1 of 36

Dabo Guan

Climate Change Economics

University College London

Disaster Footprints Modelling and Applications

2 of 36

来源：Scenarios towards limiting global mean temperature increase below 1.5 C, Nature Climate Change, 2018

全球均温持续上升

Global mean temperature keeps rising

高气候风险

High Climate Risk

中气候风险

Medium Climate Risk

升温2℃及以下

2℃ or below

全球碳排放（GtCO₂/年）

3 of 36

高气候风险对“吃货们”的影响巨大

High climate risk affects ‘your consumption’!

Sources:

Emergent constraint on crop yield response to warmer temperature from field experiments, Nature Sustainability, 2020

A bitter cup: climate change profile of global production of Arabica and Robusta coffee, Climatic Change, 2015

Vulnerability to climate change of cocoa in West Africa, Science of the Total Environment, 2016

4 of 36

Disaster Footprint

Disaster footprint is a measure of the exclusive total economic impact that is directly and indirectly caused by a disaster to the affected region and wider economic systems.

Direct economic impacts

Indirect economic impacts

5 of 36

Disaster Footprint Model

A framework to quantify and incorporate the resulting damages to physical capital and population and thereby, the economic impacts caused by the reductions in capital and labour productivity, as well as their cascading macroeconomic impacts along economic production chains.

6 of 36

Conceptual framework

Inter-Regional trades

Nation/World -wide

Percentage loss

Labour shock

Capital shock

(The dead, missing & injured)

(Damage to factories, facilities, roads, houses, etc.)

Inter-industrial links

Disaster

Footprint

Earthquakes

Floods

Droughts

Wildfires

Tsunamis

...

Indirect

Direct

(Physical damage)

Output loss over time

GDP

t

0

Pre-disaster

Disaster hit

Recovery

Road repair

Import

Labour restore

Factory reconstruction

Region n

Region 2

Region 3

Region 1

Mendoza-Tinoco, D., Guan, D., Zeng, Z., Xia, Y., & Serrano, A. (2017). Flood footprint of the 2007 floods in the UK: The case of the Yorkshire and The Humber Region. Journal of Cleaner Production, 168, 655-667.

7 of 36

Development of “Disaster Footprint Model”

Guan et al., 2020

Published in Nature Human Behaviour

Focus on global supply-chain effects of COVID-19 control measures

Mendoza-Tinoco et al., 2020

Application to the 2009 Central European floods

Zeng and Guan, 2020

Hypothetical multiple two-flood event

Zeng et al., 2019

Incorporation of ‘basic demand’ referring to the necessities of life during post-flood recovery; Prioritized-proportional rationing scheme among final consumption and reconstruction use; Flexibility in sensitivity analysis for external factors

Mendoza-Tinoco et al., 2017

Hybrid model: extension of the Adaptive Regional IO (ARIO) model (Hallegatte, 2008)

Xie et al., 2018

First integrated modelling study of ESM + Crop modelling + Disaster footprint to account for climate extreme impact to final demand and price changes, Nature Plants.

8 of 36

2015 Flooding. Flood Footprint for Yorkshire and The Humber region

Based on preliminary damage reports and a previous analysis on 2007 Floods for the same region.

2015 Flood
Region:	Yorkshire and The Humber
	Recovery time	21	weeks

Units:	GBP MILLIONS (2003)
	industrial direct damage	3,056.00
	residential direct damage	793.36
	indirect damage	3,895.90
	Ratio Indirect/Direct damage	1.01

9 of 36

10 of 36

2009 Central European floods

Direct damages: €356 million euros

Industrial capital: €238 million
Residential capital: €118 million

Indirect damages: €663 million euros

0.04% of the German annual GDP in 2009

Mendoza‐Tinoco et al (2020) Flood Footprint Assessment: A Multiregional Case of 2009 Central European Floods. Risk Analysis, 40 (8). pp. 1612-1631.

