Much subsequent research has proven this concept to be correct.  In a study of spatial and temporal patterns of land degradation in northeastern Iceland over the past 7500 years, for example, Olafsdottir and Gudmundsson (2002) found that warmth-induced increases in plant growth and development did essentially the same thing as CO2-induced increases in plant growth.  Local cooling always resulted in "deterioration in vegetation and soil cover," and during every major cold period of their entire record, land degradation was always classified as "severe."  During every major warm period, on the other hand, this condition was reversed, and soils were built up again, as vegetation cover expanded.

An entirely different consequence of atmospheric CO2 enrichment has also been demonstrated to play a major role in increasing soil stability.  In a research program carried out at the Jasper Ridge CO2 experiment site in northern California and the Sky Oaks CO2 experiment site in southern California, Rillig et al. (1999) studied belowground ecosystem responses to elevated atmospheric CO2 concentrations over a period of several years, focusing much of their attention on the growth responses of arbuscular mycorrhizal fungi that form symbiotic associations with plant roots.  They also quantified the fungal production of a glycoprotein called glomalin; and in an investigation of the consequences of this phenomenon, they examined its impact on the formation of small soil aggregates and the stability of those aggregates.

The scientists' work yielded several significant findings.  The amount of fungal-produced glomalin in the soils of the CO2-enriched treatments in all three of the ecosystems they studied was found to be greater than that observed in the soils of corresponding ambient CO2 treatments.  They also observed increases in the mass of small soil aggregates in the treatments exposed to elevated CO2; and the stability of the small soil aggregates in the CO2-enriched treatments was greater than the stability of similar aggregates in the ambient air treatments.  In addition, in the Sky Oaks study, where six CO2 concentrations ranging from 250 to 750 ppm were imposed as treatments, the authors reported that (1) "the proportion of soil mass in aggregates of 0.25-1 mm showed a linear increase along the CO2 gradient" and (2) "glomalin concentrations followed a pattern similar to that of the small aggregate size class," indicative of ever-increasing soil structure benefits with ever-increasing concentrations of atmospheric carbon dioxide.

In a similar study, Rillig et al. (2000) examined several characteristics of arbuscular mycorrhizal fungi associated with the roots of plants growing for at least 20 years along a natural CO2 gradient near a CO2-emmitting spring in New Zealand, finding that elevated CO2 significantly increased percent root colonization by arbuscular mycorrhizal fungi in a linear fashion -- and by nearly 4-fold! -- in going from a concentration of 370 to 670 ppm.  Similarly, fungal hyphal length experienced a linear increase of over 3-fold along the same CO2 gradient; while total soil glomalin experienced a linear increase of approximately 5-fold.  Likewise, Insam et al. (1999) observed that a 270-ppm increase in the air's CO2 concentration increased the presence of humic substances in the soils of nutrient-poor artificial tropical ecosystems by almost 30%.

With the significant increases in soil stability that are implied by these several studies, one would expect to see measurable decreases in soil erosion over at least the last half of the 20th century, as the air's CO2 concentration has edged ever upward.  More often than not, however, just the opposite has typically been assumed.  The "remarkable feature" of this long-held belief in continued high, or even increasing, soil erosion, in the words of Trimble and Crosson (2000), "is that it was based mostly on models."  Indeed, they state that "little physical, field-based evidence (other than anecdotal statements) has been offered to verify the high estimates," noting further that "it is questionable whether there has ever been another perceived public problem for which so much time, effort, and money were spent in light of so little scientific evidence."

In trying to get to the truth of the matter by looking at hard data from the United States, Trimble and Crosson discovered that "available field evidence suggests declines of soil erosion, some very precipitous, during the past six decades."  Likewise, in studying the Upper Mississippi River Valley, Knox (2001) determined that conversion of the region's natural landscape to primarily agricultural uses boosted surface erosion rates to values three to eight times greater than those characteristic of pre-settlement conditions, while they increased peak discharges from high-frequency floods by 200 to 400%.  Since the late 1930s, however, data indicate that surface runoff has been decreasing; and since the 1940s and early 1950s, the magnitudes of the largest daily flows have been decreasing at the same time that the magnitude of the average daily baseflow has been increasing, just as would be expected for better vegetated and more stable soils, as per the several CO2-induced phenomena discussed above, which facilitate the absorption of rainfall and the reduction of surface runoff to streams and rivers.

In conclusion, we note that solid experimental data suggest that the historical increase in the atmosphere's CO2 concentration should have significantly enhanced a number of plant-mediated phenomena that tend to stabilize soil at both the surface of the ground and throughout the plant root zone.  And we note that real-world soil and water survey data indicate that the expected decrease in soil erosion anticipated to result from these effects has indeed been observed.

References
Idso, S.B.  1989.  Carbon Dioxide and Global Change: Earth in Transition.  IBR Press: Tempe, AZ.

Insam, H., Baath, E., Berreck, M., Frostegard, A., Gerzabek, M.H., Kraft, A., Schinner, F., Schweiger, P. and Tschuggnall, G.  1999.  Responses of the soil microbiota to elevated CO2 in an artificial tropical ecosystem.  Journal of Microbiological Methods 36: 45-54.

Knox, J.C.  2001.  Agricultural influence on landscape sensitivity in the Upper Mississippi River Valley.  Catena 42: 193-224.

Olafsdottir, R. and Gudmundsson, H.J.  2002.  Holocene land degradation and climatic change in northeastern Iceland.  The Holocene 12: 159-167.

Rillig, M.C., Hernandez, G.Y. and Newton, P.C.D.  2000.  Arbuscular mycorrhizae respond to elevated atmospheric CO2 after long-term exposure: evidence from a CO2 spring in New Zealand supports the resource balance model.  Ecology Letters 3: 475-478.

Rillig, M.C., Wright, S.F., Allen, M.F. and Field, C.B.  1999.  Rise in carbon dioxide changes soil structure.  Nature 400: 628.

Trimble, S.W. and Crosson, P.  2000.  U.S. soil erosion rates - myth and reality.  Science 289: 248-250.