Three policy scenarios for CAFE

Markus Amann, Janusz Cofala, Chris Heyes, Zbigniew Klimont, Wolfgang Schöpp, Fabian Wagner

Assumptions

• CAFE baseline “with climate measures” for 2020

• Agricultural projections without CAP reform

• Further measures for road emissions taken

• Meteorology of 1997

Costs for reducing the four effectsbetween CLE and MTFR

MTFR

CLE0

10

20

30

40

0%10%20%30%40%50%60%70%80%90%100%

Gap closure between CLE and MTFR

Billion Euro/year

PM Ozone Acidification Eutrophication

*) excluding costs for road sources

Targets selected for the optimization

Ambition level

CLE Low Medium High MTFR

Years of life lost due to PM2.5 (EU-wide, million YOLLs)

137 110 104 101 96

Acidification (country-wise gap closure on cumulative excess deposition)

0% 55% 75% 85% 100%

Eutrophication (country-wise gap closure on cumulative excess deposition)

0% 55% 75% 85% 100%

Ozone (country-wise gap closure on SOMO35)

0% 60% 80% 90% 100%

0

10

20

30

40

MTFR High Medium Low CLE

Billions Euro/year

PM optimized O3 optimized Acidification optimized Eutrophication optimized

Emission control costsfor three ambition levels for the four targets*)

0

10

20

30

40

MTFR High Medium Low CLE

Billions Euro/year

PM optimized O3 optimized Acidification optimizedEutrophication optimized Joint optimization

*) excluding costs for road sources

Effects in 2000 and for CAFE medium ambition 2020

PM Eutrophication Ozone

Acid, forests Acid, lakes Acid, semi-nat.

Optimized emission reductions for EU-25of the D23 scenarios [2000=100%]

0%

20%

40%

60%

80%

100%

SO2 NOx VOC NH3 PM2.5

% of 2000 emissions

Gray range: CLE to MTFR Low ambition Medium ambition High ambition

Costs per pollutant for EU-25on top of CLE

0

5

10

15

20

25

30

35

40

45

Low ambition Medium ambition High ambition MTFR

Billion Euros/year

Road sources SO2 NOx NH3 VOC PM

Measures taken in the D23 medium ambition scenario

• SO2

– Low sulphur coal

– Low sulphur heavy fuel oil

– Flue gas desulphurization

• NOx

– Combustion modifications

– Selective non-catalytic and catalytic reduction

– NOx reduction from light- and heavy-duty diesel vehicles

• PM– High efficiency dedusters

– New boiler types in the residential sector

– Good housekeeping measures on oil boilers

– Low sulphur fuels for (national) sea traffic

Measures taken in the D23 medium ambition scenario

• Ammonia– Application of pig and cattle

manures with low ammonia application measures

– Substituting ammonium nitrate by urea

– Covers on manure storage for pigs and cattle

– Changes in feeding strategies

• VOC – Control of fugitive losses in

organic chemical industry

– Switch emulsion bitumen in road paving

– Paint application (coatings)

–

– Stage II

– Liquid fuel production (improved flare and reduction of fugitive losses)

Distribution of costs[€/person/year]

0

20

40

60

80

Aus

tria

Bel

gium

Cyp

rus

Cze

ch R

ep.

Den

mar

k

Est

onia

Fin

land

Fra

nce

Ger

man

y

Gre

ece

Hun

gary

Irel

and

Italy

Latv

ia

Lith

uani

a

Luxe

mbo

urg

Mal

ta

Net

herla

nds

Pol

and

Por

tuga

l

Slo

vaki

a

Slo

veni

a

Spa

in

Sw

eden UK

EU

-25

Total Costs (Euro/person/yr) Low ambition Medium ambition

0

20

40

60

80

Aus

tria

Bel

gium

Cyp

rus

Cze

ch R

ep.

Den

mar

k

Est

onia

Fin

land

Fra

nce

Ger

man

y

Gre

ece

Hun

gary

Irel

and

Ital

y

Latv

ia

Lith

uani

a

Luxe

mbo

urg

Mal

ta

Net

herla

nds

Pol

and

Por

tuga

l

Slo

vaki

a

Slo

veni

a

Spa

in

Sw

eden UK

EU

-25

Total Costs (Euro/person/yr) Low ambition Medium ambition High ambition

*) excluding costs for road sources

Distribution of physical benefitsMedium ambition scenario

% point improvements in total European effect indicators*), sum over four effects

*) between CLE and MTFR

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

Aus

tria

Bel

gium

Cyp

rus

Cze

ch R

ep.

Den

mar

k

Est

onia

Fin

land

Fra

nce

Ger

man

y

Gre

ece

Hun

gary

Irel

and

Ital

y

Latv

ia

Lith

uani

a

Luxe

mbo

urg

Mal

ta

Net

herla

nds

Pol

and

Por

tuga

l

Slo

vaki

a

Slo

veni

a

Spa

in

Sw

eden UK

YOLL gains Eutrophication Acidification Ozone

Sensitivity analyses

1. How would measures for ships change the outcomes?

2. Are emission reductions in the joint optimization driven by health or ecosystems targets?

3. How would alternative health impact theories change the results?

4. How would national energy and agricultural projections change the optimization outcome?

Sensitivity analysis 1:With medium ambition measures for ships [million €]

Without ship

measures

With “medium ambition” measures for ships

  Costs for land-based

sources

Costs for land-based

sources

Costs for ships

Total costs Cost difference

Low ambition

5953 5813 28 5841 -140

Medium ambition

10709 10522 28 10550 -159

High ambition

14882 14582 28 14556 -326

Sensitivity analysis 2:Are PM or ecosystems targets driving?

