Science and Policy on Global Warming : Science and Policy on Global Warming Harvey S. H. Lam
(Working with Rob Socolow)
May 5th, 2006
What is the problem? : What is the problem? We burn a lot of fossil fuels for energy.
CO2 is emitted into the atmosphere.
CO2 is a major greenhouse gas.
The average global temperature has risen, and will rise some more.
Some projected consequences are bad.
Pre-industrial Revolution Historical Data : Pre-industrial Revolution Historical Data Global atmospheric CO2 content: C-∞~600 GtC (billion tons of carbon equivalent. 1 GtC~0.48 ppmv).
Global annual CO2 emission rate: E-∞~0 GtC/year.
Global average temperature: T-∞~13.5º C.
Situation at 2000 : Situation at 2000 C(2000)~750 GtC from 600 GtC
E(2000)~7 GtC/year from 0 GtC/year
T(2000)~14.5º C from 13.5º C
And ~3.5
Historical Data T(ºC) versus C(GtC) : Historical Data T(ºC) versus C(GtC) From 1850 to 1990.
Annual average temperature ºC (1.0 ºC=1.8 ºF).
Kansas City is, on the average, 5.8ºF warmer than Chicago.
Consequences of a Warmer World : Consequences of a Warmer World Some good ...
Some bad … (how bad?).
Order of magnitude of
T~2º Celsius when C~2C-∞=1200 GtC.
Possibility of undesirable instabilities:
Thermohaline circulation, melting of ice sheets, more intense hurricanes, …, etc.
What would happen if we do nothing? : What would happen if we do nothing? For the last half of the 20th century, about half of the CO2 emitted into the atmosphere stayed in the atmosphere.
World demand for energy is certain to increase. Thus E(t) would increase.
Thus C(t) would continue to increase.
Thus T(t) would increase…
What needs to be done? : What needs to be done? Set a “ceiling” for C(t) and pledge not to breach it. Call it Css.
Must reduce future global emission rate E(t) in order to honor the pledge.
Estimate how much it costs.
Continue to think of other competitive ways of solving the problem.
What questions should we be asking? : What questions should we be asking? What Css can the world tolerate?
Can C be held fixed at a chosen Css (“forever”) while E is held at some Ess>0?
How does Ess depend on Css?
At what annual pace should E drop from its current value to Ess? Or, how many years can we spread out this effort?
How much does it cost?
The role of the scientific research community : The role of the scientific research community To understand the physics, and to provide the best possible answers to questions coming from the policymaker’s community.
To provide critiques of proposed strategies and policies.
To provide assessments of the performance (and effectiveness) of current policies.
Scientific Issues in Global Carbon Cycle Modeling : Scientific Issues in Global Carbon Cycle Modeling Atmosphere ...
Biosphere and land usage ...
Clouds, rains, snow...
Shallow oceans ...
Deep oceans ...
Other sources and sinks ….
Lots of interesting physics, chemistry and …
The role of any global carbon cycle model : The role of any global carbon cycle model For any specified future emission trajectory, E(t):
Can march forward in time to provide its prediction of the resulting response of the future atmospheric CO2 trajectory: C(t).
Some researchers can do the inverse problem.
IPCC (Intergovernmental Panel on Climate Change) Validation of Simulations : IPCC (Intergovernmental Panel on Climate Change) Validation of Simulations Given historical E(t), must recover historical C(t).
Nothing much was happening before the industrial revolution.
Most activities happened recently.
IPCC 1995 Simulations (Validated models only) : IPCC 1995 Simulations (Validated models only) C E
What did we learn? : What did we learn? We assume the IPCC predictions in the past decade (no major changes) used good science.
When C -> Css, E must drop from its 2000 value of about 7 GtC/year to some much lower number Ess.
Empirical correlation: (dimensional analysis)
Css versus Ess Correlation of IPCC Stabilization Data : Css versus Ess Correlation of IPCC Stabilization Data Amazingly, this result is not widely known.
If we want Css=1200 GtC, we need Ess~2 GtC/year.
Unlike our historical data, none of the future emitted CO2 in the steady state is predicted to stay in the atmosphere.
A Simple Physical Model : A Simple Physical Model Two tanks are joined by a big pipe, connected to a HUGE tank with a tiny pipe.
Water added to the first tank is shared “quickly” with the second tank.
Steady state (small tanks): water in=water out to huge tank.
The PACE of E Reduction : The PACE of E Reduction Needs E reduction to get E(2006) to Ess.
A vigorously fast pace is disruptive.
A very slow pace will breach Css.
The slowest pace without breaching Css is called critical pace:
The Job of the Policymaker’s Community : The Job of the Policymaker’s Community To obtain the best advice from the science community.
To recommend a sensible and practically feasible policy to deal with the problem.
To help convince the general public that the policy is the least disruptive of all options.
To provide leadership in the implementation of the recommended policy.
Policymaker’s To-Do List : Policymaker’s To-Do List Decide on a value for Css---the “stabilized” C value at large time.
Determine the magnitude of the needed total E reduction to honor the chosen Css.
Determine the slowest pace of the E reduction (i.e. the annual amount to be done) so that C will never breach Css.
Estimate the cost.
The Punch Line (using mathematics) : The Punch Line (using mathematics) Define (using dimensional analysis):
This positive number can be computed at any time.
Using mathematics: The slowest E reduction pace is to do the whole E reduction job in years---with no breaching of Css.
