day3 f siccardi

80
CIMA Research Founda2on Observe to predict, predict to prevent

Upload: sonia-calamiello

Post on 12-Feb-2017

110 views

Category:

Documents


1 download

TRANSCRIPT

Page 1: Day3 f siccardi

CIMA  Research  Founda2on  

Observe  to  predict,  predict  to  prevent  

Page 2: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

The  Founda+on    

A  few  minutes  on  the  organiza+on    

•  it  was  founded  as  the  development  of  a  previous  university  research  center  in  2007  

•  the  founders  were  the  Civil  Protec+on  Department  of  the  Presidency  of  the  Council  of  Ministers  (2  106  €),  the  regional  government  of  Liguria  (0.2  106  €)  and  the  University  of  Genoa  (use  without  charges  of  the  ac+vity  of  some  researchers  for  a  few  of  years)  

•  no  annual  monetary  support  from  the  founding  bodies.  The  Founda+on  supports  itself  on  the  basis  of  contracts  of  na+onal  and  interna+onal  research  and  consul+ng  

 

Page 3: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

RESEARCH  THEMES  •  Hydro-­‐met  applica+ons  in  Civil  

Protec+on  

•  Climate  Change  and  Disaster  Risk  Reduc+on:  Targe+ng  Extremes  

•  Marine  Biology  and  Ecosystem  Monitoring  

 •  Liability,    Responsibility  &  Governance  

of  risk  •  EO  assisted  applica+ons  •  ICT  Tools  in  support  of  research  

•  Capacity  building  and  Educa+on  from  the  interna+onal  to  the  local  dimension  

Page 4: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Page 5: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Page 6: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Repo

rt  to

 Fou

nda-

on  

the numbers of the Foundation

Page 7: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

         

Rapp

orto  a  Fon

dazio

ne  

Chairmanship

Administration Training Management

Project Leaders

Researchers

Ph.D and Post Doc Students

Repo

rt  to

 Fou

nda-

on  

Page 8: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Rapp

orto  a  Fon

dazio

ne  

Staff  composi2on  (71)  

Research staff

Administrative staffPost-DocPh.D. Students

Occasional Staff

External Consultants

Personnel Appointed on

Project

Repo

rt  to

 Fou

nda-

on  

Page 9: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Rapp

orto  a  Fon

dazio

ne  

how we map the Foundation on basic sciences

env.eng.

biologists

physicists

ICT peoplesocial scientists

Research  staff  composi2on  

env.eng.

biologists

physicists

ICT people

social scientists

Repo

rt  to

 Fou

nda-

on  

Page 10: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Rapp

orto  a  Fon

dazio

ne  

How we map Foundation on peer review Int.Journals

0123456

JMEPS

JMS

AWR

CSDAIIN

ME

QJRMS

JGR

JMB

JHM

AMS

NHESS

WRR

ESIJA

SJM

BA EFEMS

Ph.Lett

ers

Fres

enius

B.

Page 11: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Rapp

orto  a  Fon

dazio

ne  

External ConsultantsOccasional StaffPersonnel

Appointed on Project

Ph.D. Students

Post-DocStaff

Research  staff  costs  (2014:1.8  M€)  

Page 12: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Personnel

Travels

Third Institutions

Operational costs

Training Goods

Rapp

orto  a  Fon

dazio

ne  ann

o  2009  

Total  Founda2on  costs  (2014:  4.2  M€)  

Page 13: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Rapp

orto  a  Fon

dazio

ne  

our new goal for the second

decade of the third millennium

environment disasters food and

poverty

Repo

rt  to

 Fou

nda-

on  

Page 14: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

a  few  historical  notes:    Mr.  Zamberle?  in  Italy  and  Mr.Tazieff  in  France  at  the  same  -me  appoint   at   the   na-onal   scale   three   Research   Groups   on  hydrological  and  meterological  hazards,  on  eartquake  hazard  and  on  volcano  hazard.  To  assist  the  Government  to  build  up  policies  to  manage  the  risk.      In   Italy   a   Dept.   of   the   Presidency   of   the   Council   of  Ministers   is  named   as   Protezione   Civile.   The   Head   of   Civil   Protec-on   is   in  charge   of   coordina-ng   the   country   resources   for   assessing   the  risk,  diffusing   in   real  -me  proper  alert  messages,   rescue  vic-ms  and  support  affected  people.        The   Presidency   of   the   Council   of   Ministers   is   assisted   by   a  network   of   research   Centers.   CIMA   Research   Founda-on   deals  with  the  floods,  landslides,  drougts  and  forest  fires  risks.    Understanding   the   climate   change   impact   on   the   extreme  phenomena  is  a  key  issue  for  predic-ng  and  planning.  