0

10

20

30

40

MTFR High Medium Low CLE

Billion Euros/year

O3 optimized Acidification optimized Eutrophication optimized Joint optimization without PM

0

10

20

30

40

MTFR High Medium Low CLE

PM optimized Joint optimization without PM Joint optimization

*) excluding costs for road sources

Sensitivity analysis 2: Are PM or ecosystems targets driving?

0%

20%

40%

60%

80%

100%

PM

driv

en

Eco

syst

ems

driv

en

Join

t op

timiz

atio

n

PM

driv

en

Eco

syst

ems

driv

en

Join

t op

timiz

atio

n

PM

driv

en

Eco

syst

ems

driv

en

Join

t op

timiz

atio

n

PM

driv

en

Eco

syst

ems

driv

en

Join

t op

timiz

atio

n

PM

driv

en

Eco

syst

ems

driv

en

Join

t op

timiz

atio

n

SO2 NOx VOC NH3 PM2.5

% of 2000 emissions

Gray range: CLE - MTFR Low ambition Medium ambition High ambition

Sensitivity analysis 3:Uncertainties in PM health impact theories

• Alternative hypothesis: “Secondary inorganic aerosols do not contribute to health impacts, all PM effects are related to primary PM2.5 emissions”

• Sensitivity study: – Achieve same relative improvement in mortality estimated for CLE

based on “primary PM only” theory – or, expressed alternatively: – Reduce primary PM2.5 concentrations by the same percentage as total

PM2.5 would be reduced in reference case

• Two optimization runs:1. Health only

2. Multi-effect optimization

PM fractions associated with health impacts

Natural

Sec organics

Nitrates

Sulfates

Carbon

Primarynon-carbon

StandardRAINS

approachSensitivity

case

WHO advice

Primaryanthrop.particles

Secondaryanthrop.particles

Sensitivity analysis 3:Control costs for alternative impact theories

0

10

20

30

40

Standard approach Primary PM only Standard approach Primary PM only

Health only optimized Multi-effect optimization

Billion Euros/year

Low ambition Medium ambition High ambition MTFR

Sensitivity analysis 3:Reductions of

“Primary PM only” case vs. Standard approach, joint optimization

0%

20%

40%

60%

80%

100%

"Prim

ary

PM

only

"

Sta

ndar

dap

proa

ch

"Prim

ary

PM

only

"

Sta

ndar

dap

proa

ch

"Prim

ary

PM

only

"

Sta

ndar

dap

proa

ch

"Prim

ary

PM

only

"

Sta

ndar

dap

proa

ch

"Prim

ary

PM

only

"

Sta

ndar

dap

proa

ch

SO2 NOx VOC NH3 PM2.5

% of 2000 emissions

CLE - MTFR Low ambition Medium ambition High ambition

Sensitivity analysis 4:Implications of national energy and agricultural projections

• National energy and agricultural projections available for 10 countries

• However, these do not comply with Kyoto obligations

• Two questions:

– How would optimization results (“emission ceilings”) change based on the national projections?

– What about the feasibility/costs of emission ceilings, if the underlying projection does not materialize?

• Approach:

– Joint optimization with national projections for same target setting rules (gap closures and relative YOLL improvement recalculated for new CLE/MTFR)

CO2 emissions in 2020 of national and PRIMES energy projections, relative to 2000

40%

60%

80%

100%

120%

140%

Belgium Denmark Finland France Italy Portugal Sweden UK CzechRepublic

Slovenia

With climate measures No further climate measures National projection

Sensitivity analysis 4:Costs of optimized scenarios, CAFE baseline vs. national projections

0

10

20

30

40

MTFR high medium low CLE

CAFE "with climate measures" National projections

Billion €/year

*) excluding costs for road sources

Sensitivity analysis 4:SO2 emissions, CAFE baseline vs. national projections

0%

20%

40%

60%

80%

100%

120%

Au

stri

a

Be

lgiu

m

Cyp

rus

Cze

ch R

ep

.

De

nm

ark

Est

on

ia

Fin

lan

d

Fra

nce

Ge

rma

ny

Gre

ece

Hu

ng

ary

Ire

lan

d

Ita

ly

La

tvia

Lith

ua

nia

Lu

xem

bo

urg

Ma

lta

Ne

the

rla

nd

s

Po

lan

d

Po

rtu

ga

l

Slo

vaki

a

Slo

ven

ia

Sp

ain

Sw

ed

en

UK

EU

-25

CAFE baseline CLE-MTFR low med high National projections CLE-MTFR low med high

Emissions in 2000 = 100%

Conclusions

• Three cases calculated for three ambition levels: costs of 6, 11 and 15 billion €/year

• For targets on PM, eutrophication, acidification and ozone

• Resulting emission reductions are cost-effective and have equitable distributions of costs and physical benefits

• Findings from sensitivity analyses:

– Control of ship emissions decrease overall costs

– Optimization driven by health and ecosystems targets

– Multi-effect optimization increases robustness against uncertainties in health impact mechanisms

– Robustness against national energy projections needs further attention (and more robust national projections!)