A few weeks ago, my brother posted
on my Facebook page “Any thought oh you global warming people” with a link to a
Wall Street Journal opinion piece by Matt Ridley entitled “Dialing Back the
Alarm on Climate Change” (
http://online.wsj.com/article/SB10001424127887324549004579067532485712464.html?mod=iPhone).
I, rather flippantly replied “You can ask again when the report that the
reporter reportedly reports on is published.” The report in question was the
fifth assessment report by the Intergovernmental Panel on Climate Change (IPCC).
A draft report is now available online (http://www.ipcc.ch/report/ar5/wg1/#.UksXud0hG8A)
so now that I have some extra time on my hand due to the US government shutdown
I thought I would take a stab at replying to my brother’s question. At over
2200 pages, the report is a rather daunting document and I won’t pretend I have
read it all. (On the other hand, the Ridley didn’t claim he had read the whole
thing either.) In addition to the full report, there is a 36 page “Summary for
Policymakers”.
The main points Ridley makes are
that (1) the estimates of the amount of climate change in the new report are
less than in the previous report, (2) that in the next 70 years, the effects of
climate change may be beneficial, and (3) the estimates in the latest report
may still be too high.
What strikes me when comparing
Ridley’s article with the “Summary for Policymakers” is how different the
emphasis is in the “Summary for Policymakers” from what Ridley presents. Ridley
emphasizes the differences in the estimates of "equilibrium climate
sensitivity" and "transient climate response" between the most
recent report and the previous one. While these are technical measures of
climate change, they don’t really tell you that much without some context. Overall,
it seems to me that Ridley is cherry-picking; he is emphasizing minor
differences that don’t really mean much and ignoring the main message.
So what is the main message? Well
here are some of the most strongly highlighted messages from the “Summary for
Policymakers.” (I have only picked a few because I don’t have space for all of
them.)
Warming of the climate
system is unequivocal, and since the 1950s, many of the observed changes are
unprecedented over decades to millennia. The atmosphere and ocean have warmed,
the amounts of snow and ice have diminished, sea level has risen, and the
concentrations of greenhouse gases have increased.
Each of the last three
decades has been successively warmer at the Earth’s surface than any preceding
decade since 1850… In the Northern Hemisphere, 1983–2012 was likely the warmest
30-year period of the last 1400 years (medium confidence).
The atmospheric
concentrations of carbon dioxide (CO2), methane, and nitrous oxide
have increased to levels unprecedented in at least the last 800,000 years. CO2
concentrations have increased by 40% since pre-industrial times, primarily from
fossil fuel emissions and secondarily from net land use change emissions. The
ocean has absorbed about 30% of the emitted anthropogenic carbon dioxide,
causing ocean acidification.
Human influence has
been detected in warming of the atmosphere and the ocean, in changes in the
global water cycle, in reductions in snow and ice, in global mean sea level
rise, and in changes in some climate extremes… This evidence for human
influence has grown since AR4. It is extremely likely that human influence has
been the dominant cause of the observed warming since the mid-20th century.
Continued emissions of
greenhouse gases will cause further warming and changes in all components of
the climate system. Limiting climate change will require substantial and
sustained reductions of greenhouse gas emissions.
Changes in the global
water cycle in response to the warming over the 21st century will not be
uniform. The contrast in precipitation between wet and dry regions and between
wet and dry seasons will increase, although there may be regional exceptions.
Cumulative emissions
of CO2 largely determine global mean surface warming by the late 21st
century and beyond... Most aspects of climate change will persist for many
centuries even if emissions of CO2 are stopped. This represents a
substantial multi-century climate change commitment created by past, present
and future emissions of CO2.