1985  1987            1987  1999            2000  2015  

Page 15: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

         hydro-­‐meteo  risk;        earthquake  risk,                      volcanic  risk  

             CIMA  Res.  Founda-on          Univ.  Of  Genoa          Dept.  Prot.  Civile      

   Fond.EUCENTRE      Univ.  of  Pavia  

Dept.  Prot.  Civile      

   Observatory    of  Mount  Vesuvius  

     

Page 16: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

opera-onal  room  

room  staff  

Page 17: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

the  Italian  Civil  Protec+on  Department  

of  the  Presidency  of  the  Council  of  

Ministers    

Page 18: Day3 f siccardi

Presidency  of  the  Council  of  Ministers  -­‐  Department  of  Civil  Protec2on,  ROME  

Page 19: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

re-­‐building  

preven7on  

mi7ga7on  

emergency  

back  to  life  

Disaster  Disaster  Risk  Reduc-on  

Cycle  

preparedness  

PROPER  CIVIL  PROTECTION  ACTIONS  

 WHERE  DOES  THE  DEPARTMENT  OF  CIVIL  PROTECTION  WORKS  IN  THE  CYCLE  

OF  THE  RISK  REDUCTION  

..Pre-­‐disaster  ac2vi2es  that  are  undertaken  within  the  context  of  disaster  risk  management  and  are  based  on  sound  risk  analysis.  This  includes  the  development/enhancement  of  an  overall  preparedness  strategy,  policy,  ins2tu2onal  structure,  warning  and  forecas2ng  capabili2es,  and  plans  that  define  measures  geared  to  helping  at-­‐risk  communi2es  safeguard  their  lives  and  assets  by  being  alert  to  hazards  and  taking  appropriate  ac2on  in  the  face  of  an  imminent  threat  or  an  actual  disaster  (ISDR’s  defini2on)…  

   UN-­‐ISDR  

Page 20: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Presidency  of  the  Council  of  Ministers  -­‐  Department  of  Civil  Protec2on  

Steps  for  the  prepara2on  

1.  KNOW  THE  VULNERABILITY  AND  RISK  SCENARIOS  

2.  IMPLEMENTING  RISK  PREDICTION  AND  EARLY  WARNING  

3.  BUILDING  A  GOOD  PERFORMANCE  IN  BACK  TO  LIFE  ACTIVITY  

Page 21: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Presidency  of  the  Council  of  Ministers  -­‐  Department  of  Civil  Protec2on  

PREPARE  A  RISK  SCENARIO  

…figures  dendent  on                                                              space                and                    2me….  

q  Iden2fying,  zoning,  quan2fying  the  DANGER  (P)  

q  Loca-on  and  evalua-on  of  the  numbers  (N)  and  the  value,  social  and  economic,  for  different  categories  of  exposed  en--es(E)    

q  Determina2on  of  the  overall  vulnerability(V)    

q  Defini2on  of  the  expected  damage  (D)  given  the  event.  D=  E  x  V  

R = P x E x V = P x D Uff.  Previsione  e  Prevenzione  

Page 22: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Presidency  of  the  Council  of  Ministers  -­‐  Department  of  Civil  Protec2on  

An example of residual risk

the  effect  of  ac-ons  can  be  considered  in  the  defini-on  of  risk  through  indices  (Iv)  of  effec-veness  of  specific  interven-ons:example  

q  IV  interven-ons  to  reduce  vulnerability  

Uff.  Previsione  e  Prevenzione  

T=50anni  

T=200  years  

rischio  residuo  

difesa  domes-ca  da  inondazione  riduce  la  frequenza  dei  danni  

inondazione  molto  frequente,  T=5-­‐10  anni  elevata  vulnerabilità  

the  low  probability  events    remain  unchanged  –  PROCIV  early  warnings  Insurance  

