Comparison of synoptic-based preprocessor estimates

Abstract Short-range atmospheric dispersion models for environmental impact assessment usually contain meteorological preprocessors to determine boundary layer parameters based upon routine, synoptic observations. In this study, two such meteorological preprocessors are compared with field measurements from the NOPEX campaign during spring and summer at two sites in central Sweden. The two preprocessors considered are the ones used at the Finnish Meteorological Institute (FMI) and at the National Environmental Research Institute, Denmark (OML). The FMI scheme is based upon the original scheme by Holtslag and van Ulden, and the OML scheme is based on the Berkowicz and Prahm scheme. For the friction velocity, the comparisons show high correlations. However, for extremely stable atmospheric conditions, OML generates a large number of cases with near-zero friction velocities due to the roughness length dependent limiting of the Monin–Obukhov length. The sensible heat flux and net radiation is, in general, well captured by both preprocessors. Higher values tend to be underpredicted. Especially FMI exhibits an upper cut-off behavior, since net radiation was underestimated due to a lack of hourly sunshine duration data. The Monin–Obukhov length comparisons indicate that FMI gives a slower decline for the number of near-neutral situations on the stable side than OML.