representative concentration pathways. · sres only covered scenarios without climate policy 3 13...

18
1 Representative Concentration Pathways. Detlef P. van Vuuren

Upload: trinhcong

Post on 02-Aug-2018

216 views

Category:

Documents


0 download

TRANSCRIPT

1

Representative Concentration Pathways.

Detlef P. van Vuuren

13 november 2013 2

Reasons for new scenarios

Four important reasons to develop new community scenarios for climate assessment:

– 1. Need to cover a wider range of GHG concentrations (SRES only included baseline scenarios)

Moss RH, et al (2010) The next generation of scenarios for climate change research and

assessment. Nature 463:747-756.

Four important reasons to develop new community scenarios for climate assessment:

– 1. Need to cover a wider range of GHG concentrations (SRES only included baseline scenarios)

SRES only covered scenarios without climate policy

13 november 2013 3

A2

A1 B2

B1

RCP8.5

RCP6

RCP4.5

RCP2.6

SRES does not lead to correct insights with respect to 21st century warming and warming commitment.

13 november 2013 4

Reasons for new scenarios

Four important reasons to develop new community scenarios for climate assessment:

– 1. Need to cover a wider range of GHG concentrations (SRES only included baseline scenarios)

Moss RH, et al (2010) The next generation of scenarios for climate change research and

assessment. Nature 463:747-756.

Four important reasons to develop new community scenarios for climate assessment:

– 1. Need to cover a wider range of GHG concentrations (SRES only included baseline scenarios)

– 2. Need for a wider set of parameters (Climate models have become more complex; higher information need).

– 3. Need for scenarios that cover mitigation & adaptation issues (need for more collaboration between “WGs”)

– 4. Use more recent insight into trends in scenario drivers (update)

13 november 2013 5

New scenarios

Moss RH, et al (2010) The next generation of scenarios for climate change research and

assessment. Nature 463:747-756.

13 november 2013 6

Overall process

Van Vuuren, D.P. et al 2011. Representative Concentration Pathways: An overview.

Climatic Change.

Detailed, mapped data from integrated Assessment models

Earth system and climate models

Data on emissions, land and concentration

Data on climate change and functioning of Earth system

13 november 2013 7

Emissions

Van Vuuren, D.P. et al 2011. Representative Concentration Pathways: An overview.

Climatic Change.

RCPs span nicely the range of the scenario literature

2000 2020 2040 2060 2080 2100-20

0

20

40

60

80

100

120

Em

issio

ns (

GtC

O2/y

r)

13 november 2013 8

Emissions

Van Vuuren, D.P. and Carter, T 2013. Reconciling the old with the new. Climatic Change.

Reference

8.5 Wm - 2

6.0 Wm - 2

4.5 Wm - 2

2.6 Wm - 2

SRES

A2 / A1FI

B2 / A1B

B1

RCP8.5: High range emission scenario (possible development for high population numbers, high fossil/coal use)

RCP6.0: Medium range emission scenario (low-medium baseline scenario or high mitigation scenario)

RCP4.5: Medium range emission scenario (high mitigation scenario)

RCP2.6: Low range mitigation scenario 2000 2020 2040 2060 2080 2100

-20

0

20

40

60

80

100

120

Em

issio

ns (

GtC

O2/y

r)

13 november 2013 9

Emissions

Van Vuuren, D.P. et al 2013. Roads from Rio+20. PBL Netherlands Environmental

Assessment Agency.

2000 2020 2040 2060 2080 2100-20

0

20

40

60

80

100

120

Em

issio

ns (

GtC

O2/y

r)

2000 2020 2040 2060 2080 2100-20

-10

0

10

20

30

40

50

Em

issio

ns (

GtC

O2/y

r)

Peak in emissions in 2020.

Around 50% reduction in 2050

Negative emissions after 2080 (energy system alone already negative around 2070)

Is this feasible?

Model studies (e.g. FP7 AMPERE, LIMITS and RESPONSES) find reductions to be technically feasible

IAM literature indicates around 60-70% probability of achieving 2oC.

However, stringent conditions….

