GIS can be used in different phases of the disaster management cycle. In my experience, we use GIS and remote sensing in Central America and the Caribbean to improve environmental monitoring and disaster management (www.servirglobal.net).

Using Alexander's definition of the disaster cycle (Principles of Emergency Planning and Management, 2002), we can start in the mitigation phase. GIS datasets such as topography, land cover/land use, infrastructure, populations, etc., are all layers of information we need to know in order to reduce the impact of future disasters. The bio-physical datasets help us to characterize the actual hazard (e.g., flood, landslide, storm surge, earthquake) and its potential area of impact. The human/cultural datasets indicate the assets that are exposed to those hazards. Attributes of those assets can be analyzed to understand the vulnerability of certain populations to a disaster, given their age, access to roads and alternate routes if one is blocked or damaged, for example. Attributes of assets such as infrastructure can also give us insight into the vulnerability of buildings, roads, lines of power and communication, etc. All of this is possible because GIS allows us to intersect the physical hazard with the elements at risk.

This brings us to the preparation phase, which comes just before the disaster impact. Overlaying or intersecting those layers of information allows us to estimate, often in real time, populations potentially affected by a looming event. Examples could be rainfall forecasts from a numerical weather model or flood forecasts from a hydrological model, which contribute to pinpointing specific areas that may experience extreme conditions. If those areas intersect with populations, that could be recipe for disaster.

Then comes the response phase after disaster has struck. Resources are most stressed during this time as every effort is immediately directed to emergency rescue and recovery. Remote sensing can come into play here because a satellite image's perspective can really help convey the magnitude of a disaster. Using visible and near infrared channels, GIS and remote sensing analysts can quickly estimate the extent of a flood, delineate landslides, and even identify damaged buildings and roads if high enough resolution images are available. (Sidenote on satellite images, disasters and developing countries: The International Charter on Space and Major Disasters is set up for specific agencies in developing countries experiencing a disaster to access satellite images from commercial and government sources in mostly developed countries http://www.disasterscharter.org/home). Interferometry can be used to measure ground deformation after an earthquake or volcanic eruption. On the ground, emergency responders rely on GIS and GPS to most efficiently reach affected locations, keep track of impassible roads and search for alternative routes. First responders also need to know how many people might be in need of help. GIS can be used to investigate census data and can help determine where to locate temporary shelters.

The recovery phase involves restoration and reconstruction. Using past experiences of where and how severely disasters struck should guide planning efforts. This could mean land use planning to restrict development in certain places, direct reconstruction efforts in less-disaster prone areas, or engineering civil works to withstand greater natural forces. Catalogs of historic disasters provide crucial information, especially when they are georeferenced or incorporated into a GIS, because they show land planners the spatial and temporal aspect of disasters.

In my opinion, most efforts related to using GIS for disaster management are focused on identifying or characterizing the physical hazard, rather than analyzing socioeconomic patterns that describe vulnerability. This could be due to data availability. The GIS and remote sensing datasets more readily available in the developing countries I've worked in are related to biophysical characteristics and not human or cultural features. Census and socioeconomic data is much harder to come by, as agencies might not be willing to share such information publicly. Because disasters become disasters when they affect people, I believe that we must understand and address the underlying vulnerability of populations at greater risk. Understanding the physical hazard is one thing, but if people cannot remove themselves from harm or be resilient, our efforts could be in vain.

Great question, and I hope this answer may be useful!

I reccomend to you this book : "GIS in public Health Practice" edited by Massimo Craglia - CRC press - I think that in this book there are interesting ideas that could be helpful to your work

GIS can be used in different phases of the disaster management cycle. In my experience, we use GIS and remote sensing in Central America and the Caribbean to improve environmental monitoring and disaster management (www.servirglobal.net).

Using Alexander's definition of the disaster cycle (Principles of Emergency Planning and Management, 2002), we can start in the mitigation phase. GIS datasets such as topography, land cover/land use, infrastructure, populations, etc., are all layers of information we need to know in order to reduce the impact of future disasters. The bio-physical datasets help us to characterize the actual hazard (e.g., flood, landslide, storm surge, earthquake) and its potential area of impact. The human/cultural datasets indicate the assets that are exposed to those hazards. Attributes of those assets can be analyzed to understand the vulnerability of certain populations to a disaster, given their age, access to roads and alternate routes if one is blocked or damaged, for example. Attributes of assets such as infrastructure can also give us insight into the vulnerability of buildings, roads, lines of power and communication, etc. All of this is possible because GIS allows us to intersect the physical hazard with the elements at risk.

This brings us to the preparation phase, which comes just before the disaster impact. Overlaying or intersecting those layers of information allows us to estimate, often in real time, populations potentially affected by a looming event. Examples could be rainfall forecasts from a numerical weather model or flood forecasts from a hydrological model, which contribute to pinpointing specific areas that may experience extreme conditions. If those areas intersect with populations, that could be recipe for disaster.

Then comes the response phase after disaster has struck. Resources are most stressed during this time as every effort is immediately directed to emergency rescue and recovery. Remote sensing can come into play here because a satellite image's perspective can really help convey the magnitude of a disaster. Using visible and near infrared channels, GIS and remote sensing analysts can quickly estimate the extent of a flood, delineate landslides, and even identify damaged buildings and roads if high enough resolution images are available. (Sidenote on satellite images, disasters and developing countries: The International Charter on Space and Major Disasters is set up for specific agencies in developing countries experiencing a disaster to access satellite images from commercial and government sources in mostly developed countries http://www.disasterscharter.org/home). Interferometry can be used to measure ground deformation after an earthquake or volcanic eruption. On the ground, emergency responders rely on GIS and GPS to most efficiently reach affected locations, keep track of impassible roads and search for alternative routes. First responders also need to know how many people might be in need of help. GIS can be used to investigate census data and can help determine where to locate temporary shelters.

The recovery phase involves restoration and reconstruction. Using past experiences of where and how severely disasters struck should guide planning efforts. This could mean land use planning to restrict development in certain places, direct reconstruction efforts in less-disaster prone areas, or engineering civil works to withstand greater natural forces. Catalogs of historic disasters provide crucial information, especially when they are georeferenced or incorporated into a GIS, because they show land planners the spatial and temporal aspect of disasters.

In my opinion, most efforts related to using GIS for disaster management are focused on identifying or characterizing the physical hazard, rather than analyzing socioeconomic patterns that describe vulnerability. This could be due to data availability. The GIS and remote sensing datasets more readily available in the developing countries I've worked in are related to biophysical characteristics and not human or cultural features. Census and socioeconomic data is much harder to come by, as agencies might not be willing to share such information publicly. Because disasters become disasters when they affect people, I believe that we must understand and address the underlying vulnerability of populations at greater risk. Understanding the physical hazard is one thing, but if people cannot remove themselves from harm or be resilient, our efforts could be in vain.

Great question, and I hope this answer may be useful!

As Eric Anderson said, you can use GIS tool in different phases of the disaster management cycle and for different purposes. I have some experience in the usage of GIS tools to seismic vulnerability analysis at urban scale. If you are interested in this topic, you can find some practical examples in my publications, namely in the one that you can find in attach

Conference Paper Seismic vulnerability assessment of the old city centre of S...

A person in my old university department was involved with MapAction - they went in to disaster zones and used GIS to create up-to-date maps for effectively delivering aid and relief. I think this fits the bill of what you want:

http://www.mapaction.org/