency working requires that a series of potential problems
are identified, and where relevant, addressed. These may include:
x
x
x
x
x
x
x
x
x
x
x
x
x
103
lack of leadership at a level appropriate to the project to ensure that progress is real,
appropriate and not just superficially overt.
For effective utilisation to be achieved it is vital that the process of [GIS]
implementation starts from an understanding of how particular organisations
operate in practice and not an idealised notion of how they should.
(Campbell & Masser, 199526)
This is a long and rather discouraging list of potential obstacles to effective data and
information sharing, but there are concrete steps that can be taken to achieve a much more
positive experience. Successful coordination projects have been characterised by:
x
A recognition that individual agencies information needs are at times best met
through other agencies data
Demonstrating the value of GIS and an integrated approach to relevant staff, leading
to active rather than passive, participation
Preparation
Response
Recovery
The data and information requirements to support these activities are wide-ranging and
sourced from a range of category 1 and 2 responders. Some form of data and information
hub is required to manage these for applications. In its simplest form, such a hub could be a
26
Campbell, H., and I. Masser. 1995. GIS and Organisations: How Effective Are GIS in Practice?
Taylor and Francis, London.
104
single computer hard drive, although networked access to such files are more appropriate
and longer term developments towards interoperability are elaborated in section 10.
During an emergency response, data and information are often generated very rapidly and in
considerable volumes. During the initial response phase, the majority of this will come from
the emergency services. Integrating such data and information with the hub that is available
for emergency response is problematic, but an effective multi-agency response demands
that such integration is pursued. For example, an emergency services control centre might
be able to capture and display incident information and the deployment of resources from
one or more of the blue-light services. If this was a major incident, such as an atmospheric
release of radiation, the Met Office might be requested to model the diffusion of radiation. At
the present time this information would be faxed through to the control centre, as required. In
the future, however, transmission in a digital format such as GML (see Section 10) could
result in rapid and precise integration of this within the command and control system. To
start to characterise the affected population and facilities however would require data from a
hub that may be located in a local authority.
There are short and longer term issues here. In the short term, the ability to supply data and
information as required to support an effective multi-agency response must be considered,
developed and rehearsed. In this model the hub(s) is/are physical entities and transfers have
to be managed, usually through manual processes. In the longer term, interoperable
systems that can automatically fetch data and information as required have to be the
objective (see section 10).
9.2
Up to this point, the treatment of GIS in organisations has been relatively generic, drawing
on case studies of what works and what does not work, from a broad range of application
areas. This next section explicitly focuses on evaluations that have been conducted in the
aftermath of GIS applications within multi-agency emergency, and humanitarian operations
including the World Trade Center attacks of September 11th 200127 and Civil-Military Cooperation following NATO intervention in Kosovo from 1999. There is a wide range of
material available on these projects, some of which is referenced in the appendix, and the
intention here is to provide a distilled version of the key points.
There is a high level of convergence between the case studies on what the drivers for GIS
applications were, although their specific circumstances were disparate. These included:
27
The need to develop situational awareness and a Common Operational Picture that
is accessible, both literally and culturally, to a range of organisations with differing
ways of working and missions.
The increasing acceptance of, and focus on achieving, cooperative and collaborative
working in pursuit of common goals.
An increasing acceptance of the potential of GIS to serve as the conceptual basis for
information sharing, planning, operational coordination and evaluation of progress
towards common goals.
See Greene (2002) and Kevany (2003) in Appendix 2 and the very thorough treatment in Kevany,
M. (2002). GIS in the World Trade Centre Attack. Critique: What was done? What can we learn?
http://www.plangraphics.com/publications/urisa2002papers/kevany_urisa.pdf
105
The need to be prepared across the range of functions and to include GIS and
related technologies in emergency plans.
The need to combine datasets from disparate sources across functional areas. For
instance, in Kosovo, once Serb forces withdrew and refugees began to return, GIS
facilitated the integration of data on minefields and other post-conflict hazards such
as unexploded ordnance (from military sources) with data on potable water and state
of the housing stock (from humanitarian agencies) to better manage the process of
resettlement.
