generated at the eastern end of the trench were considered important. Since 

 tsunami generation is equally probable anywhere along the trench, the proba- 

 bility of generation at the eastern end of the trench was one-fifth the value 



predicted by Equation 5. (The eastern end of the trench is approximately one- 

 fifth the total length of the trench.) 



Use of Numerical Model 



27. Using the vertical permanent uplift of the seabed presented by 

 Plafker (1969), the numerical model was used to simulate the behavior of the 

 1964 Alaskan tsunami. Tsunami elevation time-histories predicted by the model 

 were saved at numerical gage locations throughout the study area. The model 

 results indicated the average runup adjacent to the source region of the 1964 

 tsunami to be 11.8 ft and its intensity to be 2.4. 



28. The uplift distribution over the rupture zone of the 1964 Alaskan 

 earthquake was used to establish a record of tectonic deformations of the sea- 

 bed. Each uplift distribution was given the same shape but a different magni- 

 tude from that of the 1964 event. To synthesize the record of uplifts in ac- 

 cordance with the linear model equations, the ratio of the uplift heights of 

 two different tsunamis was equal to the ratio of the average runup heights on 

 the coast. This ratio is equal to 2 ^ -1 2' for tsunamis with intensities 



ii and ip . Since the uplift heights and intensities were determined for 

 the 1964 event, a record of uplift heights was established by allowing tsunami 

 intensity to vary from -1.0 to 4.5 (incrementing by 0.5) and then calculating 

 the associated uplift heights. The lower limit was chosen because numerical 

 experiments indicated tsunamis having lower intensities did not affect the 

 100- and 500-year combined tsunami and tide elevations. The upper limit was 

 chosen because the largest tsunami intensity ever reported was less than 4.5. 



29. The numerical model was used to simulate propagation of the tsunami 

 caused by each of the 12 uplifts. For each of the 12 simulated tsunamis, 



24 hr of tsunami elevation time-history were saved at numerical gage locations 

 throughout the study area. The intensity of each of the 12 tsunamis was cal- 

 culated using the model results and Equation 4. The calculated intensity was 

 then used to assign probability using one-fifth the value found using Equa- 

 tion 5. 



15 



