© 2005 University of Ottawa

Proceedings of the 98th Annual Canadian Institute of Geomatics Conference

Integrating Geographic Information Systems, Spatial

Databases and the Internet: A Framework for Disaster
Management
Sam Herold, M. Sawada and Barry Wellar
Laboratory for Applied Geomatics and GIS Science (LAGGISS), Department of
Geography, University of Ottawa, Ottawa, ON K1N 6N5. www.geomatics.uottawa.ca

ABSTRACT
At the global scale, high magnitude – low frequency natural disasters are a damaging phenomenon. There is growing recognition
that successful disaster response strategies depend upon access to real-time spatial data/information that can be effectively
utilized by not only relief agencies but also by local decision makers. Currently, many local and regional authorities are
developing geographic information system (GIS) - based spatial decision support systems (SDSS) in order to improve local
disaster response and management capacity. Although such systems are useful, the evidence is that they tend to lack
interoperability, and they require substantial GIS/technical knowledge through each of the design, development and
implementation stages. Moreover, their creation and maintenance requires significant human and financial resources, and
frequently access to information derived from such systems is very limited during the disaster response phase.

Due to the different characteristics of disasters, and variations in appropriate response mechanisms and initiatives, increasing
attention is being given to open source, Internet-based SDSS capabilities that can be made accessible by internet connections of
modest bandwidth. This concept of an intuitive, user-friendly, online GIS-based SDSS, which has powerful implications for
disaster response (for the operations, planning and management functions), is yet to be fully explored.

Experience to date indicates, however, that Internet-based GIS could play a key role in the collection and dissemination of
pertinent information in a fast, relatively inexpensive and straightforward manner during various stages of a disaster life cycle.
That is, by integrating datasets from various online spatial data networks into one consistent web-enabled GIS interface, using
open source MapServer technology and Open Geospatial Consortium (OGC) standards, vital spatial information is readily
available to numerous users without requiring high-level technical skills on the hardware, software, data collection, data fusion,
and data transformation side. In this paper we outline a framework for establishing an online GIS-based SDSS for natural
disaster management on a global scale that integrates what we term "spatial reporting."

Key words: Disaster Management, GIS, Internet, SDSS, Open-Source

1. Introduction information technologies effectively often requires

The field of disaster management has greatly benefited considerable technical skills and/or training time,
from recent advancements in computers and related which, in the hours and days immediately following a
technology. In particular, GIS, remote sensing, and the high magnitude disaster, are typically in short supply.
Internet have had a significant impact, and are Creating a way for non-experts (e.g. local decision
currently being used in a variety of ways during all makers, aid workers and the public) to quickly utilize
phases of disaster management. Integrating these mapping technology would increase their capacity to
technologies to create a spatial decision support system make informed and effective decisions based on the
(SDSS) offers new possibilities for disaster most up-to-date information.
management, particularly during the initial response
phase. During this phase, access to pertinent spatial In this paper we emphasize the need, and discuss the
information/data is among the most essential basic framework, for developing a user-friendly, online
requirements (Jayaraman et al., 1997). However, a GIS-based SDSS specifically designed for non-experts.
primary challenge for disaster managers is how to It is the intention that the system be used during the
distribute appropriate, accurate information in a timely response phase following high magnitude-low
manner to the necessary parties involved (Radke et al., frequency natural disasters, which often require
2000). Furthermore, to use the aforementioned international assistance. Our proposed system is

Canadian Institute of Geomatics

Ottawa, Canada, June 13-15, 2005
 © 2005 University of Ottawa

Proceedings of the 98th Annual Canadian Institute of Geomatics Conference

designed for use particularly in developing countries provide a brief overview of natural disasters and how
which, for a variety of reasons, are most vulnerable to information technology is an integral component of
large-scale high magnitude natural disasters for reasons disaster management. Subsequent sections explain our
ranging from population distribution, socioeconomic proposed system design and its architecture and present
characteristics, and resource bases to planning and concluding considerations.
building codes. The introductory sections of this paper

