A time-dependent, one-dimensional numerical model of the boundary layer at individual stations is solved by a finite-difference technique for the St. Patrick's Day storm of 1965. Using only surface pressure and temperature gradients at 3-hr intervals as input, the model predicts the changing boundary layer wind profile at each of 225 grid points covering the central and eastern United States. Predicted winds are used to evaluate horizontal flow structure, vertical motion, three-dimensional trajectories, and flow acceleration. Modeled fields are compared to those corresponding to simple steady-state solutions and to observed data.
The solution of the equations of motion retains the local acceleration and baroclinity, omits the nonlinear advection terms, and contains a damping effect in order to control the inertial-like oscillations which have complicated previous models. With the damping included, the model reasonably predicts the location of the low-level jet core, the vertical motion distribution, and, qualitatively, the orientation of the local acceleration in the boundary layer. The model-forecast ascent is well correlated with the observed precipitation rate when adequate surface moisture is available.