Page 23: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Presidency  of  the  Council  of  Ministers  -­‐  Department  of  Civil  Protec2on  

They  say,  in  Rome,  that  the  Early  Warning  System  of  Italian  Civil  Protec-on  is  among  the  most  advanced  in  the  world:  

Understanding  and  mapping  the  danger    

Monitoring  physical  processes  and  predict  upcoming  events  

Dissemina-ng  clear  alerts  by  poli-cal  authori-es  Popula-on  undertaking  appropriate  -mely  ac-on  as  a  result  of  warnings  

1  

2  

3  

4  

Page 24: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Presidency  of  the  Council  of  Ministers  -­‐  Department  of  Civil  Protec2on  

Strategy:  integra-ng  the  tools  for  real  -me  and  the  tools  for  mi-ga-on  -me    

Policy  of  Early  Warning  

Previsioni  modellis2che  Inondazioni  storiche  -­‐  AGGIORNAMENTO  

Espos2  –  Centri  abita2  

Strategy:  public  and  private  sector  coopera-ng  in  the  collec-ng  observa-ons  and  informa-ons  to  build  scenarios  

Page 25: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Presidency  of  the  Council  of  Ministers  -­‐  Department  of  Civil  Protec2on  

 early  warning  systems  are  much  more  than  a  technological  tool,  are  a  whole  technological  system,  social,  even  cultural.  The  civil  protec-on  system  includes  ci-zens  and  their  consciousness  of  the  risk  as  ac-ve  component:  ci-zens  must  believe  on  the  effec-veness  of  an  early  warning  system  and  should  know  how  to  behave  once  alerted.  

Civil  Protec2on  System  for  Early  Warning   INSURANCE  

“Empowering  the  ci-zens”  airaverso  l’interoperabilità  fra  sistemi,  necessità  di  una  determinazione  del  rischio  sul  territorio  condivisa  (pubblico,  privato,  ciiadini)  

 the  biggest  barrier  to  the  penetra-on  of  insurance  is  the  lack  of  risk  percep-on  

CICLO  VIRTUOSO  

Page 26: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

       The  NP-­‐HFA  is  (  or  beier  should  be)  an  en-ty  that  promotes  a  more  open  and  effec-ve  dialogue  between  stakeholders  in  a  shared  interna-onally.    It  promotes  the  development  of  financial  mechanisms  and  risk  transfer,  in  par-cular  insurance  and  re-­‐insurance  against  disasters    It  encourages  the  forma-on  of  partnerships  to  increase  public-­‐private  partnerships  involving  the  private  sector  in  the  ac-vi-es  of  risk  reduc-on:  Sponsors  of  a  risk  culture  Allocates  of  resources  in  the  pre-­‐event  for  risk  assessment  and  for  the  implementa-on  of  early  warning  systems    Develops  and  promotes  alterna-ve  and  innova-ve  financial  instruments  to  deal  with  disasters.  

Na2onal  Plagorm  of  Hyogo  Framework  for  Ac2on  (NP-­‐HFA)  

DPC  è  il  coordinatore  della  piaiaforma  

Presidency  of  the  Council  of  Ministers  -­‐  Department  of  Civil  Protec2on  

Page 27: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

The  Founda+on  ac+vity  

Page 28: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Predictability  of  meteorological  extremes    Modeling  and  predic-on  of  floods  and  droughts      Observa-on  of  hydro-­‐meteorological  variables    

                                   Data  fusion  and  data  assimila-on    

                                   Modeling  and  predic-on  of  forest  fires    

                                                 Modeling  and  predic-on  of  pollutants  dispersion  in  water,                s                                                                soil  and  atmosphere  

CIMA  Research   Founda-on   ac-vity   is   financed   by   research   and   technological   innova-on  contracts  with  the  Italian  Civil  Protec-on  and  regional  governments,  with  UNDP,  UN-­‐ISDR,  UE,    NGO  and  public  and  private  companies.    

Predictive  ability  of  severe  

rainfall  events  over  Catalonia  

for  year  2008

master  thesis  reportDirectors:  Dra.  Maria  Carme  Llasat  Botija  

(UB)Dr.  Antonio  Parodi  (CIMA  Res.  

