risk and global change © pascal peduzzi, unep/grid-geneva, 2014. 1 pascal peduzzi, phd gregory...

Risk and Global Change © Pascal Peduzzi, UNEP/GRID-Geneva, 2014.1

Pascal Peduzzi, PhDGregory Giuliani, PhDAndrea de Bono, PhDChristian HeroldBruno Chatenoux

GEO Ministerial and Plenary Meetings – Side Event13 January 2014

Data access and interoperability. GAR and PREVIEW Global Risk Data Platform

Generating and sharing risk data

UNEP / GRID-Geneva


The PREVIEW Global Risk Data Platform

Presentation plan

Global level risk analysis

Who are we?

GAR 2013: new developments

Risk and Global Change © Pascal Peduzzi, UNEP/GRID-Geneva, 2014.

UNEP

DEWA DEPI DRC DGEFDCPIDTIEDEWA

UNEP/GRID-Geneva

Global Change & Vulnerability

Unit(ex Early Warning)

Dr P. Peduzzi C.Herold Dr G.Giuliani

Global Change & Vulnerability Unit

Swiss Env. Agency

University of Geneva

B. Chatenoux

Global Change & Vulnerability: a unit of the

UNEP/GRID-Europe

Dr A. De Bono


Field data collection

Image analysisStatistical analysisSpatial analysis (GIS)

Global Change & Vulnerability Unit

Maps & Info

PREVIEW

Data (SDI)


200520062006

Contribution to 12 UN reports on risk & global change2004

20072008

2009

2010

2011

2012

G

A

R2013

+ 28 Scientific papers


1. Global analysis


Exposure

Hazards

Vulnerability

Natural variability

AnthropogenicChanges

ClimateEnvironment

DEVELOPMENT

Disaster RiskManagement

Adaptation

Disaster

GHG emissions, deforestation,…

DISASTERRISK

How to generate risk data


Who generates the data?

Global Flood Model UNEP/GRID-Geneva and CIMA Foundation

Global Tropical Cyclones UNEP/GRID-GenevaProbabilistic TC model CIMNEGlobal Landslides Norwegian Geotechnical Institute (NGI)Global Tsunami Norwegian Geotechnical Institute (NGI)Tsunami events NOAA

Volcanic eruption Smithonian InstituteFlood events Dartmouth Flood Observatory (now at Colorado Uni)Earthquakes shakemaps USGSForest fires ESADrought Model IRI

Earthquakes GSHAP, CIMNE, (GEM coming)GDP WorldbankPopulation distribution Landscan

Global Exposure Model UNEP/GRID-Geneva


Tectonic Hazards

New Global Hazard Datasets created for GAR 2009


New Human & Economic exposure datasets (1 x 1 kmPopulation and GDP distribution Models made for every years from 1970 to 2010


1006 Past floods as detected by satellite sensors


Compilation of Past Earthquakes ShakeMaps

5686 events downloaded over the period 1973-2007


>6000 tropical cyclones events were processedGlobal coverage for the period 1970 to 2012.

Using central pressureMaximum windspeedLatitude …

Individual past hazardous events modeling


Nargis 2 May 2008Myanmar

Extraction of exposure and other parameters


DateIso3KilledEst. damages

Footprints Pop. exp. GDP exp.Pop.Urb exp.

GDP Urb. exp

Category

1 10,500,000 43,000,000 4,800,000 32,500,000

2 1,500,000 3,500,000 1,400,000 525,000

3 400,000 800,000 375,000 150,000

Country: MyanmarIso3: MMRDate: 02 May 2008

Preview Tropical Cyclones Database EM-DAT, CRED

DatabaseDateIso3

Killed: 138,366

VulnerabilityDatabase 43 indicators

Damages: 4,000 US$ millionsGDPcap: 1,227 US$Voice & acc.: -2.16Governance efficiency : -1.608Radio/inhabitant: 99.68%HDI: 0.592…Urban growth: 2.55%

…

DateIso3GDPcapVoice & acc.Governance efficiencyRadio/inhabitantHDI…Urban growth


1 AIDS estimated deaths, aged 0-49 (% of tot. pop.)

2 non GLC2000 bare land

3 Arable and Permanent Crops - % of non GLC2000 bare land

4 Motor vehicles in use - Passenger cars (thousand)

5 Motor vehicles in use - Commercial vehicles (thousand)

6 Physical exposure to conflicts

7 Corruption Perceptions Index (CPI)

8 Arable and Permanent Crops - Total

9 Arable and Permanent Crops - Percent of Land Area

10 Control of Corruption

11 Deforestation rate

12 % of population with access to electricity

13 Forests and Woodland (% of Land Area)

14 Gross Domestic Product - Purchasing Power Parity per Capita

15 Gross Domestic Product - Purchasing Power Parity

16 inequality (Gini coefficient)

17 Human Induced Soil Degradation (GLASOD)

18 Government Effectiveness

19 Human Development Index (HDI)

20 Per capita government expenditure on health (PPP int. $)

21 # of hospital beds per 100,000 habitants # of doctors

22 infant mortality and malnutrition (though are also factored into HDI)

23 Improved Drinking Water Coverage - Total Population

24 telecommunications (phone density per 100,000 habitants)

25 Political Stability

26 Population (Persons (in Thousands))

27 Urban Population (% of Total Population)

28 Radio receivers (per thousand inhabitants)

29 Regulatory Quality

30 Rule of Law

31 School enrollment, primary (total)

32 % of urban population living in slums / squatter settlements

33 Physicians density (per 10 000 population)

34 Under five years old mortality rate

35 Undernourished (% of total population)

36 Urban Population Growth on past 3 years

37 Voice and Accountability

38 Motor vehicles in use - Passenger cars (per inhabitant)

39 Motor vehicles in use - Commercial vehicles (per inhabitant)

40 School enrollment, primary (per inhabitant)

41 Population growth on 3 past years

42 income-consumption poverty (from WB poverty calculator also from MDG project)

43 Transport

43 indicators on:Economy, Demography, Environment, Development,Early Warning,Governance,Health,Education,…

List of vulnerability parameters considered


From hazardous events to frequency and exposure


Aggregation of human exposure at country level


Aggregation of economical exposure at country level


Landslides risk


About 2.2 million people are exposed to landslides worldwide.

55% of mortality risk is concentrated in 10 countries, which also account for 80% of the exposure.

Comoros, Dominica, Nepal, Guatemala, Papua New Guinea, Solomon Islands, Sao Tome and Principe, Indonesia, Ethiopia, and the Philippines

Landslides (modelled for both precipitation and earthquakes)


Tropical cyclones riskMultiple Risk

23


Multi Mortality Risk Index (MRI)


Floods Mortality Risk Index (MRI)Cyclones Mortality Risk Index (MRI)Earthquakes Mortality Risk Index (MRI)Landslides Mortality Risk Index (MRI)


3. PREVIEW Global Risk Data Platform


The Global Risk Data Platform

http:// preview.grid.unep.ch

Used by: in GEOSS portalUNEPUNISDR (For GAR).World BankUNHCRInform (EU/JRC)WRI (UNU)OCHAMapplecroftAnd many others

Users can visualise, interrogate, download data related to disaster risk (hazard, exposure, risk).


Fully Open Source

OGC & ISO compliant

Based on:

PostgreSQL/PostGIS,

PHP,

Geoserver,

GeoNetwork,

OpenLayers & GeoExt.

Analysis of geospatial data: ESRI ArcInfo & ArcGIS

GEO-X: Disasters Risk Reduction and Earth Observations, a GEO perspective - 13.01.2014


~220’000 visitors

~3’200’000 pages

~7’100’000 maps produced

~300 GB of data downloaded

Access x4 after Sichuan and Haiti events

Access x10 after Fukushima

GEO-X: Disasters Risk Reduction and Earth Observations, a GEO perspective - 13.01.2014

Risk and Global Change © Pascal Peduzzi, UNEP/GRID-Geneva, 2014.30GEO-X: Disasters Risk Reduction and Earth Observations, a GEO perspective -

13.01.2014

PreView Mobilehttp://preview.grid.unep.ch/mobile

Web-basedMultiplatformAccess all layers in WMS

Zoom IN/OUT, PanMulti-touch control

Search location:GeoNamesGPS

Risk and Global Change © Pascal Peduzzi, UNEP/GRID-Geneva, 2014.31GEO-X: Disasters Risk Reduction and Earth Observations, a GEO perspective -

13.01.2014


4. GAR 2015: new developments


GAR 2009-2011 probabilistic approach?

Yes No Remarque

Earthquakes Based on GSHAP 1:475 years

Landslides (Eq) Based on GSHAP 1:475 years

Tsunamis Based on GSHAP 1:475 years

CC.

FloodsBased on 100 years returning period

Trop. Cyclones Based on 1970 – 2009 detected events

Landslides (Pr) Based on 1960 – 2000 precipitations

Forest firesBased on 1997 – 2010 detected events

Drought Based on 1960 – 2000 precipitations


GAR 2013: probabilistic approach ?

Yes No Remarque

Earthquakes Based on GEM

Landslides (Eq) Based on GSHAP 1:475 years

Tsunamis Based on GSHAP 1:475 years

CC.

FloodsBased on 5 different returning periods

Trop. Cyclones Based on synthetic tracks and stochastic approachone global estimation of climate change impacts

Landslides (Pr) Based on 1960 – 2000 precipitations

Forest firesBased on burnt areas 2000 - 2011

Drought Based on FEWS methodology (6 countries)


Hazards: GAR 2015

Yes No Remarque

Earthquakes Based on GEM

Landslides (Eq) Based on GEM

Tsunamis

CC.

FloodsBased on 5 different returning periods

Trop. Cyclones Based on synthetic tracks and stochastic approach

Landslides (Pr) Based on stochastic approach

Forest firesNot yet discussed

Drought Based on FEWS methodology, more countries


2.4 Tropical cyclones global trends

Peduzzi, P., Chatenoux, B., Dao, H., De Bono, A., Herold, C., Kossin, J., Mouton, F., Nordbeck, O. (2012) Tropical cyclones: global trends in human exposure, vulnerability and risk, Nature Climate Change, 2, 289–294.


Scenarios on TC for 2030

As adapted from Knutson et al. (2010)


Global Flood Model

NEW GLOBAL FLOOD MODEL

5 returning periods


5 returning periods



5 returning periods


5 returning periods


Outputs

GED - GLOBAL EXPOSURE MODELGAR13

GED 2013 (Global Exposure Database): each record (exposed value) represents a certain building structural type of certain income level/sector in a certain urban area with a special point representation in the centroid of the 5x5 cell.

Urban Area ID

Income Level or Sector

Building type

VALFIS [USDX106]

1 Low Income S5 $ 496,646

1 Low Income C1 $ 7,449,689

1 Low Income C1L $ 7,449,689

1 Low Income C2 $ 496,646

1 Low Income C2M $ 2,483,230

1 Low Income C3 $ 7,449,689

1 Low Income M2 $ 993,292

1 Low Income UFB3 $14,899,377

1 Low Income UCB $ 4,966,459

1 Low Income UNK $ 2,979,875

1 Middle Income S5 $ 822,740

1 Middle Income C1 $12,341,105

1 Middle Income C1L $12,341,105

1 Middle Income C2 $ 822,740

1 Middle Income C2M $ 4,113,702

1 Middle Income C3 $12,341,105

1 Middle Income M2 $ 1,645,481

1 Middle Income UFB3 $24,682,210

1 Middle Income UCB $ 8,227,403

1 Middle Income UNK $ 4,936,442

1 High Income S5 $ 45,026

1 High Income C1 $ 675,384 Capital stock distribution on a 5x5 km grid: map shows aggregate values for resident buildings.


GEDGlobal Exposure Model

GAR13Andrea de Bono (GRID) Miguel Mora (CIMNE)

GAR 2013 / 2015

A


GED Thematic components


Produced capital and urban land

Building structure class(WAPMERR)

Demographic

Socioeconomic

Building type

Assets value

People living in urban areas

Built-environment

Income, employment, health, education


Built-environment and urban population


Built-environment

4) populate “urban areas”

extract

Urban areas mask: from remote sensing (MODIS 500m)

Population: number people per cell (Source Landscan)

Urban population: nb. people per cell

1


Capital stock estimation


GAR 2013

We use the World Bank’s “comprehensive wealth” methodology*.

* World Bank (2011). The changing wealth of nations : measuring sustainable development in the new millennium

•Produced Capital using the Perpetual Inventory Method for machinery and structures, based on Gross Capital Formation data, and layers on urban land as a proportion of this.

Capital stock data are at national level. The downscaling to cell is done using GDP at subnational scale as proxy


Than youhttp://www.grid.unep.ch/GCV

http://preview.grid.unep.ch

risk and global change © pascal peduzzi, unep/grid-geneva, 2014. 1 pascal peduzzi, phd gregory...

Documents

risk global change

global change pascal

global analysis slide

global coverage

risk data slide

peduzzi c

bono slide

agency university of