The Critical Pace : The Critical Pace The guideline for the future average annual E reduction is:
This is the recommended average annual amount for the next years.
Faster or Slower? : Faster or Slower? A faster than critical pace will have a slower C(t) rise toward Css. But a faster pace is expected to be more disruptive.
A slower than critical pace will allow C(t) to breach Css.
Once Css is breached, the decay of C(t) back to Css is very, very slow.
The 2006 Numbers for Css=1200 GtC (doubling) : The 2006 Numbers for Css=1200 GtC (doubling) We have
and
Remember: dE/dt~+0.14 GtC/year2 now at 2006.
What does that mean? : What does that mean? If we want never to let C to exceed 1200 GtC, we have to reduce E(t) from 7 GtC/yr to 2 GtC/yr.
The critical pace spreads the needed E reduction job over 260 years.
The average annual replacement of existing coal-fired power plants by new emission free power plants is 0.75 Three Gorges Dam (per year) for the next 260 years.
What if we procrastinate? : What if we procrastinate? E(t) is expected to increase because of rising world energy demand.
The value of E(t)-Ess will increase.
C(t) will continue to increase by approximately E(t)/2 each year.
The value of decreases by about 3 each year. So, ….
Some Useful Numbers : Some Useful Numbers If nothing is done, for Css=1200 GtC can become negative within this century.
1 GtC of CO2 is emitted by coal-fired power plants generating 700 GW of useful energy (Pacala and Socolow)---the amount of electricity generated by 39 Three-Gorges-Dam; each costs 25 billion US dollars (total=975 billion).
Pacing CO2 Mitigation (Socolow and Lam, 2006, in preparation) : Pacing CO2 Mitigation (Socolow and Lam, 2006, in preparation) Procrastinate for 75 years ….before doing it with critical pace at 2081… E C t t
How drops : How drops t
How Was the Critical Pace Determined? : How Was the Critical Pace Determined? A credible global carbon cycle model is needed.
Instead of deriving another model, we exploit the published results of IPCC models of the past decade (1995-2006).
An emulator of IPCC simulations was obtained.
The emulator equation can be solved exactly.
A Global Carbon Cycle Emulator : A Global Carbon Cycle Emulator Elevation of 1st tank=C, 2nd tank=B, huge tank=C-∞. Historical data suggests B~C.
E(t) C B C-∞
How Good is the Emulator? : How Good is the Emulator? is a slowly increasing function of time, rising from 200 to 300 in about 400 years.
The emulator with ~300 years can reproduce the IPCC data to 20%.
Comparison with HILDA (HILDA data from Bryan Mignone, *06) : Comparison with HILDA (HILDA data from Bryan Mignone, *06) The simple emulator is respectable for 2-3 centuries. ( << 1)
For 10 centuries, replace C-∞ by H: =0.37 E(t) C(t) t
Useful Remarks : Useful Remarks The emulator equation is a first order linear ordinary differential equation. Its exact solution C(t) for arbitrary E(t) is easily obtained.
as defined can be deduced by dimensional analysis. It was “derived” using the emulator equation with mathematics.
The empirical relation between Ess and Css-C-∞ can also be deduced by dimensional analysis.
More Useful Remarks : More Useful Remarks The physics underlying is clear. It is related to the time scale of deep ocean turnover. No observational validation is credible unless it spans O( ) years.
C-∞ is expected to eventually rise from its pre-industrial revolution value of 600.
Ess is inversely proportional to .
Dimensional Analysis, Physics and Mathematics : Dimensional Analysis, Physics and Mathematics Dimensional analysis:
Physics:
Mathematics+dimensional analysis+….. :
; and
The Best Possible Css (2006) : The Best Possible Css (2006) Suppose we do the best we can, starting now in 2006.
Find Css from: Css
Critiques of Current Discourse : Critiques of Current Discourse No consensus on Css. Immediately “stopping” the rise of C(t) is not a viable option.
No public recognition of the magnitude of the E reduction job (50% reduction is not enough for Css=1200 GtC) or the meaning of .
No public appreciation of the relation between pace and breaching of Css.
Tools for Policymakers : Tools for Policymakers Magnitude of the E reduction job:
Current status of the GW world:
Guideline for E reduction pace:
The pace parameter P : The pace parameter P P is defined by:
Right now, P~ -7. Must drive P toward +1.
Whenever P<1, the future is more difficult.
A message from Socolow : A message from Socolow Attention in the policy community is shifting from the proper stabilization level, Css, to the best pace, P. The Framework Convention on Climate Change (Rio, 1992), which the U.S. did sign AND ratify, imagined that the next step would be a global discussion of Css, the level that would (in the Framework’s words), “prevent dangerous interference with the climate system.” For the past decade, Css was the focus, but in the last year or two this has changed.
Socolow’s message, continued : Socolow’s message, continued The scientists are saying, we won’t play this game. We can’t define “dangerous.” We prefer to say the level is “dangerous” already.
So, the discussion moves to pace. Let’s reduce emissions as quickly as we can sensibly, and the meaning of “sensibly” is becoming the focus.
Difficult policy questions : Difficult policy questions Is there a lower limit to E(t) (because of the special needs of the transportation sector)?
Is there an upper limit to annual E reduction effort?
With a credible cost estimate in hand, what other options are competitive?
How far ahead should we plan? lam@princeton.edu