LEARNING OBJECTIVES
▪ How emergency planning relates to GIS
▪ GIS tools for understanding vulnerability, hazards and exposure
▪ Ways in which GIS and spatial analysis can be used in different aspects of emergency planning
Readers might ask, why devote an entire chapter of a social sciences GIS textbook to emergency planning? At first glance this appears to be a topic falling solidly within the area of expertise of practitioners and policy makers, not necessarily students or researchers in the social sciences. Upon further consideration, however, emergency planning emerges as virtually the quintessential social science and GIS combination – risks and impacts of emergencies are spatial, but they are also related to population characteristics, information transmission, government structures and a host of other societal factors. As the chapter title suggests, a raft of possibilities exists for actual GIS applications in emergency planning. More than that, though, emergency planning provides an exemplar for those working in other areas: many of the methods, concepts and spatial perspectives employed here could easily be retooled for other subjects. Finally, emergency planning provides a blueprint for thinking more generally about how the social sciences and GIS can combine to provide information and understanding about ‘real-world’ issues.
The phrase ‘emergency planning’ is deceptively simple. It conjures images of disaster response teams moving into a location struck by earthquake, flooding, hurricane or tornado, providing evacuation assistance, health care or food and water. In fact, each word, ‘emergency’ and ‘planning’, represents a range of activities, all informed by spatial data and GIS research. Emergencies, covered in more depth below, might be environmental in origin but this is not necessarily the case: terrorist attacks or disease outbreaks are also emergencies. Emergency planning also refers to the provision of emergency services such as police and fire services in order to respond to more frequent emergencies such as domestic fires (see again the discussion in Chapter 7). Planning suggests an element of forethought and prearrangement. Although the emergency might come as a surprise, all levels of government aim to be prepared for the unexpected. This preparation indeed includes consideration of evacuation routes and knowledge about population centres and areas of potential demand during an emergency. However, it also involves providing up-to-date coverage of disaster impacts and eventual recovery efforts. In fact, planning activities can be thought of as an umbrella, under which proactive, active or real-time and reactive actions fall.
Anticipation of disasters or emergencies means understanding (and even imagining) the types of emergencies that might occur, as well as their likelihood, location and areal impact. Simulations can also help expose unintended consequences and spill-over effects, and assist with response planning. For example, simulating fires along different locations of a subway line can help estimate access to the site and response times depending on where the fire occurs. Should an emergency occur, often multiple immediate responses will be possible. Advance planning helps consider alternatives and identify best options under various circumstances. Emergency planning is equally important during and after a disaster, however. As a crisis is occurring, responses must be recalibrated or redirected according to experiences on the ground or unexpected challenges. Finally, in the aftermath, coordinating clean-up and recovery measures necessitates not only advance planning but also in-the-moment planning that incorporates new information, stakeholder needs and other factors.
Underpinning a good part of before, during and after emergency planning is access to appropriate data and effective location and organisation of emergency services. Locations of fire stations, for example, must meet the day-to-day needs of the local population but must also be equipped to respond quickly to the unexpected emergency. High-quality and recent population and infrastructure data are also key. It is not sufficient to know where an emergency has occurred: effective planning means also knowing how many are likely to be impacted and what the toll on infrastructure might be. Where will be the heaviest hit locations and how many people will be involved? Will they be able to leave the area unassisted or will assistance be required? Will evacuation via road be possible? Will communication via mobile phones be possible?
At every stage of preparation or response and in every aspect of management, emergency planning is inherently spatial. Emergencies, of course, occur somewhere and the extent of the event interacts with the people and things within that area – this is core GIS territory. Research and planning around movement in and out of an afflicted area (evacuation for instance) also lends itself naturally to a GIS environment. Even visualisation in ‘real time’ of an emergency almost demands the use of a GIS. GIS and a spatial approach provide the necessary tools and techniques to engage in emergency planning. They also enhance understanding of the data inputs required for such planning – GIS fundamentals such as spatial scale, accessibility and interaction help determine the efficacy of any emergency planning. The necessary integration of information and viewpoints is also well suited to a GIS environment. Regardless of the methods used in disaster planning, an enormous assortment of data must be assembled and integrated in such a way that it can be analysed and interpreted efficiently. A GIS offers this capability.
