The primary problem is the simple fact that most observational evidence does not support the model predictions of reduced soil carbon storage under elevated temperatures.  Fitter et al. (1999), for example, evaluated the effect of temperature on plant decomposition and soil carbon storage, finding that upland grass ecosystem soils artificially heated by nearly 3°C increased both root production and root death by equivalent amounts.  Hence, they concluded that in these ecosystems, elevated temperatures "will have no direct effect on the soil carbon store."  Similarly, Johnson et al. (2000) warmed Arctic tundra ecosystems by nearly 6°C for eight full years and still found no significant effect of that major temperature increase on ecosystem respiration.  Furthermore, Liski et al. (1999) showed that carbon storage in soils of both high- and low-productivity boreal forests in Finland actually increased with warmer temperatures along a natural temperature gradient.

Perhaps most impressive of all, Giardina and Ryan (2000) analyzed organic carbon decomposition data derived from the forest soils of 82 different sites on five continents, reporting the amazing fact that "despite a 20°C gradient in mean annual temperature, soil carbon mass loss ... was insensitive to temperature."  And in the same issue of Nature in which Giardina and Ryan's work was published, a group of thirty other scientists weighed in with yet another important contribution to the developing science (Valentini et al., 2000).  This group of researchers had collected data on net ecosystem carbon exchange in fifteen European forests; and they reported that their results "confirm that many European forest ecosystems act as carbon sinks."  Their data also demonstrated that the warmer forests of southern Europe annually sequester far more carbon than the cooler forests of northern Europe, again in direct contradiction of the claims of those who have long touted global warming as a sure-fire recipe for forest carbon loss (Pearce, 1999).

In yet another discussion of the subject, Grace and Rayment (2000) present still more evidence that refutes the claim that new forests in a rapidly warming world "will swiftly become saturated with carbon and begin returning most of their carbon to the atmosphere."  Specifically, they cite a number of additional studies that "show quite clearly," as they put it, "that old undisturbed forests, as well as middle-aged forests, are net absorbers of CO2."  They also note that these real-world observations mean that "forests are serving as a carbon sink, providing a global environmental service by removing CO2 from the atmosphere and thus reducing the rate of CO2-induced warming."  Commenting further on the work of Giardina and Ryan (2000) and Valentini et al. (2000), Grace and Rayment unequivocally state that "the results from these two papers should send a powerful message to those working with models of global vegetation change," namely, "that the doomsday view of runaway global warming now seems unlikely."

Why the big discrepancy between model predictions and reality?  Thornley and Cannell (2001) hypothesized that the physico-chemical processes involved in attaching organic materials onto soil minerals, or bringing them together into aggregates that are less subject to decomposition, require a large amount of activation energy; and they suggested that higher temperatures can provide that energy.  In fact, they developed a dynamic soil model in which they demonstrate that under such circumstances "long-term soil carbon storage will appear to be insensitive to a rise in temperature, even if the respiration rates of all [soil carbon] pools respond to temperature as assumed by [most models]," which is, in fact, what experimental and real-world data clearly indicate to be the case.

Building upon this concept and extending it even further, Agren and Bosatta (2002) developed what they call the continuous-quality theory, which is based on the premise there is a wide-ranging continuous spectrum of soil organic carbon "mini-pools," each of which possess differing degrees of resistance to decomposition.  It states that soils from naturally higher temperature regimes will have soil organic matter continuous quality distributions that contain relatively more organic matter in carbon pools that are more resistant to degradation and are consequently characterized by lower rates of decomposition, which has been observed experimentally to be the case by Grisi et al. (1998).  In addition, their theory states that this shift in the distribution of soil organic matter qualities, i.e., the higher-temperature-induced creation of more of the more-difficult-to-decompose organic matter, will counteract the decomposition-promoting influence of the higher temperatures, so that the overall decomposition rate of the totality of organic matter in a higher-temperature soil is either unaffected or actually reduced.

Moving from warmer to colder climates, another crack in the once-seemingly-sound hypothesis of warming-induced carbon loss from natural ecosystems appeared when Oechel et al. (2000) published the results of their long-term measurements of net ecosystem CO2 exchange in wet-sedge and moist-tussock tundra communities of the Alaskan Arctic, which indicated they were gradually shifting from being carbon sources to becoming carbon sinks.  The ultimate transition occurred between 1992 and 1996, at the apex of a regional warming trend that culminated with the local climate experiencing the highest average summer temperature and surface water deficit of the previous four decades.

