Printer Friendly

Reasons for scepticism about greenhouse warming.

Predictions of global warming due to the greenhouse effect are being used to justify expensive socioeconomic action by Canada and other nations, but direct evidence in support of the hypothesis is hard to find

Is there a causal connection between an increase in global temperature (properly, global-average annual-mean surface-air temperature), increasing levels of atmospheric C|O.sub.2~ (partial pressure of C|O.sub.2~ in the atmosphere), and increased C|O.sub.2~ emissions (primarily from combustion of fossil fuels and manufacture of cement)?

All these show a positive trend over time since ca. 1860 which is taken to be evidence in favor of the "global warming from increases in greenhouse gases" (GW) hypothesis (most recently reiterated by Houghton et al. 1992). We shall examine this proposition from a regional climatological standpoint, with the overall perspective that the role of the oceans has not been adequately considered in current climate models.

Global record of temperature

Global and hemispheric air temperature records are considered to be one of the primary pieces of evidence available to test the GW hypothesis. Researchers who have assembled such data-sets have encountered problems that are especially serious because they are attempting to detect a weak signal (global warming) within very noisy data-set.

Records come from meteorological stations that are mostly (75 %) on land in the Northern Hemisphere (NH), but most of the Southern Hemisphere (SH) record comes from the ocean (most of the SH is ocean). The stations are not uniformly distributed geographically and become fewer in number as we go back in time (in 1860, less than 10% of the NH; by 1900, no more than 30%).

Many stations are in or near cities that generate more heat now than when instrumental records began. When the urban effect is eliminated from the climatic record of stations located in the contiguous United States (1895-1987), no warming trend, but a cooling over the last 50 years, emerges.

Ocean data are even less reliable than land-based. Historical sea surface temperatures (SST) used for global temperature compilations were neither collected nor analyzed in a consistent manner. SSTs were initially measured on seawater caught in a bucket (wooden, then leather, then metal), later, as engine cooling-water intake temperatures. Correction for these changes in methodology is difficult, and only after 1945 can SSTs be used with any real confidence to compute global and regional temperatures.

The resultant NH surface air temperature record, Fig. 1, shows a warming trend of a few tenths of a degree Celsius over the past 110 years that is claimed to be broadly consistent with predictions of GW climate models. It is also consistent with the interpretation that modern instrumental records began in a world emerging from the well-documented Little Ice Age (1450-1850; Grove 1988) when temperatures were lower (by about 1.5 |degrees~ C in the NH) and is now returning to more "normal" values.

For those proposing that the rise in global temperature is evidence in favor of GW, there is a serious discrepancy in the timing of the warming. The global temperature plot shows a rise in the early years of record followed by no increase from 1945 to 1975, with a subsequent increase to present, Fig. 2.

The greatest percentage of the total increase in global temperature (70%) occurred before 1940, but the increase in industrial emissions to that time was only 20% of the total to date; most emissions occurred after 1940. This increase in C|O.sub.2~ from fossil fuel combustion and cement manufacture cannot be held responsible for a global warming that preceded it. The warming of the late 19th and early 20th centuries must, therefore, have been from other (most probably natural) causes that, obviously, can be of the same magnitude as the postulated GW effects.

It has been suggested that if the rate of rise of global temperature can be shown to exceed that of the past, this will constitute detection of the GW effect. Indeed, much has been made of the fact that the three of the warmest years on record occurred in the 1980s. However, on a decadal time-scale, the warming trend of about 0.3 |degrees~ C from 1981-1990 was equalled by warming that occurred in the decades 1891-1900 and 1921-1930. The most recent decade did not, therefore, show unprecedented rates of warming.

Greater warming at high latitudes

Climate simulations by general circulation models (GCMs) predict greater warming at high latitudes, roughly double that at low latitudes, especially in winter. However, high latitudes in the NH have cooled in recent decades.

A group of U.S. and Russian researchers reported recently not only on the absence of evidence for warming over the central and western Arctic Ocean in the past 40 years, but that surface temperatures in the western Arctic showed significant cooling. The central and western North Pacific Ocean has also become colder by more than 0.75 |degrees~ C since 1977.

We ourselves have analyzed mean annual surface air temperatures at coastal and island stations in the North Atlantic Ocean between 44 |degrees~ N and 75 |degrees~ N, and have found either no statistically significant trends over the period of record, or a significant cooling trend at individual stations.

The above results are not of merely regional relevance; in total, they cover 70% of the NH. The similarity between temperatures at our longest-running stations with the NH record, Fig. 1, is very close from 1880 to 1975. Only in the last 15 years has the trend in the overall NH record diverged significantly from that at our N. Atlantic stations.

