Heartland Policy Study
Climate Change 95: An Appraisal
by Vincent Gray, M.A., Ph.D.
September 10, 1997
Vincent Gray lives in Wellington, New
Zealand. He holds a Ph.D. in physical chemistry from Cambridge
University, UK, and has worked in industrial, government, and
university research laboratories in UK, France, Canada, New Zealand,
and China in the fields of colloid science, petroleum, plastics, coal,
timber, adhesives, building materials, and forensic science. He has
published over 100 scientific papers. He has recently specialized in
the study of the greenhouse effect and is currently a peer reviewer
for the Intergovernmental Panel on Climate Change. This paper was
first published in the New Zealand Science Review 53(4) (1996) and is
being distributed in the U.S. by The Heartland Institute with
permission of the editor and the author.
It has been seven years since the
Intergovernmental Panel on Climate Change (IPCC) published its first
scientific report: Climate Change: The IPCC Scientific Assessment (henceforth
Climate Change 90). That report was the basis for the Rio de
Janeiro "Earth Summit" of 1992 and the resulting Framework
Convention on Climate Change, by which the nations of the world
committed themselves to reductions in greenhouse gas emissions. Two
supplementary reports, Climate Change 1992 and Climate
Change 1994, updated some aspects of the earlier report--but not
until the release of Climate Change 1995: The Science of Climate
Change (henceforth Climate Change 95) had there been a
full revision of that initial influential report.
The new report is hardly easy to digest.
There are 572 pages and no index. Although the chapter headings are some
guide, it is difficult to find any particular subject, especially if it
is addressed in several chapters. The report has 11 chapters and 7
appendices; 80 lead authors, and 376 contributors (with some
duplication). Drafts were circulated to at least 540 scientists in 40
countries, including 142 from the United States and 72 from Australia.
Summaries of the report are needed, and
there is a nestled hierarchy of these. Some paragraphs have individual
summaries, as does each chapter. There is also a Technical Summary.
Finally, there is a Summary for Policymakers: the only part of
the report that was approved line-by-line at the November 1995 meeting
of Working Group I in Madrid.
A foreword to Climate Change 95
requests that each citation to the report include references to the lead
authors of the chapter concerned. This study, regretfully, has not found
it possible to abide by that request without greatly increasing the
length and complexity of the references. Chapter 2 of Climate Change
95, for example, has 27 lead authors and 49 contributors. For
simplicity, citations in this study will be to Climate Change 90,
Climate Change 92, Climate Change 94, and Climate Change 95.
Organization of the Study
Part 1 of this Heartland Policy Study
evaluates the latest IPCC report on the basis of two "hallmarks of
sound science": how the report addresses the issue of consensus,
and how precise are its terms and data. Though Climate Change 95
is full of scientific information, it presents this information in a
disturbingly imprecise and sloppy form.
Part 2 considers the report's greenhouse gas
science. Estimates of emissions and concentrations of carbon dioxide,
methane, and nitrous oxide all are slanted to support exaggerated
climate trends. Several graphs are included in this section to document
how Climate Change 95 diverges from real science on these
issues.
Past temperature trends, and the IPCC's
predictions of future trends, are discussed in Part 3. I explain how the
biases reported in Part 2 are compounded to give scenarios that exceed
current, or even possible, rates of climate change and economic change.
It is clear that climate models--though made more plausible in Climate
Change 95 by having been modified for the effects of sulphate
aerosols--have not yet been sufficiently validated against current and
past climates to be reliable for future projection.
A summary and concluding remarks aimed
directly at policy makers constitute Part 4. I conclude that the new
IPCC report fails to find evidence of significant or harmful climate
change resulting from human activities, but merely "suggests"
that human activities have a "discernible" effect. The
evidence for this pronouncement is unconvincing. When bias is
discounted, Climate Change 95 provides no evidence to support
drastic or economically damaging measures to control greenhouse gas
emissions. If current climate and economic trends continue, global
warming over the next century is unlikely to be above 1°C.
Part 1
Hallmarks of Sound Science
Consensus
Climate Change 95 shows evidence of
at least a partial retreat from consensus: a term that belongs
in the realm of politics, not science. Science seeks to discover truth,
or at least the closest approximation to it given current knowledge and
understanding. Science progresses by discussion and argument, not by
seeking unanimity. Many scientific arguments in the past have been won
by small minorities.
In Climate Change 90 it was stated
(page 353) that some scientists "formed a minority opinion which
could not be reconciled with the larger consensus." That
"minority opinion" was not permitted to be expressed in the
1990 report. In Climate Change 95, by contrast, there is no
mention of "consensus," and some critics are given a fair
hearing.
I, for example, submitted thirty pages of
amendments and comments to the first draft of Climate Change 95.
