Forecasting landscape-level carbon sequestration using gridded, spatially adjusted tree growth

We apply a method for forecasting carbon sequestration at the landscape-level, accounting for spatial and temporal scaling issues and develop formulae to incorporate spatial variability in growth and senescence functions. The effect of environmental variability was modelled by: (a) using a relationship between stand height at age 50 years and environmental characteristics; (b) adjusting the apparent age for different environments to reflect effects on growth rate; (c) adjusting age-dependent volume to reflect effects on potential biomass. Carbon sequestration for the next 250 years was forecast for a Eucalyltus regnans-dominated water catchment reserve in Australia; stand ages ranged from 20 to 450 years. In the absence of fire or succession after 2003, the total carbon sequestered in the E. regnans stands increased by 130( 65) t-C/ha (10.3%) to a peak of 1275( 130) t-C/ha in the year 2130( 50), followed by a net efflux of carbon. However, with fire or species succession, the mixture of young and old stands maintains a long term, stable amount of sequestered carbon. A gauge of the magnitude of the bias in the landscape-level carbon accounts arising from spatial averaging of the model's input data was obtained. Comparison of results using different parameter settings for characteristics such as senescence and understorey biomass, revealed where more field data would allow both the timing and size of the maximal carbon sequestration, and the size of the ensuing net efflux of carbon, to be determined more accurately. (C) 2004 Elsevier B.V. All rights reserved.