Within applied geography and GIS, the field of emergency planning or management is healthy and even growing. Accessing current research and practical guides on the use of GIS in this area is easy, requiring the simplest of internet searches (a few recommendations for further reading are also provided at the end of this chapter). Because research in this area encompasses so many applications and includes research in ‘preparedness, response, recovery, and mitigation’ (Emrich et al., 2011: 321), the conceivable uses of GIS or spatial analysis in emergency planning are myriad. Armed with basic concepts and understanding of common GIS applications, students and researchers should be equipped to engage with the academic literature and also contribute to planning in the world of practice.
This chapter has two purposes. The first is to provide the building blocks of good emergency planning research and investigation using GIS, from key concepts to data considerations to common techniques and methods. The second purpose is to illustrate the utility of GIS for emergency planning through examples that emphasise the application, rather than individual techniques or data inputs. As with much research in the social sciences, little is accomplished with just one method. Rather, the goal is to answer questions and provide solutions or recommendations. Thus this section of the chapter emphasises application and shows how methods introduced in Part I of the book can be combined in novel ways to create new information.
A simple definition of an emergency is that it is an unexpected and harmful or dangerous event or occurrence. Using this definition, everything from an auto accident to a house fire to a hurricane is considered an emergency. Within the realm of emergency planning, then, some additional criteria must be met. These criteria are related to scale, impact and, perhaps, some aspect of predictability. Scale conveys an impression of the magnitude of the event. Events qualifying as emergencies or disasters are typically of a larger scale, measured in terms of area affected, the response required, length of time the area experiences and recovers from the event, or the number of people or properties affected. Magnitude can also be measured or thought of in terms of impact. An auto accident might impact a household, a hurricane an entire city or region. Impact can be assessed in terms of human lives affected or lost, property damage, range of resources to be mobilised or, again, some time component.
Defining emergencies in terms of predictability is a bit more challenging. As science and documentation of past experience improves, our ability to anticipate when and where a disaster might occur (and the quickest and most effective ways of responding) should also improve. That said, an element of unexpectedness or suddenness is often attributed to emergencies. We might know that an area is prone to flooding or earthquakes, but we are unable to predict exactly when they might occur. This means that resources must be allocated and plans readied, without knowing precisely when or where they will need to be mobilised. This limitation can lead to important data challenges, which are discussed below. One way in which planning is accomplished is by way of assessing exposure, vulnerability and risk (see Table 10.1).
Table 10.1 Some key concepts and terminology
Below are several terms that, although also used in common language, take on specific meaning within the fields of emergency planning or hazards research. These concepts form the backbone of much published research in this area. Hazard: A potential source of danger or incapacitation. Hazards with spatial implications can be natural or biological in origin, but also human-induced. A hazardous situation, such as a tropical storm or earthquake, becomes a disaster or emergency when its occurrence has societal impacts – i.e. on people, property, or some aspect of the environment that society depends upon. Hazards can be characterised by their location of course, but also their likely area of impact, duration and the frequency with which they occur (van Westen, 2013). Exposure: A measure of who and what might be affected by a hazard. At a very basic spatial level, this might be thought of as the area (and the people and things within that area) likely to be impacted by a particular hazard. Vulnerability: The characteristics of people and places that make them more likely to be affected by a hazard, should it occur. In many cases, exposure is a good proxy for vulnerability. In addition, however, particular population characteristics (e.g. age or poverty levels) and place characteristics (e.g. type of government or quality of infrastructure) may exacerbate vulnerability. Risk: The combination of the characteristics of the hazard, exposure and vulnerability provides a sense of the probability that a hazard will have deleterious effects. In many cases, hazards cannot be moved, but through efforts in terms of exposure and vulnerability its eventual impacts can be lessened. Resilience: The ability of a population or area to recover from or adapt to emergencies or disasters. |
Hazards and eventual emergencies are typically classified as natural, biological or human-made. Natural emergencies may be weather-related, such as blizzards, hurricanes or tornadoes, or geologic in origin, such as earthquakes or volcanoes. Biological emergencies are typically diseases or illnesses and human-made emergencies could be terrorist attacks but also, for example, oil spills. Many hazards or emergencies are combinations of two or more of the above: mud slides, for example, might start from a storm or earthquake but are also due to human impacts on the landscape from economic development and building. Figure 10.1 shows hazards the city of New York has judged to be important (and relevant) to the city, as well as the myriad and complex impacts these hazards can have. Note the wide range of impacts each hazard can have.