Many of my thoughts were incorporated into the final document. For
example, I pointed out that no climate model has been
"validated," in the sense that it has given a satisfactory
representation of climate and its changes over an extended period.
Although both Climate Change 90 and Climate Change 92
devoted chapters to the validation of climate models, the IPCC changed
the word "validation" to "evaluation" no less than
fifty times from the first draft of Climate Change 95, thus
conceding my point. However, several other comments were not addressed,
and thus they are raised again in this report.
Sloppy Data
Those who are accustomed to seeing
scientific data presented using measures of precision--such as mean,
standard deviation, correlation coefficient, confidence interval, or
level of significance--will be disappointed with Climate Change
95. Instead, we have average, estimate (including best
estimate), range, guess (including best guess), uncertainty,
several, or so, consistent (including broadly
consistent), tend towards, could be largely due to, suggest, is
discernible, essentially the same as, not uniform, of the same magnitude
as, comparable to, essentially unchanged, about, ~, and many other
imprecise and unscientific terms.
As an example of this sloppiness, Table 6.1
on page 298 of Climate Change 95 lists the results from 16
current computer models of the climate (Transient coupled
atmosphere-ocean general circulation models). The characteristic
calculation of these models is climate sensitivity: the
equilibrium temperature rise at the tropopause from a doubling of carbon
dioxide in the atmosphere. The range found for this quantity in the
listed models is 2.1-4.6°C, with a mean of 3.2°C. Yet the IPCC
persistently claims that the range is 1.5-4.5°C, and that the Best
Estimate is 2.5°C. There is no current model that calculates a
climate sensitivity below 2.1°C. Even this figure is suspect, as it is
found by only one model, and the second lowest figure is 2.5°C. How a Best
Estimate of 2.5°C could have been decided defies rational
explanation. The decision has had important consequences, however, as
this Best Estimate is used to indicate the most likely of the
future projections made by the IPCC.
Then there is the sloppiness over dates.
Levels of greenhouse gases are said to be measured from
"pre-industrial times" to "the present day."
"Pre-industrial times" appear to start in 1765 in most of the
graphs. But on page 3 of Climate Change 95 they are "about
1750"; on page 69 they become "the late 18th
century"; and, on page 76, "1800." Figure 3 on page 19 of
Climate Change 94--a diagram of radiative forcing (the change
in mean radiation at the earth's surface) due to different climatic
influences--applies from "1850 to the present day." The
identical diagram in that same report, Figure 4.8 on page 195, applies
to the period "from 1850 to 1990." Slightly modified, the
diagram appears in Climate Change 95 as Figure 2 (page 16),
applying to "pre-industrial times to the present (1992)," . .
. except for the change in solar output, which applies from "1850
to the present." The same diagram appears on page 117 as Figure
2.16, where it now applies "from pre-industrial times to the
present day," with the change in solar output applying from
"1850 to the present day."
Much is made of the claim that the
calculated figure for radiative forcing of the main greenhouse
gases--2.45Wm-2 (watts per square meter) from
"pre-industrial times" to the "present day"--is
"unchanged" from that calculated in 1990, but little regard
seems to be paid to the interpretation of "the present day."
Does it mean 1990, 1992, 1994, or 1995? Are we to assume that there was
no change in radiative forcing of the main greenhouse gases between 1990
and 1995? Although supposedly "unchanged" at 2.45Wm-2,
a different figure, 2.62Wm-2, is used for the radiative
forcing of these gases since "pre-industrial times" to
calculate projections of future climate for the IS92 emissions scenarios
on page 320.
This lack of precision in presenting the
climate data and the results of calculations makes it easier to
exaggerate the extent of climate change, and to claim unjustified
agreement of the climate data with the models.
Part 2
Greenhouse Gas Science
Carbon Dioxide
There is no paragraph, let alone a chapter,
in Climate Change 95 addressing the supposed cause of the
greenhouse effect: the emission of carbon dioxide by the combustion of
fossil fuels. One unattributed graph showing emissions from 1850 to 1990
appears as part of an insert to Figure 1 (page 16). Added to the end of
paragraph 2.1.3 (page 83) is my suggested mention that the emission
figures for 1991 and 1992 are essentially the same, at 6.1GtC/yr, as
those for 1990 . . . although the chapter's authors then added the
caveat, "although the slow-down may be temporary."
A footnote to Table 2 on page 17 states that
emissions in 1994 were 6.1GtC/yr, the same figure as 1989-1992. This is
untrue, as the latest figures from the Carbon Dioxide Information and
Analysis Center at Oak Ridge, Tennessee (presented in Figure 1 below)
give 6.2GtC as the emission for 1994, and 6.06GtC for 1993. Figure 1
indicates that carbon dioxide emissions have increased by only 0.132Gt
(0.026GtC/yr) between 1989 and 1994, a full five years. A recent
estimate of 6.25GtC for 1996, if correct, would show that this low rate
of growth is being maintained, giving a probable projected figure for
2000 of 6.35GtC. The objectives of the Framework Convention on Climate
Change to stabilize emissions at 1990 values by the year 2000 have thus
almost been achieved--a fact that has received no mention in the media.
