abstract
-
Nitrogen (N) controls on carbon and water exchanges were analyzed in a 70-year old eastern temperate conifer forest in Ontario, Canada from 2003 to 2007 using a newly developed nitrogen (N) cycle coupled model -- CLASS-CTEMᴺ⁺. This process-based model incorporates sunlit and shaded big-leaves for C3 and C4 photosynthesis and semi-mechanistic canopy conductance formulation for dynamic plant-functional-types. Recently, key soil and plant N cycling algorithms have also been included (e.g., biological fixation, atmospheric N deposition, fertilization, mineralization, nitrification, denitrification, leaching, soil nitrous dioxide (N₂0) emissions, root N uptake, plant N allocation and N controls on plant photosynthesis capacity). The simulated values of soil-plant N contents and processes rates including N₂0 fluxes were generally in agreement with observations.
Comparison of default non-N and C&N-coupled model simulations clearly revealed N controls on photosynthetic uptake and water loss. Predictions of daily gross ecosystem productivity (GEP), ecosystem respiration (Re), net ecosystem productivity (NEP) and evapotranspiration (ET) showed better agreement with eddy covariance (EC) flux measurements when using the N-coupled model (RMSE of 1.97, 0.73, 1.44, 0.92; and MAE of 1.48, 0.55, 1.01, 0.60 for GEP, Re, NEP, and ET, respectively; n=1825) as compared to the non-N model simulations (RMSE of2.95, 1.35, 1.93, 1.03; MAE of2.38, 1.15, 1.55,0.71 for GEP, Re, NEP, and ET, respectively; n=1825) over 5 years (2003-2007). Annual values of N-coupled model simulated NEP were 134, 195, 183, 225 and 255 g C m⁻² yr⁻¹ for 2003-2007, as compared to non-N model simulated annual NEP values, which were 535, 562, 507, 540, and 535 g C m⁻² yr⁻¹ for respective years. These values were compared to measured NEP values of 220±67, 126±67, 33±67, 142±67 and 102±67 g C m⁻² yr⁻¹ for the years 2003-2007, respectively. The difference between N-coupled model simulated and EC measured annual variations of carbon exchanges was largely due to specific extreme weather events (e.g. drought, spring warming) during certain years. Overall, the impacts of N limitations on carbon fluxes were more pronounced during early spring, late autumn and winter seasons. This newly developed model will help to evaluate the response of terrestrial vegetation ecosystems to N variations under different scenarios for future climate change.