11 of 36

Spatial distribution of direct and indirect damages (flood footprint)

Overall most affected region: Vienna in Austria

Fig. 3(a) depicts the distribution of direct losses of industrial capital. Austria was the most affected country, as it experienced 38% of all losses in this category (ca. €91 million). Within Austria, Vienna (the darkest region) was the most strongly affected region, accounting for 32% of direct industrial losses. Fig. 3(b) shows the distribution of direct losses caused by residential capital damage. Again, Austria was the most affected country, as it experienced 44% of the total losses in this category (ca. €52 million). The three most affected regions are localised within Austria: Vienna (the darkest region), Niederösterreich (Lower Austria) and Oberösterreich (Upper Austria) experienced 32%, 21% and 20% of the national residential losses, respectively. The indirect losses caused by constraints in labour and industry are shown in Fig. 3(c), which constitute two-thirds (65%) of the total flood footprint. The most affected country is Austria, with 31% of total indirect losses (€205 million), while the most affected region is Oberbayern in Germany, which accounts for 36% (€63 million) of national indirect losses. The geographical distribution of the flood footprint is presented in Fig. 3(d). This footprint shows that Austria experienced the largest proportion of losses, accounting for over one-third of the total flood footprint (€347 million).

12 of 36

Flood footprints: integrated with climate models and hydrological and hydraulic models

The integrated modelling framework used in this study:

Hazard assessment: a global hydrological model called multi-regional HBV is driven by climate data derived from observations (for baseline period 1961-1990) and outputs from five global climate models (GCMs) (for future period 2086-2115) at six warming levels from 1.5 to 4 oC; the runoff simulated by the hydrological model is then used to force a global hydraulic model called CaMa-Flood to simulate flood depth, extent, probabilities, etc.

Exposure assessment: socio-economic information derived from World Bank (for population) and European Space Agency Climate Change Initiative (ESA CCI) land cover data (baseline) and the Shared Socio-economic Pathways (SSP2) (future scenarios) is combined with the flood hazard maps to estimate the exposure of population and cropland and urban land area.

Damage assessment: country and sector specific depth-damage functions are employed to estimate the direct flood damages to four different land use sectors.

The estimated direct damages are input into an economic model to further calculate the indirect economic losses.

When these assessments are performed for floods with different probabilities (here, exceeding the baseline 100-year floods), the flood risks can be expressed as the expected annual exposure or damage.

13 of 36

CC experiment: million US$/yr

CC+SE experiment: billion US$/yr

Direct and indirect fluvial flood damages

Damages increase with increasing warming levels
In CC: direct damages dominate total losses
In CC+SE:

direct and indirect damages increase significantly
Indirect losses surpass direct damages

This figure presents estimates of direct and indirect economic damages for each country, and climate scenario, under the CC and CC+SE experiments.

Damages in panel A are expressed in million US$/yr for the CC experiment and in panel B in billion US$/yr for the CC+SE experiment. Bars represent the model ensemble average, with whiskers indicating the ensemble maximum and minimum.

The results reflect the underlying data provided from the flood hazard model, highlighting increasing economic damages, above the baseline, in line with the increasing warming scenarios.

India, China, Brazil suffer the largest absolute damages.

Under the CC (climate change only) experiment: direct damages dominate total losses

In CC+SE (climate change and socio-economic development) experiment:

the magnitude of direct damages increase significantly for all countries, reflecting the increasing population and economic assets at risk.

the magnitude of indirect damages not only increase but also surpass the direct damages, because the CC+SE experiment further captures indirect losses due to the continuous slowdown in economic growth from its potential trajectory

This figure also illustrates that there is a large range in uncertainty, shown as the ensemble maximum and minimum values, which also increases under higher warming levels. This reflects the variance seen in the flood model outputs, representing differences in climate change patterns projected by the five GCMs.

14 of 36

Indirect damage as a share of national GDP

Under both CC and CC+SE: Egypt suffers the largest reductions to national GDP, reaching 2.3% and 3.0% of GDP under 4.0°C warming, respectively.
Under CC+SE: for China and Ethiopia, losses decline from the baseline when socio-economic development is included; socio-economic growth makes the economy more resilient
Under CC+SE: for Brazil, Ghana and India, while losses initially decline at lower warming levels, increases are seen from 2.5°C or 3°C onwards; suggesting a tipping point where increasing flood risk outweighs any relative benefits of socio-economic development

This figure shows the average annual indirect economic damage as a share of national GDP (%) (model ensemble average) caused by fluvial flooding for the baseline and future scenarios in the six countries.

