TCS Daily

Hot Soil?

By Willie Soon - July 1, 2003 12:00 AM

The May 23 science update from the online journal Nature highlighted a paper just published in the American Geophysical Union's Geophysical Research Letters by climate researchers from the prestigious U.K. Hadley Centre. The update noted that the "[h]olistic model hints [the] next century could get even hotter than we thought" because as the future climate warms, the land biosphere that now locks up more carbon than it emits will suddenly turn into a potent emitter.

Is there really cause for additional alarm?

Fact: Terrestrial carbon is stored in both soil and vegetation, and the soil ecosystem is the larger reservoir. Indeed, the top meter of soil is estimated to contain twice as much carbon as there is in the atmosphere in the form of carbon dioxide (CO2).

Fact: The soil and atmosphere exchange carbon continuously. The amount of carbon shifted between the soil and atmosphere in a year is about 10 times larger than the current annual emission of carbon resulting from human activities.

Now, regarding the Hadley Centre paper, there are three main aspects to keep in mind:

First, the common assumption in climate models is that there will be about 750 parts per million (ppm) of CO2 in the air by 2100, most of that produced from human burning of fossil fuels. The paper in Nature predicts 980 ppm of CO2 in the air, with the additional carbon dioxide emission having come from the soil under the scenario of a very warm Earth.

But why is that increase in soil emission into the air predicted to happen? Well, it is based on the vegetation reservoir turning from acting as a carbon sink to a carbon source starting around 2050-2060. Why? Because "as the climate warms [there will be a] dieback of the Amazon forest due to regional rainfall reduction." Thus, it is the death of forests in the Amazon predicted by the models that would lead to a net increase in carbon emissions to the atmosphere.

(We will not discuss the issue here, but please consult this previous ChartiFact regarding how unreliable the current generation of climate models is in making predictions of future precipitation, especially with regard to the claim that the Amazon will dry out and disappear.)

Secondly, according to the paper, global surface temperature is supposed to warm by 5.5 degrees Celsius in 2100. And the land is expected to warm more dramatically -- by an average of 8 degrees C in 2100. This degree of global warmth will be truly unprecedented. For example even during the well-notable Mid-Holocene Warm Period of 6,000 years ago, the summer temperatures around the high northern latitude were estimated to be no more than 2 to 4 degrees C warmer than today.

Finally, the paper claims to be the first to account for "both interactive carbon and sulphur cycles ... in a GCM climate change projection with a full set of natural and anthropogenic (human caused) climate forcing factors [including anthropogenic sulphate aerosols, ozone, solar and volcanic forcings]."

All of this is meant to explain the term "holistic modeling." Basically, it means that regardless of how many unknowns and uncertainties there are in each factor affecting the climate, scientists will try to model as many of these factors into the model so it can be "holistic" rather than partial.

But that raises a basic question: How accurate can a model so constructed actually be? Only if it is accurate can it really inform policy makers' decision-making on issues of climate change. And several details suggest the results from the models to be uninformative.

The first two disappointments with the model involve its handling of the effects of sulphate aerosols, which come from release of sulphur dioxide (SO2) by the burning of coal and oil. The direct effect of sulphate is to cool the Earth as more of it would tend to reflect more sunlight. But there are also indirect effects.

One indirect effect of human-released sulphate is included by the U.K. Hadley Centre's researchers -- and this is what the paper says -- "using a non-interactive method in which cloud albedo perturbations are imposed based on output from a set of preliminary sulphur cycle runs." In plain English, that means changing the reflective power of clouds somewhat artificially. But this approach may already be unrealistic because this first indirect effect of sulphate, rather than being handled interactively as claimed, instead has been determined from other models. The mechanisms of cloud formation and the formation's interaction with sunlight involve very fast processing times, so an interactive modeling of the sulphur cycle, as the U.K. researchers promised, is necessary for realistic results.

Next, "the second indirect sulphate effect [through changes in precipitation efficiency] ... has been estimated to be significantly smaller ... and is excluded for simplicity." The problem with this assumption is that it contradicts the United Nations' Intergovernmental Panel on Climate Change (IPCC) Third Assessment Report (TAR). The IPCC report found that "[m]odels which include the second indirect effect [of sulphate aerosols] find that it increases the overall indirect [radiative] forcing by a factor from 1.25 to more than a factor of two." (p. 292 of TAR)

The real scientific consensus about the second indirect sulphate effect is that nobody really knows. There are large uncertainties and unknowns associated with the estimate of this indirect sulphate aerosol forcing. But that is distinct from claiming the effects to be small, as the Hadley researchers would indicate.

Further, as the IPCC TAR continued, "[p]recipitation changes could be important to climate change even if their net radiative forcing effect is small, but our ability to assess the changes in precipitation patterns due to aerosols is small." (p. 292 of TAR) In other words, the second indirect effect of sulphate aerosols is potentially important and cannot be neglected outright for "simplicity" if the "holistic" modeling approach is to be seriously tested.

A Stanford University researcher emphasized that there are up to 47 species of aerosols that need to be accounted for in order to achieve any comprehensive understanding of climatic forcings -- changes forced upon climate -- by both natural and anthropogenic aerosols.

One species of aerosol whose climate role seems to be increasingly important is black carbonaceous aerosols (or soot aerosols) that have been linked to the newly discovered phenomenon of the "Asian Brown Clouds."

The authors of the new Hadley Centre research paper also estimate solar forcing factors for climate only up until 2000 which then "is kept constant after 2000," while a volcanic forcing factor affecting climate is specified "up to the present day, and assuming it is zero thereafter." The notion of a constant sun, while appealing in poetry, is not a scientific reality, and the idea that there will be no volcanic eruptions for a century for the first time in world history may be necessary for academic study, but likewise is unrealistic.

Even more significant is the fact that field and laboratory data contradict the key assumption in the model about long-term sensitivity of soil respiration to global warming.

The model essentially, without support, extrapolates that more and more carbon stored in the soil will be emitted into the atmosphere as the climate warms. But researchers from the University of Hawaii and U.S. Forest Service studied 82 sites over five continents and found little change in soil respiration as temperature increases.

Those researchers explained that over decades the enzyme action of microbes, which ultimately releases carbon dioxide from decomposing soil, is not sensitive to temperature. Instead, it depends more on factors like the availability of nutrients, moisture, soil clay content, and quality of the mineral soil itself.

Additionally, as hypothesized by two distinguished UK ecologists, a warming of the air "increases the rate of physico-chemical processes which transfer organic carbon to 'protected,' more stable, soil carbon pools." Hence, instead of emitting more to the atmosphere as temperatures rise, more carbon is stored in the soil ecosystem.

Furthermore, a group of Finnish forest and plant ecologists, in a publication in Global Change Biology, recently provided this insight that "at later stages, decomposition may become more tolerant of climate ... and decomposition of old soil organic matter has been found to be rather insensitive to temperature."

Professor Stephen Long, the Robert Emerson Professor of Plant Biology at the University of Illinois, recently remarked, "[O]ne of the biggest challenges [now] is to gain more certainty about how plants and ecosystems are going to respond in the long term to elevated carbon dioxide and elevated temperature. We know a lot about the short term, and we, of course, have models of the longer term, but we really need experiments that are going to give us definitive answers."

Despite the problems cited here, the ambitious strategy of the new "holistic [climate] modeling" trend is actually a good one. For science to advance, though, the assumptions in such models need to continue to be clearly stated so they can be honestly confronted with realities.

TCS Daily Archives