Foundation)Albert  Comellas  Prat

Page 29: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

1-­‐Climate  change  impacts  on  the  organiza-on  of  the  date  base  of  events  severity,  which  is  a  cri-cal  issue  for  predic-ng  and  planning  

Page 30: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

2-­‐Climate  change  impacts  on  the  rate  of  transforma-on  of  the  territory,  which  is  a  cri-cal  issue  for  adapta-on  

Page 31: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

3-­‐Climate  change  impacts  on  the  legisla-on  for  risk  mi-ga-on,    which  is  a  cri-cal  issue  for  social  responsibility  

Italian  direc-ve  2004   European  direc-ve  2008  

1992  Italian  Civil  Protec-on  legisla-on  

1993-­‐98  Regional  Civil  Protec-on  regula-ons  

Page 32: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Page 33: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

modern  predic-on  and  communica-on  of  the  ground  effects  greatly  improves  the  social  response      

tradi-onal  modeling  

DEWETRA – Real Time fields of rain intensity, Temperature, …..

RT- Rain Intensity mapTime range

distributed  complex  modeling  

Page 34: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

The  Founda+on  and  the  role  of  remote  sensing  

Page 35: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

new  modelling  with  up  to  date  DEM  and  data  observed  by  sensors  on  board  of  satellites  do  deserve  for  improving    ground  processes  predic-on  

DEWETRA – Risk AssessmentAn example .

RISICO model gives a Wildfire risk index which represents the potential fire linear intensity (kW/m). It also estimate the wildfire risk index forecast for the next 72 hours.

Page 36: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

december  30  2009  –  inunda-on  of  Massaciuccoli  area,  reclaimed  by  Medici  family,  1570,  Pisa-­‐Italy    Elabora-on  from  the  SAR  data  flying  on  the  Cosmo  SkyMed  fleet  of   the   Italian  Space  Agency  with   the  system   DEWETRA   of   the   Civil   Protec-on-­‐Cima  Research  Founda-on.    

Page 37: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

January  9  2010  –    inunda-on  of  Skutari  area,  reclaimed  by  Venice,  1550,  Skutari-­‐Albania      Elabora-on   from   the   SAR  data  flying  on   the  Cosmo  SkyMed  fleet  of  the  Italian  Space  Agency  with  the  system  DEWETRA  of  the  Civil  Protec-on-­‐Cima  Research  Founda-on.    

 

Page 38: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

July  2010,  Indo  river  flooding,  water  depth  near  Peshawuar      Elabora-on  from  the  SAR  data  flying  on  the  Cosmo  SkyMed  fleet  of  the  Italian  Space  Agency  with  the  system  DEWETRA  of  the  Civil  Protec-on-­‐Cima  Research  Founda-on.  

Page 39: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Cosmo/Skymed  images  acquired  near  Scutari  (Albania)  in  Stripmap  mode  (pixel  resampling  at  10  meters),  in  descending  configura7on  with  right  look  angle    Delle  Piane  et  al.,  FR4.L07.5,  Fr.  17:00      Pierdicca  et  al.,  FR4.L07.1,  Fr.  15:40  

CIMA    RESEARCH  FOUNDATION    

Page 40: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

2007-­‐2012  Demonstra-ve  pilot  project  of  ASI  (Italian  Space  Agency)  and  DPC  (Department  for  Civil  Protec-on)  for  EO-­‐based  applica-ons  

Mul--­‐mission,  focus  on  COSMO-­‐Skymed  

Elena Angiati3, Giorgio Boni2, Laura Candela1, Fabio Castelli4, Silvana Dellepiane3, Fabio Delogu2, Fabio Pintus5, Roberto Rudari2, Sebastiano B. Serpico3, Stefania Traverso2, Cosimo Versace6.  1Italian  Space  Agency,  Unità  Osservazione  Della  Terra,  CGS,  Contrada  Terlecchia,  75100  Matera  (Italy)  2CIMA  Research  Founda-on,  Savona  University  Campus,Via  Armando  Maglioio  2,  I-­‐17100  Savona  (Italy)  3University  of  Genoa,  Dept.  of  Biophysical  and  Electronic  Eng.  (DIBE),Via  Opera  Pia  11a,  I-­‐16145,  Genoa  (Italy)  4University  of  Florence,  Dept.  of  Civil  and  Environmental  Eng.  (DICEA),  via  S.  Marta,  3  -­‐  50139  Firenze  (Italy)  5ACROTEC  S.r.L.,  Via  Armando  Maglioio,  2  17100  Savona  (Italy)  6CONSORZIO  COS  (OT),  Via  Casalnuovo,  86,  75100  Matera  (Italy)      