Yet assumptions of a continuous increase from 1990 persist, and IPCC
emissions scenario IS92a (considered to be "central") assumes
emissions will be 7.0GtC/Yr in the year 2000.
The Carbon Dioxide Information
Analysis Center, funded by the U.S. Department of Energy, was
established in 1982 at the Oak Ridge National Laboratory, Tennessee. The
Center collects, collates, and publishes climate data from around the
world, covering not only carbon dioxide but also other trace gases and
aerosols, temperature, and precipitation. This research organization,
and the data it has collected, has been all but ignored by the IPCC. It
is not even mentioned in Climate Change 90 or Climate
Change 92; its publication Trends 91 (but not the latest, Trends
93) has a single reference in Climate Change 94 and Climate
Change 95.
The concentration of carbon dioxide in the
atmosphere, its variability, and its rate of increase are fundamental to
an understanding of the greenhouse effect. Trends 93 lists 69
sets of figures for monthly mean concentrations of atmospheric carbon
dioxide from measurement stations around the world. Yet the IPCC insists
on selecting only two of these, at Mauna Loa (Hawaii) and the South
Pole, justified by the unsupported and typically vague statement (on
page 78, Climate Change 95): "Data from the Mauna Loa
station are close to, but not the same as, the global mean." Then
why not use the global mean figures, as supplied by the U.S. National
Oceanic and Atmospheric Administration Climate Monitoring and
Diagnostics Laboratory (NOAA/CMDL), whose latest results (from their
Internet site) are presented below in Figure 2.
The global average CO2
concentration increased from about 340ppm in 1981 to 358ppm in 1994, an
increase of about 5.3 percent in 14 years. It is evident from Figure 2
that the atmospheric carbon dioxide concentration fluctuates
considerably from hour to hour, month to month, year to year, and decade
to decade. The year-to-year variation in global CO2
concentration has been as little as 0.5ppm and as much as 2.5ppm. This
variation leads the IPCC to give a variety of contradictory figures when
characterizing the "recent" or "current" rate of
change. Consider the following statements, all of which appear in IPCC
documents:
-
"The high growth rate of the late
1980s, the low growth rate of the early 1990s, and the recent upturn
in the growth rate are all apparent" (Climate Change 95,
caption to Figure 2.2, page 81);
-
The rate of increase from
1990-2025, for emissions scenarios IS92a, b, d, e, and f, appears to
be 1.8ppmv/yr (0.5 percent/yr) (Climate Change 95, Figure
6.19 (a), page 321). The same figure, 1.8ppmv per year from
1990-2025, is the basis for the "stabilization scenarios"
(Climate Change 95, Figure 2.5, page 84).
As the lower portion of Figure 2 clearly
shows, the annual variation in CO2 concentration (excluding
seasonal variations) has never been as high as 1.8ppm for more than two
consecutive years, and in most years was below 1.5ppm. The average
annual change during the 1980s was 1.1ppm, or about 0.4 percent per
year. The average annual change from 1990 to 1994 was 0.08ppm, or about
0.3 percent per year.
It seems evident that the recent
"upturn" is a fluctuation similar to that of 1982-83, rather
than being one likely to provide "growth rates currently comparable
to those averaged during the 1980s," as the IPCC authors claim in Climate
Change 95 (page 3). The choice by the IPCC of "the 1980s"
for their carbon budget is an unfairly high sample of the whole curve.
The feeble excuse given for ignoring trends since 1990 is the absence of
recent estimates for tropical forest exchange (Climate Change 95,
page 78).
Nearly all of the computer climate models
assume that starting in the year 1990, carbon dioxide in the atmosphere
will increase by 1 percent per year, compounded. This is 2½ times the
measured rate of carbon dioxide increases in the 1980s and 1990s (0.4
percent). However, it is stated in Climate Change 90 that the 1
percent figure refers to equivalent carbon dioxide--that is,
the rate of increase of all the greenhouse gases, treated as if they
behaved like carbon dioxide.
The rate of increase of equivalent
carbon dioxide between 1980 and 1990, calculated from the radiative
forcing figures in Climate Change 90, was 0.89 percent per
year, linear. But if Climate Change 92 figures are used, the
average was 0.74 percent per year from 1980 to 1992. When sulphate
aerosols are included (calculated from the radiative forcing in Figure
6(a), page 24 of Climate Change 95 and my Figure 1 above), it
works out at 0.46 percent per year. For the years 1990-95, the rate of
change of equivalent carbon dioxide fell to 0.33 percent per year,
assuming constant aerosols. IS92a, b, d, e, and f scenarios assume that
carbon dioxide in the atmosphere will increase from 1990-2025 at a
linear rate of 0.5 percent a year, whereas the actual rate from 1980 to
1990 was 0.4 percent a year, and from 1990 to 1995 below this. The IPCC
report authors also assume an "effective" carbon dioxide
increase of 0.85 percent a year between 1990 and 2025, when the actual
"effective" rate between 1980 and 1992 was 0.74 percent a year
(according to figures reported in 1994), and it is falling.
