As discussed in great deal in previous sections throughout this project, the Greater Vancouver Regional District (GVRD) is exposed to the highest levels of seismic risk in Canada, and dense populations and aging infrastructure in Vancouver puts this region at an even greater risk for a potential disaster in the event of an earthquake. By using Geographical Information Systems (GIS) we have modeled a method of analysis to determine high and low risk areas (in the event of an earthquake) based on several factors: potential slope failure, proximity to transformer stations and large water pipeline systems, coastal regions, low elevations and areas subject to liquefaction based on underlying soil material. The GVRD has been expecting significant seismic activity for quite some time now. Based on geological evidence and research large scale seismic activity such as a devastating earthquake happens every 300 to 500 years, and the last record of this type of seismic activity in this region was between 400 and 500 years ago. Therefore the GVRD is due for a potentially devastating earthquake, and it is in the best interest of the City of Vancouver and the University of British Columbia to plan for these events now. This is where we drew our inspiration from in the choice of this topic for our final project.
On November 4th, 2011 our group was invited for a tour of the E-Comm building (the emergency communications center for southwest British Columbia) and a meeting with Kevin Wallinger, Director of Emergency Management for the City of Vancouver and Daniel Stephens, Manager of Emergency Planning. This meeting was critical to our understanding of what results we were hoping to produce and what type of GIS analysis that would achieve these results. Both Kevin and Daniel identified local community centers as locations around the city where civilians would be able to have access to shelter, food, running water and various supplies. They explained that there are unmarked steel storage containers at every community center that contain basic supplies including cots, blankets, propane, food, water, first aid supplies etc… In the event that there was a devastating earthquake, the general population is urged to return to their residence. Part of emergency planning involves preparing households with food and water. If the residence is inhabitable for any reason, the public is then urged to go to community centers.
We used several basic shapefile data setsin our GIS analyses, they included: a Vancouver outline, underlying geology, roads, community center locations, major hospital locations, slope, transformer station locations, and water pipeline systems. The first set of maps we created was the hospital and community center ‘Thiessen’ polygon maps. We determined that creating traditional circular buffers around each community center and hospital would not satisfy our needs. The Thiessen polygon tool builds buffer polygons around each location in a puzzle-piece layout – dividing the entire city into polygons. In our first attempt we discovered the polygons did not cover the entire city of Vancouver boundary so we needed to create four ‘null’ points in the shape of a rectangle around the city, encompassing everything. Then using the Geoprocessing-clip we clipped the polygons using the city of Vancouver boundary. The idea of the polygons is to indicate which community center or hospital you would go to in the event of an emergency, given your location.
Our second set of maps included two Cost Path/MCE (Multi Criteria Evaluation) maps that indicate routes to the nearest hospital and community center based on a random location. Our objective with these maps were to include several tools and processes that we are had used in the labs this term. There are several components to these maps; we did a Multi Criteria Evaluation that took into consideration slope, underlying geology subject to liquefaction (landfill, peat, silt and clay), and proximity to transformer stations and water pipeline systems. Without going into great detail, our final MCE layer has three colours.
We predict that green areas have relatively low danger of mass destruction based on the factors listed above, meaning these areas have zero ‘dangerous factors’. The yellow indicates areas that have two dangerous factors, and subsequently red areas indicate the most dangerous areas, that have two or more dangerous factors. From there we could create a land cost layer that weights the green areas less than the red areas. Then using the Cost Path tool we created a path from an arbitrary location to the closest hospital and then the closest community center. The purpose of this was to demonstrate how our analysis could be put to use in ‘real-life’ in the event of an earthquake. When we met with Kevin Wallinger, we discussed a previous project that had been undertaken by a group of Simon Fraser University students. They had met with Kevin and Daniel and they designed an iPhone app called QuakeAware. Today most people carry around smart phones with built in GPS systems and oddly enough this app was missing a map component. Kevin and Daniel suggested we create an analysis that could be adapted to a real world use. Given more time and resources our analysis could be added into a Google Maps type system. If an earthquake was to happen and assistance was needed, someone could launch their iPhone app that could determine their exact location and could draw the safest route to the closest community center or hospital. People work, live, and study in different places geographically therefore this app could potentially save lives. The existing QuakeAware app contained information of simple first aid procedures and other life-saving techniques. Our mapping component could build on this idea, and obviously be expanded upon given more time and resources.
Overall we felt that our analysis was successful. Our objective was to use a method of GIS analysis that we had been exposed to in the labs this semester, and to perform an analysis that could potentially save people’s lives in the event of a natural disaster such as an earthquake. Kevin Wallinger and Daniel Stephens were a tremendous source of guidance and information in our project. We have met the objectives of this project, but given more time and more resources (data, money) our analysis could go much deeper and become more advanced. We understand that underlying geology, slope, proximity to transformer stations and waterline systems are not the only or true indicators of potential danger areas. We had wanted to include age of infrastructure including all buildings in our analysis, but that data was not available. We considered including a population analysis – looking at potentially how many people in a given area could visit a community center or hospital, but the reality most of us are constantly on the move and an earthquake can strike at any moment and census population would not reflect the daytime vs. night time populations. UBC for example has a substantially higher population during the day than what is reflected in the census. More extensive research and understanding into underlying geology would have benefitted our analysis of our project. This type of analysis is a project that would be taken on by a city, with thousands of dollars and more sensitive data at their disposal. We would have liked to map many more examples of more arbitrary locations (where you could be caught during an earthquake) to more community centers and hospitals do demonstrate our analysis, but unfortunately time did not permit. Nonetheless we have met the project objectives and hopefully one day our analysis could be expanded upon and possibly a collaboration project could be established between this idea and the QuakeAware app.