The need for an electronic base map to become the cartographic basis for all
agencies operations. This relates to the pre-digital certainty that when things do go
wrong, it is going to be at the boundary of four (paper) map sheets and that different
responding agencies will have (a) different versions, (b) different scales or (c) no
versions of the base map for the area.
The need to anticipate needs for GIS applications, data and information requirements
through a familiarity with the cyclical, and to a degree predictable, nature of
emergency operations.
Key organisational issues to be addressed to realise the potential of GIS to support the
information requirements of emergency managers include:
x
The need to ensure that key decision-makers within and across agencies understand
what capacities exist that they can draw on, the range of standard information
products they can produce and the toolsets they have access to.
The need to develop a strategic information plan addressing the following issues:
Establish the data needs to meet the information requirements for IEM within
the context of the Community, Regional and National Risk Registers
Determine priorities: data and information sharing is complex and timeconsuming and a timed, staged and prioritised approach to establishing a
sustainable data and information sharing framework is required
106
28
Develop a database of who is doing what and where28 this relates in part
to the concept of a Common Operational Picture, although it can also be
advantageous in respect of potentially accessing emergency staff backup in
the event of a serious emergency.
Improve response time to supply information products at the point of need. The
extremely slow rate of map production during the Columbia recovery operation in the
US during 2003 has previously been identified in Box 9.
Standard products have both supply and demand side advantages: producers
can work to a standard template and users have a continuity of experience
they know what to expect and what to do with it.
Electronic media are the most efficient mode of information distribution for a
range of map users, but there will always be a strong demand for paper maps
and this needs to be met.
Paper map production (printing) has frequently been a bottleneck in the past
and needs to be adequately resourced.
The need for a request tracking system to be established: experience suggests that
the demand for maps and map- based information can be high from the outset and
then accelerate to unforeseen levels during a major incident. Meeting all requests
rapidly may be out of the question, so a prioritisation and tracking system needs to
be established, in part to ease pressure of the GIS staff themselves.
Ensure resilience under field constraints. Although a primary office facility may be
adequately resourced, backup sites or field locations may be below standard in
respect of office space, power availability, internet and network connectivity and
staffing. In respect of staffing, a (potentially reciprocal) system to draw appropriate
staff from other departments, agencies or agencies external to the region should be
agreed and established.
Off-site backup of data and resources is critical in the event that the
designated operations centre is affected in the emergency. Where data are
networked and accessible from multiple locations, it is critical that the data
themselves are backed up and accessible in an alternative manner.
Communications are vulnerable and alternative plans for voice and data
communications in the event of loss of primary systems is critical. This should
of course be part of the wider emergency planning process.
Key technical issues to be addressed to realise the potential of GIS to support the
information requirements of emergency managers include:
x
Develop common standards for data, metadata and information recording and
reporting.
Develop a comprehensive and consistent baseline dataset for the potential area of
operations.
Address the cross-referencing of frameworks, both spatial-spatial and attributespatial. For instance, during the WTC recovery operations GPS signals at Ground
Zero were weak and relatively unreliable due to the surrounding high buildings which
obscured and degraded the satellite signal. For this reason, the Fire Department
generated a 75 x 75 grid over the rubble pile for the recording of debris and human
remains and associated finds. Clearly, this spatial framework needed to be (and was)
linked to the wider spatial framework used in the Emergency Mapping and Data
Centre. Also during this operation, significant problems were experienced at the
outset of the operation as the digital outlines (footprints) of buildings were not linked
to the Building Identification Numbers (BINs), which were themselves used to
associate a wide range of attributes relating to those buildings.
Assess security implications relating to individual data layers and sensitivities relating
to the aggregation of data layers.
This is a long list of considerations, but given the significance of information in decision
making, the need to embed data access, management and information creation and
provision in emergency planning more broadly, these are all worthy of careful attention.
108
Section Ten
Working across boundaries: the significance of
interoperability
Summary
Interoperability describes the ability of systems to work seamlessly together.
