Horizontal variability and boundary-layer modeling

Abstract

Micrometeorologists have traditionally set aside consideration of horizontal variability and have studied boundary-layer structure with horizontal homogeneity. The numerical forecasting of boundary-layer structures, over normally varying terrain and including normal disturbances such as fronts, requires selection of an ‘appropriate’ horizontal scale.

A simple analysis of steady-state balance between horizontal advection and vertical diffusion provides estimates of the vertical scale (or depth) of surface-induced features. The scale height is a function of the horizontal scale of the variations. Models neglecting important terrain scales of length below ~ 1000 km can predict down to levels of ~ 0.5 to 1 km while those that neglect important terrain scales below ~ 100 km can predict down to ~ 0.2 to 0.6 km. Below these levels, any predicted features will be dominated by the vertical diffusion so that they are solutions of a one-dimensional boundary-value problem.

The boundary-induced advection effects dominate free atmosphere advection effects in the lowest few hundred meters as well. This means that if mesoscale advections are resolved and terrain influences are strong, the predictions in the layer ~ 0.2 to 0.8 km can provide mesoscale detail without mesoscale initial conditions above the surface, because the surface forcing will dominate the solution.