Upload
phamdiep
View
229
Download
0
Embed Size (px)
Citation preview
A geodetic matched-filter search for slow slip with application to the Mexico subduction zone
Baptiste Rousset, Michel Campillo, Cécile Lasserre, William Frank, Nathalie Cotte, Andréa Walpersdorf, Anne Socquet & Vladimir Kostoglodov
Cargèse workshop 2017
Slow Slip Events in Mexicodi
spla
cem
ent (
m)
From Radiguet et al., Nat. Geo., 2016
série de SSE
Series of Mw 7.6 Slow Slip Events in the Guerrero Area with ~ 3 years recurrence intervals.
Low Frequency Earthquakes in Mexico
LFEs detected along the MASE profile clustered in two spots :
- the sweet spot - the transient zone
From Frank et al., EPSL, 2014
Inter-SSE bursts of LFEs
Bursts of LFEs during the inter-SSE period, on the transitional
zone
From Frank et al., EPSL, 2014
-0.015
-0.01
-0.005
0
0.005
0.01
0.015
0.02
Dis
plac
emen
t (m
)
Average short-term SSE
From Frank et al., GRL, 2015
Processing of the GPS time series
Processing of the GPS time series
5.2
5.4
5.6 PINO North
−0.05
0
0.05
−0.05
0
0.05 Residuals
2000 20102005 2015
Disp
lace
men
t (m
)
1. Raw data processed with Gamit / Globk
4. Take the temporal derivative
3. Removal of slow-slip periods and linear inter-SSE trends
time (yrs)
disp
lace
men
t (m
)
2. Removal of interseismic signal, linear term, co-seismic step and post-seismic signals
The synthetic templates
Processed GPS Time Series
Library of synthetic signals
104oW 102oW 100oW 98oW 96oW 94oW 15oN
16oN
17oN
18oN
19oN
20oN
21oN
ACAPACYA
ARIG
AYUT
CALC
CAYACOYU
CPDP
DEMA
DOAR
HUAT
IGUA
LAZAMARO
MEZC
MRQL
OAX2OAXA
OMTP
OXAC
OXEC
OXES
OXGU
OXLP
OXMA
OXNC
OXPE
OXTHOXTU
OXUM
PAPA
PINO
SLUISMRC
TCPN
TOL2 UNIP
YAIG
ZIHP
1 mm
Parameters : - Location - Amplitude of slip - Sliping area - Duration
From Rousset et al., JGR., 2017
104oW 102oW 100oW 98oW 96oW 94oW 15oN
16oN
17oN
18oN
19oN
20oN
21oN
ACAPACYA
ARIG
AYUT
CALC
CAYACOYU
CPDP
DEMA
DOAR
HUAT
IGUA
LAZAMARO
MEZC
MRQL
OAX2OAXA
OMTP
OXAC
OXEC
OXES
OXGU
OXLP
OXMA
OXNC
OXPE
OXTHOXTU
OXUM
PAPA
PINO
SLUISMRC
TCPN
TOL2 UNIP
YAIG
ZIHP
1 mm
Processing of the GPS time series
Library of synthetic signals
Dis
plac
emen
t (m
m)
Dis
plac
emen
t (m
m) OXAC - North Component
OXAC - East Component
0 100 200 300 400 500 600
-5
0
5
0 100 200 300 400 500 600
-5
0
5
Synthetic time series White and coloured noise
+ transient events
Synthetic tests
Library of synthetic signals
Processing of the GPS time series
Library of synthetic signals
Weighted Correlation
(1) Computation of the cross correlation for each individual GPS time series
Dis
plac
emen
t (m
m)
Dis
plac
emen
t (m
m) OXAC - North Component
OXAC - East Component
0 100 200 300 400 500 600
-5
0
5
0 100 200 300 400 500 600
-5
0
5
(2) Sum all the correlation functions, weighted by the amplitude of the synthetic template
⇤
Weighted correlation function
Weighted correlation function
Processing of the GPS time series
Library of synthetic signals
Detection: Weighted Correlation
Normalised correlation function, stacked over the East and North time series, and weighted by the amplitude of the synthetic template
From Rousset et al., JGR., 2017
Time (days)
Cor
rela
tion
Coe
ffici
ent
Dis
plac
emen
t (m
m)
Dis
plac
emen
t (m
m) OXAC - North Component
OXAC - East Component
0 100 200 300 400 500 600
-5
0
5
0 100 200 300 400 500 600
-5
0
5