Page 41: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

And  now  the  NASA  puzzle  (courtesy  of  NASA-­‐restricted  circula-on):why  NASA  distributes  such  a  trick?  At  the  Founda-on  we  feel  there  is  a  basic  misunderstanding.      

§    The   Founda-on  unit   dealing  with   satellite   sensors  for  environmental  monitoring   shares   the   interest  on  satellites  images  but  our  paradigm  is  slightly  different  then  NASA  

§   Not  from  data  to  images,  but  from  data  to  models  and   possibly,   when   needed   for   communica-on  purposes,   from  models   to   images:   in   the   analysis   of  environmental   transforma-ons     and   environmental  disasters,  images  are  not  enough.      

Page 42: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

The  Founda+on  and  the  new  paradigms  in  modelling  

 Con+nuous  chains  from  

meteorology  to  hydrology  and  hydraulics  

Page 43: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

hip://www.drihm.eu/  

Page 44: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  Why?  

Page 45: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

NASA  EARTH  Observatory.  Image  acquired  December  30  2004  by  the  European  Space  Agency  astronaut  Alexander  Gerst.  On  upper  les  corner,  the  orography    

Page 46: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

9   October   2014.   Mean   sea   level   pressure   (Pa)   at  00UTC   from  ECMWF  re-­‐analysis  on  October  9  2014  run   at   00UTC.   A   pressure   gradient   of   the   order   of  4%0   in  a  space  not  exceeding  100  km  is  established  between  the  western    Po  floodplain  and  the  Ligurian  sea.    

9   October   2014.   Daily  mean   T2m   temperature   provided  by   the   Italian   Civil   Protec-on   Department   ground  network.  Sea  Surface  Temperature  provided  by  the  Global  1-­‐km  Sea  Surface  Temperature  data  set  produced    by  the  JPL   Regional   Ocean   Modelling   System   (ROMS).   A  temperature  gradient  of  8-­‐9  C  is  established  between  the  western  Po  floodplain  and  the  Ligurian  sea.    

Page 47: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

observed  reflec-vity  field  evolu-on  at  eleva-on  z=  3000  m  amsl  (Seiepani  meteo  radar)  from  01:30UTC  to  07:00UTC    

Page 48: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  Horizontal  wind  and  ver2cal  thermals  at  01:30,  02:00  and  02:30  UTC  

Page 49: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

The  Founda+on  and  the  uncertainty  in  predic+ng  

Page 50: Day3 f siccardi

understanding  the  uncertainty  is  a  ques-on  of  

swans  I’ll  open  this  part  of  the  lecture  with  a  quite  exhaus-ve  example.  The  example  is  aimed  to  introduce,  without  lengthy  use  of  the  concept  of  probability,  a  discussion  on  the  uncertainty  in  the  hydrometeorogical  predic-ons  and  its  social  relevance.    In  fact  the  decision  maker  of  Civil  Protec-on,  in  any  of  his  levels,  from  municipal  to  suprana-onal  agencies,  is  confronted  with  uncertainty.    

 

Page 51: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Taleb's  thesis  "Simply,  we  cannot  predict"  is  the  -tle  of  the  second  part  of  his  book  The  Black  Swan.  The  impact  of  the  highly  improbable              Mediocristan:  a  world  in  which  the  experts  are  able  to  measure  the  uncertainty  of  future  observa-ons  from  the  observa-ons  of    the  past  Extremistan:  a  world  where  in  some  cases  the  future  eludes  the  measurements  of  the  experts  and  catches  them    Where  we  live?  How  we  learn  from  the  past?  Which  errors  are  possible?  Which  is  our  responsibility  when  we  do  wrong?  The  society  is  equipped  to  respond  to  the  impact  of  the  highly  improbable?