Figure 10.1 A selection of hazards and hazard impacts as identified by the New York City government
Source: www.nyc.gov/assets/em/downloads/pdf/hazard_mitigation/nycs_risk_landscape_chapter_4_intro.pdf
Hazards, and subsequent emergencies, can take on varied spatial signatures. Some, like volcanoes or tsunamis, occur in locations that can be very specifically delineated. Others, such as tornadoes or hurricanes, only occur in certain type of areas, but these areas can be quite large and the likelihood of occurrence in any one area might be quite small. Still other types of emergencies, such as disease epidemics, could start almost anywhere; the transmission of the illness will be an important characteristic of the emergency and the subsequent response. Finally, there are emergencies such as terrorist attacks, for which risk at any one particular location is difficult to assess, such that planning is necessarily more abstract and therefore more difficult.
Although this chapter is aimed at those wanting to learn more about GIS and emergencies, it is worth noting that the approaches discussed here – everything from data to techniques to applications – are partially or completely applicable for any sort of impact analysis. That is, there is a status quo, some measurable landscape of people or property or firms, and we would like to conduct a ‘what-if?’ analysis. In the case of emergencies, the question is ‘what if this hazard occurs?’ The question might also easily be adapted to ask about any sort of shock to the existing landscape (e.g. what if a water main breaks in a certain part of the city or what if a bridge closes due to storm damage?).
The nature of emergency planning is such that data inputs can be quite demanding. Poor data lead to poor results that do not accurately reflect vulnerability, risk or best response. Also, because emergency planning is multi-faceted (involving not only advance preparation but also response and recovery) and the nature of emergencies is multi-layered (requiring information about the hazard but also exposure and vulnerability) the range of data needed can be vast. A complete inventory of data requirements would be impossible here. Instead, this section highlights some main types and characteristics of data needed for GIS applications in emergency management.
At a minimum, emergency planning within a GIS framework (but also elsewhere) will require information about the hazard and about exposure to the hazard. Both types of information are spatial; we need locations of hazards, as well as non-spatial attributes of these hazards. Often, additional spatial modelling will take place that estimates the spatial impact of any hazard. This modelling, which combines subject matter and GIS expertise, results in maps and data that indicate the likely severity of any hazard for a given location. The hazard risks becoming an emergency or disaster when it has some societal impact. To judge the potential societal impact, we need information about the people, property, infrastructure and environment in the area of the hazard. Hazards such as possible disease outbreaks require intermediate information about modes of transmission. Combining these layers of information allows us to measure vulnerability. In some cases, the hazard analysis might already have been completed and can be used as a finished GIS layer. The same can be said for vulnerability. On other occasions, raw data will first need to be manipulated and analysed. When visualised, the above data – hazard and vulnerability – give a sense of potential areal, economic and demographic impact of an emergency. Response, management and recovery phases will require additional data about transportation and communication networks, as well as mechanisms for on-the-ground updates.
Table 10.2 lists some of the more common types of data required for GIS use in emergency planning and shows a GIS interface with a website monitoring outbreaks of influenza. In every case, scale – temporal and spatial – is of paramount importance. Temporal scale refers to the time period the data reference. Unfortunately, although some types of hazards might not change a great deal over time, the inputs to vulnerability assessments likely do. Population size and distribution changes and the characteristics of those living in an area might shift over time. Thus, recent, complete data are important. The spatial scale of the data is also important. Hazards can impact very small areas; vulnerability data collected at too large a scale will render any locational targeting impossible. The key is to have all data organised and ready to use so that when an emergency occurs, response-related analysis can commence immediately without the need for new or updated information.