0 100 200 300 400 500 600-0.2
-0.1
0
0.1
0.2
0.3
Stack of the GPS time series
Processing of the GPS time series
Library of synthetic signals
Detection: Weighted Correlation
Stack of displacement GPS
time series
0 100 200 300 400 500 600Time (days)
-4
-3
-2
-1
0
1
2
3
4
5
Dis
plac
emen
t (m
m)
Raw StackWeighted Stack
Sum of the GPS time series, weighted by the amplitude of the synthetic template
From Rousset et al., JGR., 2017
Estimation of the events duration
Processing of the GPS time series
Library of synthetic signals
Sum on stations and components
Extraction of the magnitude and the
duration10 20 30 40 50 60 70 80 90 100
0
0.5
1
1.5
2
2.5
3
3.5
Time (Days)
Dis
plac
emen
t (m
m)
10 20 30 402.9
3.1
3.3
3.5
Template Duration (Days)
Resid
uals
(mm
)
0
Detection: Weighted Correlation
Estimation of the events magnitude
Processing of the GPS time series
Library of synthetic signals
Sum on stations and components
Extraction of the magnitude and the
duration
10 20 30 40 50 60 70 80 90 1000
0.5
1
1.5
2
2.5
3
3.5
Time (Days)
Dis
plac
emen
t (m
m)
5.5 6 6.5 7 7.5
Magnitude
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Res
idua
ls (m
m)
5
1 patch
9 patches
25 patches
Detection: Weighted Correlation
Application to the Guerrero GPS data
Aug 2005 Feb 2006 Aug 2006 Feb 2007 Aug 2007 Feb 2008 Aug 2008 Feb 2009 Aug 2009 Feb 2010 Aug 2010 Feb 2011 Aug 2011 Feb 2012 Aug 2012
0
0.02
0.04
0.06
0.08
0.1
10 -4
10 -3
Recc
uren
ce in
terv
al (s
)
Trem
or e
nerg
y (m
2 .s-1)
0
5
4
3
2
1
10-11
Corr
elat
ion
coeffi
cien
t
2006 SSE 2009 / 2010 SSE
Detection of 28 events from 2005 to 2013From Rousset et al., JGR., 2017
Example of stack an transient event
-10 0 10 20 30 40 50Time (days)
-40
-20
0
20
Disp
lace
men
t (m
m) Weighted Stack - Event 8
2007 2007.1 2007.2 2007.3 2007.4 2007.5
0
20
40
60
80
100
120
disp
lace
men
t (m
m)
North Component
ACAP
ACYA
CAYA
COYU
CPDP
DOAR
IGUA
LAZA
MEZC
YAIG
2007 2007.1 2007.2 2007.3 2007.4 2007.5
0
20
40
60
80
100
120
disp
lace
men
t (m
m)
East Component
ACAP
ACYA
CAYA
COYU
CPDP
DOAR
IGUA
LAZA
MEZC
YAIG
From Rousset et al., JGR., 2017
Locations of the detected events
1
2
3
4
0.02
0.050.1
0.15
Mw 7.1
Mw 6.8
Mw 6.5
Mw 6.2
5
67
10
11
13
14
15
12
8
9
16
17
19
20
21
2223
25
2627
28
18
24
0.15
0.05
0.1
0.02
0.2
0.25
0.15
0.1
0.05
SSE
2006
SSE
2009
/ 20
10
2006
2007
2008
2009
2010
2011
2012
2013
2014SSE
2014
16 N
17 N
18 N
19 No
o
o
o
102 W 101 W 100 W 99 W 98 W 97 Wo oo o o o
16 N
17 N
18 N
19 No
o
o
o
16 N
17 N
18 N
19 No
o
o
o
102 W 101 W 100 W 99 W 98 W 97 Wo oo o o o 102 W 101 W 100 W 99 W 98 W 97 Wo oo o o o
16 N
17 N
18 N
19 No
o
o
o
From Rousset et al., JGR., 2017
Conclusion
We developed a method: (1) (1) to detect short-term transient aseismic events (2) (2) to characterise their location, duration and magnitude
Compare to other methods, like the NIF: - - The detection threshold of the geodetic matched filter is lower, because it
takes into account the redundancy of information in a network of stations.- - It is particularly suited for short-term transients. - - It doesn’t allow neither to characterise the spatial extent of slow slips, nor
any propagation.
Applied to the Mexico subduction: - We have detected 28 new transients events (Mw 6.3 to Mw 7.1) from 2005 - 2014.- They are located at the down-dip edge of the large slow-slip events.