To a large degree this is a set of technical issues which are covered, but the
most significant issues and problems to be addressed concern issues around
the meaning of data and information (semantics) and ensuring the political
will to ensure convergence around the standards upon which interoperability
is founded.
10.1
Introduction
The main focus in this document has been on the provision of appropriate information in
support of effective decision making. Much of this, it has been shown, depends on agencies
working together. In the context of information sharing, interoperability has been defined as
the ability to exchange and use information across different hardware and software without
special effort29. So, interoperability describes the ability of different organisations systems to
work effectively, indeed seamlessly, to achieve this. The Oxford English Dictionary definition
is Interoperable able to operate in conjunction, derivative interoperability.
As an important element of this, users
should also be able to search for data and
information and have an enhanced
understanding of what is available and for
what it might be used. To use an analogy,
this turns a set of unknown buildings with
books in them into a referenced library and,
to stretch the analogy, this library should
ideally have a catalogue that is searchable
over the internet and reciprocal lending
arrangements should exist.
The objective of this final section is to establish what interoperability is, why it is important
and establish what organisations need to consider now, to ensure future systems
interoperability.
29
Rose, M., Gabriel, M. and Jones, N. (2004). Achieving Interoperability and Information Sharing,
Report on Environment Agency Data and Information Exploitation Unit Seminar on Interoperability
and Information Sharing, February 2004.
109
Data should be collected once and maintained at the level where this can be done
most effectively;
It must be possible to seamlessly combine spatial data from different sources and
share it between many users and applications;
It must be possible for spatial data collected at one level of government to be shared
between all levels of government;
Spatial data for good governance should be available on conditions that do not
restrict its extensive use; and
It should be easy to discover which spatial data are available, to evaluate their fitness
for purpose and know which conditions apply for their use.
In respect of enhancing access to information these principles are important, and the role of
the internet in supporting portals for accessing metadata and data is at the heart of the
directive, both for public-private-government data flows and also for intra-government data
mobility. The Commission adopted the proposal for a Directive on INSPIRE in July 2004, yet
110
on both practical and political levels there will be many steps in the future to realise the
vision behind the directive itself.
10.2
Obstacles to Interoperability
The INSPIRE initiative defines principles rather than a description of reality, and there are
four essential obstacles to interoperability at the present time:
1. a great many local solutions have been developed for locally defined problems.
These solutions, however, are sub-optimal within the evolving context for effective
integrated working within and across agencies;
2. Interoperability between Commercial Off The Shelf (COTS) GIS software is limited,
although improving;
3. Communications interoperability and data transfer is problematic, although data
bearers such as TETRA (TErrestrial Trunked RAdio) promise the means to wirelessly
and digitally transfer both voice and data between systems in an operational
environment;
4. The significance of a shared understanding of data, its definition, limitations and
intended and potential applications is a major shortcoming, and a very serious issue,
although one that is introduced in less than serious style in box 13.
Do emergency planning, category one and category two agencies and their staff need to
consider these kinds of issues? To answer this with another question, do these agencies
and their staff need to be able to transfer information about hazards, resources,
vulnerabilities and emergency incidents and their consequences? The answer to this is yes,
and the Civil Contingencies Act establishes the significance of information transfer. It is
however too easy for such agencies to regard interoperability as the software developers
problem and carry on sharing (or not sharing) problematic data through manual means such
as a CD in the post.
The following kinds of problems are commonly observed in a GI/S context, and all of them
are symptomatic of a lack of interoperability30:
x
An inability to share maps over the web or other networks utilising web-browsers as
the viewer;
The absence of a common language to talk about spatial data and the attributes of
spatial data;
These may be problems with a strongly technical flavour, but their operational consequences
may be profound.
10.3
The objective of this section is to establish that the responsibility for working towards
interoperability rests with all agencies that have a role in preparing for, responding to and
recovering from emergencies.