1.1 Natural disasters: an overview

The December 26th, 2004, tsunami in the Bay of Bengal resulted in a large-scale natural disaster, and provides recent
evidence of the severe devastation and disruption that natural hazards can cause (see Figure 1). Each year, some 210
million people are affected by natural disasters (IFRCRCS, 2001). Although disasters are so complex that they defy
easy classification, Smith (2004) describes them as: “an event, concentrated in time and space, in which a
community experiences severe danger and disruption of its essential functions, accompanied by widespread human,
material or environmental losses, which often exceed the ability of the community to cope without external
assistance.” Natural hazards can be understood as unpredictable acts of nature, characterised by extremes in physical
processes (Zerger and Smith, 2003).

Figure 1: Geography of December 26th, 2004, tsunami affected region

Although natural disasters occur globally, their impact is more severe on the population and economy of developing
countries, which lack the capability to deal with such events for many reasons, including poverty, illiteracy and the
inability to undertake mitigation and preparedness initiatives. About 90% of impacts from natural disasters are felt
by developing countries, and loss of GNP due to disasters is 20 times that of the developed nations (Alexander,
1995). To exacerbate the situation, there is mounting evidence that climate change could increase the frequency of
such atmospheric hazards as major storms and floods (Smith, 2004). This vulnerability, coupled with an increased
potential for natural hazards, provides reason for concern. The designation by the United Nations of the 1990's as the
International Decade for Natural Disaster Reduction (IDNDR) recognizes the impact of past disasters and the
impending consequences of future natural hazards.

1.2 Disaster management from an information technology perspective

Although natural disasters are increasing in frequency as global population rises, so are the ways in which we
manage the disaster process and consequences. Disaster management can be understood as a cycle (Figure 2) which
includes an effort to mitigate against, prepare for, respond to, and recover from a disaster (Montoya, 2002).
Effective disaster management requires assimilation and dissemination of real-time information/data to various
decision makers. This requirement can be considerably assisted with the use of GIS and remote sensing technology.

Canadian Institute of Geomatics

Ottawa, Canada, June 13-15, 2005
 © 2005 University of Ottawa

Proceedings of the 98th Annual Canadian Institute of Geomatics Conference

Many studies (Rivereau, 1995; Simonovic, 2002; Mansor et al., 2004; Becking, 2004) highlight the intelligence
gathering potential of satellites (e.g. SPOT, IKONOS, RADARSAT) to dramatically assist all phases of disaster
management by providing critical earth observation information. Similarly, GIS are useful tools in the disaster
management process for displaying, modeling and/or integrating data and information derived from satellites, and
other spatial data sources (Kumar et al., 1999; Gunes and Kovel, 2000).

Figure 2. Disaster management cycle (from Alexander, 2000)

Recently, the Internet is gaining popularity as a mechanism that can facilitate the exchange of information/data
(spatial and non-spatial) relevant to a disaster. Initiatives such as the Global Disaster Information Network (GDIN)
provide evidence of the importance and the value of disaster-related information, as well as the need to be able to
obtain and share it effectively. The aim of GDIN is to “provide the right information, in the right format, to the right
person, in time to make the right decision” (GDIN, 2005). However, to make a decision implies that there are
choices. Spatial decision support systems, commonly considered as application-specific software solutions (Rinner,
2003), are used in solving complex spatial problems where alternative decisions require consideration. Wellar
(1990) and Crossland et al. (1995) showed that the use of GIS as a type of SDSS reduced the decision time and
increased the accuracy of individual decision-makers, while Peng and Tsou (2003) emphasize that the Internet
provides an ideal platform for non-experts to realize the power and benefits of GIS. Integrating these technologies in
an on-line GIS-based SDSS has the potential to increase the use and accessibility of spatial data, as well as the
accuracy and efficiency of decision-making, thereby improving the effectiveness of disaster response.

However, developing countries lag behind in terms of their technological use, and highly technological solutions are
too expensive to acquire and maintain in most cases (Alexander, 1991). Harnessing the Internet to create a simple,
user-friendly, map-based system would allow local decision makers in developing countries to utilize beneficial data
and technology they otherwise would not have the access to, nor the expertise to implement. Accessibility is a
prerequisite to use and so accessibility to mapping technologies will increase the local capacity for disaster
management, and reduce reliance on international aid. To date, the authors are unaware of any similar system
created and maintained by the international community that is designed specifically for use in developing countries
following large-scale high magnitude natural disasters.