Another difference I note is that
Ridley expects the climate change to be beneficial. He states that climate
change “would extend the range of farming further north, improve crop yields, [and]
slightly increase rainfall (especially in arid areas).” Warming would indeed
extend the range of farming further north but I am not clear where he gets the
idea that it would increase rainfall in arid areas. The Summary for
Policymakers states that “the contrast in precipitation between wet and dry
regions … will increase.” That directly contradicts Ridley. A lot of our
agriculture is in areas that don’t get all that much rainfall. If they get
less, they will be able to grow less. Thus the idea that a warmer climate would
mean more food is speculation on Ridley’s part. It is not something that is backed
up by science. It seems to me that a more likely scenario is that agricultural
productivity would increase in Canada and Russia due to warmer temperatures but
would decrease in the United States and China due to less rainfall. However,
this is speculation on my part. It is not backed up by any more science than
Ridley’s speculation. I am reminded of a quote from Mark Twain that ends “There
is something fascinating about science. One gets such wholesale returns of
conjecture out of such a trifling investment of fact.” (For the full quote, see
http://todayinsci.com/T/Twain_Mark/TwainMark-Quotations.htm.)
Anybody can speculate. Backing up those speculations with science is a lot more
difficult. That’s why you end up with reports that are over 2200 pages long.
Finally, Ridley emphasizes possible
benefits in the next 70 years but ignores what happens after that. In my
opinion this is cherry picking again but with the added fault that it is based
more on speculation than science.
Thus, I find Ridley’s article
unconvincing. To me, it doesn’t seem like an unbiased treatment of the
evidence. On the other hand, should I blindly accept the conclusions of a big
long report that I haven’t read and which I might not understand if I did read?
I think a lot of us are in that position these days not only about this issue
but about a lot of things. How do we judge whether something is true if we
don’t really understand it?
A few years ago, I was a juror in a
civil trial. One thing the judge told us in our instructions seems relevant
here. He told us that in making our decision we did not have to ignore our life
experiences. We could use our life experiences, for example, in judging the
credibility of the witnesses.
A key part of the issue of climate
change is the predictions that various models make regarding future climate.
Now while, I don’t know much about climate models in particular, I do know a
bit about models in general – their strengths, their weaknesses, what they are
good for, and how they can go wrong. I work with groundwater models all the
time and I think some of what I know about groundwater models is applicable to
models in general including climate models. Thus, while I do not have the background
to fully judge the validity of the climate models, I do have some knowledge
that is applicable and I can use that to make a judgment.
First of all, why use a model at
all. My colleague, Ken Belitz, had a good explanation for that. He said (and
I’m paraphrasing) “We model what we can not measure. If we could measure everything
we would not need to model.” If we knew, really knew, what the future climate
was going to be, we would not need climate models. We don’t know what the
future climate will be. However, we still need to make decisions now that
depend on what we expect for the future. We make such decisions about weather
all the time. Are we going to take an umbrella with us today? If we expect it
to rain we will. How do we judge whether it will rain or not? We look at a
weather report and see what it predicts. When I was growing up, there were lots
of jokes about weathermen and how bad their predictions were. It’s been a long
time since I’ve heard one of those. That’s because these days, weather
predictions are usually pretty good. If you check the weather report in the
morning and it says it is going to rain that day, then it is very likely that
it will, indeed, rain. Those weather reports are based on computer models and
today those computer models are pretty good. I don’t understand the details of
those weather models but their track record gives me confidence in them. So one
lesson I take from this is that computer models can be useful.
I think computer models can be
harmful too. We have a recent example of that too. One of the causes of the
recession that began in 2008 was the large numbers of credit default swaps that
had been sold. The companies, such as AIG, that sold them used computer models
to estimate the risk associated with them. Those models were flawed. As I
understand it, the basic fault was that they assumed that the risk of someone
defaulting on an obligation was independent of the risk of someone else
defaulting. For example, if the owner of the house next door defaults on their
mortgage, that doesn’t mean I will default on my mortgage too. (This is almost
certainly a simplification of the actual situation.) Unfortunately, for them
(and us) that wasn’t correct. For example, if a big company closes a big
factory in a small town, a lot of people in the town will be out of a job and a
lot of them might default on their mortgages. Basically two of the seven deadly
sins were big players here: greed and pride. The people who created and used
the financial models were too confident about how well they understood the
financial system and in order to get as much money as they could, the took
unwarranted risks.