It would seem that the cooling trends in coastal and oceanic data-sets are being totally negated by a very intense (but not very extensive) terrestrial warming that completely dominates the NH mean. Where is it? Not in the well-sampled contiguous U.S. (as mentioned), nor on the Atlantic coast of Canada or in western Europe (own data). The majority of NH (winter) warming is apparently centered in two regions: northwestern North America and north central Eurasia. This is a very small portion of the Earth's surface upon which to rely for confirmation of a global greenhouse warming signal.

Direct satellite measurements of a microwave emission of oxygen that is proportional to temperature have been made since late 1978. No significant trend in global temperature has been detected, although the precision of these satellite measurements (0.01 |degrees~ C monthly) is quite sufficient to reveal a trend of one or two tenths of a degree per decade (model predictions run from + 0.2 to + 0.5 |degrees~ C per decade), if one existed. There is no indication from this data-set that any global warming was in progress, let alone that "... the hottest decade since temperatures were first written down ..." (Gore 1993) had just come to an end.

Oceans' role in global temperature

Modellers acknowledge that existing global climate models do not adequately include the role of the oceans. Despite popular misconception that the greenhouse gases are limited to C|O.sub.2~, C|H.sub.4~, CFCs, etc., water vapor is the most abundant and important of all the greenhouse gases (responsible for 90% of global heat retention).

The effects of other gases that absorb at the same wavelengths as water (e.g. in the 15 ||micro~meter~ region for C|O.sub.2~ absorption) are unimportant over the half of the globe where humidity is uniformly high, i.e. the tropical oceans. Heat must be moved from the tropics to latitudes where the infrared opacity of the atmosphere is lower before it can be re-emitted (through water vapor "windows", e.g. 8-12 ||micro~meter~) to outer space. Convective transport of heat from the tropics to high latitudes is, therefore, a fundamental cooling process of the planet.

Atmospheric processes alone are not responsible for redistributing this heat; oceanic, as well as atmospheric, circulation is crucial. Even if our information on the atmosphere were sufficient (which is not the case), the data base for the 70% of the globe covered by the oceans is inadequate.

Only the most recent climate models permit the ocean to play any role other than that of a passive energy store. In reality, ocean currents carry large volumes of water (and associated heat) from one part of the globe to another.

It has been calculated that without altering greenhouse gas concentrations the full range of temperatures experienced on Earth over the past 180 million years could have been generated by no more than a factor-of-two variation in ocean heat transport. So, the predictions of GCMs that lack a mobile ocean must be viewed with healthy scepticism; the same criticism can be made of the lack of adequate representation of clouds in current GCMs.

Oceans as a sink for C|O.sub.2~

Only since 1958 have direct measurements of the partial pressure of C|O.sub.2~ in the atmosphere become available from the Mauna Loa Observatory, Hawaii (and from other locations, including some Canadian stations, since 1975). The 1990 annual mean atmospheric concentration of C|O.sub.2~ was 353.95 ||micro~mol~ C|O.sub.2~ per mole of dry air (usually expressed as 354 ppmv).

At the beginning of the Mauna Loa record it was taken to be 315.83 ppmv. This gives an increase of 38 ppmv over the period of record that converts to 78 gigatonnes of carbon (GtC). During that same time, industrial emissions are estimated to have contributed 131 GtC to the atmosphere. The non-trivial discrepancy between these two (53 GtC) is the so-called "missing C|O.sub.2~", Fig. 3, and it must be accounted for in the climate models.

The interpretation of this discrepancy has gone through an interesting evolution over the last 30 years (Munk 1991). Initial calculations (from tracer-calibrated models) required the oceans to be the natural sink for it. Then concerns about C|O.sub.2~ produced by burning tropical forests required the oceans to take care of an additional 2 gigatonnes per year (GtC/a). Recent realization that increased storage of carbon in a terrestrial sink (forests in the NH) can account for this 2 GtC/a, means that there is no further need for it to be buried at sea -- it can be lost in the woods.

We agree with Munk that: The ocean plays three roles in this game: it serves as a reservoir of carbon; it serves as a reservoir of heat; and most of all, it serves as a reservoir for ignorance."

No matter how much C|O.sub.2~ is annually consigned to the ocean, the small fraction deposited in marine sediments is the only portion that is removed from the global carbon cycle for meaningful lengths of time. Present estimates of rates of marine sedimentation are far from robust, and we are currently incapable of quantitatively assessing any possible increase in the rate of storage of C|O.sub.2~ in marine sediments that might have occurred since industrialization.