Table 10.2 Some of the more common types of data required for GIS use in emergency planning
1 Demographic and household-level: These data often come from national or regional statistics offices and can be complemented with local data and knowledge. Important variables are total population counts, but also information related to potential vulnerability, such as age, income and education levels, and possibly other demographic characteristics. The size of spatial unit should match the scale of analysis. So, for instance, if the goal is to identify regions susceptible to hurricanes, data at the province or county level might be sufficient. If the goal is to estimate the number of people likely to be affected by hurricane-related winds or flooding, the unit will need to be much smaller in order for results to be useful. These GIS data are, almost without exception, in vector format. 2 Economic and civic: This information – firm locations and characteristics, school locations, tourist sites or land parcel information, for example – could come from national statistics offices, but is more likely to be collected at the appropriate scale at the municipal or regional level. These data are usually in vector format. 3 Environmental: Raw data on natural hazard locations and past history of events (e.g. hurricane and tornado paths and locations) are often available via national natural resources agencies or academic institutions charged with aggregating and assimilating the necessary raw information. Regions and cities will also undertake hazard assessments. Increasingly these data are available in spatial format, as GIS gains currency as a primary emergency planning tool. These data may be in vector or raster format. 4 Transportation and other infrastructure: The spatial component of these data is generally available at a range of spatial scales (nation to region to city). Information about flows and use of different types of infrastructure can be more difficult to find or access. 5 Social media: As increasing numbers of people around the world use social media such as Twitter or Facebook to communicate, interest is growing in using these platforms to both collect information about emergencies and also communicate with areas affected. Twitter feeds on a particular topic, for instance, can be downloaded, often with locational information, either latitude and longitude or city. In the case of epidemics, diffusion of awareness as well as illness can potentially be tracked. For example, Flu Near You (https://flunearyou.org/) uses the internet and smartphone applications to help keep track of flu occurrence and symptoms in the United States. You have the opportunity to make use of social media data in Practical D. |
Figure 10.2 An example of user-contributed information: Flu Near You
Source: https://flunearyou.org/
Aside from the data issues discussed above, other aspects of emergency planning within a GIS framework also demand attention. Primary among these are challenges related to dynamic and changing situations and those stemming from user characteristics. In spite of advanced analysis and simulation, emergencies when they occur will involve innumerable unexpected elements. In addition, as response and recovery set in, the need for new, updated data grows. In short, the unpredictable and dynamic nature of emergencies and disasters means that facts on the ground are constantly changing and the need for timely data is high.
Another aspect of emergencies is the large number of actors involved at multiple levels of government and geography. GIS offers a unique environment for sharing data and coordinating responses. Challenges can arise, though, when not all are conversant with GIS analysis and when many users are contributing and using information. As both practitioners and the general population become more conversant with spatial data and as use of smartphones and internet increases, new opportunities for data collection and sharing appear. Social media are increasingly viewed as opportunities for collecting and disseminating information about emergencies (see Figure 10.2) but other innovations are also taking place. Figure 10.3 shows Google’s Person Finder, a web application initially launched after the 2010 Haiti earthquake, providing a database of ‘missing’ persons in the immediate aftermath of a major disaster, enabling friends, relatives, press and other agencies to help people reconnect post-disaster.
One final note about data. Emergency planning is one of the few areas covered in this book that typically requires a combination of vector and raster data (see Chapter 1 for a refresher). Moreover, the combination of data types, scales and sources can mean that a great deal of cleaning and editing are necessary before the various layers will align properly.
Examples: GIS for emergency planning
As discussed in the introductory section of this chapter, the use of GIS in both applied and academic research in emergency management and planning is vast and growing. The remainder of this chapter provides a few selected examples of how GIS informs emergency planning. For particulars on data and techniques, as well as best practices, published literature on a specific topic (e.g. determining evacuation routes) is best consulted for state-of-the-art information.