So what about climate models?
Personally, I don’t think they are nearly as good as the weather models but I
doubt they are as bad as the financial models turned out to be. Climate models
don’t have the sort of track record that weather models have. Every day we get
a new weather report and we can compare that report to the weather that
actually occurs. We can’t do that with climate models. Climate models make
predictions over much longer time scales and the predictions they make are
harder to compare to reality.
First of all, it is hard to measure
the climate. We can’t go outside and measure the average temperature. We can
only measure the temperature at that time we make the measurement. We have to
make lots of measurements to get an average temperature.
Second, the long time scales
involved in climate models means that we haven’t had the opportunity to compare
the models to reality to the same degree that we have with weather models. That
means if there are important errors in the models, we might not yet have had a
chance to see them.
Twenty years ago, I was of the
opinion that global warming was likely but unproven. (I also thought that global
warming was sufficiently likely and serious that we should try to do something
about it.) The reason I thought global warming was unproven was because of a
conflict in the data. The ground-based measurements indicated that the
temperatures were increasing but the satellite data showed no such trend. It
wasn’t clear which was correct. The ground-based data might have been affected
by local effects. One hundred years ago, a weather station that is now in an
urban area might have been in a rural area because a city grew up around it. It
is well known that cities are heat islands. They have warmer temperatures that
the surrounding rural areas because people heat up their houses in winter and
they build black roads that absorb heat in summer and probably because of other
reasons that I don’t know about. Such effects could have introduced a bias into
the ground-based data. You could try to eliminate it by only using readings
from weather stations that were not affected in this way but you would have to
be careful in how you did it.
It turned out that it was the
satellite data that were wrong or at least wrongly interpreted.
Measuring temperatures of the earth
from space is a bit more complicated than measuring it on the ground. You don’t
measure temperature directly. Instead you measure something else like infrared
light and convert that to a temperature reading. One of the things that wasn’t
adequately understood at first was that the distance from the earth to the
satellite affects the temperature reading. That wouldn’t matter much if the satellite’s
orbit never changed. Unfortunately, the orbits do change. Friction with the
upper parts of the earth’s atmosphere causes the orbits to decay and
eventually, the satellite crashes to the ground, burns up in the atmosphere, or
is boosted into a higher orbit. Because the satellites were getting closer and
closer to the Earth, their temperature measurements got progressively worse.
Once this effect was understood, it was possible to correct for it. When the
corrections were made, the ground-based and satellite-based temperatures both
showed a warming trend.
The problem with the satellite
temperature measurements was a problem with the computer model used to convert
what the satellite actually measured to a temperature reading. The model did
not include an important process (change in the satellite orbit) in its
calculations and hence gave an incorrect result.
With groundwater models, there have
been similar problems. In the textbook, Applied Groundwater Model, by Mary P.
Anderson and William W. Woessner, the authors discuss a study in which the
predictions made by four groundwater models were compared with what actually
happened several years later. In all four cases, the model predictions were
wrong. The reasons why they were wrong varied and I don’t remember exactly what
they were. In at least one case, the model had failed to include an important
process. If I remember correctly, another one was wrong because they had to
guess how much people would attempt to pump out of the aquifer and that
estimate was wrong. In any case, one lesson from that study was that with
predictions over long time scales there is a lot more opportunity for the
predictions to go awry and one of the ways they can go wrong is by failing to
include an important process. Another lesson is that some things that are
important inputs for the model (the pumping rate) may themselves need to be
predicted and that predicting them may be difficult or impossible. In the
context of climate change, the timing and magnitude of volcanic eruptions is a
similar sort of input. By injecting sulfates and other chemicals into the
atmosphere, volcanos can affect climate but nobody can predict a volcanic
eruption long enough in advance to be useful for climate modeling. The climate
modelers are certainly aware of the possibility of errors due to important
processes being misrepresented because one section of the IPCC report
specifically addresses the known limitations of the models in representing
certain processes (page 9-60).
