Abstract Research and operational numerical weather prediction models rely on bulk-layer parameterization techniques - primarily, the Monin-Obukhov Similarity theory - to compute vertical turbulent fluxes of momentum within the atmospheric surface layer (ASL). In this way, the aerodynamic roughness length and consequently the turbulent drag over land is assumed to be an intrinsic property of the surface, ignoring characteristics of the overlying flow. Although recognized to be invalid near heterogeneous surfaces, to date, no suitable alternatives have been developed for ASL parameterization near coastal environments. In these regions, drastic spatial gradients in surface thermal and roughness properties drive cross-coastal flows, leading to phenomena that directly contradict bulk-flux assumptions. Here, we define a flow-dependent, local, inland coastal aerodynamic roughness length Z₀₂ for onshore flow conditions. Analysis of observations collected from a cross-shore array of inland flux towers anchored at the Monterey Bay, CA coastline from June to October 2021 during the Coastal Land-Air-Sea Interaction (CLASI) campaign reveals significant departures in Z₀₂ from the expected homogeneous values for increasing wind speeds and inland fetches within 8 kilometers of the coast. These findings inform development of a physical framework describing a non-dimensional Z₀₂ as a function of the residence time of the inland flow, a reference height, and a representative homogeneous roughness length. We explore these relationships using large-eddy simulations of a coastal onshore flow scenario to achieve general understanding of the spatial variability in Z₀₂. Finally, we present a baseline empirical relationship for Z₀₂ based on the CLASI dataset under near-neutral, onshore flow conditions.
Hlywiak et al. (2026) studied this question.