The exaggerated rates of change in carbon
dioxide and "equivalent" carbon dioxide assumed by the models
and scenarios have important consequences. It seems to be universally
believed by the media that carbon dioxide in the atmosphere will double
"by the middle of the next century" (1 percent a year
compounded gives 70 years). If only carbon dioxide were considered, the
1980s' rate of 0.4 percent would double in 174 years, if compounded, or
250 years if linear. If the average for the past decade of 1.1ppmv/yr
persists, it will take 226 years for the concentration to double, if
compounded, and 325 years if linear.
Methane and Nitrous Oxide
The changes in atmospheric methane and
nitrous oxide concentrations are similarly overestimated by the IPCC.
Figure 2.9 on page 87 of Climate Change
95 shows trends in methane concentration, both global and for one
site (Mould Bay), as well as corresponding changes in the rate of
increase, from 1983 to 1994. Figure 3 below shows an updated version of
this graph (from NOAA/CMDL). It is apparent that despite fairly large
fluctuations, there has been a downward trend in the growth rate of
atmospheric methane over the period, indicating an average growth rate
for 1995 of about 5ppbv/yr. Yet the IPCC states (Climate Change 95,
page 87) that "in 1994 global methane growth rates recovered to
about 8ppbv/yr, close to the range of rates observed throughout the
period 1984 to 1991 (13-9 ppbv/yr)." This statement is clearly
refuted by Figure 3 and ignores the obviously falling growth rate.
The IPCC authors show the 1994 growth rate (Climate
Change 95, Table 1, page 15) as 10ppbv/yr, as averaged over the
decade beginning 1984, again ignoring the decline. Moreover, they assume
in the IS92a scenario methane growth rates from the year 1990 of
9.8ppbv/yr for 1990-1995 (in defiance of the observations) and 12.2 ppbv/yr
for 1995-2000. Similarly inflated figures for methane concentrations are
assumed in the other IS92 scenarios. Figure 3 above shows that the
"upturn" is turning down again, and that the decline in global
growth rate of methane in the atmosphere continues, such that it may
become zero by the year 2000.
Similarly inaccurate interpretations of
nitrous oxide growth rates are used by the IPCC in its various
scenarios. The rate of growth of nitrous oxide in 1993 was 0.5ppbv/yr,
yet the figure assumed by all IS92 scenarios for 1990-1995 is
0.8ppbv/yr.
The following footnote appears on page 319
of Climate Change 95:
For both N2O and CH4,
1990 emissions specified in the IS92 scenarios do not accord with
emission estimates based on observed concentration data and current
lifetime estimates. . . . We therefore use these latter values to
ensure a balanced contemporary budget, and assume that the scenario
values are a valid representation of the changes from 1990, rather
than correctly specifying absolute emissions.
In other words, they admit distorting the
data!
Part 3
The Temperature Record
Annual average global surface temperature
anomalies constitute the principal evidence for global warming over the
past century. The latest version--global figures from 1851-1996, and
hemispheric figures for 1856-1996--are shown in Figure 4 on the
following page. This compilation is the only one that provides a
combined surface and ocean record.
Though there is no doubt as to the excellent
work done by the team at the University of East Anglia, there must be a
question as to what extent an alternative team might exactly match their
results. The exercise involves a great deal of judgment in the choice or
rejection of records, and in their "correction," however
assisted by statistical guidelines. There is particular difficulty with
marine records, as much supporting information in the older
records--such as type of bucket collector, size of ship, or wind speed
on deck--simply has to be guessed. It is admitted that, in doubtful
cases, there is a tendency to match the marine records with the land
records. The baseline for these results has recently been altered, from
1951-1980 to 1961-1990. The reason given is to improve accuracy for the
most recent years--but this must surely reduce accuracy for the earlier
years. The most recent figure for 1996 is 0.23°C above the 1961-1990
average. That figure is the same as it was in 1944, 1983, and 1987,
indicating a leveling off of global temperature since 1990.
Climate Change 95 says of this record (page
138):
Global surface temperatures have increased
by about 0.3 to 0.6°C since the late 19th century, and by about 0.2
to 0.3°C over the last 40 years (the period with the most credible
data).