30
So, up to this point we have established some of the cornerstones of how Category one and
two responders and other responsible agencies should regard, and start to work towards,
interoperability. At the risk of being repetitive, these are:
x
Benefits of interoperability are broadly the same as those which can be realised
through organising, indexing, permitting searching and preventing duplication of
records within an organisation it is inconceivable that an organisation would not
attempt to systematise its own internal approach in this way;
There are costs at the local, agency-specific level in addressing issues around
interoperability, but the costs of failing to address these issues are much more widely
felt and strong leadership is needed to take these steps;
Integration is not the same as interoperability, although the ability to integrate data is
a foundation;
112
Communications
Information
Data
Web Services: true systems interoperability requires the internet (or intranet or
extranet) as a framework for data transfer, applying standards for data interpretation
and application and developing open systems for processing data and reporting
information. This is collectively termed web services.
Metadata Standards: without metadata (information about data see 8.4.2), datasets
are just masses of records without any conceptual or technical application-relevant
details. Metadata permits users and would-be users to understand what a dataset is
about, what it can be used for and what potential weaknesses or drawbacks there
may be. When discovery-level metadata are replicated in a searchable form that is
independent from the dataset itself, this contributes to a powerful dictionary of
available data and is in itself a significant driver towards consistency for
interoperability. At a very basic level you have to know about a problem to be able to
address it, and publicising metadata can be very valuable in identifying, for instance,
differential quality standards, semantic inconsistencies and variable geo-referencing
approaches.
Semantic Interoperability: if two datasets contain a field that has the same
header/descriptor and all the records appear to be comprised of common categories
(e.g. extreme, high, medium, low, negligible) they would appear to be consistent.
However, if the two datasets derive from different agencies, each of which has
different ideas about, and thresholds between these severity classes, they are
clearly, and significantly inconsistent. Working towards semantic consistency is a
significant foundation in achieving interoperability.
113
Commercial Off The Shelf (COTS) Software: although the GIS marketplace is heavily
concentrated in favour of a relatively small number of software suppliers, a significant
number of projects in the past have seen the development of proprietary, or
customised software. COTS packages, in contrast, have a wider user-base, derive
from a larger development budget and have had to respond to pressure from users
for data format compatibility with other suppliers. As a general principle, COTS
software should be considered most appropriate for any GIS project unless there are
overriding reasons to the contrary.
Early development on mainframe computers placed them out of wide reach for
reasons of cost and complexity;
Development of Workstation and PC-based desktop systems reduced the cost, and
with the advent of Windows software, the interface complexity;
The basic toolkit of GIS, in common with other tools such as word-processing,
databases and spreadsheets, has grown in power and efficiency over the years;
Local and Wide Area Networks (LANs and WANs) enabled users to manage and
share data and work in teams more efficiently;
The advent of the Internet built on the gains of networking, but enabling wider
searches and links to be made between individuals, agencies and communities of
interest. Critically, information served over the internet was platform independent, so
it could be accessed irrespective of the hardware, operating system and software
profile of your computer;
The development of the Internet saw the advent of distributed computing whereby
not only data could pass between users over the Internet, but the use of remote
processing resources (i.e. other peoples computers) could be achieved, through the
appropriate protocols;
The development of Open Systems or Open Source Software which are wholly
transparent and may be developed or embedded into other applications without
license or copyright issues has been very significant in permitting the development of
systems that can effectively relate to each other;
The shift from physical networks to wireless networking capability enabled data flows
between, for example, a field worker who is surveying structural damage following a
storm and a base office which could receive data entered onto a PDA (Personal
Digital Assistant) and transferred wirelessly through a GSM phone or even a satellite
link;
The technical enablers identified above have driven increased expectations of realtime data flows. These may include reports from automatic chemical release sensors
at a known location or on-board train fire detection systems which combine GPS with
status indicators. If the data from such sensors is flowing at predetermined intervals
into GIS then the current status (the lag time is effectively the separation period
between reports from the device) can be mapped and decisions made on the basis of
an unfolding situation.
Thus, there has been a progression away from big computers that were stuck in rooms,
towards more powerful, user-friendly and portable computers, and also a shift away from
114
computers that could not talk to any other computer, through data and information transfer
over a limited network, to a global internet which can enable the seamless transfer of data
and information, and indeed facilitate the remote processing of data and the serving of
resultant information back to a wirelessly networked PDA or other mobile device in a field
environment. Web services is the term that describes this linking over the internet of
information systems and business processes through web-based protocols.
