WP6

1
WP6

• Monitoring Earthquakes and Related Environmental
  Hazards

2
WP6 Monitoring Earthquakes and Related
Environmental Hazards

• Participants
• CDCS J.Kurths
• ENS M.Ghil
• IGRA D. Zugravescu, L.Asimopolos, N.Cadicheanu,
  C.Diacopolos, M.Ivascu, A.Marin, M.Popescu,
  D.Stanica, M.Tatu.
• IPGP J.-L.Le Mouël, M.Mandea, C.Narteau
• MITPAN V.Kossobokov, G.Molchan, O.Novikova,
  L.Romashkova, P.Shebalin, A.Soloviev, I.Vorobieva
  
• ULG M.Ausloos, F.Petroni
• IGPP V.Keilis-Borok, I.Zaliapin

• Constituents of the Work Package
• Models of seismicity
• Observational
  
• Data set network
• Earthquake Real-time
• prediction earthquake
• algorithms prediction

3
WP6 Data Set (MITPAN, IGRA)

• Vrancea (Romania) is a geographical region
  between Eastern and Southern Carpathian
  Mountains. The region is characterized by a
  rather high level of seismic activity mainly at
  intermediate (up to 200 km) depths. These
  intermediate-depth earthquakes occur between
  45o-46oN and 26o-27oE. The shallow earthquakes
  are dispersed over much broader territory.
• The dataset on seismic activity has been compiled
  using several available sources. This dataset
  contains the earthquake catalogue for the Vrancea
  region that covers period 1900-2005. Since 1962
  the catalogue is complete for magnitude M 3.0
  and since 1980 it is complete for M 2.5.
• The comparative analysis of earthquake catalogues
  available for Vrancea region has been performed
  aiming at compilation of a data set, as much
  complete and homogeneous as possible, which,
  hopefully, will be used for prediction of strong
  and possibly moderate earthquakes in the region.
  The two catalogues under study are 1) Global
  Hypocenter Data Base catalogue, NEIC and 2) local
  Vrancea seismic catalogue.
• This data set is continued by the RomPlus
  earthquake catalogue that is compiled at the
  National Institute of Earth Physics (Magurele
  Romania). Starting from April 2006 the IGRA
  provides monthly update of this catalogue.

4
Earthquakes in Vrancea and vicinities reported in
GHDB, 1962-2004All events
Events at depths of 60 km or more
5
The annual number of earthquakes from the local
Vrancea catalogue, 1930-2005
6
Gutenberg-Richter graphs for different versions
of the Vrancea catalogue
7
WP6 Earthquake Prediction Algorithms (MITPAN,
IGPP)

• The intermediate-term earthquake prediction
  algorithm M8 are improved and modified to apply
  it to the Vrancea region for prediction of
  earthquakes with magnitude 6.5 or higher.
• The new short-term prediction algorithm based on
  chains of earthquakes as short-term precursors of
  strong earthquakes and the earthquake prediction
  methodology named Reverse Detection of
  Precursors (RDP) is adapted for prediction of
  Vrancea earthquakes.

8
WP6 Real-time Earthquake Prediction (IGRA,
MITPAN, IGPP)

• The real-time prediction experiment by means of
  M8 algorithm has been launched in the region.
• Alarm for a strong (M 6.5) earthquake has been
  declared for the five-year period starting from
  July 2006.

9
WP6 Observational Network (IGRA)

• Specific pattern recognition of the test sites
  for continuous monitoring of the short-term
  precursory phenomena of the tectonic activity
  (earthquakes, active faults and landslides
  associated) has been made.
• The adequate equipment has been installed in two
  relevant geodynamic test sites (Surlari
  Observatory for short term precursory parameters
  and Provita de Sus landslide test site).
• Continuous monitoring of geophysical fields in
  order to reveal the short-term precursory
  phenomena of the tectonic activity (earthquakes,
  active faults and landslides associated) has been
  launched.

