Tsunami Early Visualization and Warning System

Date: 4/25/2019

Time: 06:00 PM

Room: Grand Ballroom

In the moments following a potentially tsunamigenic event chaos reigns as scientists scramble to gather and assess any sensor data that might be available to help them evaluate the threat. The first notification generally comes from a seismic analysis center, but then what? There is a plethora of data that can be gathered, combined and ranked in importance to make that all important decision – is a tsunami possible? Clearly the initial hypocenter is not enough to make that determination, and the in the next hour duty scientists search for and consider many different kinds of sensor data as the situation unfolds. The first supporting information is likely to be GNSS and hydroacoustic waveforms followed by moment tensors, DART (Deep-ocean Assessment and Reporting of Tsunamis) and tide gauge data as they become available. Much rides on the correctness, timeliness and accuracy of the situational assessment.

This talk demonstrates a prototype for a real-time, cognitive visualization (dashboard) that automatically searches for pertinent data and presenting it to the scientists for assessment. Computers do what computers do best and let people do what people do best to make sure that decision makers and first responders have the best information available as timely as possible. The key element is a multimodal data fusion engine - part of the real-time xQuake earthquake analysis system that was developed for the Air Force for combining multiple forms of data for detecting and characterizing clandestine nuclear testing. Based on the open source xGraph technology, it is both a program and a graph database that can be configured while operational. Data sources can be added (or deleted) and seismic network detectors can be changed to keep the system current in an age when budgets are tight and whatever data that is available must be utilized as effectively and efficiently as possible.

Presenting Author: Caryl Johnson

Authors