2. System Design
Our system falls under the broad category of ‘Internet GIS’, or IGIS which, since its infancy in the early 1990s, has
greatly evolved to comprise a wide range of services and applications. Peng and Tsou (2003) define IGIS as
network-based geographic information services that can utilize wired or wireless internet to access geographic
information, spatial analysis tools and GIS web services. While there are many GIS software vendors that offer
packages designed for IGIS, there is increasing awareness of the many benefits of the open source software

Canadian Institute of Geomatics

Ottawa, Canada, June 13-15, 2005
 © 2005 University of Ottawa

Proceedings of the 98th Annual Canadian Institute of Geomatics Conference

movement. A main advantage of open source software, aside from being free, is that it can easily be customized to
meet a wide variety of end-user requirements.

Our current prototype utilises MapServer, an open source development environment for constructing spatially
enabled Internet applications (University of Minnesota, 2003). MapServer supports several scripting languages,
including PHP, Perl and Python, and its notable features include the following:

• fully customizable with template driven output

• feature selection by item/value, point, area or another feature (basic spatial query)
• supports tiled raster and vector data
• map element automation (scalebar, reference map, and legend)
• scale dependent feature drawing and application execution
• on-the-fly projection

MapServer also supports several Open Geospatial Consortium (OGC) web specifications. OGC is an international
industry consortium of 278 companies, government agencies and universities participating in a consensus process to
develop publicly-available interface specifications (OGC, 2005). OGC specifications support interoperable solutions
that ‘geo-enable’ web, wireless and location-based services. These specifications empower technology developers to
make complex spatial information and services accessible and useful with a variety of applications. A limitation of
many present GIS-based web services it that they lack interoperability and do not comply with web mapping
specifications.

Our system also utilizes another open source web mapping program called Chameleon (v.2.0). Chameleon is a
distributed, highly configurable environment for developing Web mapping applications (DM Solutions Group,
2003). It is built on MapServer as the core mapping engine, and works with all MapServer supported data formats.
An advantage of Chameleon is that it incorporates the ability to quickly set up new applications using a common set
of core functions called ‘widgets.’ Widgets are defined through special Chameleon Web Mapping Components
(CWC) tags in standard HTML Web pages.

Since Chameleon was designed and developed to support evolving OGC standards, it can be used to incorporate any
remote data sources that publish data consistent with these standards. Another benefit of this type of application is
that is utilizes a thin client/thick server model, where data and software are concentrated on a single machine (the
server). This type of architecture is generally cheaper, easier to update, and uses computer power more efficiently
(Plewe, 1997). GIS and mapping are computationally-intensive tasks, and by using a powerful server less load is put
on the user’s computer, which can have a wide variety of capabilities, particularly in the developing world. In
addition, and perhaps the largest advantage of our system, is the ability to provide mapping capability to anyone
who can run a web-browser, even in low-bandwidth conditions. Gigabytes of spatial data can be manipulated on the
server-side and only small compressed images (usually < 100 Kb) are sent to the client. This is in stark contrast to
scalable vector graphics (SVG) web-based mapping which must send large text files containing data across the
internet for maximum detail. The Internet bandwidth required to transfer between the server and client is an
especially important feature, given that most regions in developing countries lack wide-scale accessibility to high-
speed Internet connections.

2.1 Spatial reporting

An important feature of our system is the implementation of a “spatial reporting” component. This allows the
creation of spatial features (points, lines and polygons) on the map, and the association of particular attributes with
these features, updating the server-side database. For example, a user could input an area recently established as a
distribution center for critical supplies, such as food and water; or input a point with a note attached detailing how
particular infrastructure, such as a bridge, has been affected by the disaster. Providing functionality for users to
dynamically update maps with new data/information increases ability to provide key spatial information in a useable
form to those who need it most.