On page 4, Climate Change 95
reports: "Recent years have been among the warmest since
1860." The IPCC thus isolates the only features of the global
temperature record that suggest a possible agreement with greenhouse
theory, but ignores those features that cast doubt on the theory, such
as:
-
the fact that 1851-1919, together with
the earlier 1800s, was the coldest period in the proxy past record (Climate
Change 95, Figure 3.20, page 175) since the late 17th
century, thus exaggerating the significance of an increase over the
past century;
-
the fact that the greatest temperature
rise of this century (about 0.65°C) took place between 1910 and
1945. Atmospheric carbon dioxide rose only 3 percent over this
period. Climate Change 90 (page 233) said, "The rather
rapid changes in global temperature seen around 1920-1940 are very
likely to have had a mainly natural origin";
-
the fact that the period from 1945 to
1978, 33 years, saw a fall in temperature of 0.2°C, at a time when
atmospheric carbon dioxide rose by 9 percent;
-
the fact that the rise "over the
last forty years" of 0.3°C took place (as a rise of about 0.4°C)
over the short period of 1978-1996, only 18 years. The temperature
since 1990 has shown signs of leveling off, even if it was "the
warmest since 1860."
-
the fact that there is no consistent
difference between the hemispheres. But recently, the Southern
Hemisphere has been cooler than the Northern Hemisphere, in conflict
with the theory of the influence of sulphate aerosols.
None of those features can be explained by
the greenhouse theory.
Climate Change 95 admits that the
surface temperature record is subject to error: from poor sampling
(particularly over the oceans), from instrumental and operational
differences, and from urban development around land measurement
stations. The earlier measurements had poor land coverage and very poor
ocean coverage. The later land-based figures may have been
insufficiently corrected for urban warming. There should, therefore, be
high regard given to radiosonde measurements, which have been made since
1958, and the NASA satellite measurements (using Microwave Sounder Units
(MSU)) of mean lower troposphere temperature, which have been operating
since 1978. As can be seen in Figure 3.7 on page 148 of Climate
Change 95 and the updated version of the MSU measurements (Figure 5
below), there appears to be no evidence of a global temperature trend
over the past 37 years if the radiosonde measurements are considered, or
over 18 years if the satellite measurements only are considered.
The radiosonde measurements are dismissed by
Climate Change 95 (page 147). It reports, "Although the
radiosonde coverage was adequate from 1958 in the Northern Hemisphere,
it has only been adequate in the Southern Hemisphere since 1964. 'Global
values . . . before 1964 are therefore somewhat suspect because of the
reduced geographical coverage." But the same can be said of the
surface data before 1940, which the IPCC does not discount.
The IPCC admits (also on page 147) that
"The global MSU tropospheric trend from 1979 to May 1995 was -0.06°C
per decade, and that for the seasonal radiosonde data for the same
period was -0.07°C per decade." The IPCC attempts to explain away
this contradiction of its conclusions, however, by noting that "if
the transient effects of volcanoes and the El Niño Southern Oscillation
are removed from the various time series, positive trends become evident
(e.g. 0.09°C per decade for MSU) in closer accord with surface
data." They give as support for this view a reference to a paper by
Christy and McNider, Nature, January 1994, which was based on
data only to December 1993. More recent data (to January 1997) do not
confirm "an upward trend" to the MSU data, however
"corrected."
The Long-Term Trend
According to the IPCC, any climate change,
however short, that agrees with greenhouse theory is a long-term
trend. Any climate change, however extensive, that does not agree
with greenhouse theory is either ignored (like the global drop in
temperature between 1940 and 1978); too short to be representative; or
an anomaly that has to be explained.
Thus, for all calculations of the carbon
cycle, the period 1980-1989 is chosen by the IPCC, as it gives the
highest rates of increase in atmospheric carbon dioxide. (See Figure 2
on page 7 above.) The period after 1989 is an anomaly, since
the rate of increase has fallen. The fluctuation upwards in only two
years (1994-1995) is a recent upturn, even when it is not
sustained.
The emissions scenarios similarly assume a
rise in atmospheric methane concentration from 1990 despite the observed
fall since 1990. (See Figure 3 on page 9 above.) The low
fluctuation in 1992 was, of course, an anomaly. The subsequent
fluctuation upwards in 1993 is a return. The upward
fluctuations of carbon dioxide, methane, and nitrous oxide over only two
years enable the confident remark (Climate Change 95, page
3)--now proved to be wrong--that "recent data indicate that the
growth rates are currently comparable with those averaged over the
1980s."
The satellite record of lower troposphere
temperature, which shows no temperature trend over 18 years, is not long
enough to establish a trend. The record of stratospheric
temperature made by the same satellites for exactly the same period is
long enough, however, because it shows a decrease in temperature,
as predicted by greenhouse theory. When the results from the
stratosphere measurements are plotted, however, the only significant
temperature decline was for 1993-1996, a period of only three years.
