LINCOLN COUNTY TSUNAMI EVACUATION ANALYSIS
Introduction
Tsunamis are one of the most powerful natural disasters on Earth. An estimated 430,000 people have lost their lives since 1850 as a result from various tsunamis around the world. Tsunamis depend on seismic events such as earthquakes, which occur on a daily basis, which presents a risk to anyone living on a coastline, especially if it is near a subduction zone. The Cascadia subduction zone is long overdue for a large earthquake, which in effect would develop a large tsunami in the Pacific Ocean. The damage along the coast would be devastating but nothing is more costly than human lives, which is why it is important to study tsunami evacuation so injuries and casualties are prevented. GIS has important applications and one of the most important is displaying an analysis of data to support the process of environmental decision-making. Often decisions are choices between alternatives, different actions, objects, events, or locations. J.R. Eastman defines multi-criteria evaluation in GIS as being “concerned with the allocation of land to suit a specific objective on the basis of a variety of attributes that the selected areas should possess” (Eastman 493). Multi-criteria evaluation is a suitable method for cases involving tsunami evacuation because it uses and evaluates multiple criteria such as elevation, local rivers, streams, and water bodies, then displays the best locations for evacuation buildings that meets all the criteria. Lincoln County is a primarily coastal county in Oregon. Many locals are aware that they are at risk of distant tsunamis and local tsunamis produced by the subduction zone but are they a bit tranquil about their preparation for “the big one?”
Methods
First, I had to collect and organize the data that included every river, stream, road, and body of water within Lincoln County. I also gathered the county DEM, or digital elevation model, so I had another type of criteria to use. Once I had all of the criteria defined, I needed to determine areas at risk of being affected by a large tsunami. When picking evacuation routes to safe locations, The National Tsunami Hazard Mitigation Program recommends choosing routes or locations “100 feet (30 meters) above sea level or as far as 2 miles (3 kilometers) inland, away from the coast.” In general, every step you take upwards and inland makes a difference. NOAA also suggested the same ideal distance and elevation. However, I found various inundation maps on the Oregon Geology website that suggest tsunamis have not reached 1 kilometer inland yet in Lincoln County, as seen in Figure 1: Local Source (Cascadia Subduction Zone) Tsunami Inundation Map Lincoln City North, Oregon. Therefore I needed to use a value in between both recommendations that conformed with the next step in the process, which involved human travel times. I needed an estimation on how far the average human can travel on foot before a tsunami makes landfall. Local tsunamis are the quickest to reach shores with an estimated 15-20 minutes after the earthquake so I used 15 minutes for the time part of the equation. The average human can run about 12-15 mph at short distances and walk 4mph, with a 15-minute mile time. Using 15 minutes and roughly an average speed of 5 mph, I estimated that the study site should be at about 1.25 miles (2 kilometers) away from the ocean since most people would be able to reach that distance in the 15 minutes prior to landfall even if they were right on the beach. The next step was ruling out water features that would potentially be affected by the tsunami. Running on roads would be the best decision for evacuators so I created a path distance raster for roads and a cost distance raster that represented the cost traveling from tsunami risk zones to the safe zones. Once all of the raster data was complete, I standardized the variables for each of the criteria. Finally, I ran a model that calculated the criteria, analyzed the results, and displayed the best locations for evacuation routes and safe places (Figure 2). Once, the results from that calculation were mapped, I ran a second weighted linear combination (Figure 3). After that step, I made one last alter to the weighting scheme and calculated those results in Figure 4.
Figure 1:
Figure 2:
Figure 3:
Figure 4:
Results
My results were a sum of each decision-making variable in the raster calculation. Figure 5 shows the ocean, streams, roads, and rivers in Lincoln County, as well as a base map. This figure also included the calculated results from the equation in Figure 2. According to the six criteria used in the evaluation, the ideal evacuation safety zones are located in the red areas on the map, while the most at risk areas are yellow. That scale was a result of the first weighting calculation, which placed more emphasis on the distance away from the water, higher elevations, and proximity to roads, than the other criteria. As you can see, all of the red zones at the furthest reaches of the study site at distances around least two kilometers inland, and away from water. That was the outcome that I wanted to see displayed on my map, which is why I gave more weight and influence to those variables. After I altered my weighting scheme the first time, there was not much change in the tsunami safe zones. Figure 6 shows that the only difference is a very slight shift to more reds and oranges around the greater Newport area suggesting that the criteria was a hair more flexible. The weighting in my second calculation was more evenly distributed between each variable than in the first calculation. Once again the most suitable locations for tsunami safe zones were on the outskirts of the study site and well away from the ocean and large bodies of water. Figure 7 displays the results from my third and final calculation, which conforms to Figure 4. This time I put the most emphasis on the cost distance traveling on the surface and path distance from roads, while giving a small emphasis to elevation and distance from water, as well as a very small emphasis to distance from rivers and streams. The results were much different than in the previous two calculations. Safe zones were closer to shore. There was hardly any yellow, unsuitable locations, hence why you see an abundance of red and orange throughout the study site. I recommend that Lincoln County uses my first multi-criteria evaluation results for their tsunami evacuation zones. Figure 5 shows consistency and the most accuracy in terms of the zoning. There should be a safe zone a little west of Toledo to Toledo, north of Toledo towards Siletz, northeast of Lincoln City towards Otis, southeast of Kernville, and east of Waldport towards Tidewater which is not a coincidence regarding the town’s name.
Figure 5:
Figure 6:
Figure 7:
Conclusion
In the beginning of this project, I thought ideal tsunami evacuation zones might not be realistic when comparing research with my buffers and layers on the map. Apparently, humans cannot run as fast as I would expect, well short of the median 2-mile run times in my high school PE class, and therefore most humans on the beach would be at great risk if a large, local Cascadia-triggered tsunami was scheduled to make landfall within 15 minutes. By doing this lab, I realized that the local and statewide inundation maps may be using dangerous areas for evacuation safe zones in terms of the distances from water and elevation. Multi-criteria evaluation is a fantastic method for cases involving tsunami evacuation because of the way it calculates more than one variable such as elevation, local rivers, streams, water bodies, and then displays the best locations for evacuation zones that meet all of the criteria. Using different weighting schemes often lead to different results as long as the data and approach are not too narrow. By using three different weighting schemes, I was led to the map of best fit in terms of suitable tsunami evacuation safe zones. If people are unsure where exactly to run when they hear the tsunami warning sounds, just run as fast as you can inland and upward if possible. Others will know that they need to run quickly towards Toledo, southeast of Kernville, Otis, Siletz and Tidewater.
References
Eastman, J.R. (2000). Multi-criteria evaluation and GIS. Retrieved from http://www.geos.ed.ac.uk/~gisteac/gis_book_abridged/files/ch35.pdf.
National Oceanic and Atmospheric Administration. What’s Your Tsunami Preparedness? Retrieved from http://www.noaa.gov/features/tsunami/preparedness.html.
National Tsunami Hazard Mitigation Program. Tsunami Awareness and Safety. Retrieved from http://nws.weather.gov/nthmp/tpw/tsunami-factsheet.html.
Oregon Geology. Local Source (Cascadia Subduction Zone) Tsunami Inundation Map Lincoln City North, Oregon. Retrieved from http://www.oregongeology.org/pubs/tim/Linc01_LincolnCityNorth_Plate1_onscreen.pdf