2.2 Interface design

Canadian Institute of Geomatics

Ottawa, Canada, June 13-15, 2005
 © 2005 University of Ottawa

Proceedings of the 98th Annual Canadian Institute of Geomatics Conference

Since our system is intended for use by people in developing countries, who may be unfamiliar with GIS
technology, important consideration must be given to the system functionality and interface design. It is important
that the interface be designed in an intuitive and user-friendly fashion, and that the functionality does not exceed
what users can comprehend (Wellar, 1995). An example of a web-browser based mapping application can be seen in
Figure 3. Attractive button graphics enhance the recognition of the various user tools and increase the overall
aesthetics of the application interface, hence, making it more user-friendly.

Figure 4, another screen shot, shows an example of how the spatial reporting feature can increase the information
content of the map. In this example the point location of the CIG conference (The Westin Hotel, Ottawa, Ontario)
was added to the map, along with a label. The development of this addressed-based geocoding functionality was
done at LAGGISS to test the effectiveness of spatial reporting within the context of another web-based mapping
project for the Ontario Early Years Centers in Ontario.

Figure 3. Example of a Chameleon-based web application interface for the disaster web-mapping portal (DMapP-
http://mapsherpa.com/tsunami/) that was developed for the tsunami devastated regions in December 2004 by DM
Solutions Group in conjunction with the Laboratory for Applied Geomatics and GIS Science (LAGGISS).

Figure 4. Spatial reporting feature (addressed based geocoding functionality added to MapServer at LAGGISS)

Canadian Institute of Geomatics

Ottawa, Canada, June 13-15, 2005
 © 2005 University of Ottawa

Proceedings of the 98th Annual Canadian Institute of Geomatics Conference

3. Conclusion
Disaster management in developing countries exists within a complex political, social and economic environment,
where a coherent and coordinated approach can be difficult to implement. As such, large-scale high magnitude low-
frequency disasters will continue to overwhelm local disaster managers, prohibiting effective management,
particularly during the response phase. Although most natural disasters cannot be prevented, disaster losses can be
minimized when appropriate actions are taken which utilize the latest technology and best available spatial
data/information. Through appropriate applications of GIS and remote sensing we can move a significant way
toward making data/information available, and answering questions concerning the spatial and temporal dimensions
of disaster management.

Our system integrates GIS, satellite imagery and spatial data networks to form a web-based SDSS. Maps are the
primary output of this system which, when displayed on computer screens, are more dynamic and versatile than hard
copy versions (Alexander, 1991). This type of system complies with Coppock (1995) who points out that
technological developments intended for use in developing countries must be kept simple, considering the skills
(both technical and bureaucratic) and resources available.

Realizing that in many developing countries spatial data may be difficult to obtain, and that spatial data is a
prerequisite for the effective use of the proposed system, it is hoped that future improvements in satellite sensor
technology accompanied by the willingness of various agencies to share and create necessary spatial data will help
the system realize its full potential.

Finally, computer-based disaster management systems reduce some of the pressure of having to organize
information and data directly after a disaster and in the midst of a time of crisis. They provide decision makers with
the information that they require to develop and maintain their understanding of the spatial characteristics of a
natural disaster, and its impacts over a range of time. As a result, they will improve the capacity of developing
countries to provide more effective disaster response, and thus reduce their overall vulnerability to the apparently
increasing number of natural hazards.

Acknowledgments
This paper is based on a M. Sc. research project that is being carried out at the Laboratory for Applied Geomatics
and GIS Science (LAGGISS), Geography Department, University of Ottawa. A beta-version of the proposed system
will be implemented over the next year. The addressed based geocoding and spatial reporting functionality added to
MapServer at LAGGISS was supported by the University of Ottawa Interfaculty Research Grants programme. We
thank Dr. Jason Morrison, Dr. Michelle Driedger and Mr. Poopalasingham Elenkesewaran who are working with us
on the IMAP technology development. This research is supported by an NSERC, CFI and OIT grant to Dr. M.
Sawada. We also acknowledge the outstanding humanitarian contributions of DM Solutions Group Inc. who have
dedicated considerable resources to this project over the past year.