How did it happen, this dramatic and unexpected biological transformation?  The answer of the scientists who observed and documented the phenomenon was that it was really nothing special -- no more, as they put it, than "a previously undemonstrated capacity for ecosystems to metabolically adjust to long-term changes in climate."

Yes, just as people can change their behavior in response to environmental stimuli, so can plants.  And this simple ecological acclimation process is only one of several newly-recognized phenomena that have caused scientists to radically revise the way they think about global change in Arctic regions.

The most recent of the still-evolving new work in this area comes from Camill et al. (2001), who studied (1) changes in peat accumulation across a regional gradient of mean annual temperature in Manitoba, Canada, (2) net aboveground primary production and decomposition for major functional plant groups of the region, and (3) soil cores from several frozen and thawed bog sites that were used to determine long-term changes in organic matter accumulation following the thawing of boreal peatlands.

In direct contradiction of earlier thinking on the subject, but in confirmation of the more recent findings of Camill (1999a,b), the authors of the new study discovered that aboveground biomass and decomposition "were more strongly controlled by local succession than regional climate."  In other words, they determined that over a period of several years, natural changes in plant community composition generally "have stronger effects on carbon sequestration than do simple increases in temperature and aridity."  In fact, the authors' assessments of peat accumulation over the past two centuries showed that rates of biological carbon sequestration can almost double following the melting of permafrost, in harmony with the findings of Robinson and Moore (2000) and Turetsky et al. (2000), who found the rates of organic matter accumulation in other recently-thawed peatlands to rise by 60-72% in newly-warmed climatic regimes.

Another relevant cold-region study was conducted by Neilsen et al. (2001), who collected samples of soil from a northern hardwood forest in New Hampshire, USA. These samples, from nearly pure stands of sugar maple and yellow birch, were placed in small vessels and either maintained at the normal laboratory temperature of 20-25°C or subjected to mild and severe freezes of -3 and -13°C, respectively, for ten days, after which all samples were kept at the normal laboratory temperature for 23 additional days.  The evolution of CO2 from the soils was measured at the beginning and end of the full 33-day period and at three other times during the course of the experiment.

Freezing had a significant effect on CO2 evolution from the soils.  Cumulative 33-day totals of respiration (in units of mg carbon per kg of soil) for the soil samples taken from the maple stand were 1497, 2120 and 3882 for the control and -3 and -13°C temperature treatments, respectively, which numbers represent carbon loss enhancements (relative to that of the control) of 42 and 159% for the -3 and -13°C treatments, respectively, or an increased carbon loss of 13 ± 1% for each degree C below freezing.  For the soil samples taken from the birch stand, the corresponding respiration numbers were 1734, 2866 and 5063, representing carbon loss enhancements of 65 and 192% for the -3 and -13°C treatments, respectively, or an increased carbon loss of 18 ± 3% for each degree C below freezing.

It can be readily appreciated how these results relate to the subject of global warming effects on soil carbon sequestration.  As temperatures gradually warm over the course of many years and different climate zones move poleward in latitude and upward in elevation, regions that experienced many hard freezes in the past will experience less of them in the future.  Other regions will experience a shift from hard freezes to mild freezes.  Still other regions that experienced mild freezes in the past will experience fewer - or none - in the future.  And in all of these situations, together with every permutation that falls somewhere in their midst, there will be a tendency for less carbon to be released to the atmosphere as the climate warms.

Another important fact to remember in relation to the subject of temperature effects on carbon sequestration is that an increase in atmospheric CO2 concentration is expected to accompany any increase in air temperature.  Indeed, an increase in the air's CO2 content is even more assured than is an increase in its temperature.  Hence, we briefly report on two relevant experiments in which we were personally involved that bear upon this issue.