(See Figure 6 below.)
Forecasting the Future
World Dynamics (1971) and Limits
to Growth (1972), sponsored by the Club of Rome, predicted that the
world would run out of "resources" by 1990, and there would be
a population decline by 2020. Those predictions were based on computer
models that attempted to quantify such notions as "resources,"
"quality of life," and "pollution."
The IPCC, which has relied heavily on
computer models of the climate, has given up the task of attempting
computer models of the world economy, but resorts instead to
"educated guesses" of future economic and population growth,
and of energy usage.
The IPCC emissions scenarios--IS92a, b, c,
d, e, and f--were first described in Climate Change 92, and in
a supplementary document. They comprise a range of assumptions on
greenhouse gas emissions, energy usage, and economic and population
growth, between 1990 and 2100. The IPCC has consistently pointed out
that these scenarios are not predictions, but rather projections,
based on specific assumptions. They claim that they do not favor any
particular scenario, but in practice they tend to emphasize IS92a, which
is referred to as the "central" scenario, a modification of
scenario SA90 of Climate Change 90, then referred to as
"Business as Usual." Climate Change 95 has modified
all of the scenarios to take into account assumptions of the effects of
sulphate-based aerosols. As there are two different assumptions for
aerosols (constant aerosols and increasing aerosols), we now have two
different possibilities for each scenario, both different from the same
scenarios as described in Climate Change 92.
Most of the assumptions of the scenarios
seem reasonable until they are examined in detail. Each assumption
usually exaggerates slightly; when they are combined, they give
scenarios that are unrealistically exaggerated.
The upwards bias in the projected future
emissions and atmospheric concentrations of carbon dioxide, methane, and
nitrous oxide, as discussed in Part 2 above, is built into all
scenarios. In order to justify the assumed increases in carbon dioxide
it is necessary to postulate large increases in world coal production.
IS92a, the "central" scenario, assumes a sevenfold
increase between 1990 and 2100, and IS92e assumes an elevenfold
increase over the same period. Yet between 1990 and 1995 world coal
production fell by 4½ percent.
These scenarios appear to assume that oil
supplies will be depleted by mid-century, and that 1930s' technology
will be used to make oil from coal on a colossal scale.
Population trends persist over a long
period. Recent figures are falling below the World Bank's "medium
population growth" assumed for scenarios IS92a, b, and e, and
population growth could descend to the "UN Medium-Low" case
assumed by IS92c and d. The UN Medium-High Case assumed by IS92f would
appear to be an impossible extrapolation of current trends, so the whole
scenario should be rejected.
World economic growth of 2.9 percent for the
period 1990-2025, and 2.3 percent for the period 1990-2100, is assumed
for IS92a and b; rates of 2.0 percent for 1990-2025 and 1.2 percent for
1990-2100 are assumed for IS92c; 2.7 percent for 1990-2025 and 2.0
percent for 1990-2100 are assumed for IS92d; 3.5 percent for 1990-2025
and 3.0 percent for 1990-2100 are assumed for IS92e; and 2.9 percent for
1990-2025 and 2.3 percent for 1990-2100 are assumed for IS92f.
World economic growth has been well below
all of these figures for the period 1990-1995, and they recovered only
slightly in 1996. Individual countries may achieve growth rates over
several years of as much as 12 percent, but the world average seems very
unlikely to reach the 3.5 percent over 35 years assumed by IS92e. That
scenario, therefore, should also be rejected.
Having rejected both IS92e and IS92f, IS92a
is thus no longer the "central" scenario, but rather the high
one.
Climate Change 94 included a
section on the evaluation of the emissions scenarios. It contained the
priceless statement (page 252):
"Scenarios
deal with the future,
so they cannot be compared with observations."
Of course, the future has an awkward habit
of becoming the present, and then the past. The scenarios all begin in
1990, so we have now had six-and-a-half years of observations to check
their plausibility. When this is done (see Table 1 below), it is evident
that all of the IS92 scenarios exaggerate what has happened over the
past five years, whichever set of assumptions is chosen.
It can be seen that the only scenarios with
any shred of plausibility are IS92c and IS92d, and even those exaggerate
likely greenhouse gas and temperature trends, as well as coal
production. IPCC projections to the year 2100 can therefore be taken
seriously only for scenarios IS92c and IS92d, with a probability that
the true figure will be below both of these. After adjusting for the
exaggerated rates of increase in CO2 and methane, we can
expect the maximum temperature rise between 1990 and 2100 to be
1°C.