In the absence of any convincing evidence for increased C|O.sub.2~ storage in the oceans (or anywhere else for that matter), the "missing C|O.sub.2~" is still missing and the fate of a large (and apparently increasing, Fig. 3) proportion of industrial emissions remains unknown.

Evidence from the past

What is the natural background of past variation in concentrations of greenhouse gases against which anthropogenic inputs may be compared? As direct measurements of these gases in the atmosphere have been made for only a century or so, we need retrospective measurements for longer time-studies, such as those made on ice taken from glaciers.

Determinations of C|O.sub.2~ in air inclusions in ice-cores have been taken to be truly representative of C|O.sub.2~ concentration in the air at the time the ice consolidated (some years after the original snow fell). A typical result of this analysis is in the Antarctic where the Vostok station glacial record provides the strongest evidence that the greenhouse gases C|O.sub.2~ and C|H.sub.4~ experienced concentration changes during the last major climatic cycle.

However, the resolution of the time-scale for the ice-core is not fine enough to say whether increasing atmospheric C|O.sub.2~ preceded higher ambient temperatures, only that variations of the two are positively correlated. At the beginning of the deglaciation, C|O.sub.2~ increase is either in phase with the rise in Antarctic temperature, or lags it by about 1000 years. A cause-and-effect relationship can be inferred from these results.

One disconcerting feature of the ice-core records is that the Medieval Climate Optimum (950-1350) and the Little Ice Age do not even stand out as major features. During the Medieval warm phase -- when Vikings sailed through ice-free waters to establish settlements in Greenland and voyaged on to Newfoundland -- the NH was certainly warmer than today, but greenhouse gas concentrations were no higher. Enhanced atmospheric C|O.sub.2~ is not, therefore, a necessary prerequisite for a warmer world.

Recently, Norwegian and Japanese researchers have severely criticized the methods used in ice-core work. Their claim that "glaciological studies are not able to provide a reliable reconstruction of either the C|O.sub.2~ level in pre-industrial and ancient atmospheres or paleoclimates" (Jaworowski et al. 1992), if correct, negates all conclusions drawn to date from ice-core records.


In the past 130 years, the content of greenhouse gases in the atmosphere has been increasing, most likely as a result of human activities. The longest temperature records do suggest global warming, but none that can be directly attributed to greenhouse gas increase. Any malignant influence of industrial emissions upon global climate has yet to be proven. The current algorithm that "... there really is no remaining dispute ..." about greenhouse warming, does not compute.

In order to progress in understanding global climate change we need not only more and better data (we always ask for that) but also models that better explain what we already know to have happened on Earth. Only then can we hope to (cautiously) predict the future. All involved in the public debate -- scientists, industrialists, environmentalists, politicians -- would do well to bear in mind these wise words attributed to Niels Bohr: "Prediction is very difficult, especially about the future...".


Boden, T.A., Sepanski, R.J. and Stoss, F.W. (eds.), 1991. Trends '91: A Compendium of Data on Global Change. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, Oak Ridge, TN, USA.

Gore, A., 1993. Earth in the balance: ecology and the human spirit. Penguin, New York, NY, USA.

Grove, J.M. (1988). The Little Ice Age. Methuen, London, U.K.

Houghton, J.T., Callander, B.A. and Varney, S.K. (eds.), 1992. Climate Change 1992: The Supplementary Report to the IPCC Scientific Assessment. Intergovernmental Panel on Climate Change (IPCC), Cambridge University Press, Cambridge, U.K.

Jaworowski, Z., Segalstad, T.V. and Ono, N., 1992. Do glaciers tell a true atmospheric C|O.sub.2~ story? Science of the Total Environment, 114, 227-284.

Munk, W.H., 1991. Doherty Lecture: Is there time to measure (not speculate) ocean warming before making policy? MTS Journal, 25, 53-57.

Roger Pocklington, FCIC, and Kenneth Drinkwater are research scientists in Physical and Chemical Sciences (Department of Fisheries and Oceans) at the Bedford Institute of Oceanography in Dartmouth, NS. Richard Morgan is a consultant meteorologist, also resident in Dartmouth. For further information on their research, call Roger at 902-426-8880.
COPYRIGHT 1993 Chemical Institute of Canada
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:evidence for increase in global temperature lacking
Author:Pocklington, Roger
Publication:Canadian Chemical News
Date:Oct 1, 1993
Previous Article:Technologists: a front line look at ISO 9000.
Next Article:Radionuclides in the marine environment - uses and concerns.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters