S4.Grp2.3 Mongolia, AAL presentation, lecture 2, UPDATED Lecture... · 2020-07-08 · Microsoft PowerPoint - S4.Grp2.3_Mongolia, AAL presentation, lecture 2, UPDATED Author: SakollerdL

Understanding and Estimating Disaster Risks Concepts and methods

Madhurima Sarkar-SwaisgoodICT & DRR – UNESCAP

Expert consultation: Addressing the transboundary dimensions of the 2030 Agenda through Regional Economic Cooperation and Integration in Asia and the Pacific

21-22 November 2019

FORMERLY AN INFRASTRUCTURE INVESTMENT

Estimated global investment in infrastructure 2015 - 2030 approximately USD 90 trillion

Unless this investment is risk informed it represents a major investment in future disasters

Complete/partial destruction of immovable assets and of stock (including final goods, goods in process, raw materials, materials and spare parts).

The flows of goods and services that will not be produced or rendered over a time span that begins after the disaster and may extend throughout the rehabilitation and reconstruction periods

Knock-on impacts on production e.g. through supply chain or medium/long run productivity effects from a natural disaster

Reflect the manner in which disasters modify the performance of the main economic variables of the affected country

Direct

Indirect

Wider

DISASTER AND CLIMATE IMPACTS

Stock

Flow

DIRECT LOSSES

Peru: Coastal ENSO event 20173000 km roads damaged, 323 bridges destroyed. Direct losses approximately USD 7 billion equivalent to 72% of the public investment executed in 2016

WHY DO WE NEED TO ADOPT A PROBABILISTIC APPROACH

PROBABILISTIC RISK MODELS

Probabilistic risk models can provide robust estimates of risk for countries and specific infrastructure sectors

Metrics include Annual Average Loss (AAL) and Probable Maximum Loss (PML)

PROBABILISTIC RISK METRICS

LOSS EXCEEDANCE CURVE (LEC)

[#/a

ño]

Loss

Exce

ed

an

ce r

ate

[#

/year]

AVERAGE ANNUAL LOSS (AAL)

PROBABLE MAXIMUM LOSS (PML)

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

35.0%

40.0%

45.0%

0.00%

0.50%

1.00%

1.50%

2.00%

2.50%

3.00%

3.50%

4.00%

4.50%

0 2344

Pérdidas por evento

Pérdida acumulada

Return period

PM

L

THE GLOBAL RISK MODEL: ANNUAL AVERAGE LOSS (AAL)

Global Annual Average Loss (AAL) = USD 293 billion (2017) in built environment for rapid-onset hazards

Represents 5% of expected annual global investment in infrastructure.

DISASTER LOSSES- MONGOLIA, KAZAKHSTAN, KYRGYZSTAN

Kazakhstan Kyrgyzstan Mongolia

EXTENSIVE RISK INCREASESTHE AAL

Exposed value Retrospective AAL (extensive risk)

Prospective AAL (intensive risk) Hybrid AAL

US$ million US$ million ‰ US$ million ‰ US$ million ‰

Lao PDR 45.475 104,67 2,30 325,96 7,17 410,98 9,04

Sri Lanka 99.813 169,63 1,70 63,52 0,64 408,99 4,10

IMPLICATIONS FOR SUSTAINABLE DEVELOPMENT

The higher the ratio of AAL to social expenditure or capital investment, the lower the sustainability of development.

AAL/capital investment:Myanmar 27%Philippines 13%

AAL/Social expenditureMyanmar 120%Philippines 94%

MULTI-HAZARD AAL (EARTHQUAKES AND FLOODS)

Country EXPOSED VALUE AAL AAL/Capital stock AAL/GFCF AAL/Social

expenditureMillion US$ Million US$ [‰] [%] [%]

Kazakhstan 734.310 750,46 1,02 1,73 4,10Kyrgyzstan 18.467 92,68 5,02 3,90 8,20Mongolia 36.588 34,87 0,95 1,20 2,22

Only direct losses in the built environment

Country AAL AAL/GDP AAL/Capital stock AAL/GFCF AAL/Social

expenditureMillion US$ [%] [‰] [%] [%]

Kazakhstan 1876,15 1,02 2,55 4,33 10,26Kyrgyzstan 231,7 3,47 12,55 9,75 20,51Mongolia 87,175 0,74 2,38 3,00 5,55

Estimating possible direct and indirect losses

IDENTIFY RISK LAYERS AND DEFINE RISK APPETITE

As hazard increases in intensity it becomes more difficult and costly to reduce risk.

Defining a reasonable level of resilience

100 year return period or 1% probability per year

[#/a

ño]

Loss

Exce

ed

an

ce r

ate

[#

/year]

Current risk

Loss

Exce

ed

an

ce r

ate

[#

/year]

Resilience target

Risk reduction

Identify risk management alternatives Structural:Adoption of standards for new infrastructure investments

Engineering assessments to retrofit critical infrastructure

Financial:Design of risk transfer instruments

Estimation of costs of risk transfer

Response:Contingency plans

Early warning systems

Reduces Direct Loss

Reduces Net Loss and

accelerates recovery

Improves system performance and reduces the Indirect Loss.

ASSESS COSTS FOR EACH LEVEL OF RESILIENCE

A hypothetical example: flood mitigation infrastructure:

• Currently 1m flood defense wall

• 2m wall – cost $30M reduction in AAL = $10M

• 2.5m wall - cost $60M reduction in AAL = $15M

• 3m wall cost $120M reduction in AAL = $20M

RISK METRICS FOR RESILIENT INFRASTRUCTURE

National multi-hazard risk profile

High resolution infrastructuresector risk model Risk management strategy

LEC/ AAL / PML Extensive riskIdentify financing gapsImplications for sustainable development

Define resilience targetsand strategyIntegrate into public and private investment planning

AAL / PML values for each sector or region (energy, transport etc.)Identification of critical risks in networks and nodesModelling of indirect losses

Identification of risk layers. Define risk appetite

Adoption of standards for new infrastructure investments

Retrofit critical infrastructure

Risk transfer

Contingency plans and early warning systems

Finance and planning ministries

Sector line ministries and sub-national governments

Cost-effective risk management strategies

Climate change scenarios

Utility companies, infrastructure operators

CONTRIBUTION OF AGRICULTURE TO GDP

Value Added Agriculture, Forestry and Fishing –VA AFF

*http://www.fao.org/faostat/en/#data/MK

KAZAKHSTAN KYRGYZSTAN MONGOLIAGDP 184.388 6678,18 11749,6

VALUE ADDED 8.686 939 1569,3VALUE ADDED/GDP (%) 5 14,06 13,35

GFCF/VA (%) 15 2 22

PROBABILISTIC DROUGHT RISK

ASSESSMENT

HAZARD

Simulated weather time seriesSimulated time series are generated stochastically from the historicalinformation.

The objective is not to forecast future weather conditions, but to generatefeasible combinations of drought conditions, such as low precipitation andhigh temperature.

Historical Simulated(not a forecast)

HAZARD

Identification of regional droughts using indexes

DROUGHT HAZARD

Collection of scenariosHazard is represented as a set of stochastic scenarios.

Mutually exclusive

Collectively exhaustive

These scenarios (events) are assumed to be:

They allow probabilistic representation:

Occurrence frequency (temporal probability)

Gridded statistical moments (spatial probability)

Time series, at any location, of weather variables (precipitation and temperature)

EXPOSURE

Location

Crop characteristics

Exposed elements database

Crop valuation

Georeferenced data, area

Geographical distribution

Production cost

Inpu

tsO

utpu

ts

Type and seasonality

VULNERABILITY

Crop development

Crop transpiration

Soil water balance

Biomass production

Yield (Y) from biomass (B)

Decreased hydro-power potential in drought-prone regions

DROUGHT RISK

Relative AAL to agriculture production, PML curves, probability exceedance loss in the next year and next event in a Central American country for current and future climate with different representative concentration pathways (RCP 2.6, 4.5, and 8.5)

Probable Maximum loss

Exceedance loss probability in the next year

Exceedance loss probability in the next event

DROUGHT RISK

Relative Average Annual Loss to the agriculture production by provinces in a Central American country for current and future climate with different representative concentration pathways (RCP 2.6, 4.5, and 8.5)

DATA SOURCES

UN Digital Library

Global Assessment Report on Disaster Risk Reduction (GAR) Risk Atlas, 2015

ESCAP statistics

Asia Information Superhighway, 2018

ESCAP Asia-Pacific Energy Portal 2018

ESCAP Transportation Data 2018(c)

ESCAP Statistical Database

Socio-economic – World Bank, IMF, ADB, ESCAP

NASA’s Earth Database

DATA SOURCES

Copernicus Open Access Hub

WMO

NOAA

IPCC

Humanitarian Data Exchange

THANK YOU!

Madhurima Sarkar-Swaisgood, PhDEconomic Affairs OfficerICT and Disaster Risk Reduction Division, UNESCAPEmail: sarkar-swaisgood@un.org