BACKGROUND

The conversion of land from its native state to an agricultural use commonly results in a significant loss of soil carbon (Mann, 1985; Davidson and Ackerman, 1993). Globally, this loss is estimated to account for as much as 1/3 of the net CO2 emissions for the period of 1850 to 1980 (Houghton and others, 1983). Roughly 20 to 40 percent of original soil carbon is estimated to be lost as CO2 as a result of agricultural conversion, or "decomposition enhancement". Global models use this estimate along with land conversion data to provide agricultural contributions of CO2 emissions for global carbon budgets (Houghton and others, 1983; Schimel, 1995).

Soil erosion rates are significantly (10X) higher on croplands than on their undisturbed equivalents (Dabney and others, 1997). Most of the concern over erosion is related to diminished productivity of the uplands (Stallings, 1957; McGregor and others, 1969; Rhoton, 1990) or to increased hazards and navigability of the lowlands in the late 1800's to early 1900's. Yet because soil carbon is concentrated at the soil surface, with an exponential decline in concentration with depth (Harden et al, 1999), it is clear that changes in erosion rates seen on croplands must also impact soil carbon storage and terrestrial carbon budgets as well.

As yet, erosional losses of carbon are not included in global carbon budgets explicitly as a factor in land conversion nor implicitly as a portion of the decomposition enhancement. However, recent work by Lal and others (1995) and by Stallard (1998) suggests that significant amounts of eroded soil may be stored in man-made reservoirs and depositional environments as a result of agricultural conversion. Moreover, Stallard points out that eroding soils have the potential for replacing part of the carbon trapped in man-made reservoirs. If true, then the global carbon budget may grossly underestimate or ignore a significant sink term resulting from the burial of eroded soil.