Download Two Tsunamis on the BC Coast

A Look at Two Tsunamis on the BC Coast with
the Neptune-Canada Tsunami-meters
Richard Thomson (PI-IOS)
Isaac Fine (Modelling-IOS)
Alexander Rabinovich (Long Waves-IOS)
Martin Heeseman (PGC)
Earl Davis (PGC)
Maxim Krassovski (IOS)
NEPTUNE Science DMAS Team
Steven Mihaly (IOS)
NEPTUNE TSUNAMIMETER
3 BPR array in 2600m surrounding
the BPR at ODP Borehole 1026;
the NE leg not deployed.
ODP 889:
1260m
Barkley Upper Slope:
400m
Folger Deep:
100m
Failed: Folger 27-Nov; NW 21-Feb
BPR (Bottom Pressure Recorders)
PGC in collaboration with Bennest Enterprises
developed a novel way to process data from
Paroscientific Digiquartz pressure sensors
substantially enhancing pressure resolution.
This, along with the Cabled observatory enabled
high temporal resolution gives us an unprecedented
look at both seismic and oceanic waves in real-time
1 sec sampling, 0.4Pascal (~0.04mm ocean depth),
5microK.
DART: 15min -> 15s acoustic modem-> satellite
Major Tsunamigenic Earthquakes in the
Pacific since Deployment
Samoan Earthquake
Chilean Earthquake
200 km south of the Samoan Islands
Offshore Maule, Chile
Tuesday, September 29, 2009,
17:48:11 UTC
Saturday, February 27, 2010, 06:34:14
UTC
Mw=8.0
(indicating 1/16th of the energy
released in the Chilean earthquake)
Mw=8.8
(6th strongest Earthquake since 1900,
highest Tsunami wave energy in
Pacific since 1964)
Tsunami travelled 8300km to arrive at
the Neptune array 11 hours later,
averaging 755km/h
Tsunami travelled 10,650km to arrive
at the Neptune array 16 hours later,
averaging 666km/h
Well defined narrow band (9-13min)
wavetrain with trough to crest height
~5cm
Broad band wave energy (5-150min)
max trough to crest wave height
~6cm
BPR Antenna
SAMOAN TSUNAMI
• Primary arrival consisted of 4
~5cm waves with decreasing
times between wave crests of
12min 50s, 10min 54s and 9min
54s, following waves were
reduced to 2-3cm.
•With waves traveling in this
direction the arrival times at the
array nodes were only separated
by ~30s - 1min. The third leg
arrival time would be 4.5min
later and thus greatly improve
accuracy. It would also enable
error estimation, or possibly test
the plane wave assumption.
•Using the three BPR array and
shallow water wave dynamics,
wave characteristics could be
estimated. (100km wavelength
travelling 170km/h at 56 deg
azimuth)
Northwest BPR
CORK BPR
South BPR
10km
IOS Regional Tsunami Numerical
Model
Comparison between observations and model
Data
Model
•Driven by the BPR tsunami data;
model initialized using the first
observed tsunami waves transposed
to the boundaries
•A no-frills model optimised by Isaac
Fine to run quickly based on the
linear shallow water equations
20 min
cm
•Grid spacing 180m x 160m, very
sensitive bathymetric accuracy
Neptune-Canada BPR array
80km
26 min
120km
140km
38 min
10 Day Section of Tide-removed filtered pressure from
BPR 1026 S
0.1
0.08
4.5h data
gap resulting
in unreliable
data for ~day
Seismic
energy
decibar
0.06
0.04
Pre-event
wave energy
0.02
0
-0.02
-0.04
02/24 02/25 02/26 02/27 02/28 03/01 03/02 03/03 03/04 03/05
18h Zoom from 27-Feb 20:46:43 to 28-Feb 14:36:46, 2010
CHILEAN TSUNAMI
DATA
(DMAS Related Issues)
• Improvements: most BPRs were
recording, so that pre-event pressure
spectra can be generated; seismic waves
were recorded.
• Demerits: array no longer functional, NW
BPR stopped recording 6 days prior to the
event, giving rise to less accurate wave
representation for the Regional Tsunami
model; data gaps greatly affect processing
WAVE CHARACTERISTICS
0.03
• Marginally larger (6 cm amplitude) waves
compared to Samoan tsunami; ringing for
half a day.
0.02
decibar
0.01
0
• Elevated energy for at least 3 days
-0.01
-0.02
-0.03
140min
15min ticks
• Complex Wave train with wave groups of
5-6, 9-13 min (as in Samoan) waves
superposed on much longer 120-140 min
waves
IOS Regional Tsunami Model
• Both the complexity of the wave and the loss of BPRs in the array limited our ability to
extrapolate the observed pressure at 1026B to a wave at the model boundaries
• The modelled tsunami has spatially
variable skill such that one pressure
record at neighbouring pairs can be well
represented whereas the other not; likely
a result of inaccurate bathymetry.
Comparison of the two Tsunamis at Tidal Stations
•Although open ocean wave heights for the two
events were similar, the coastal responses were
functions of the specific features of the two tsunamis
•The BC coast acts
as an active
“transfer function”
for incoming
tsunami waves
110min
Vertical Scale
3:1
• Note: no data gaps
in 20 records!
50min
Port Alberni
(eigenfreq=120min)
120min
50 cm
150min
25 min
Prince Rupert
FIN