Rapid accumulation of soil carbon and nitrogen in restored prairie is not predicted using a chronosequence approach

Background/Question/Methods

Prairie restoration following agricultural abandonment can lead to the accumulation of soil carbon (C) and nitrogen (N).  Understanding the rate and patterns of C and N accumulation is important for determining how quickly these systems recover from agriculture and the capacity for restored prairies to act as a carbon sink.  Studies on the dynamics of C and N accumulation following prairie restoration are generally based on chronosequence studies.  However, few studies have compared the results of using a chronosequence to direct measures of accumulation over time.   We present the results of a 15-year study to determine rates of C and N accumulation in restored tallgrass prairies in the Cowling Arboretum of Carleton College, Northfield, Minnesota, USA.  Restored prairies were established annually from 1995-2007. In 2000 and again in 2010, we sampled for plant community composition, soil C and N content and soil bulk density at different soil depths. In 2000 we sampled both restored prairies and those planning to be restored, but at the time still in agriculture. Thus, we are able to determine both the effects of prairie age using the chronosequence of restored fields as well as the change in accumulation rates of C and N through succession.

Results/Conclusions

We observed a 16% average increase in soil C and a 27% average increase in N over the 10-year period.  In both 2000 and 2010, there was no effect of field age across the chronosequence on C and N pools at either the 0-10 or 10-20 cm soil depth. However, rates of soil C and N accumulation per year were significantly related to prairie age.  Both C and N accumulation rates decreased with restoration age, suggesting rapid recovery of soil C and N in the years immediately following agricultural conversion and a decrease in accumulation rates over time.  Accumulation rates of C or N were not predicted by the cover of C₄ grasses or legumes, as both functional groups increased in cover through successional development.  We expect the rapid turnover of annual and biennial species that dominate plant biomass in the first three years following conversion from agriculture contribute significantly to this early increase in soil C and N pools.  This is supported by a significant increase in δ¹³C values with increasing prairie age, suggesting a greater contribution by C₃ species to soil C pools early in succession and a shift to inputs by C₄ species in older prairies.