Idso et al. (1995) measured the net photosynthetic rates of leaves of sour orange trees growing out-of-doors in clear-plastic-wall open-top chambers maintained at 400 and 700 ppm CO2 under the searing summer sun of Phoenix, Arizona.  It was so hot during the experiment, in fact, that most of our measurements were made at temperatures above the trees' optimum growth temperature.  As a result, foliage net photosynthetic rates dropped lower and lower as the air temperature climbed higher and higher into the middle of each afternoon.  In fact, they dropped so low that at a leaf temperature of 47°C (117°F), the net photosynthetic rates of the leaves on the trees growing in air of 400 ppm CO2 dropped all the way to zero and actually became negative thereafter, as the temperature rose higher still.  In contrast, leaves on the CO2-enriched trees continued to exhibit positive rates of net photosynthesis until a leaf temperature of 54°C (129°F) was reached.  Thus, the extra 300 ppm of CO2 to which the CO2-enriched trees were exposed allowed them to continue to remove CO2 from the air until it had warmed by an additional 7°C (12°F).

An even more dramatic example of the plant "heat relief" provided by atmospheric CO2 enrichment is described in another publication of ours (Idso et al., 1989).  Throughout the summers of 1985 and 1986 -- again in Phoenix, Arizona -- we grew floating mats of tiny water ferns on the surfaces of sunken metal stock tanks filled with water and located out-of-doors within clear-plastic-wall open-top chambers maintained at atmospheric CO2 concentrations of ambient and ambient plus 300 ppm CO2.  In both years, plant growth rates (assessed as weight gains per week) in the ambient CO2 enclosures first decreased, then dropped to zero, and finally became negative, when the air temperature rose above 30°C.  In the CO2-enriched enclosures, however, the debilitating effects of high air temperature were significantly reduced.  In one experiment the plants exhibited a much less severe negative growth rate; in another they experienced only a short period of zero growth rate; and in a third instance they suffered no ill effects at all, in spite of the fact that the plants growing in ambient air all died!

Midway through one of these experiments, we measured rates of net photosynthesis every hour of the day approximately one week after we first began to detect high-temperature-induced reductions in plant growth rates in the ambient CO2 enclosures.  Our data revealed that the net photosynthetic rates of the plants growing in ambient air went from negative to positive at 8 o'clock in the morning, peaked at approximately 10 o'clock, and dropped back to zero at noon, becoming negative thereafter.  Consequently, with only four hours of positive net photosynthesis during a 24-hour period, the plants growing in ambient air lost a significant portion of their biomass each day.  In the CO2-enriched plants, on the other hand, net photosynthesis went from negative to positive just after 7 o'clock in the morning, peaked at about 11 o'clock, and -- being better able to withstand the high afternoon temperatures -- did not decline to zero until just before 5 pm.  Averaged over a full 24-hour period, these CO2-enriched plants took up about as much CO2 during the day as they gave off at night, enabling them to maintain their biomass during this stressful high-temperature period that saw dramatic weekly weight losses in the plants growing in ambient air.

These experiments on tiny water ferns and large sour orange trees vividly demonstrate the ability of atmospheric CO2 enrichment to enable plants to better withstand the physiological ravages of high temperatures, which for some plants occur both daily (in the afternoon) and seasonally (in the summer) in nearly all parts of the world where plants grow.  In fact, they demonstrate that elevated levels of atmospheric CO2 can sometimes mean the difference between life and death itself.  And if one is concerned about carbon sequestration, it doesn't take much gray matter to realize that dead plants have done all they'll ever do in the way of removing CO2 from the atmosphere.  One of the important keys to greater carbon sequestration, therefore, is to keep plants both living and growing as long as possible; and in this regard, elevated levels of atmospheric CO2 seem to be just what the plant doctor ordered.

In conclusion, it is clear from these many real-world studies that rising temperatures will not result in greater losses of carbon from earth's natural ecosystems in either hot or cold parts of the world, and that rising CO2 concentrations will actually lead to increases in carbon sequestration.

References
Agren, G.I. and Bosatta, E.  2002.  Reconciling differences in predictions of temperature response of soil organic matter.  Soil Biology & Biochemistry 34: 129-132.

Camill, P.  1999a.  Patterns of boreal permafrost peatland vegetation across environmental gradients sensitive to climate warming.  Canadian Journal of Botany 77: 721-733.

Camill, P.  1999b.  Peat accumulation and succession following permafrost thaw in the boreal peatlands of Manitoba, Canada.  Ecoscience 6: 592-602.