Under both CC and CC+SE: Egypt suffers the largest reductions to national GDP, reaching 2.3% and 3.0% of GDP under 4.0°C warming, respectively.

Under CC+SE: for China and Ethiopia, losses decline from the baseline when socio-economic development is included. This reflects the different underlying baseline and future projections of socio-economic growth in Ethiopia and China, which makes the country appear more resilient, when viewed in relative terms, to the costs of fluvial floods under future projections of climate change.

Under CC+SE: for Brazil, Ghana and India, while losses initially decline at lower warming levels, increases are seen from 2.5°C or 3°C onwards; suggesting a tipping point where increasing flood risk outweighs any relative benefits of socio-economic development.

15 of 36

Climate change and your

diet security

16 of 36

Global Food Consumption

(kg/capita/yr)

Source: FAOSTAT

0.8%/yr

17 of 36

Source: FAOSTAT

Source: KIRIN

18 of 36

Research Method Framework

ESMs

Climt. Chng. Shock

Extreme Shock

Barley Yield Shock

Beer Drinking

DSSAT

GTAP

Rest of World

Production

GTAP

Beer

Livest.

Oth.

Demand

Price

Production

Supply

Price

Barley

Beer

Imp.

Yield

Shock

Emission

scenario

Initial

field

Climate change

(tmp, pre, …)

ESMs

Extreme climate

Barley

phenophase

Barley

planting

region

Simulation

Extreme event year

↓

Climatic conditions

Extra Shock

DSSAT

Climate Data

Soil Conditions

Crop Data

Management

Simulation

Distributions of

Yield

Consumption

Supply

Rest of World

Exp.

Imp.

Exp.

19 of 36

Changes in beer consumption and price under increasingly severe drought-heat events

20 of 36

Global supply-chain effects of COVID-19 control measures

Analyse the supply-chain effects of a set of idealized lockdown scenarios, using the latest global trade modelling framework coupling with the disaster footprint model.

Guan, D., Wang, D., Hallegatte, S., Davis, S. J., Huo, J., Li, S., . . . Gong, P. (2020). Global supply-chain effects of COVID-19 control measures. Nature Human Behaviour, 4(6), 577-587. doi:10.1038/s41562-020-0896-8

21 of 36

21

Introduction of research design - scenario settings

Number of countries affected

Spatial spread

Number of months that lockdown measures are in place.

Duration

Percentage by which labour availability and transportation capacity are reduced
Impact-to-labour multipliers

Level of exposure to the virus (that is, the degree and proximity of in-person interactions)
Essential or lifeline sectors (such as electricity)
Option of performing work from home (for example, education)

Strictness

22 of 36

Scenario settings

36 + 3 scenarios
3 recovery scenarios:

New normal scenario: go-slow lifting restrictions
Recurrent pandemic scenario with global cooperation
Recurrent pandemic scenario without global cooperation

23 of 36

Timing and spread

Our first scenario set, China only, assumes that the outbreak of COVID-2019 is only in China, with epidemic control measures from 22 January 2020 onward.
In the second set of scenarios (Europe and the United States), we assumed that regions with the current severe epidemic situation have taken measures from the eleventh week (11 March 2020) to control their epidemic.
The labor and transportation restrictions are consistent with the settings of the scenario set China only, and take the China only 80%–2-month scenario as the default in China, which matches with the reality shown in the Baidu big data.
In the global set of scenarios, we assumed that, in addition to mainland China and the economies in the scenario set Europe and the United States, other economies in the world also began to take measures to control the epidemic in the fifteenth week (8 April 2020).

24 of 36

Restricting labor supply

Isolation measures have different effects on labor supply in different sectors. We set a specific impact to labor multipliers for each sector on the basis of three factors:

exposure level of the sector’s work
whether it is an essential sector
and whether it is possible to work at home

For example, 0.5 for wheat production as the level of exposure is low; 0.1 for electricity and gas supply as essential activities; 0 for health related activities.