10
WP6 Models of Seismicity (CDCS, ENS, IPGP,
MITPAN, ULG, IGPP)

• cellular automata that include long-range, as
  well as nearest-neighbour interactions
• scaling organization of fracture tectonics
  (SOFT)
• colliding cascades model
• model of block-and-fault system dynamics
• accounting for the fracture growth process and
  interactions between different faulting
  mechanisms in the determining the seismogenic
  potential of active tectonic zones

11
Study of Vrancea seismicity

• Temporal properties of seismicity in Vrancea have
  been studied in order to estimate the hazard rate
  distribution of the largest seismic events.
• The average return period of the largest events
  has been estimated on the basis of Generalized
  Extreme Value techniques. Then, scaling
  properties of recurrence times between
  earthquakes have been studied in appropriate
  spatial volumes.
• It has been found that the seismicity is
  temporary clustered, and the distribution of
  recurrence times is significantly different from
  a Poisson process even for times largely
  exceeding corresponding periods of foreshock and
  aftershock activity.
• Modelling such a clustering by gamma distribution
  of recurrence times gives possibility to estimate
  hazard rates now depending on the time elapsed
  from the last large earthquake.

12
Simplified SOFT model

• Strain accumulation rates inferred from
  aftershocks have been studied.
• The Limited Power Law (LPL) model of the
  aftershock decay derived from the SOFT model
  suggests that the delay of the power-law decay
  of aftershocks depends on the heterogeneity of
  the stress field the larger stresses exist, the
  shorter is this delay. This delay has been
  estimated by considering stacked aftershock
  sequences from moderate main shocks in some
  volume (for example, a system of faults,
  different time intervals).
• A strong correlation of this delay with the focal
  mechanism (rake angle) of the main shocks has
  been found.

13
A new model of rupture

• A hierarchical model of rupture incorporating
  healing and faulting has been developed.
• The seismicity obtained in the model demonstrates
  the main characteristics of the observed
  seismicity Gutenberg-Richter law for the
  frequency-magnitude relationship, Omori law for
  the aftershock decay rate, clustering of major
  events, swarms of earthquakes, seismicity of
  creeping segment and seismic noise.
• A new multiscale cellular automation model of
  rupture has been designed to reproduce structural
  patterns observed in the formation and evolution
  of a population of strike-slip faults.

14
BLOCK MODEL OF THE VRANCEA REGION
KINEMATIC MODEL (after V.I.Mocanu)
BLOCK STRUCTURE
15
COMPARISON BETWEEN MODEL AND OBSERVED SEISMICITY
MAP OF VRANCEA SEISMICITY
MODEL EPICENTERS
16
New Version of the Block Model for Vrancea

17
WP7
Socio-Economic Barometer
18
WP7 Socio-Economic Barometer

• Participants
• CDCS J.Kurths
• CEPS G.Schaber, P.Liégeois, P.Van Kerm
• ENS M.Ghil
• MITPAN I.Kuznetsov, A.Mostinskii, M.Rodkin,
  A.Soloviev, T.Tseplinskaya, F.Winberg
• ULG M.Ausloos, F.Petroni
• IGPP V.Keilis-Borok

• Objective
• Development of a socio-economic barometer that
  is the methodology and relevant algorithms and
  software for forecasting crises (or stable
  development) in socio-economic systems.
• The forecast will be based on analysis of
  relevant socio-economic indicators, which are the
  input for the algorithms
• The output of the algorithms is a forecast
  whether a crisis is or is not approaching the
  forecasts will include probabilities of false
  alarms and failures to predict
• Different algorithms will be needed for
  prediction of different crisis types, but for a
  given crisis type we plan to develop a
  self-adaptive algorithm and relevant software
  that can be applied without readaptation in
  different territories.

19
WP7 Approaches

• At each moment an algorithm indicates whether a
  crisis should or should not be expected within
  subsequent ? months, ? being duration of alarm.