All the scenarios can be summarized by
Figure 6(a) on page 24 of Climate Change 95, which shows the
net radiative forcing (change in mean radiation at the earth's surface)
from the year 1765 to 2100, adding together the effects of greenhouse
gases and changes in solar radiation, and subtracting the effects of
sulphate aerosols. Up to the year 1990, the measures plotted on the
graph were obtained from observations. Although the radiative forcing of
greenhouse gases alone is 2.45Wm-2, the total, after the
cooling effects of sulphate aerosols are considered, gives a radiative
forcing of 1.32Wm-2 between 1765 and 1990. During the last
part of this period, the twenty years from 1970 to 1990, the rate of
increase was linear, with a measured slope of. 0.0291Wm-2 per
year.
From 1990 are plotted the assumed
increases in radiative forcing for the six IS92 scenarios. The rate of
change of radiative forcing is linear from 1990 to 2025 for all the
scenarios except IS92c. As measured, at 0.0365Wm-2, it
represents a sudden increase of 25 percent over the rate for 1970-1990.
The actual rate of increase in
radiative forcing for the period 1990 to 1995 can be calculated by
adding the contributions from each of the greenhouse gases, using the
formulae given in Climate Change 90 and the data in Climate
Change 95 . When this is done, the average rate of increase of
radiative forcing from 1990-1995 comes to 0.0280Wm-2 per
year. This is 29 percent below the rate for 1970-1990, and 43 percent of
the value assumed for IS92a, b, d, e, and f for the period 1990 to 2025.
It is even below IS92c. Whatever may be claimed about the
"unusual" radiative forcing circumstances for the past five
years, a plausible set of scenarios should encompass all
possibilities, including reality. The IS92 scenarios all
exaggerate the current climate and economic trends.
Few professional economists would be
prepared to make forecasts beyond fifty years, and most would be
reluctant to go beyond twenty years. The main reason is the difficulty
in predicting future changes in technology. One only has to consider the
situation in 1896, when Arrhenius published the first calculations of
the greenhouse effect. The principal environmental hazard from vehicles
at that time was the proliferation of horse dung, and nobody could have
imagined today's motor vehicles.
Nevertheless, the IPCC makes no apology for
projecting one hundred years into the future, and for assuming that in
the year 2050 we will be using the 1930s' Bergius process to make oil
from coal. In its Stabilization Calculations, the IPCC goes one better,
presenting graphs showing the emission controls required to achieve
different levels of atmospheric stabilization of carbon dioxide
concentrations to the year 2400. Those projections have been
widely accepted, although they must surely be categorized as science
fiction, not science. The models upon which those projections are based
were constrained to upwardly biased estimates of the rate of change of
carbon dioxide concentrations. Moreover, those projections rely on the
carbon cycle figures for 1980-1989, when rates were highest. Frankly,
those projections may not be plausible even for ten years, let alone
four hundred.
Discernible Human Influence
IPCC scientists are under substantial
pressure from those who believe that human activity is having a harmful
effect on the global climate. It is to the credit of the IPCC that it
has largely resisted this pressure. There is nothing in any of the IPCC
reports to support the claim that human activity has a harmful effect on
the climate . . . . although the absence of that supporting information
has not stopped some organizations and publications from claiming that
such a harmful effect has been confirmed.
Humans are undoubtedly influencing the
climate. The increase in greenhouse gases is but one of many indicators.
The important point, though, is whether the influence is significant,
and whether it matters. The IPCC provides no answers to those two
important questions. Instead, it strives to answer a trivial,
unimportant question: Can a human influence be detected?
Much has been made by the media of the one
advance of Climate Change 95 over its predecessors: the
pronouncement that "the balance of the evidence suggests a
discernible human influence on the climate." Note that this is only
a suggestion, and that the nature and extent of the influence are not
specified.
An entire chapter of Climate Change 95,
and a great deal of research money, is devoted to the essentially
trivial pursuit of detecting a human influence on climate change. Two
arguments in support of such an influence are presented. The first is
based on the use of models to simulate "natural variability."
Those models intend to show that actual climate variability without
human influence agrees with "natural variability," but that
actual climate variability has exceeded the amount that could be
"natural" during the past century, when greenhouse gas
emissions have increased.
It has been noted earlier that none of the
climate models has been validated, in the sense that they have been
shown to simulate successfully an extended period of climate change.
However, a number of models have been "fitted" to a single
climate situation by using a set of " fudge factors" called
"flux adjustments."
So how much "natural variability"
occurred before greenhouse gases could have been a significant
influence? Climate records before 1940 are defective. Whole continents
and oceans were omitted. The further back we go, the less reliable
become the instruments and the observers. Proxy records, such as those
from ice cores or tree rings, are subject to much uncertainty; trying to
validate the results of unreliable models with unreliable past data
becomes very doubtful. As we have seen earlier in this Part 3, even the
recent temperature record is subject to question. If the surface
temperature rise over the last century were at the lower end of the
estimates (0.3°C), then even the "suggestion" of
"discernibleness" of the greenhouse effect would disappear.
