Evaluation of some daytime boundary-layer forecast techniques

Gareth Berry

University of Reading

Simplified theories of the convective boundary layer under clear skies suggest a temperature profile that changes from an approximate isothermal to adiabatic state during the course of a day, in response to surface heating. This theory is used by some forecast models to predict convective boundary layer evolution. The performance of two forecast techniques is assessed by comparing model predictions to observations.

The first technique uses an empirical method to predict the maximum daytime temperatures from a pre-dawn temperature profile. It was found to give a mean forecast error of 0.06 degrees Celsius, which verifies that the underlying assumptions are of practical use.

The second technique examined was a numerical model used to predict the vertical extent of the convective boundary layer. The results were of the same order of magnitude as the observations, but no definite conclusions could be made due to large errors associated with the implementation of the model.