Page 52: Day3 f siccardi

supponiamo  che  esistano  diecimila  universi  tu?  uguali  

Page 53: Day3 f siccardi

each of them with identical Thirrenian seas, with the same topography and hydrography of the coastal area

Genoa

Page 54: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

the stream

the soccer stadium

with ten thousand identical cities of Genoa with the same urban planning developed in the same way

the stream covered

the stream mouth

the  stream  

the  soccer  stadium  

Page 55: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

with  ten  thousand  iden-cal  streams  covered  by  an  en-re    monumental  district  designed  by  the  archistar  of  the  fascist  period  

Page 56: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

the  urban  development  from  1400  to  1937,  five  years  aser  the  stream  was  covered  

the  event  theater  

Page 57: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  57  

full  bank  level  

flooding  level  

when an extreme flood is carried to the sea the space under the cover is not enough. The water level rises and touches the inner surface of the cover. The flow under the cover is suddenly reduced and a wave of reflux propagates back. Suddenly the excess discharge inundates the streets on the two enbankments.

Page 58: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  58  

Sunday it will rain for at least twelve hours, and the rain depth will possibly be equal or exceeding 200 mm

now let us suppose that out of the ten thousand universes a fall extreme storm is announced in, say, three hundred of them

Page 59: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  59  

Sunday,  october  xx      199y    

0

50

100

150

200

250

mm

A  

let us choose now, at random, the universe A, one out of the three hundred universes in which Sunday it will rain a lot. Describe the event. It starts raining at noon and it rains until midnight, but more than two third of the total depth fall continuously at the beginning of the event.

Page 60: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  60  

.

teams had to play their first league match in the soccer stadium near the stream

capacity of 40,000. Heavy rains. Only 10,000 soaked fans under their umbrellas waiting to see if the match will start. Three o’ clock sharp. The referee comes out and throws the ball to see if it bounces over the green. SPLASH. Again SPLASH and again SPLASH.

. the referee whistles: game postponed. Ten thousand people leaving, walking along the streets on the embankments. The stream exceeds the full bank flow. The water touches the cover.The reflux wave explodes back. The river inundates the two roads. The water drags pedestrians and cars

Page 61: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Page 62: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Sunday,  xx      oiobre  199y    

B  

0

50

100

150

200

250

1 2 3 4 5 6 7 8 9 10 11 12

let us continue to examine the three hundred universes. Call B the second one. It starts raining at noon and it rains until midnight. More than 200 mm. Less than half in the three hours at the beginning of the event and the most at the end.

Page 63: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  63  

teams had to play their first league match in the soccer stadium near the stream

capacity of 40,000. Heavy rains. Only 10,000 soaked fans under their umbrellas waiting to see if the match will start. Three o’ clock sharp. The referee comes out and throws the ball to see if it bounces over the green. SPLASH. Again SPLASH and again SPLASH.

. the referee whistles: game postponed. Ten thousand people leaving, walking along the streets on the embankments. The stream does not exceed the full bank flow. 10000 fans reached back their destinations soaked under their umbrellas.

Page 64: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  64  

about nine o'clock in the evening it starts again to rain hard. It rains for three hours continuously. The  stream  exceeds  the  level  of  full  bank  flow.  Same scenario that we've seen in the universe A. The difference is that in the universe B it happens shortly after midnight. Subways are closed. No one in the streets. Only parked cars. The buses are in the night garages. Many damages. No victims.

Page 65: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

The  event  in  the  universe  B,  that  did  happen  late  in  the  night,  when   the  Genoeses  were   all   home,   is   almost   as   unlikely-­‐or  likely-­‐  as  that  of  the  Universe  A.    And   so,   among   others,   two   professors   of   Civil   Engineering  are  s-ll  teaching  at  the  University  of  Genoa.      They   went   to   watch   the   match   between   Sampdoria   and  Milan,  and  survived.    Why?  Because  they  were  in  the  universe  B,  so  we  discovered  aser  the  event.    Do   you   perceive   how   thin   is   the   physical   role   of   the  uncertainty  and  how  large  is  the  social  one?  