Figure 10.3 Google Person Finder: an example of how the internet and social media allow new ways of sourcing information, both spatial and aspatial
Source: http://google.org/personfinder/global/home.html
GIS for emergency visualisation
Visualisation – seeing data – is useful at every stage of emergency planning (see Table 10.3). Two different examples are presented below. First, proactive hurricane vulnerability and evacuation information prepared for the general New York City public and, second, estimates of impacts of Hurricane Katrina, which hit the US city of New Orleans and the surrounding area in 2005.
In late October 2011 Hurricane (or Superstorm) Sandy hit New York City, destroying homes and infrastructure and killing dozens. The disaster was a combination of a strong storm system and an enormous urban agglomeration, New York City. Much of the impact of the storm was due to its unexpected strength and effect on city infrastructure and housing. Having learned the lesson that advanced education of the public is valuable, New York City now provides interactive maps (Figure 10.4) that allow users to assess the vulnerability of different areas of the city, as well as locations of evacuation centres. The information provided in Figure 10.4 does not emphasise evacuation routes or the characteristics of those living in particular storm zones. Rather, its purpose is to encourage preparation and awareness in the general population.
Similar to Sandy, Hurricane Katrina, which struck the city of New Orleans, Louisiana, in late August 2005, was a historic example of a confluence of hurricane-related weather events, including wind and especially flooding, affecting a poorly prepared, densely populated area. The historic nature of the storm and its impacts has resulted in an unprecedented amount of research in the social sciences on inequality and how, at virtually every stage – hazard mitigation, preparation, response and recovery – impact and handling of the emergency varied by socio-economic group. Vulnerability within the context of New Orleans and Hurricane Katrina is discussed in more detail below. This section highlights a different aspect of Katrina: the use of maps and cartographic visualisation to assess the extent and type of damage, as well as the unequal impacts. Figure 10.5 shows one element of research done by Frederick Weil at Louisiana State University. This particular map shows the area of flooding in the city and compares it to the extent of resulting housing damage. Other ways of presenting the information, as tables or figures, might also effectively show the magnitude of damage and flooding; the map shows the reader how location mattered.
Table 10.3 Useful GIS techniques and methods
As made (hopefully) amply clear above, GIS is used at every stage of emergency planning, from evaluating potential hazard impacts and vulnerability to evacuation plans to response management. Most applications will combine data sources discussed above and will apply a wide range of GIS techniques. A few relevant methods are highlighted here, alongside details of where to look in this book for more information. Cartography and visualisation (Chapter 3) is simple but powerful. Seeing the location of hazards, the areal extent of vulnerability and even evacuation routes is among the most powerful modes of communication to the general public but also to policy makers and emergency responders. Visualisation of emergencies at every stage, from preparation to recovery, can help provide common ground for multiple stakeholders with a range of knowledge about a subject and a range of perspectives. It can also help focus attention on priority areas. Network analysis (Chapter 4), which can be used to find the best route out of a place or the closest emergency shelter, is used not only during the emergency preparedness and response phases of planning, but also in estimating vulnerability. Just as individuals might use a network to leave an area, some locations are more vulnerable to impacts of hazards, given their connectivity to a network (think, for example, of international transmission of disease, via aeroplane routes or waterborne illness via water mains or fresh water systems). Spatial analysis (Chapter 2) covers an enormous number of methods, but can be summarised as tools that consider the overlay of layers as well as those that are concerned with proximity to a feature or set of features. Together it is these types of tools that enable researchers to estimate the spatial impact of hazards, exposure to them and underlying vulnerability. |
Figure 10.4 Map showing storm vulnerability and location of evacuation centres in New York City
Source: http://maps.nyc.gov/hurricane/
Figure 10.5 Type of damage and flooding extent post-Hurricane Katrina in New Orleans, Louisiana, US
Source: Frederick Weil’s research on recovery from Hurricane Katrina. Available from: http://lsu.edu/fweil/KatrinaMaps/index.htm
GIS for evacuation
A core component of disaster or emergency response is evacuation – the process of removing individuals from affected areas. Evacuation planning requires a GIS, not only for planning of efficient routes, but also for locating evacuation centres or temporary housing and assessing how the emergency might impact transportation routes. Underlying population distribution and characteristics, and even time of day, are important components to include in evacuation planning. Efficient routes, measured by distance or travel time, might not be nearly as desirable if all households in an area employ that route. Thus, in terms of effective evacuation planning, information dissemination and beforehand assessment of alternatives are important.
