Jmh. Knops et D. Tilman, Dynamics of soil nitrogen and carbon accumulation for 61 years after agricultural abandonment, ECOLOGY, 81(1), 2000, pp. 88-98
We used two independent methods to determine the dynamics of soil carbon an
d nitrogen following abandonment of agricultural fields on a Minnesota sand
plain. First, we used a chronosequence of 19 fields abandoned from 1927 to
1982 to infer soil carbon and nitrogen dynamics. Second, we directly obser
ved dynamics of carbon and nitrogen over a 12-yr period in 1900 permanent p
lots in these fields. These observed dynamics were used in a differential e
quation model to predict soil carbon and nitrogen dynamics. The two methods
yielded similar results. Resampling the 1900 plots showed that the rates o
f accumulation of nitrogen and carbon over 12 yr depended on ambient carbon
and nitrogen levels in the soil, with rates of accumulation declining at h
igher carbon and nitrogen levels. A dynamic model fitted to the observed ra
tes of change predicted logistic dynamics for nitrogen and carbon accumulat
ion. On average, agricultural practices resulted in a 75% loss of soil nitr
ogen and an 89% loss of soil carbon at the time of abandonment. Recovery to
95% of the preagricultural levels is predicted to require 180 yr for nitro
gen and 230 yr for carbon. This model accurately predicted the soil carbon,
nitrogen, and carbon : nitrogen ratio patterns observed in the chronoseque
nce of old fields, suggesting that the chronosequence may be indicative of
actual changes in soil carbon and nitrogen.
Our results suggest that the rate of carbon accumulation was controlled by
the rate of nitrogen accumulation, which in turn depended on atmospheric ni
trogen deposition and symbiotic nitrogen fixation by legumes. Our data supp
ort the hypothesis that these abandoned fields initially retain essentially
all nitrogen and have a closed nitrogen cycle. Multiple regression suggest
s that vegetation composition had a significant influence on the rates of a
ccumulation of both nitrogen and carbon; legumes increased these rates, and
C-3 grasses and forbs decreased them. C-4 grasses increased the C:N ratio
of the soil organic matter and thereby increased the rate of carbon accumul
ation, but not nitrogen accumulation.