Page 66: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

A   few  weeks   aser   the   presenta-on   of   such   an   example   in  Rome   an   extreme   event   hit   the   area   I   just   described.   Very  similar  flooding,   around  one  o’   clock   in   the   asernoon,  with  pupils   leaving   the   schools.   Six   casual-es.   Warnings   were  issued  the  day  before.    I   had   to   explain   to   newspaper   and   tv   people   that’s  impossible  to  predict  weeks  before  the  event.  That  my  tale  at  the  conference  was  not  a  forecast.    Because  to  predict  where  and  how  it’s  easy,  but  to  forecast  when  and  how  much  is  quite  another  maier.    That’s  the  reason  why  the  2011  event   in  Genoa  was  chosen  as  a  study  case  for  Founda-on.    

Page 67: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Why  e-­‐infrastructures  for  Civil  Protec-on?    Thirty  years  ago  I  began,  with  a  few  friends,  the  construc-on  of  the  system  of  Italian  Civil  Protec-on.  We  put  every  effort  in  so-­‐called   non-­‐structural  measures,   those   useful  measures   to  alert   the   authori-es   and   ci-zens   when   a   paroxysm   of  meteorology   could   bring   water   to   their   homes,   and   kill   and  destroy  proper-es  and  means  of  produc-on.      So   they   could  be   ready   to  accept   resctric-ons  on   the  use  of  the   land   and   proper-es,   but   also   so   that   they   could   take  simple  temporary  protec-ve  measures.      Because   we   can   -­‐   it   always   has   been   -­‐   live   in   flood   prone  areas.  The  paroxysmal  events  are  rare  and  do  not  strike  in  the  same  place.    

Page 68: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Dealing  with  uncertainty.  By  the  use  of  ensemble  predic-on  in  meteorology  and  by  the  use  of  disaggrega-on  of  predicted  rainfall  fields  from  meteo  to  hydro  scales.          The   Italian   system   for   predic-ons   is   now   distributed   into   a  number   of   technical   groups   of  meteo   and  hydro   experts   at  the  local  scale  and  a  coordina-ng  group  at  the  na-onal  scale.  More  than  one  hundred  skilled  people.    The  procedures  are  extending  to  the  whole  Europe  as  a  best  prac-ce   for   the   European   Civil   Protec-on,   presently   under  transforma-on  and  strengthening.    That’s  the  reason  why  I  think  that  DRIHM  e-­‐infrastructure  is  a  very  promising  hot  spot  in  hydrometeorological  research.  

Page 69: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

S-ll  research  needs?  Oh,  yes.    Nassim   N.   Taleb,   published   in   2007   "The   black   swan:   the  impact  of  the  highly   improbable".  The  book  created   intense  controversy  in  mathema-cal  circles.      It   deeply   revises   the   paradigms   of   the   forecast   of   future  states  of  a  system,  based  on  the  observa-on  of  past  states.    The   Taleb   thesis,   in   essence,   is   that   the   human   condi-on,  which   learns   from  experience,     forces   into   a  mental   tunnel  the  predic-ons  of  what  might  happen.      The   predic-on   tunnel   is   formed   by   the   experience   of   past  events,  among  which  the  highly  unlikely  event  almost  never  appears   because   it   is   very   rare   and   therefore   almost   never  belonged  to  the  experience.    

Page 70: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

I  did  try  to  offer  the  example  in  a  way  that  I  hope  is  readable.  I  did  it  in  order  to  avoid  that  e-­‐infrastructures  are  perceived  as   the   saving   solu-on   to   all   the   problems   of   dealing   with  uncertainty.      e-­‐infrastructures   allow   the   operators   to   operate   repeated  simula-ons   of   reality   much   faster   than   before   and   so   give  the  operators  -me  to  think.    The  basic  problems,  why  we  have   to   think   the  members  of  the   ensembles,   true   ensambles   or   poor   man   ensembles,  equiprobable,   or   why   we   have   to   think   members   of   the  rainfall   field   disaggrega-on   indipendent   on   the   terrain  orography,   are   s-ll   there   as   food   for   the   minds   of   young  researchers.    

