MEJORAS A CRITERIOS DE SELECCIÓN Y MANEJO DE LA INCERTIDUMBRE EN EL SISTEMA INTEGRADO DE PREDICCIÓN Y ALERTA DE TSUNAMIS (SIPAT)
Catalogado desde la version PDF de la tesis.
At present, Chile has the ¨Integrated System of Prediction and Alert of Tsunamis-(SIPAT).This system consists of a seismic database and pre-calculated tsunami.The methodology ofoperation is that at the occurrence of an earthquake, the database is consulted with thepreliminary seismic parameters (location of the epicenter and magnitude), the purpose isgenerating a risk assessment of the tsunami in a sectorized way along the whole coast ofthe territory of Chile. Given the uncertainty inherent in the seismic rupture process, SIPATinitially adopted a conservative process, Which needs to be better evaluated. At presentand depending on the magnitude of the reported earthquake,can be given overestimates orunderestimates of the peak tsunami amplitudes.The objective of this work is to improve the management of uncertainty regarding thelocation of the epicenter in the area of rupture and the uncertainty due to the random distributionof slip. These are faced by SIPAT from two perspectives, the uncertainty regardingthe location of the epicenter is faced with the application of the maximum risk method alsocalled location filter. While the slip distribution is handled by the magnitude filter appliedto the scenarios contained in the database.In order to improve the management of the uncertainty related to the location of theepicenter, a sensitivity analysis of the areas of uncertainty used for the maximum risk methodin SIPAT was performed using Tsunami Maximum exceedance Curves (CEAMT). For this,four different areas were analyzed, one of which coincides with the area used by the JMA.The old system considered an area of uncertainty 2L x 2W, but the present work concludesthat an area of L by W reduces the forecast of tsunami amplitudes, and obtains alert levelscloser to the actual measurements of the levels stations of sea, for recent tsunamis.In the second instance and as a way to improve the handling of uncertainty due torandom slip distribution, a method of interpolation is proposed. This method provides abetter criterion for selection of candidate scenarios, which includes a reliable and flexibleparameter to calibrate the system. The method was successfully validated by comparing themaximum amplitudes generated by the method with the tsunamis modeled with their actualfocal parameters. As a result it was obtained that when considering an equivalent magnitudeequal to Mw +Mw in the interpolation method, where Mw was considered equal to 0.3 according to the results of Mueller et al. (2015), overestimates the risk.As future research, it is recommended perform a sensitivity analysis or analysis by stochasticslip models in the rupture area to calibrate the Mw parameter for the Chileancoast.