Canadian Forest Service Publications
Evaluating the impacts of climate variability and disturbance regimes on the historic carbon budget of a forest landscape. 2013. Chen, B.; Arain, M.A.; Khomik, M.; Trofymow, J.A.; Grant, R.F.; Kurz, W.A.; Yeluripati, J.; Wang, Z. Agricultural and Forest Meteorology. 180: 265-280.
Year: 2013
Issued by: Pacific Forestry Centre
Catalog ID: 35435
Language: English
Availability: PDF (request by e-mail)
Available from the Journal's Web site. †
DOI: 10.1016/j.agrformet.2013.06.002
† This site may require a fee
Abstract
Landscape-level understanding of forest carbon (C) dynamics is required to quantify the net contributionof forest biomes to the global C cycle and to help forest managers to understand the impacts of forestmanagement activities on the C sequestration in forests. In this study, the effects of interannual climatevariability, carbon dioxide (CO2) fertilization, and disturbance regimes on the C dynamics of an old-growth Pacific Northwest temperate conifer forest landscape (2500 ha) were studied from 1920 to 2005,using a process-based land surface model, known as the Carbon and Nitrogen coupled Canadian Land Sur-face Scheme (CN-CLASS). The model was parameterized with ecological, forest inventory and historicalland-use data, and run using historical meteorological observations. Before performing landscape levelsimulations, model results were evaluated against eddy covariance flux tower observations. Simulatedmean annual net ecosystem productivity (NEP) over the flux tower footprint area was 340 g C m−2yr−1from 1998 to 2005, while the measured value was 293 ± 20.5 g C m−2yr−1. When two anomalous weatheryears, corresponding to El Ni˜no (1998) and La Ni˜na (1999) events, were excluded while performing sta-tistical analysis, measured and simulated fluxes were highly, but negatively, correlated to both annualmean air temperature and annual precipitation (R2= 0.69 and 0.60, respectively). Interannual variabilityof simulated and measured NEP over the flux footprint area, calculated as deviations of the respectiveannual NEP values, was 143 g C m−2yr−1and 61 g C m−2yr−1, respectively. On the landscape-level, priorto disturbance in 1920, simulated C fluxes indicated that the forest landscape was close to C neutral,with annual net biome productivity (NBP) of 0.8 g C m−2yr−1. However, during the intense disturbanceperiod from 1938 to 1944, landscape-level NBP reached about −5083 g C m−2yr−1. Then from 1951 to1997, when there were no major disturbance events, NBP gradually recovered to about 365 g C m−2yr−1.At the end of the study period, in 2005, the landscape again became C source, due to harvesting of secondgrowth stands that occurred from late 1990s to 2000s.
The regression of age-detrended variations in the simulated annual C fluxes to mean daily maximumair temperature over the peak growing season (July–September), during an undisturbed period from1963 to 1984, indicated that summer temperature was the dominant climatic control on landscape-levelC fluxes. Higher temperatures caused a decrease in gross primary productivity at almost twice the rateof increase in ecosystem respiration (i.e. 27 g C m−2yr−1C−1versus 15.7 g C m−2yr−1C−1, respectively).A sensitivity analysis to evaluate the impacts of climate variability and disturbances showed that therelative effect of disturbance on carbon stocks was greater than the effect on carbon fluxes. Overall CO2fertilization effects were minor. Disturbance type and severity, represented by the standard deviation inNBP, as described in the model, determined the magnitude of the simulated C losses to the atmosphere.This study enhances our understanding of the impacts of future climate change and forest managementon landscape-level C dynamics in forests.