Blending of Surface and Rawinsonde Data in Mesoscale Objective Analysis

Abstract : This report describes a simple technique to blend surface and rawinsonde observations in the planetary boundary layer (PBL) using a Barnes- type objective analysis. This goal is to improve the mesoscale detail of boundary layer input fields for a mesoscale numerical weather prediction model. The technique uses both surface and rawinsonde data in the boundary layer, where it is known that mesoscale details are inadequately resolved by the rawinsonde network and where the influence of the ground surface is strongest. Our challenge is to blend the signals contained in the denser surface observations that capture at least some mesoscale detail with those contained in the denser surface observations that capture at least some mesoscale detail with those contained in the boundary layer rawinsonde data. This is accomplished simply by comparing two analyses at the ground, one containing both rawinsonde data-resolvable and surface-data-resolvable information and the other, only the rawinsonde scale information. Differences between these two fields are assumed to be the signals undetected by the rawinsonde network. These mesoscale signals are then incorporated into the PBL analyses as a function of the distance from the ground. Results show a positive impact on boundary layer analyses due to the blending of surface and rawinsonde data, demonstrating that he technique is useful for providing improved mesoscale detail at and near the ground. Also discussed is the relative impact of data density and reference topography on the analysis of model parameters on terrain-following surfaces.