25 of 36

Reductions in final demand

The epidemic not only affects the global economic system from the supply side, but also affects economic output through its impact on consumer demand.

26 of 36

Reductions in final demand

Owing to a lack of data, we assumed that the final demand for recreation, accommodation, food, and other services in the affected area fell close to 90% during the duration of the lockdown.

27 of 36

28 of 36

The first insight from the model is that the global cost of the pandemic depends foremost on the number of affected countries, and then on the required duration of lockdown policies; by contrast, the strictness of these policies is comparatively less important.

29 of 36

The second insight from the model is the importance of propagation through global supply chains—even countries that are not directly affected by the virus experience large losses, and low- and middle-income countries are more vulnerable to indirect effects.

countries with close supply-chain relationships with China

countries

with a dominant economic sector

hashed area represents direct losses due to containments and the solid area represents the propagation

Fig. 2 | Direct and indirect value-added losses of COVID-19 in selected countries under 9 scenarios. a–i, Economic loss (measured by the percentage

of value-added losses) in the following nine selected countries: China (a), New Zealand (b), United States (c), Vietnam (d), Nigeria (e), Malaysia (f),

Kazakhstan (g), Jamaica (h) and Mongolia (i). The top row country includes China (affected in the China only scenario), developed countries such as

the United States (affected in the Europe and US scenario) and New Zealand (only affected in the global scenario). The middle row includes countries

(affected in the global scenario) that have close supply-chain relationships with China to assess propagation effects. The bottom row includes countries

with a dominant economic sector. Each plot contains three selected scenarios from the three scenario sets (12 per figure as indicated). The shades of blue,

green and red indicate a duration of 2, 4 and 6 months, respectively. The hashed area in the bars represents direct losses due to containments and the solid

area represents the propagation. The results of other selected countries are provided in Supplementary Figs. 4–11.

30 of 36

The third insight is that specific country sectors are quite vulnerable to impacts that are propagated through global supply chains, even in scenarios in which COVID-19 does not spread globally.

Supply-chain impacts to the German automobile industry. Length of bars show the industries’ relative production losses compared with the original capacity. Cohesion is measured by the trade volume between the sector and the German automotive sector

China only, 80%–2 month (a,b);

Europe and the United States, 60%–4 months (c,d);

and global, 40%–6 month (e,f)

Fig. 4 | Supply-chain impacts to the German automobile industry under three main scenario sets. a–f, The economic impacts to supply chain upstream of German automobile industries (a,c,e) and the economic impacts from the perspective of downstream supply chain (b,d,f) under three scenario sets (China only, 80%–2 month (a,b); Europe and the United States, 60%–4 month (c,d); and global, 40%–6 month (e,f)). The colour and bar area are as described in Fig. 3. The length of bars in a–f show the industries’ relative production losses compared with the original capacity under different scenario sets. The colors of the bars represent the cohesion level of the particular sector to the German automotive sector from blue (weak) to red (strong), which is measured by the trade volume between the particular sector and the German automotive sector.

31 of 36

Finally, we designed and modelled three post-pandemic scenarios of recovery�

Pandemic as a new normal scenario: starting in January 2020, China only placed 80% strictness for 2 months, which was then reduced to 20% for 12 months. EU and the United States placed 60% strictness for 4 months, which was then reduced to 20% strictness for 12 months. Global placed 40% strictness for 6 months, which was then reduced to 20% and gradually relaxed to 0% over a period of 12 months.
Recurrent pandemic scenario with global cooperation: starting in January 2020, each country’s lockdown (that is, China, 80%–2 month; Europe and the United States, 60%–4 month; all other countries, 40%–6 month) was first relaxed to 0% strictness over a period of 2 months, followed by a 3-month period of no restrictions and then another round of strict (80%), 2-month global lockdown starting in January 2021.
Recurrent pandemic scenario without global cooperation: starting in January 2020, each country’s lockdown (that is, China, 80%–2 month; Europe and the United States, 60%–4 month; all other countries, 40%–6 month) was first gradually relaxed to 0% strictness over a period of 2 months, followed by a 3-month period of no restrictions, and then another round of the same less strict, longer lockdowns starting in January 2021, as the first round.