The UK government has established that interoperability for the sharing of information and
co-ordination of activity amongst public sector bodies can be achieved through the media of
web services, and this also holds for geo-spatial technologies. It would be nave, however, to
suggest that this represents the short term objective for all emergency planners and
category one and two responders in the UK. It has to be acknowledged that the route to
interoperability is at best unclear, certainly with specific reference to GIS applications in IEM.
The drive to ensure interoperability between systems that support effective Command,
Control, Co-ordination and Communication will proceed, but currently there are few tight
guidelines to influence interim developments. What appears here are a set of principles and
issues that should be observed and considered in developing GIS applications in IEM.
Interoperable web-services can take users from the ability to instruct systems to read these
data to read these data, carry out some sophisticated analysis, send them onto another
service, undertake further processing then post them onto another system to use31. This
critically depends on systems being able to interpret the data in a consistent way. XML, the
eXtensible Markup Language, is a development of HTML (HyperText Markup Language)
and is what is termed a metalanguage, that is a language that describes other languages.
In the same way that HTML uses tags to instruct a web browser how to display text and
images, tags in XML describe the data in such a way that data in XML format is selfdescribing. XML schemas are in effect languages which describe data and in line with the
principle of transparency and openness that these are widely published and the best of
breed are in effect sponsored by bodies such as the UK Office for National Statistics or
Office of the e-Envoy so that they become standards for application in a given area.
The key idea here is that of data which are self-describing and as such as highly mobile,
and meaningful, between systems.
10.3.2 Metadata Standards
Information production [is] growing at about 50% a year yet the amount of time
people spend consuming [information] is growing by only 1.7% each year a critical
task ahead will be to stop volume from simply overwhelming value.
Brown, J.S. and Duguid, P. (2002). The Social Life of Information,
Harvard Business School Press
A GIS is a tool for generating information. Data is the fuel that drives that tool. Users require
GIS to help them define solutions to their problems. As a prerequisite for this they require
data and this usually requires a search of some description. In the absence of any sort of
signposts to the right data (see section 8.4 for a discussion of quality issues and what makes
a given dataset right) this search could be frustrating, time consuming, involve a lot of
queries to already busy people and may be ultimately unsuccessful. Metadata provides the
required signposts.
The definition of metadata is information about data. Consider a basic example: if you
receive a CD in the post, which of the following options would be preferable?
31
i)
ii)
iii)
iv)
v)
Nebert, D.D. (2004). Developing Spatial Data Infrastructures: the SDI Cookbook (Version 2.0),
Technical Working Group, Global Spatial Data Infrastructure.
33
http://www.iggi.gov.uk/publications/index.htm
116
This reference information establishes the ability for geometric interoperability, something
that was introduced in Box 4 (Integrating disparate datasets using a spatial key); unless
locations can be related to each other data cannot be spatially integrated. If problems do
exist with spatial frameworks (for example the Columbia Space Shuttle recovery operation
see Box 9) they can have serious and time-consuming consequences. Due to consistent use
of Ordnance Survey referencing systems and products this is not often a severe problem in
the UK, but where users need to share data or have access to information of a consistent
quality the appropriateness of different addressing and geo-referencing frameworks and
standards needs to be considered and recorded in full in metadata.
10.3.4 Semantic Interoperability
Semantics define the meaning of records within a dataset. Think back to the last time you
heard someone say what Im trying to say is - usually that person is struggling to find a
way of expressing themself that will also make sense to you. Their idea of how big a fire,
how serious a hazard, how widespread a flood, how large a crowd, how steep a slope or
how large an area may be different to yours (see Box 13).
