SPM2:: A simulation model for slash pine (Pinus elliottii) forests

A simulation model for closed-canopy slash pine forests (SPM2) includes extensions to a previously described model. Additions to SPM2 include incorporation of major understory species, addition of site water balance components, and replacing the daily time-step decomposition and growth respiration fluxes with hourly outputs. The model extensions are intended to increase generality and to provide output that can be directly compared with eddy covariance measurements. Sensitivity analysis was performed using a Latin Hypercube sampling scheme with parameter ranges of 1-5% around the nominal values. The Latin Hypercube method divides the range of each input parameter into equal probability class intervals. This approach ensures that the entire range of each parameter will be evenly sampled and the distribution of each parameter will be better represented with fewer samples. Of the 69 parameters tested, 29 contributed 5% or more to model output variable sensitivity. Foliage mass parameters affected 16 output variables, including gas exchange quantities and tree biomass variables. Sampling parameters at the +/-5% level resulted in only one additional parameter (initial soil water content) influencing output above the threshold Value of 5%. Simulated annual net pine canopy assimilation was 1236 g C m(-2), annual pine maintenance respiration was 523 g C m(-2), and pine growth respiration was 91 g m(-2). Annual understory assimilation was 14% of the pine value. SPM2 simulations of all combinations of a range of atmospheric CO2 concentrations from preindustrial to 750 ppm, temperature increases of 0-5 degrees C, and precipitation changes of +/-40% were used to generate response surfaces for the model. Although there was a positive annual stem increment at all combinations of conditions tested, the labile carbon pool (starch and soluble sugars) was completely depleted by the end of the simulated year under many combinations of increased temperature and reduced or normal precipitation. (C) 2000 Elsevier Science B.V. All rights reserved.