While the possibility of errors in
the models may mean that the predictions of the models are imprecise or even
wrong, that is not a valid reason to believe that climate change is not
occurring. Climate change is not a model prediction; it is an observation. What
the models are good for is for giving us some understanding, however imperfect,
of why climate change is happening and the direction of climate change.
If the climate models are
imperfect, does that mean we don’t know enough to make decisions that involve
climate change now? I believe we can make such decisions based on things that
we do know.
1. We
know that elevated levels of carbon dioxide can lead to warmer temperatures. We
know that, for among other reasons, because that’s why Venus has such a high
surface temperature. Venus is hotter than Mercury even though Mercury is closer
to the sun.
2. We
know that the levels of carbon dioxide in the Earth’s atmosphere are
increasing. We know that because we have measured it.
3. From
geologic measurements, such as measurements of carbon dioxide from Antarctic
ice cores, we know that the levels of carbon dioxide in the Earth’s atmosphere
are higher now than they have been for a very long time. According to the
Summary for Policymakers, it has been at least 800,000 since carbon dioxide
levels were this high. For context, the earliest fossil record of modern humans
is from about 200,000 years ago.
4. From
the fossil record, we know that during most of Earth’s history, glaciers
covering entire continents (such as Antarctica) have been uncommon and that
consequently sea levels have been higher than they are now.
5. We
know that glaciers are glaciers have retreated worldwide because we have
measured the retreat.
6. We
have a good estimate of the rate at which ice is being lost because we can
measure the change using satellite data.
7. We
have measured the rate at which sea level is rising and most of that rise can
be explained by glacial melting.
From the above, we can make a
decision about whether or not we can expect sea level to rise in the future. We
can expect it to rise in the future.
If our infrastructure, such as subways in New York City, is vulnerable to such
a sea level rise, we need to decide whether we should do something to protect
that infrastructure. It is important to note that our infrastructure doesn’t
need to be below sea level to be vulnerable to sea level rise. To be
vulnerable, a structure just needs to be below the highest likely storm surge.
When sea level rises, the elevation of that storm surge rises too.
1. We
know that in many regions, the timing of snowmelt is important for water
supply. This is particularly true in the western United States.
2. With
rising temperatures, we can expect the timing of snowmelt to change.
From the above, we can make a
decision about whether we need to plan for a change in the water supply in
regions where snowmelt is important for water supply. We do need to make such plans.
The biggest thing we need to make a
decision about is whether or not we need to reduce or limit carbon dioxide
emissions from the burning of fossil fuels. There are big risks to it. If we
choose to limit carbon dioxide emissions, we risk wasting a lot of money, time,
and effort on something that we may not need. If choose not to limit carbon
dioxide emissions, we risk changing the climate in ways that we don’t understand
and with effects that we do not anticipate. Furthermore, no matter what
decision we make, we will probably never know the full outcome of our decision.
(Our descendants might know but we probably will not.) I think we have enough
information to make a decision about this.
1. We
know that the climate has been changing in the sorts of ways that the climate
models predict.
2. We
understand why the climate is changing though we may not be able to predict the
extent or rate of change as well as we would like.
3. Carbon
dioxide in the atmosphere has been increasing and has increased to levels
unprecedented in human existence.
4. The
increase in carbon dioxide in the atmosphere is a major cause of climate
change.
5. The
increase in carbon dioxide in the atmosphere is due to the burning of fossil
fuels.
6. We
can not easily reverse climate change.
In my opinion, the prudent thing to
do would be to limit carbon dioxide emissions because the consequences of
failing to do so are too severe.