Changes in net ecosystem productivity and greenhouse gas exchange with fertilization of Douglas fir: Mathematical modeling in ecosys

Abstract

The application of nitrogen fertilizers to Douglas fir forests is known to raise net ecosystem productivity (NEP), but also N 2 O emissions, the CO 2 equivalent of which may offset gains in NEP when accounting for net greenhouse gas (GHG) exchange. However, total changes in NEP and N 2 O emissions caused by fertilizer between times of application and harvest, while needed for national GHG inventories, are difficult to quantify except through modeling. In this study, integrated hypotheses for soil and plant N processes within the ecosystem model ecosys were tested against changes in CO 2 and N 2 O fluxes recorded with eddy covariance (EC) and surface flux chambers for 1 year after applying 20 g N m -2 of urea to a mature Douglas fir stand in British Columbia. Parameters from annual regressions of hourly modeled versus measured CO 2 fluxes conducted before and after fertilization were unchanged (b = 1.0, R 2 = 0.8, RMSD = 3.4 mol m -2 s -1), indicating that model hypotheses for soil and plant N processes did not introduce bias into CO 2 fluxes modeled after fertilization. These model hypotheses were then used to project changes in NEP and GHG exchange attributed to the fertilizer during the following 10 years until likely harvest of the Douglas fir stand. Increased CO 2 uptake caused modeled and EC-derived annual NEP to rise from 443 and 386 g C m -2 in the year before fertilization to 591 and 547 g C m -2 in the year after. These gains contributed to a sustained rise in modeled wood C production with fertilization, which was partly offset by a decline in soil C attributed in the model to reduced root C productivity and litterfall. Gains in net CO 2 uptake were further offset in the model by a rise of 0.74 g N m -2 yr -1 in N 2 O emissions during the first year after fertilization, which was consistent with one of 1.05 g N m -2 yr -1 estimated from surface flux chamber measurements. Further N 2 O emissions were neither modeled nor measured after the first year. At the end of the 11 year model projection, a total C sequestration of 1045 g C m -2 was attributed to the 20 g N m -2 of fertilizer. However, only 119 g C m -2 of this was sequestered in stocks that would remain on site after harvest (foliage, root, litter, soil). The remainder was sequestered as harvested wood, the duration of which would depend on use of the wood product. The direct and indirect CO 2 -equivalent costs of this application, including N 2 O emission, were estimated to offset almost all non-harvested C sequestration attributed to the fertilizer. Copyright 2010 by the American Geophysical Union.