Agriculture Reference
In-Depth Information
Conversion of forest and grassland soils to agriculture reduces rates of soil
CH
4
oxidation by 80-90% (Mosier et al. 1991, Smith et al. 2000, Del Grosso
et al. 2000). In the MCSE, CH
4
oxidation in the Conventional system is about
20% of the rate in the Deciduous Forest (Robertson et al. 2000, Suwanwaree and
Robertson 2005). Much of this suppression appears to stem from greater N avail-
ability in cropped soils rather than N fertilizer per se or tillage-induced changes
in soil structure: oxidation was equally low in the unfertilized Biologically Based
system, and fertilizing Deciduous Forest plots immediately reduces oxidation
rates for the period that inorganic N pools are elevated, while tilling them has no
discernible effect (Fig. 12.4; Suwanwaree and Robertson 2005). Gulledge and
Schimel (1998) showed that much of the effect of N appears related to the com-
petitive inhibition of CH
4
oxidation enzymes by ammonium ions. A longer time
period of measurements of GHG fluxes from KBS soils shows, however, some
recovery of CH
4
oxidation in the Biologically Based, Alfalfa, and Early succes-
sional systems 20 years after establishment (Table 12.6), despite relatively high
N availability.
Nitrogen availability alone also does not explain the very slow recovery of CH
4
oxidation rates in abandoned cropland or in cropland converted to unfertilized
perennial crops. After 10 years, there was no recovery of oxidation rates in either
the Poplar system or in the Early Successional community (Robertson et al. 2000)—
two systems in which soil NO
3
−
levels and NO
3
−
leaching rates are vanishingly low
Figure 12.4
. The reduction of methane (CH
4
) oxidation upon soil disturbance and ammo-
nium nitrate fertilization (100 kg N ha
-1
) in the No-till (planted in corn), Mid-successional,
and Deciduous Forest systems of the MCSE. Vertical bars are standard errors of the mean
(SE, n = 3 sites × 7 sampling dates). Different uppercase and lowercase letters represent sig-
nificant treatment differences (p < 0.05) among and within sites, respectively. Modified from
Suwanwaree and Robertson (2005).