References
Alexander D (1991) Information technology in real-time for monitoring and managing natural disasters. Process in
Physical Geography 15(3): 238-260
Alexander DE (1995) A survey of the field of natural hazards and disaster studies. In: Carrara A, Guzzetti F (eds)
Geographical Information Systems in Assessing Natural Hazards. Netherlands, Kluwer Academic, pp 1-19
Becking I (2004) How space can better support emergency management and critical infrastructure protection.
Ottawa, Public Safety and Emergency Preparedness Canada
Coppock TJ (1995) GIS and natural hazards: an overview from a GIS perspective. In: Carrara A, Guzzetti F (eds)
Geographical Information Systems in Assessing Natural Hazards. Netherlands, Kluwer Academic, pp 21-33
Crossland MD, Wynne BE, Perkins WC (1995) Spatial decision support systems: An overview of technology and a
test of efficacy. Decision Support Systems 14: 219-235
DM Solutions Group (2003) Chameleon. Available at: http://www.dmsolutions.ca/technology/chameleon.html
GDIN (Global Disaster Information Network) (2005) GDIN Homepage. Available at: gdin.org

Canadian Institute of Geomatics

Ottawa, Canada, June 13-15, 2005
 © 2005 University of Ottawa

Proceedings of the 98th Annual Canadian Institute of Geomatics Conference

Gunes AE, Kovel JP (2000) Using GIS in Emergency Management Operations. Journal of Urban Planning and
Development 126(3): 136-149
IFRCRCS (International Federation of Red Cross and Red Crescent Societies) (2001) World Disasters Report 2001.
Geneva, International Federation of Red Cross and Red Crescent Societies
Jayaraman V, Chandrasekhar MG, Rao UR (1997) Managing the natural disasters from space technology inputs.
Acta Astronautica 40(2): 291-325
Kumar V, Bugacov A, Coutinho M, Neches R (1999) Integrating Geographic Information Systems, Spatial Digital
Libraries and Information Spaces for conducting Humanitarian Assistance and Disaster Relief Operations in
Urban Environments. 7th ACM Symposium on Advances in Geographic Information Systems, Kansas City,
USA, November 5th-6th
University of Minnesoda (2003) MapServer Homepage. Available at: http://mapserver.gis.umn.edu/
Mansor S, Shariah MA, Billa L, Setiawan I, Jabar F (2004) Spatial technology for natural risk management. Disaster
Prevention and Management 13(5): 364-373
Montoya L (2002) GIS and Remote Sensing in Urban Disaster Management. 5th AGILE Conference on Geographic
Information Science, Palma (Balearic Islands, Spain) April 25th- 27th
OGC (Open Geospatial Consortium) (2005) About Us. Available at: www.opengeospatial.org/about/
Peng ZR, Tsou MH (2003) Internet GIS. New Jersey, John Wiley & Sons, Inc
Plewe B (1997) GIS Online: Information Retrieval, Mapping, and the Internet. USA, OnWord Press
Rinner C (2003) Web-based Spatial Decision Support: Status and Research Directions. Journal of Geographic
Information and Decision Analysis 1(7): 14-31
Rivereau JC (1995) Spot Data Applied to Disaster Prevention and Damage Assessment. Acta Astronautica 35(7):
467-470
Simonovic SP (2002) Role of Remote Sensing in Disaster Management. Institute for Catastrophic Loss Reduction
(ICLR), Research Paper Series No. 2
Smith K (2004) Environmental Hazards: Assessing Risk and Reducing Disaster, Fourth Edition. New York,
Routledge
Wellar B (1990) Science, Applications, Coherence and GIS: Seizing the Moment. GIS/LIS ’90 Proceedings,
Volume 2, California, November 7th-10th
Wellar B (1995) Geomatics education and training, 1995-2000: trends, issues, opportunities and challenges.
Geomatica 49(3): 336-340
Zerger A, Smith DI (2003) Impediments to using GIS for real-time disaster decision support. Computers,
Environment and Urban Systems 27: 123-141

Canadian Institute of Geomatics

Ottawa, Canada, June 13-15, 2005