13 november 2013 10

Emissions

FP7 projects AMPERE/LIMITS and others

2000 2020 2040 2060 2080 2100-20

-10

0

10

20

30

40

50

Em

issio

ns (

GtC

O2/y

r)

Negative emissions: implies reliance on “carbon dioxide removal technologies” (mostly Bio-energy and CCS)

Full participation in climate policy

Sustainable bio-energy 150 EJ/yr ? ~ -10 GtCO2 /yr - but only if everything

is used for BECCS

13 november 2013 11

Emissions

Van Vuuren, D.P. et al 2011. Representative Concentration Pathways: An overview.

Climatic Change.

But even stronger emission reductions 2030-2050 -5% p.a.

More negative in 2100

Some delay possible (e.g. pledges)

2000 2020 2040 2060 2080 2100-20

-10

0

10

20

30

40

50

Em

issio

ns (

GtC

O2/y

r)

13 november 2013 12

Emissions

Van Vuuren, D.P. et al 2011. Representative Concentration Pathways: An overview.

Climatic Change.

Alternative no negative emissions…

But requires more immediate reductions.

2000 2020 2040 2060 2080 2100-20

-10

0

10

20

30

40

50

Em

issio

ns (

GtC

O2/y

r)

Annual capacity additions

13 november 2013 13

Average annual

capacity additions

(history and short to

medium term future)

for various fossil-

based and low-

carbon energy

technologies in the

RefPol-450

scenario.

RCPs has provided a strong link between WG1 – WG3 work.

RCPs allow for a set of policy-relevant conclusions in WG1 – and at the same updating of WG-3 type models.

RCP2.6 allowed for a strong research focus on the feasibility of the 2oC target

Further research needed on climate impacts of RCPs (needs also assessment of socio-economic conditions)

Conclusions

RCPs did not span full literature range

13 november 2013 15

SO2 emissions NOx emissions

All groups aimed to include best estimates of environmental policy

Maybe not a problem for climate research, but not attractive for people looking into air pollution (and the need for policy)

13 november 2013 16

Story about too high S emissions

Research on reaching RCP2.6

WILL CHANGE SLIDE TO INCLUDE RCP2.6 AND REMOVE CATEGORIES. THUS WILL SHOW RCP2.6 VS RECENT SCENARIOS THAT INCLUDE OR DO NOT INCLUDE NEGATIVE EMISSIONS

2000 2020 2040 2060 2080 2100-20

0

20

40

60

2000 2020 2040 2060 2080 2100-20

0

20

40

60

2000 2020 2040 2060 2080 2100-20

0

20

40

60

80

100

120

140

160

Role of policy

assumptions

Role of negative

emissions

Cat6

Cat5

Cat4

Cat3Cat2

Em

issio

ns (

GtC

O2/y

r)

Cat1

All scenarios

No negative

emissions

More rapid

reductions in the

short run

More stringent

reduction in the

long term

Rapid reductions

in the medium

tem

Em

issio

ns (

GtC

O2/y

r)

Category 1

Optimal

Delayed

Category 2

Optimal

Delayed

Category 3

Optimal

Delayed

Less short

term emission

reduction

Em

issio

ns (

GtC

O2/y

r)

Category 1

No neg. emissions

Neg. emissions

Category 2

No neg. emissions

Neg. emissions

Categroy 3

No neg. emissions

Neg. emissions

P(2o): Cat 1.: 60-80% Cat.2 : 40-60% Cat.3 : 20-40%

13 november 2013 18

Emissions

Van Vuuren, D.P. et al 2011. Representative Concentration Pathways: An overview.

Climatic Change.

Reference

8.5 Wm - 2

6.0 Wm - 2

4.5 Wm - 2

2.6 Wm - 2

SRES

A2 / A1FI

B2 / A1B

B1

RCP8.5: High range emission scenario (possible development for high population numbers, high fossil/coal use)

RCP6.0: Medium range emission scenario (low-medium baseline scenario or high mitigation scenario)

RCP4.5: Medium range emission scenario (high mitigation scenario)

RCP2.6: Low range mitigation scenario