Defining Boundary Layer Height from Aircraft and Rawinsonde Measurements

The atmospheric boundary layer height (BLH) is frequently estimated from measurements as well as from model outputs. From measurements, the most commonly used data for obtaining BLH is the vertical profiles from rawinsonde measurements. Within mesoscale or regional scale models, BLH is diagnosed based on the local bulk Richardson number, which is also based on mean profiles of the forecast variables. Here we evaluate the limitations of several frequently-used approaches for defining BLH from vertical profiles of the boundary layer mean properties and identify the optimal threshold value for each method if vertical profiles of mean quantities are the sole source of measurements. Aircraft measurements from four field projects are used, representing a variety of BL conditions including the stable boundary layer (SBL), convective boundary layer (CBL) and cloud-topped boundary layer (CTBL) over different underlying surfaces. The BLHs detected from these methods were validated against the ‘true' BLH from aircraft sounding profiles where the BLH is defined as the top of the layer with significant turbulence. Our results suggest that the Richardson number method is inadequate or not performing as well as others in most of the cases even at low vertical resolution. Although the detection rate and the adequate detection criteria vary for different type of boundary layers, the temperature gradient method seems to provide good results for most types of the boundary layers. The optimal range of the detection criteria is given for all boundary layer types examined in this study.