Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
Use of Scenarios in the Special Report on Renewable Energy of
the IPCC Fifth Assessment Report Leon Clarke
May 25, 2011
Global Technology Strategy Program Technical Workshop
Joint Global Change Research Institute
There was a long discussion about using scenarios across chapters in the SRREN.
Energy-economic or integrated assessment models
Bottom-up assessments of technology cost and potential
The interaction should go both ways.
0
1
2
3
4
5
6
0 200 400 600 800Energy Production (EJ/yr)
Res
ourc
e G
rade
RESOURCE SUPPLY CURVE
Scenarios of scale, timing, location.
ENERGY SUPPLY SCENARIO
Supply curves and related aggregate representations
Vetting scenarios from aggregate models
For Chapter 10, we took on a thorough review of existing scenarios. ! Driving Questions:
! What sorts of future levels of renewable energy deployment are consistent with different CO2 concentration goals; or, put another way, what is the linkage between CO2 concentration goals and the deployment of renewable energy?
! Which classes of renewable energy will be the most prominent energy producers and how quickly might they expand production?
! Where would an expansion in renewable energy occur? ! What is the linkage between the costs of mitigation and an expansion
of renewable energy? ! The value of the dataset
! Latest scenarios, including delayed participation scenarios and limited technology scenarios.
! Collected data on renewable energy at a more detailed level than in previous assessments.
There is a broad spread of mitigation levels in the dataset.
Imagine how much stronger the story is going to be when we have thousands of scenarios.
Although there is some correlation between renewable deployment and mitigation level, there is enormous uncertainty.
Primary energy is not all that linked to stabilization level.
Fossil energy is tightly correlated to fossil and industrial CO2 emissions.
Low-carbon energy is far less correlated, because of energy demand.
Looking only at renewable energy adds the uncertainty of competition between nuclear and CCS.
Maybe the deployment of renewable energy is just a function of whether nuclear and CCS are available.
But that is clearly not the whole story.
The increase in costs from limiting renewables is comparable to the increase from limiting other technologies.
The increase in costs from limiting renewables is comparable to the increase from limiting other technologies.
There is no correlation between carbon prices and renewable energy deployment levels
The growth rates for some technologies are quite large.
One of the main lessons is that much of the growth in renewable energy takes place in the developing world.
One of the main lessons is that much of the growth in renewable energy takes place in the developing world.
In the SRREN, a small set of scenarios were used to for a more detailed exploration.
In the SRREN, a small set of scenarios were used to for a more detailed exploration.
In the SRREN, a small set of scenarios were used to for a more detailed exploration.
Chapter 10 used these specific scenarios to construct cost curves.
Chapter 10 also produced mitigation by technology for these scenarios.
Chapter 10 also produced mitigation by technology for these scenarios.
There was a long discussion about using scenarios across chapters in the SRREN.
Energy-economic or integrated assessment models
Bottom-up assessments of technology cost and potential
The interaction should go both ways.
0
1
2
3
4
5
6
0 200 400 600 800Energy Production (EJ/yr)
Res
ourc
e G
rade
RESOURCE SUPPLY CURVE
Scenarios of scale, timing, location.
ENERGY SUPPLY SCENARIO
Supply curves and related aggregate representations
Vetting scenarios from aggregate models
These sorts of ranges were passed to the chapter with the goal that they would try to explain what might need to happen to get to the higher deployment levels.
2020 2030 2050
050
100
150
200
250
300
350
Global Biomass Primary Energy SupplyG
loba
l Bio
mas
s Pr
imar
y En
ergy
Sup
ply
[EJ/
yr] N=137
BaselinesCat. III + IV (440-600 ppm)Cat. I + II (<440 ppm)
CO2 Concentration Targets
2020 2030 2050
020
4060
8010
012
0
Global Wind Primary Energy Supply
Glo
bal W
ind
Prim
ary
Ener
gy S
uppl
y [E
J/yr
]
N=152
2020 2030 2050
050
100
150
Global Solar Primary Energy Supply
Glo
bal S
olar
Prim
ary
Ener
gy S
uppl
y [E
J/yr
]
N=156
2020 2030 2050
010
2030
40
Global Hydro Primary Energy Supply
Glo
bal H
ydro
Prim
ary
Ener
gy S
uppl
y [E
J/yr
]N=164
2020 2030 2050
010
2030
4050
60
Global Geothermal Primary Energy Supply
Glo
bal G
eoth
erm
al P
rimar
y En
ergy
Sup
ply
[EJ/
yr] N=122
2020 2030 2050
010
2030
4050
6070
Global Wind and Solar PV Electricity Share
Glo
bal W
ind
and
Sola
r PV
Elec
trici
ty S
hare
[%] N=149
Where are we headed with AR5?
! Lots and lots and lots and lots of scenarios ! AMPERE, EMF 24, RoSE, LINKS, AME
! Imagine how much more illuminating this graph will be with ten times more scenarios!!
Where are we headed with AR5?
! But there are some particularly important new dimensions of coordination of scenarios: ! Technology, particularly with limits on the deployment of particular
technologies, for example, nuclear energy and CCS. ! Policy, particlularly idealized policy scenarios, for example the
EMF 22 and EMF 24 scenarios. ! Linkage to RCPs. ! Some scenarios with coordinated socioeconomics.
! Remaining to be worked out ! How will scenarios be used within WGIII? ! How will the RCPs be used to link across working groups?
Questions