A multilayer biochemical dry deposition model 2. Model evaluation

Abstract

The multilayer biochemical dry deposition model (MLBC) described in the accompanying paper was tested against half‐hourly eddy correlation data from six field sites under a wide range of climate conditions with various plant types. Modeled CO2, O3, SO2, and H2O (latent heat) fluxes were compared with measurements. Model outputs have good correlations with measurements at all locations. Correlation coefficients between model outputs and measurements at all sites range from 0.72 to 0.96 for CO2 flux, from 0.84 to 0.98 for H2O flux, from 0.77 to 0.95 for O3 flux, and from 0.36 to 0.86 for SO2 flux. Model sensitivity analyses were conducted to investigate the variation of model outputs due to measurement errors of input variables and to changes of environmental conditions such as changes in weather conditions. The MLBC outputs were also compared with outputs from the Multilayer Model (MLM) model [Meyers et al., 1998] and the Regional Acid Deposition Model (RADM) [Wesely, 1989] at two typical sites. Comparisons show that the MLBC performs better than the other two models. The model is suitable for use in nationwide dry deposition networks, for example, the Clean Air Status And Trends Network (CASTNet). It can be used to assist in describing total pollutant loadings to major ecosystems. With some modifications, the model may also be suitable for inclusion in region (meso‐) scale numerical models, for example, the Community Multiscale Air Quality (CMAQ) model.