On the Computation of Boundary-Layer Height Using a Vriable Critical Bulk Richardson Number

Bulk Richardson number method is one of the most commonly used methods for the parameterization of boundary-layer height. The critical bulk Richardson number, Ri_bc, as a key parameter, is usually chosen as a universal value. However, the latest studies showed that Ri_bc is not a constant, and proposed a linearly stability-dependent Ri_bc and a relationship between Ri_bc and the surface Rossby number in stable boundary-layers, respectively. This study aims to improve the variable Ri_bc algorithm by combining impacts of atmospheric stability and surface roughness, and thus the accuracy of the boundary-layer height parameterization. Experimental data from four field campaigns are used to explore the variability of Ri_bc. The boundary-layer heights of three different thermally-stratified boundary layers (i.e., strongly stable boundary layers, near-neutral boundary layers, and unstable boundary layers) from the four field campaigns are determined using the turbulence method, the potential temperature gradient method, the low-level jet method, or the modified parcel method. The improvement of the Ri_bc algorithm focuses on strongly stable boundary-layers. For near-neutral and unstable boundary layers, the optimal values of Ri_bc are 0.31 and 0.39, respectively, obtained through statistical error minimization methods. The variable Ri_bc scheme proposed by this study will be substituted into Community Earth System Model to compare to the original scheme using a single universal Ri_bc in simulating boundary-layer heights.