32 of 36

Fourth insight: relaxing lockdown restrictions gradually over a long period of time (in our ‘new normal’ scenario, 12 months) results in substantially lower supply-chain effects than lifting restrictions quickly if it means avoiding another round of strict lockdowns in the coming year.
If the pandemic does recur, stricter and shorter lockdowns (which may depend on global coordination) greatly reduce losses (11% globally) in our estimates.

33 of 36

Summary

We find that supply-chain losses that are related to initial COVID-19 lockdowns are largely dependent on the number of countries imposing restrictions and that losses are more sensitive to the duration of a lockdown than its strictness.
Earlier, stricter and shorter lockdowns can minimize overall losses. A ‘go-slow’ approach to lifting restrictions may reduce overall damages if it avoids the need for further lockdowns.
Regardless of the strategy, the complexity of global supply chains will magnify losses beyond the direct effects of COVID-19. Thus, pandemic control is a public good that requires collective efforts and support to lower-capacity countries.

34 of 36

Enjoy your sweet water, but please think differently… about our consumption

Thank you!

Dabo Guan: guandabo@hotmail.com

35 of 36

Six countries: Brazil, China, Egypt, Ethiopia, Ghana, India
Global warming scenarios: <1.5 ^oC, <2.0 ^oC (aiming to stay below 1.5 ^oC and 2.0 ^oC in 2100, respectively, with 66% probability), 2.5, 3, 3.5, 4 ^oC
Five Global Climate Models (GCMs) from CMIP5 used in this study

Model ID	Modelling Centre/Group	Spatial resolution (arc degree)	References
gfdl_esm2m	Geophysical Fluid Dynamics Lab, USA	2.0°×2.5°	Dunne et al. (2012)
mohc_hadgem2es	Met Office Hadley Centre, UK	1.25°×1.875°	Collins et al. (2011)
ipsl_cm5alr	Pierre Simon Laplace Institute, France	1.9°×3.75°	Dufresne et al. (2013)
nies_mirocesmch	JAMSTEC/AORI/NIES, Japan	2.8°×2.8°	Watanabe et al. (2011)
ncc_noresm1m	Norwegian Climate Center, Norway	1.9°×2.5°	Tjiputra et al. (2013)

36 of 36

Settings of model experiments

Baseline: 1961-1990; Future: 2086-2115 (30-year average, model ensemble mean)
Flood hazard data (daily):
hydrological model (multi-regional HBV): runoff (0.5^o), driven by observations (baseline) and output from five GCMs from CMIP5 (future)
hydraulic model (CaMa-Flood): flood depth and area (0.25^o)
Selection of flood events: greater than baseline 100-year floods
Two sets of model experiments:

Climate change only (CC only): only climate variables change, while socio-economic conditions are kept constant at 1992
Climate+socio-economic changes (CC+SE): both climate and socio-economic input data change
Socio-economic conditions refer to each country’s population, labour force, national gross domestic product (GDP), capital stock, land cover, and economic structure.

A global hydrological model called multi-regional HBV is driven by climate data derived from observations (for baseline period 1961-1990) and outputs from five global climate models (GCMs) (for future period 2086-2115) at six warming levels from 1.5 to 4 oC; the runoff simulated by the hydrological model is then used to force a global hydraulic model called CaMa-Flood to simulate flood depth, extent, probabilities, etc.

To differentiate the impacts of climate change and socio-economic development, two sets of model experiments are conducted.

Firstly, a “climate change only” experiment (CC), in which socio-economic conditions are kept constant at the baseline level for the six warming scenarios, reflects the impact of climate change alone on future flood risks.

The second, “climate change and socio-economic development” experiment (CC+SE) considers both climate change and socio-economic growth in parallel. The differences between the estimates of these two experiments can reflect the effect of socio-economic development alone on future flood risks.

Here, socio-economic conditions refer to each country’s population, labour force, national gross domestic product (GDP), capital stock, land cover, and economic structure.