Impacts of Planetary Boundary Layer Processes and Land-Air-Sea Interaction on the Winter Temperature Simulation with a Regional Climate Model

Abstract

Systematic errors in simulated 2 m air temperature (SAT)—a critical indicator of climate change and a variable that comprehensively reflects the prognostic temperature errors—can lead to inaccurate representations of atmospheric circulation, air quality, and climate feedback mechanisms linked to SAT. Therefore, prioritizing an understanding of RCM systematic errors in climate reproduction experiments, which use reanalysis data as large-scale forcings (i.e., ideal boundary conditions), is essential. This study plays a significant role in identifying and addressing systematic errors in weather research and forecasting (WRF) and SNURCM (WRF's predecessor) model, particularly focusing on high-resolution RCM errors associated with the East Asian Winter Monsoon (EAWM) during boreal winter. The models exhibited more significant cold biases in Manchuria during strong East Asian monsoon years, influenced by large circulation (e.g., land-sea pressure contrast and low-level wind). So, this study attempted to understand the impacts of the PBL processes and land-atmosphere-ocean interactions on systematic temperature errors and to apply several alternatives to mitigate these biases. Firstly, three planetary boundary layer (PBL) parameterization schemes were assessed: Mellor-Yamada-Janjić (MYJ), Yonsei University (YSU), and Asymmetric Convective Model version 2 (ACM2). Results revealed mechanically-driven PBL turbulent mixing, influenced by wind shear and overestimated synoptic variables, emerged as a critical factor affecting lower-layer air temperature, SAT errors, and land surface processes. Weaker PBL mixing reduced nighttime SAT errors by approximately 21% (representing a more realistic circulation), skin temperature errors by around 19% over MC, and nighttime SAT errors decreased by 26% on cold wave days. Second, a regional ocean-atmosphere coupled model (COAWST) was applied to investigate the impact of air-sea interactions on SAT biases over the MC. The coupled atmosphere-ocean model, which simulates SSTs reflected atmosphere-ocean interactions, improved the simulation of SAT (~1K) by balancing energy realistically, thereby reducing the overestimated large-scale circulation caused by unrealistic prescribed SSTs in the atmospheric model. Third, improved soil moisture initial condition on land mitigated cold SAT errors (~1K) due to the reduced latent heat flux released from the surface to the atmosphere at night, highlighting the significance of soil moisture initialization. Improved RCMs would contribute to generating more reliable future climate data, aiding policymakers in formulating effective climate adaptation strategies, although the limited consideration of errors stemming from land surface models and other physical processes like radiation, as well as an incomplete scientific understanding informed by observations.