20
WP7 Predicting the End of an Economic Recession

• The problem of predicting the end of an American
  economic recession by analysis of macroeconomic
  indicators within the recession period has been
  considered. The study is a technical analysis
  that is a heuristic search of phenomena preceding
  the recession end. The methodology of pattern
  recognition of infrequent events is used. The
  goal is to identify by an analysis of
  macroeconomic indicators a robust and rigidly
  defined prediction algorithm of the yes or no
  variety indicating at any time moment, whether
  the recession end should be expected or not
  within the subsequent months. A specific
  premonitory pattern of six macroeconomic
  indicators that may predict algorithmically the
  recession end has been found for six economic
  recessions in the U.S. between 1960 and 2000. The
  ends of all six recessions under consideration
  are preceded within 5 months by this pattern that
  appears at no other time. The end of the last
  recession occurred in 2001 has been also
  predicted.

21
WP7 Analysis of Time Series Behaviour

• The problems that arise in data processing (i)
  what kind of change of the examined time series
  should be treated as a non-random and (ii) what
  kind of change (events) could be tried to be
  forecasted were studied. The approaches to
  development of some formal procedure(s) to
  distinguish random and non-random features and
  random and non-random time-frequency domains in
  the regime of the time series under examination
  have been analyzed.

22
WP7 New approaches to prediction of extreme
events

• The algorithms use background activity that is
  introduced for the systems under consideration
  the economy (the U.S. economy) and the megacity
  (Los Angeles or New York). Economic recessions in
  the U.S., homicide surges in Los Angeles and New
  York City, and episodes of a sharp increase in
  the unemployment rate in the U.S. are considered
  as the extreme events. Events forming background
  activity are determined from monthly series of
  the U.S. industrial production index (in the case
  of recessions), of the grand total of all actual
  offences in Los Angeles or New York City (in the
  case of homicide surges), and of index of help
  wanted advertising (in the case of episodes of a
  sharp increase in the unemployment rate).

23
Definition of events of the background activity

• Monthly series of Index f is considered.
• Let the general trend of f before the extreme
  events characterised by (a) decline, µf(m/s,u)
  Kf(m-s-u,m-s) - Kf(m-s,m), (b) surge, µf(m/s,u)
  Kf(m-s,m) Kf(m-s-u,m-s). Dots show monthly
  values of index f, straight lines are the linear
  least-square regressions on segments (m-s-u, m-s)
  and (m-s, m).

24
Transformation of scaling relation before extreme
events

• Extreme event the start of American economic
  recession. D-periods periods before extreme
  events.
• Scaling relation is defined for events of the
  background activity determined by means of
  monthly industrial production
• N(M) is the number of events with µ M.

25
Scheme of a prediction algorithm
A(m/s,u,µ,v) is the number of events with
µf(m/s,u) µ occurred within a time window of v
months m-v1, m
26
Economic recessions in the U.S. f industrial
production index
27
Episodes of a sharp increase in the unemployment
rate in the U.S. f index of help wanted
advertising
28
Periods of the homicide surge in Los Angeles f
the grand total of all actual offences
29
Periods of the homicide surge in New York City
30
WP7 Prediction of the rise of unemployment rate
in the US civilian labour force

• Precursory trends of the indicators used for
  prediction of the rise of unemployment rate.
• Brown curves show smoothed indicators S-
  short-term interest rate on 90-day U.S. treasury
  bills, at an annual rate L - long-term interest
  rate on 10-year U.S. treasury bonds, at an annual
  rate IP - total U.S. industrial production
  defined in to 2002.
• Red bars show time intervals when the trend of an
  indicator was large (i.e. above the respective
  threshold).
• An alarm for the rise of the unemployment rate is
  declared for 6 months after each month when the
  trends of all three indicators are large
  (regardless of whether this month belongs or not
  to an already determined alarm).
• A yellow bar shows the period of the alarm
  starting in May 2007 and reported in September
  2006.

31
Unemployment rate in USA, 199909-200801
conforming the prediction

• A thin blue curve shows the data of Bureau of
  Labor Statistics, U.S. Department of Labor
  (http//data.bls.gov), and a thick curve the
  smoothed data. Yellow bars indicate the alarm
  periods, vertical red lines starting months of
  the rise of unemployment rate (the rise of
  unemployment started in December 2006 has
  confirmed the prediction), a grey bar the
  period of the unemployment growth in 2000-2003.

32
Unemployment rate in the U.S., 1964-2007