The second argument depends on pattern
studies--a "qualitative" consistency of regional patterns of
climate change with predictions of models. Here the trick is to find
patterns that agree, and to ignore those that do not. There is now
fairly strong evidence that volcanoes influence climate, at least for
the single instance of the Mount Pinatubo eruption in June 1991.
Similarly, there is evidence for the assumption that increases in
sulphate-based aerosols have a regional cooling effect on the climate,
although, as noted earlier, the predicted large temperature difference
between the hemispheres has not materialized. What has not been
done is to show that increases in greenhouse gases have affected the
climate.
The "final" draft of Climate
Change 95 had the following passage (page 8.17): "When will an
anthropogenic effect on the climate be identified? The best answer is
'we do not know.'" Inexplicably, this passage was deleted from the
published report.
The opinion of most climate change
scientists is probably best stated by the team responsible for compiling
the surface temperature record (Parker, Jones, Folland, and Bevan) in a
paper published by the Journal of Geophysical Research (1994):
The fact of global warming in the past
century is beyond dispute, even though the precise amount is certainly
not. On theoretical grounds a likely contributory cause of this
warming is the rise in greenhouse gas concentrations, but despite some
similarities between the recent oceanic surface temperature anomalies
and those modelled . . . It is definitely premature to ascribe all or
most of the warming to this particular cause.
Part 4
Conclusion
There is nothing in the latest IPCC report
to support the introduction of drastic or economically damaging measures
to control greenhouse gas emissions.
Disentangling the bias, the absence of
precision, and the special pleading is difficult, but Climate Change
95 does present most of the facts on current climate change, and it
is possible to present some balanced conclusions once those facts are
teased out:
-
Global temperatures have not altered
significantly for the past five years, and there is persuasive
evidence (from satellites) that they have not changed for 18 years.
Changes over the past century, before recent increases in greenhouse
gases, could have been largely due to natural variability factors
such as volcanic eruptions or changes in solar radiation or ocean
circulation.
-
Computer models of the climate, based on
greenhouse theory, do not explain past climate changes to a degree
that gives confidence in their ability to project future change,
even after the inclusion of the cooling effects of sulphate
aerosols.
-
If radiative forcing continues to
increase at a rate below that of all the IPCC scenarios, then
temperature rises over the next century will be below all of those
projections, and so below 1°C by the year 2100. The Framework
Convention on Climate Change will require rethinking.
References
Arrhenius, S., April 1896. "On the
Influence of Carbonic Acid in the Air upon the Temperature of the
Ground." Philosophical Magazine 41 (251),
pages 237-276.
Christy, J.R. and R.T. McNider, 1994.
"Satellite greenhouse warming." Nature 367,
page 325.
Climate Change: The IPCC Scientific
Assessment (Climate Change 90). J.T. Houghton, G.J. Jenkins, and
J.J. Ephraums (editors). Cambridge University Press, 1990.
Climate Change 1992: The Supplementary
Report to the IPCC Scientific Assessment (Climate Change 92). J.T.
Houghton, B.A. Callander, and S.K. Varney (editors). Cambridge
University Press, 1992.
Climate Change 1994: Radiative Forcing
of Climate Change (Climate Change 94). J.T. Houghton, L.G. Meira
Filho, J. Bruce, Hoesung. Lee, B.A. Callander, E. Haites, N. Harris, and
K. Maskell (editors). Cambridge University Press, 1995.
Climate Change 1995: The Science of
Climate Change (Climate Change 95). J.T. Houghton, L.G. Meira Filho,
B.A. Callander, N. Harris, A. Kattenberg, and K. Maskell (editors).
Intergovernmental Panel on Climate Change and Cambridge University
Press, 1996.
Conway, T.J., P.P. Tans, L.S. Waterman, K.W.
Thoning, D.R. Buanerkitzis, K.A. Masarie, and Ni Zhang, 1994. Journal
of Geophysical Research 99 (D11), pages
22,831-22,855.
Forrester, J.W., 1971. World Dynamics.
Wright-Allen Press, Inc., Cambridge, Massachusetts.
Meadows, D.H., D.L. Meadows, J. Randers, and
W. W. Behrens, 1972. Limits to Growth. Universe Books, New
York.
Parker, D.E., P.D. Jones, C.K. Folland, and
A. Bevan, 1994. Journal of Geophysical Research 99 (D7), pages
14,373-14,399.
Pepper, W., J. Leggett, R. Swart, J. Wasson,
J. Edmonds, and I. Mintzer, May 1992. Emissions Scenarios for the
IPCC: An Update. Supporting document to Climate Change 92.
Trends '93: A Compendium of Data on
Climate Change. T.A. Boden, D.P. Kaiser, R.J. Sepanski, and F.W.
Stoss (editors). Carbon Dioxide Information Analysis Center, Oak Ridge,
Tennessee, September 1994.
|