Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

The WARN module within Oceans 2.0 provides subscribers with early notifications of tsunamis and earthquakes as part of the Smart Oceans BC initiative. It utilizes some infrastructure already in place for existing Oceans 2.0 data acquisition infrastructure while incorporating some new instruments and new event detection algorithms and software. Tsunami detection uses data from pressure sensors such as the existing Bottom Pressure Recorders (BPR) and some planned coastal radar devices.  Earthquake detection uses new sensors providing p-wave data, that were deployed in late 2014 but will be upgraded in early 2015. While WARN was implemented as a research/prototype with limited sensors and select early adopters for now, it is being designed to be expandable to greater numbers and types of instruments (such as coastal radar for tsunami detection) and subscribers.

...

Further details are available in the attached presentation:

WARN Overview

 Earthquake Earthquake Detection

Earthquake detection is done using a number of accelerometers located on land and on the seafloor west of Vancouver Island. Normally they only output a characteristic amplitude value, the Japanese Meteorological Agency's (JMA) instrumental intensity, which is not used at the moment. When an earthquake is detected, driver software Instead, WARN's driver software, which runs on a computer collated with the accelerometer, analyzes the accelerometer data on all three axes and looks for the signature of a "P-wave", which is the initial compression wave emanated by the earthquake. This P-wave does not cause damage but travels much faster than the "S-wave" that comes later and causes the damage. If a P-wave is detected the driver outputs the time of detection of the P-wave. Then the driver analyzes the first few seconds of motion and determines the maximum displacement of the vertical component of the acceleration, which we call Pd, and the maximum period of the vertical component, which we call Tau. The Pd and Tau are then sent out by the driver located beside the accelerometer computer to the Oceans 2.0 software running at the University of Victoria.

The P-wave time, Pd, and Tau are analyzed by two algorithms to determine the epicenter and time of origin, and two different algorithms to determine the magnitude (the JMA algorithm is not used at this time) - see diagram below.

The epicenter is determined provided at least three accelerometers have reported a P-wave detection. The P-wave detection times are analyzed using a Direct Grid Search algorithm where a pre-defined region is divided into 9,800 cells of equal size (9,800 for the current area, configurable). Each cell is analyzed by triangulation, knowing the location of the center of the cell and the relative detection times of the P-wave at each of the sensors, plus the location of the sensors. The cells are first analyzed on a coarse grid basis to determine the cell with the highest probability of being the epicenter of the earthquake. This cell is then further analyzed on a fine grid using the same method. In the end the center of the most probable fine grid cell is used as the epicenter latitude and longitude, with a resolution of +/- 0.05 degrees. If the quality of the solution is considered high enough, an earthquake is declared by the Earthquake Correlator and the results are passed to the Event Notification component.

...

The magnitude of the earthquake is estimated using two methods. One method compares the Tau (period) values reported by the accelerometers, which is expected to be the same at all sensors. The magnitude is proportional to the Tau the Tau on a logarithmic scale. The other method uses the Pd or maximum vertical acceleration reported by each of the sensors. The Pd is attenuated by the distance to the sensor on a logarithmic scale. So after factoring in the attenuation due to distance, the magnitude is logarithmically proportional to the Pd.

The epicenter coordinates, time of origin, and magnitude are passed to the Event Notification software. 

 

 Tsunami Tsunami Detection

Tsunami detection is done using Bottom Pressure Recorders (BPRs) which are located on the seafloor at various locations west of Vancouver Island. These are highly sensitive instruments that can measure the weight of the water (and atmosphere) above them and can detect changes in the water depth of as little as 1mm in some cases. Although tsunamis can be quite large at the shoreline, they are quite a bit smaller in the deep ocean.

...

The second algorithm is called STA/LTA, which stands for Short Term Average over Long Term Average. It computes the average of the most recent 200 (squared) detided samples divided by the average of the most recent (squared) 2400 detided samples. This improves the signal-to-noise ratio. The output is compared with a threshold.

...

For this particular BPR (only), a tsunami is declared if any 2 out of 3 tsunami detection algorithms exceed their thresholds within a certain time window AND and the Seismic Detector DOES NOT does not detect a seismic wave at the same time. This is the function of the Watcher.

...

  • Time of detection by the third BPR
  • coordinates of all the BPRs that made the detection
  • (additional information is contained in the logs including wave heights)

 Software Software in the device receiving the message (eg. client application or WARNAlert iPhone application) can further process this information. By knowing the GPS coordinates of the client the software can determine the distance to the epicenter and hence predict the arrival time of the S-waves at that location. The client software can also predict the level of shaking at the client location.

Anchor
WARNLicencing
WARNLicencing
 

Licencing

No licencing is required to become a subscriber of WARN events. However at the present time subscription to earthquake and tsunami events is only possible by invitation and only to those with an account to Oceans 2.0.