Semantic consistency demands that the representation of reality is done in a consistent
fashion. This is less of problem for operations within a local area of responsibility with
partners who have a common appreciation of the meaning of data and information. If the
person referred to above was able to point back to a common experience in the past, and
say this is almost the same as that one we dealt with in November 2005, then some
commonality will have been achieved. However, this does not work with people who have no
common ground, and it is a poor basis for introducing rigorous common standards. It is a
fine example of local solutions that are sub-optimal at higher levels and/or over wider areas.
There is a need to work towards commonality between agencies in the way that phenomena
are represented. At a (literally) basic level this has been done with the way in which
geographical objects are represented. GML (Geographical Markup Language) is a variant of
XML which defines, in universally appreciable terms, the key spatial characteristics of
geographical features. Spatial features can be defined by lines of code that define what kind
of basic object it is (area, line, point) and its coordinates.
There are technical part-solutions to the communication of what objects are and some of
their core attributes, the main example being SVG, Scaleable Vector Graphics. SVG is a
vector graphics language written in XML which describes two-dimensional graphical objects.
As such it can be used to determine how users see and can interact, albeit at a relatively
basic level, with maps in a web browser. Maps can be annotated, re-scaled and mouse
actions such as clicking to determine attributes and rolling grid coordinates with the
movement of the mouse can be set up, all with the gains of transparency and transferability
that XML and OSS brings. However, although GML can define spatial features and their core
attributes, and SVG can define the visual representation of the data, the meaning of what
they represent depends upon semantic consistency and this lacks standardisation.
At present there are differences between key agencies such as Police Forces, Fire Services,
Social Services Departments and Ambulance Services in the way that they classify
incidents. There are examples of standards such as the World Health Organisations
International Classification of Diseases. Some examples of these are illustrated in Table 7,
although it is clear that full consistency requires the semantics of contingent categories such
as residential institution and trade and service area to be realised.
117
ICD-10
Code
X1.1
X1.2
X1.3
X1.4
X1.5
X1.5
Meaning
Exposure to smoke, fire and flames - Residential institution
Exposure to smoke, fire and flames - School, other institution and public
administrative area
Exposure to smoke, fire and flames - Sports and athletics area
Exposure to smoke, fire and flames - Street and highway
Exposure to smoke, fire and flames - Trade and service area
Exposure to smoke, fire and flames - Industrial and construction area
Table 7: example of semantic consistency in coding from health
This is an area where GIS applications are currently very weak but such issues need to be
addressed, and not just at the local level.
10.3.5 Spatial Data Infrastructures
Spatial Data Infrastructures (SDIs) are described by Longley et al. (2005) as one of the big
ideas in GIS. This is not the place for a detailed discussion of the subject, but the basic
concept is that data sharing is difficult, and without the development of partnerships to
address the issues it will remain difficult. The higher the level at which partnerships are
developed, the greater the momentum will be to develop semantic, data and metadata
standards, and the means to search for and access data over the internet. SDI is a term that
describes a coordinated and partnership-based environment for producing, managing,
disseminating and using spatial data.
So, an SDI is not a physical infrastructure in the way that a railway system is, with track,
rolling stock, stations, timetables, management structures, consumer representation and
service standards. Rather it is much more conceptual, and Longley et al. (2005) observe that
at a high level, there are few who dispute the merits of achieving data sharing, reduction of
duplication, and risk minimisation through the better use of good-quality GI. But how to make
it happen for real is a different matter (p.458).
The previous sections have established that the standards, frameworks and issues that need
to be addressed for an SDI to develop over time. SDIs are fundamental and integral to
issues around interoperability.
10.3.6 Commercial Off The Shelf Software
Transparency is the most significant aspect of OSS: in contrast to much COTS Software and
many proprietary systems, the code is freely available and there are no Intellectual Property
Rights withheld in applying or developing open systems. Open Specifications provide
software engineers and developers information as well as specific programming rules and
advice for implementing the interfaces and/or protocols that enable interoperability between
systems34.
As a basic principle, proprietary systems that are not based on open specifications are
counter to basic principles of e-government and the wider pursuit of interoperability. With
many COTS packages the development emphasis is on creating extensions that permit the
seamless integration of diverse data formats, although the core software itself does not
conform to open specifications.
