Analysing Trade-Offs between Safety from Tsunamis Risk and Views of Ocean Water Using an Optimal Residential Area Model

Disaster risk reduction has become an increasingly prominent concern in urban planning due to recent catastrophic disasters such as the Great East Japan Earthquake in 2011. Building levees or relocating to higher ground are measures used to reduce the risk of tsunamis. However, if those measures are implemented too extensively, they may obstruct views of coastal water that benefit residents. Environments where residents can visually access the waterfront are crucial for promoting awareness of river and disaster risk reduction and repairing the way we interact with nature. However, in Japan, safety has frequently been prioritized by ignoring the views of coastal water that may be lost.

This study developed an optimal residential area model for analysing the trade-offs between safety from tsunamis and views of coastal water (hereafter, ocean views), which will be able to support detailed urban planning. The model comprises weighted multicriteria, that is, the total tsunami risk and ocean views with controlling optimal allocations of population. Here we optimized “Improved Potential Achievement (IPA).” This indicated the extent to which the respective optimal value achieved has been achieved against the value (improved potential) of the two objectives being optimized alone as a baseline. 

We used the viewshed analysis to quantify ocean views. The analysis used the elevation value of each cell of the digital elevation model (DEM) to determine the visibility of a particular point of the ocean from a specific residential mesh. Using the visibility between specific locations, we conceptualized the index of “the ocean view presence” and “the width of the ocean view”. The ocean view presence expresses how many locations in a particular residential mesh have an ocean view. Meanwhile, the width of the ocean view expresses whether people have a panoramic view of the open ocean or whether they can only see a small area of ocean. We quantified ocean views using these indices.

We applied the model to Kuroshio, a tsunami-prone area along the Nankai Trough in Japan and the optimal residential area is calculated for each 500-meter mesh. The results of the sensitivity analysis that changed the weight β (0<β<1) of the safety from tsunami criteria showed trade-offs in which the more safety from tsunami risk is weighted, the more the view of ocean water in the target area is reduced. If weight β is larger than 0.7, ocean views decreases steadily. This is a case study of a specific area and such results are not spatially consistent in all areas. However, similar trade-offs are likely to be obtained in areas with the ocean and mountains in close proximity. This analytical technique is likely to be useful in pre-disaster recovery planning that explores induction-encouraged residential areas that benefit safety from tsunamis and ocean views.