Page 71: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Why  e-­‐infrastructures  for  DRR?    For   climate   change   applica-ons   the   Taleb’s   effect   stays  upstream   of   the   future  meteorological   possible   states.   The  uncertainty   is   absorbed   into   the   construc-on   of   future  clima-c  systems.  Their  effects  are  highly  unlikely  events  per  se.    Contrary  DHIHM  e-­‐infrastructure  plays  the  essen-al  role  of  a  specific  tool,  a  quite  powerful  tool,  to  inves-gate  the  effects  at   small   scale,   i.e.   the   scale   of   the   impacts,   condi-onal   on  possible  meteorological  states.      It’s   the   tool   for   evalua-ng   the   effects   of   disaster   scenarios  through   repeated   simula-on   experiments   that   the   e-­‐infrastructure  easily  allows.      

Page 72: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Why  DRIHM,  in  essence?    There   is   no   doubt   that   the   e-­‐infrastructure   is   the   most  appropriate   tool   to   facilitate   the  work   of   forecasters   in   the  field  of  civil  protec-on  and  simplify  the  role  of  risk  managers  or  planners  in  the  field  of  disaster  risk  reduc-on.    However,   as   I   hinted,   here   and   there,   there   is   s-ll   a   lot   of  food  for  the  mind.    I  wish  you  a  long  career  of  reflec-ons  and  successes.      Like  I  had.      Thanks  for  your  aien-on.    

Page 73: Day3 f siccardi

Monitoring  the  effec2veness  of  the  Italian  Civil  Protec2on  System:  Decision  making  in  a  overcau2ous  jurispruden2al  environment  

A  few  word  more  on  trial  environment  

Page 74: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

20  mm  

July  2nd,  2006  

200  mm  

July  3rd,  2006  

Page 75: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Our  Observatory  

>200  casual2es  

0  

2  

4  

6  

8  

10  

12  

14  

16  

18  

20  

2004   2005   2006   2007   2008   2009   2010   2011   2012   2013   2014  

Num

ber  o

f  Proceed

ings  

Years  

#  procedimen-  Proceedings  

>50  pending  proceedings    

>150  duty  holders  involved  

17/20  Regions  involved  

 

Page 76: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Observed  cri-cali-es  

Civil  Protec-on  effec-veness  

0  

50  

100  

150  

200  

250  

300  

350  

400  

450  

500  

0  

2  

4  

6  

8  

10  

12  

14  

16  

18  

20  

2004   2005   2006   2007   2008   2009   2010   2011   2012   2013   2014  

Num

ber  o

f  Events  

Num

ber  o

f  Proceed

ings  

Year  

#  procedimen-  #  even-  

Proceedings  Events  

Page 77: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Performance  paradox  in  CP  

0%  10%  20%  30%  40%  50%  60%  70%  80%  90%  

100%  

1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16   17   18   19   20  

Soios-ma  -­‐  Mancate  Allerte  Sovras-ma  -­‐  False  Allerte  Correie  

6%  10%  84  %  1%  60%  39  %  

Performan

ce  

Regions  

Underes-ma-on  –  missed  alerts  Overes-ma-on  –  false  alerts  Correct  

Observed  cri-cali-es  

Page 78: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

•  Increased  number  of  alerts;  •  Increased  level  of  alerts;  •  Adop-on  of  restric-ve  measures  (i.e.  evacua-ons,  mobility  limita-ons…);  

•  Preven-ve  shutdown  of  public,  private  produc-ve  buildings.  

Ac-ve  “defensive  behaviour”:  

•  Resigna-on  from  appointments  (lowered  professional  level  of  CP  operators);  

•  Fragmenta-on  of  mandates  (nobody  wants  to  take  decisions  in  an  uncertain  world);  

•  Suppression  of  services.  

Passive  “defensive  behaviour”:  

•  Fears  of  dutyholders;  •  Inflexibility  of  the  system.  

Impossibility  of  valorising  mistakes:  

Observed  Cri-cali-es  

Page 79: Day3 f siccardi

Observe  to  predict,  predict  to  prevent  

Access  to  informa2on  

Consulta2on  of  available  documenta2on  

Par2cipatory  approaches  for  emergency  planning  

Res

ilien

ce a

nd

resp

onsi

bilit

y of

co

mm

unit

ies

One  (out  of  many)  possible  way  foreward  

Page 80: Day3 f siccardi

CIMA  Research  Founda2on  thank  you  for  your  pa7ence  

on  behalf  of