34
119
Section Eleven
Future Developments
Summary
In such a fast moving field it can be difficult, even foolhardy, to try and identify
future developments with any degree of certainty. This short section identifies
those areas where developments are likely to be the most significant for
emergency planners and responders.
11.1
Location-Based Information
Data sources
The exponential growth in data has previously been identified. These are not just created
through the same sources, as data are expanding not just in volume but also in diversity.
Some of these are what might be termed value-added products, created through the
integration of existing sources, and the example of geo-demographics is given here. Others
are data from sources that are well established at a generic level (remotely sensed data is
the example) but where refinements are creating data in much greater volumes and with an
ever increasing range of potential applications.
11.2.1 Geodemographics
Geodemographics describes data and information that profile areas on the basis of
composite indicators of consumer behaviour. Typically they are commercial datasets,
provided by a small number of businesses, which are based in small-area census data, but
have added value to these through the integration of other data sources such as surveys
and consumer records (Electronic Funds Transfer At Point Of Sale - EFTPOS - links
together what someone has bought together with who they are and where they live if they
also use a loyalty card). These might seem of very limited relevance to emergency planning
and management, but they provide ever richer information on the wider characteristics and
behaviour of the population, often down to the level of individual postcodes that is of
relevance to risk assessment and emergency planning.
11.2.2 Remotely sensed data
As has previously been identified, remotely sensed data (satellite images and aerial
photographs) are raster data. There are two dimensions to resolution in such data: spectral
range and spatial resolution. Any form of photography, of which remote sensing are variants,
is the capturing of energy being given off by a series of objects. For satellite images this is
usually solar radiation that is reflected or emitted from the earths surface. Such radiation is
120
121
given off across a wide range of wavelengths, and early satellite-based sensors could only
measure a small number (three would be typical) of these wavelengths. The capacity to
measure a wider range of wavelengths means the ability to more accurately discriminate
between characteristics of the earths surface. Modern sensors have the ability to measure a
much wider range, thereby supporting a more diverse range of applications.
The spatial resolution of satellite images and aerial photographs have also been increasing
with time. Until recently the highest resolution (smallest pixel size) satellite images that were
commercially available to general users were around 20m x 20m. The IKONOS satellite,
launched in 1999, dramatically changed this with a spatial resolution of 90cm, followed in
2001 by imagery from the QuickBird satellite, which is reported to have a resolution of 62cm.
Figure 69 shows an panchromatic (black and white) IKONOS image of the centre of Rome at
a spatial resolution of approximately one metre.
Figure 70 illustrates the real colour imagery available from the Quickbird satellites, showing
in this instance a freighter breaking up off the Aleutian Islands in 2004. It is clear from this
that the quality of the image is comparable with aerial photography, and as it is a commercial
product there is no need to commission such images although far from cheap, they just
need to be purchased from the back catalogue.
11.2.3 Real time data
Real-time describes data that reflect the situation as it is at the present time, potentially with
a short delay for processing and display. The use of Automatic Vehicle Location Services,
such as those illustrated for Durham Constabulary (see 7.4.6), are an example of this which
permits decision makers to see the distribution of resources at any given time. A range of
potential devices such as traffic or river flow meters, Automatic Number Plate Recognition
Systems or radiation or atmospheric pollution sensors have the potential to issue alerts if
measurements exceed defined parameters or another trigger is identified. Where such
sensors or devices are integrated with GIS the spatial location of the anomaly and its
potential consequences can rapidly be assessed, analysed and visualised. Digital CCTV can
also be integrated with GIS, as flows of imagery that are associated with given locations, and
which can then be accessed as video feeds for diagnostic or confirmatory reasons. The
integration of CCTV with GIS-enabled Command and Control systems in Police control
rooms is a good example of this.
The number and range of sensors feeding real time or near real time data in a form that is
compatible and accessible with GIS will increase over time with significant implications for
emergency management. One example of such sensors feeding information in real time (or
near real time i.e. with a short delay due to processing or transmission) to a web-site is the
facility of the Centre for Environment Fisheries and Aquaculture Science (CEFAS), illustrated
in Figure 71.
