Estimating cloud water deposition to subalpine spruce-fir forests—II. Model testing

Abstract

A modified version of a previously published steady-state model for computing cloud water deposition to a subalpine balsam fir ( Abies balsamea) forest was tested using water throughfall data collected in a red spruce ( Picea rubens) forest on Whitetop Mountain, Virginia. Detailed wind data were collected in two distinctly different spruce stands to define airflow conditions within the forest canopy. Other meteorological and canopy structure data were also collected for use as inputs to the deposition model. Model simulations of cloud deposition during 11 cloud events in the two forest stands revealed that the model performed best when site-specific wind speed profiles and droplet size spectra were used along with an experimental droplet collection efficiency scheme that treats the densest portions of trees as bulk collectors (as opposed to modelling collection efficiency for individual tree components). An analysis of residuals indicated that model errors were most strongly correlated with cloud liquid water content ( W), a model input. It is speculated that the correlation with W was due to a combination measurement bias when clouds were thin or intermittent and a model computational bias when the potential (defined by the model) for the vertical turbulent flux of cloud water was high. Overall, computed values (using the optimally-configured model) of net cloud water flux tended to exceed measured throughfall rates by 20–30%, and the model explained 38–68% of the variance in throughfall rate. A comparison between the mechanistic model and a simpler empirical model indicated that the mechanistic model performed no better than its empirical counterpart.