Figure 10.6 Snapshot of USGS’s pedestrian evacuation tool capability
Source: Jones et al. (2014) and Wood and Schmidtlein (2013)
Rather than provide a summary of research or real-world examples of evacuation plans, this section provides an example of a tool developed by the United States Geological Survey (USGS). A federal agency, the USGS is responsible for information, data and policy relating to natural resources and the environment, but also natural hazards. Within this latter area, the agency has developed a tool – essentially an add-on or extension for ESRI’s ArcGIS software – that assists with the development of pedestrian evacuation plans for natural hazards such as volcanoes or floods. Figure 10.6 provides an overview of the tool’s capabilities. From a pedagogical standpoint, the USGS’s tool is useful; it provides a sense of how a variety of spatial data is integrated, including hazard areas, elevation, land use and networks, to find paths (and distances) to areas of sufficiently high elevation. Tool inputs make clear that evacuation plans are not solely least-cost routes, but must also acknowledge hazard-specific constraints. In the case of flash flooding, for example, evacuation must be speedy but also from lower to higher ground. The USGS extension can also incorporate population data, which can be used to estimate how many people are in various time bands for evacuation.
GIS for vulnerability assessment
Vulnerability measures the ‘combination of physical, social, economic, and political components that influence the degree to which an individual, community, or system is threatened by a particular event’ (Wood and Schmidtlein, 2013: 1604). The measurement of vulnerability is important: not only is it used to estimate impacts of a hazard, but also to focus scarce resources for education or awareness, as well as potential mitigation efforts. Vulnerability can also affect a group’s ability to evacuate an area (Wood and Schmidtlein, 2013) as well as recover from a disaster (Flanagan et al., 2011). Vulnerability assessments, like the two other examples of emergency planning applications above, form their own, large literature. This section focuses on two aspects of vulnerability assessment that are important for any researcher to consider. The first is the measurement of the socio-economic aspects of vulnerability. The second is the spatial – how will individuals with a given set of characteristics respond to an emergency? Specifically, how will their characteristics affect the ability to evacuate?
Figure 10.7 Vulnerability and drowning deaths from Hurricane Katrina
Source: Flanagan et al. (2011)
Vulnerability is typically measured along socioeconomic and demographic lines by developing composite indices that weight a variety of factors, including age, race/ethnicity and income. A recent paper by Flanagan et al. (2011) uses US census data at the tract level (small administrative units that typically contain around 4,000 people) to develop an index that considers, simultaneously, socio-economic status, household characteristics, minority status and language spoken, housing and transportation. The development of the index is straightforward and aspatial. That is, there is no need for a GIS to calculate the measure, although the units themselves are spatial. The results, though, can be mapped to show variation in vulnerability and how one location compares to others. Moreover, locations of other variables can be compared to vulnerability scores. Figure 10.7, taken from Flanagan et al. (2011), shows the elderly vulnerability at the tract level in New Orleans and compares it to drowning deaths that occurred. This information would not be helpful as an emergency was under way, but provides baseline information for future emergencies and an assessment of why outcomes (e.g. drownings) were shown to vary over space.
A recent paper by Wood and Schmidtlein (2013) shows how the spatial variation in vulnerability, however measured, also plays into ‘evacuation potential’. Their paper, which uses two counties in Washington – areas prone to tsunamis – uses evacuation times to help estimate exposure to potential tsunamis. The argument here is that characteristics might matter, but ability to leave an area, which is mediated by charac-teristics but also distance to safety, is an important component of exposure. A nice aspect of this study is that it emphasises the importance of both spatial and aspatial data in evaluating exposure and vulnerability.