The CEFAS website reports data from the DEFRA strategic wave monitoring network for
England and Wales, which is a network of wave buoys located in areas at risk from flooding.
According to CEFAS Data from this network will be used to improve the management of
flood and coastal erosion risk for which DEFRA has policy responsibility and the data will be
used by Flood Managers, Local Authorities, Consultants, and other stakeholders in order to
assess flood risk, and on a longer timescale will be to help design improved flood defence
schemes and to provide data for climate change studies.
122
Figure 71: The CEFAS Wavenet website, providing access to real time wave data around
the UK (http://www.cefas.co.uk/wavenet/default.htm)
11.3
Mobile Technologies
The linkage of mobile phones with location-based information has been mentioned
previously. Just as GI is becoming increasingly embedded in other forms of information
capture, management and reporting systems, the decreasing size and cost of GPS and their
integration with devices such as mobile phones, digital radios and PDAs means that
outgoing messages from mobile devices can include a locational identifier. The increasing
processing power of such devices also means that mapping technologies for information and
navigation can be accessed from almost anywhere, with clear implications for warning and
informing the public. Although it is a seemingly mundane issue, the most significant
drawback in the use of such mobile technologies is limited battery life a clear resilience
issue.
11.4
GIS applications in emergency planning and management have mirrored the development of
GIS more broadly: early developments were carried out on single machines within separate
agencies, and data and information sharing was partial and difficult to achieve. Recent
developments have seen statutory, technological and organisational shifts towards the
integration of practice and systems that support that; legislation has clarified rights, roles,
responsibilities and requirements to share data; partnerships and protocols have promoted
this and the increasing mobility of data within systems and organisations, underpinned by
transfer formats, common standards, open systems, metadata and web-services have
permitted the development of integrated information for decision-making. Future
developments will build on integration of data, systems and processes in the development of
interoperable data, systems and processes that effectively remove the need for manual
123
interventions to transmit data and information between organisations. The principles, both
technical and operational, of interoperability are established (and embedded in the principles
of e-government in the UK) and the technical enablers are all proven. It will now require
vision and leadership to realise the gains.
124
Backdrop Mapping
Maps such as those at 1:250k, 50k, 25k or 10k from the Ordnance Survey
which are used in a GIS for the purposes of context and orientation rather
than any direct analytical applications. Such maps are usually in raster
format.
Framework
Mapping
Geo-referencing
Hotlinking
This is the process of linking files to locations. Usually it refers to files other
than the standard table of attributes, for instance graphics files, external
databases, word documents, hyperlinks or movie files. An example of a
hotlinked file could be a jpeg file which illustrates the feature under normal
operating conditions. A field worker could access this to ascertain whether any
changes or alterations had taken place.
Metadata
A large scale map is one of a relatively small area that shows a large amount
of detail. Some people find this confusing, expecting a large scale map to be
one of a large area, but this is not the case they are termed large scale as
1:5,000 is a larger fraction of 1 than, for instance, 1:250,000.
Metadata are information about data. See section 8.3.
Raster data
Resolution
A small scale map is one of a relatively large area that shows a limited amount
of detail. Some people find this confusing, expecting a small scale map to be
one of a small area, but this is not the case they are termed small scale as
1:250,000 is a smaller fraction of 1 than, for instance, 1:5,000.
Thematic Maps
Thematic maps can be distinguished from backdrop maps, as they are maps
of a specific dataset rather than general topographic maps for the purposes of
context and orientation. A thematic map, for example, could be census data
describing the distribution of children aged less than 14 in an area or the
distribution and attributes of COMAH sites in a District.
Vector Data
Spatial data that are stored as points, lines or areas. See Box 5.
125
126
Kevany, M.J. (2003). GIS in the World Trade Center attacktrial by fire, Computers,
Environment and Urban Systems, 27, 571-583
Zerger, A. and Smith, D.I. (2003). Impediments to using GIS for real-time disaster decision
support, Computers, Environment and Urban Systems, 27, 123-141
127