Camill, P., Lynch, J.A., Clark, J.S., Adams, J.B. and Jordan, B.  2001.  Changes in biomass, aboveground net primary production, and peat accumulation following permafrost thaw in the boreal peatlands of Manitoba, Canada.  Ecosystems 4: 461-478.

Fitter, A.H., Self, G.K., Brown, T.K., Bogie, D.S., Graves, J.D., Benham, D. and Ineson, P.  1999.  Root production and turnover in an upland grassland subjected to artificial soil warming respond to radiation flux and nutrients, not temperature.  Oecologia 120: 575-581.

Giardina, C.P. and Ryan, M.G.  2000.  Evidence that decomposition rates of organic carbon in mineral soil do not vary with temperature.  Nature 404: 858-861.

Grace, J. and Rayment, M.  2000.  Respiration in the balance.  Nature 404: 819-820.

Grisi, B., Grace, C., Brookes, P.C., Benedetti, A. and Dell'abate, M.T.  1998.  Temperature effects on organic matter and microbial biomass dynamics in temperate and tropical soils.  Soil Biology & Biochemistry 30: 1309-1315.

Idso, S.B., Allen, S.G., Anderson, M.G. and Kimball, B.A.  1989.  Atmospheric CO2 enrichment enhances survival of Azolla at high temperatures.  Environmental and Experimental Botany 29: 337-341.

Idso, S.B., Idso, K.E., Garcia, R.L., Kimball, B.A. and Hoober, J.K.  1995.  Effects of atmospheric CO2 enrichment and foliar methanol application on net photosynthesis of sour orange tree (Citrus aurantium; Rutaceae) leaves.  American Journal of Botany 82: 26-30.

Johnson, L.C., Shaver, G.R., Cades, D.H., Rastetter, E., Nadelhoffer, K., Giblin, A., Laundre, J. and Stanley, A.  2000.  Plant carbon-nutrient interactions control CO2 exchange in Alaskan wet sedge tundra ecosystems.  Ecology 81: 453-469.

Liski, J., Ilvesniemi, H., Makela, A. and Westman, C.J.  1999.  CO2 emissions from soil in response to climatic warming are overestimated - The decomposition of old soil organic matter is tolerant of temperature.  Ambio 28: 171-174.

Neilsen, C.B., Groffman, P.M., Hamburg, S.P., Driscoll, C.T., Fahey, T.J. and Hardy, J.P.  2001.  Freezing effects on carbon and nitrogen cycling in northern hardwood forest soils.  Soil Science Society of America Journal 65: 1723-1730.

Oechel, W.C., Vourlitis, G.L., Hastings, S.J., Zulueta, R.C., Hinzman, L. and Kane, D.  2000.  Acclimation of ecosystem CO2 exchange in the Alaskan Arctic in response to decadal climate warming.  Nature 406: 978-981.

Pearce, F.  1999.  That sinking feeling.  New Scientist 164 (2209): 20-21.

Robinson, S.D. and Moore, T.R.  2000.  The influence of permafrost and fire upon carbon accumulation in high boreal peatlands, Northwest Territories, Canada.  Arctic, Antarctic and Alpine Research 32: 155-166.

Thornley, J.H.M and Cannell, M.G.R.  2001.  Soil carbon storage response to temperature: an hypothesis.  Annals of Botany 87: 591-598.

Turetsky, M.R., Wieder, R.K., Williams, C.J, and Vitt, D.H.  2000.  Organic matter accumulation, peat chemistry, and permafrost melting in peatlands of boreal Alberta.  Ecoscience 7: 379-392.

Valentini, R., Matteucci, G., Dolman, A.J., Schulze, E.-D., Rebmann, C., Moors, E.J., Granier, A., Gross, P., Jensen, N.O., Pilegaard, K., Lindroth, A., Grelle, A., Bernhofer, C., Grunwald, T., Aubinet, M., Ceulemans, R., Kowalski, A.S., Vesala, T., Rannik, U., Berbigier, P., Loustau, D., Gudmundsson, J., Thorgeirsson, H., Ibrom, A., Morgenstern, K., Clement, R., Moncrieff, J., Montagnani, L., Minerbi, S. and Jarvis, P.G.  2000.  Respiration as the main determinant of carbon balance in European forests.  Nature 404: 861-865.