GIS and social media and crowd-sourced data
A small but growing source of information in emergency planning, especially response and recovery, comes from social media and crowd-sourced data. Social media data typically refer to posts and tweets coming from Facebook or Twitter, for example. Crowd-sourced data can come from social media sources but are also the result of volunteers creating useful data products from raw information. Recent large-scale natural disasters – the Haiti earthquake (2010), the Japan earthquake and subsequent tsunami (2011) and Hurricane Sandy (2011) – all had in common the increased use of these non-traditional data types. The Haiti earthquake is often cited as the first major disaster for which these sorts of data took on prominence and this is partly due to timing (Katrina was pre-Facebook, pre-Twitter) and also due to characteristics specific to Haiti, specifically the lack of existing spatial data to aid the response and recovery efforts (Zook et al., 2010). This section is brief and its goal is simply to draw attention to a new source of information that is likely to take on a more central role in the future. To date, applied examples of Twitter’s effectiveness in disaster response is unproven. Likewise, although technological capability and knowledge are increasing, crowd-sourced data are becoming increasingly important but are still secondary to traditional data sources.
Concepts such as emergency, vulnerability, resilience and risk can all be viewed and studied through a spatial lens. Indeed, some aspects of each of these are inescapably spatial. Whether assessing or responding to disasters that have already occurred, or planning for response to future emergencies, GIS has much to contribute. A key aspect of research in this area is data input requirements; the methods and techniques that are used are often also employed in other subject areas and you are likely to be exposed to those throughout this textbook. Emergency planning itself is a broad topic. Those working in this area are advised to seek out further resources in their particular subfield (see below for some suggestions and a practical exercise).
This chapter is accompanied by Practical 7: Emergency planning – fire station location, which provides an opportunity to work with a road network data set in the context of fire station coverage in the Australian state of Victoria. The practical builds on previous network analysis practicals and walks you through the process of identifying areas of ‘risk’ of domestic fires. You will import and geocode data related to fire station locations and work with the road network to assess the provision of fire services and their accessibility to the population ‘at risk’. Practical D (Part I) is also related to the content of this chapter and uses social media data from Hurricane Sandy to pre-process, import and visualise social media data related to a natural disaster.
All website URLs accessed 30 May 2017.
Emrich, C. T., Cutter, S. L., & Weschler, P. J. (2011) GIS and emergency management, in T. Nyerges, R. Macmaster, & H. Couclelis (eds) The SAGE Handbook of GIS and Society, SAGE Publications Ltd, London, 321–343.
Flanagan, B. E., Gregory, E. W., Hallisey, E. J., Heitgerd, J. L., & Lewis, B. (2011) A social vulnerability index for disaster management. Journal of Homeland Security and Emergency Management. DOI: https://doi.org/10.2202/1547-7355.1792.
Jones, J. M., Ng, P., & Wood, N. J. (2014) The pedestrian evacuation analyst: geographic information systems software for modeling hazard evacuation potential: U.S. Geological Survey Techniques and Methods, book 11, chap. C9, 25pp.
van Westen, C. J. (2013) Remote sensing and GIS for natural hazards assessment and disaster risk management, in J. F. Schroder & M. P. Bishop (eds) Treatise on Geomorphology (Remote Sensing and GIScience in Geomorphology Vol. 3), Academic Press, Elsevier, San Diego, CA, 259–298.
Wood, N. J., & Schmidtlein, M. C. (2013) Community variations in population exposure to near-field tsunami hazards as a function of pedestrian travel time to safety. Natural Hazards, 65(3), 1603–1628.
Zook, M., Graham, M., Shelton, T., & Gorman, S. (2010) Volunteered geographic information and crowdsourcing disaster relief: a case study of the Haitian earthquake. World Medical & Health Policy, 2(2), 7–33.
