Premium
Tropical Pacific Air‐Sea Interaction Processes and Biases in CESM2 and Their Relation to El Niño Development
Author(s) -
Wei HoHsuan,
Subramanian Aneesh C.,
Karnauskas Kristopher B.,
DeMott Charlotte A.,
Mazloff Matthew R.,
Balmaseda Magdalena A.
Publication year - 2021
Publication title -
journal of geophysical research: oceans
Language(s) - English
Resource type - Journals
eISSN - 2169-9291
pISSN - 2169-9275
DOI - 10.1029/2020jc016967
Subject(s) - climatology , mixed layer , entrainment (biomusicology) , sea surface temperature , environmental science , western hemisphere warm pool , pacific ocean , atmospheric sciences , oceanography , geology , physics , rhythm , acoustics
Coupled processes and associated subsurface dynamics near the eastern edge of the Indo/western Pacific (WP) Warm Pool are important for air‐sea interactions involved in tropical Pacific dynamics. We seek to shed light on the physical mechanisms governing air‐sea interactions in the region and the impacts of their biases in models. In this study, we use the Ocean ReAnalysis System 5 (ORAS5) to identify mean‐state biases in the National Center for Atmospheric Research Community Earth System Model version 2 (CESM2) with a particular focus on upper ocean properties and air‐sea interaction processes. We show that the CESM2 has warm and fresh surface biases in the tropical Pacific Ocean, a barrier layer that is too thin in the WP, and an isothermal layer depth (ILD) that is too deep in the eastern Pacific (EP). These biases impact air‐sea interaction processes involved in El Niño development. We compare the strong El Niño events in ORAS5 and CESM2 and show that biases in barrier layer thickness in the WP and in ILD in the EP are significant before the onset of the El Niño events. These biases then influence vertical mixing and entrainment processes, resulting in mixed layer cooling biases. Biases in the sea surface temperature seasonal cycle in the CESM2 also influence the development of the El Niño. We emphasize how the El Niño progression in models can be influenced by its sensitivity to the mean state biases in both subsurface ocean structure and seasonal cycle through local as well as the large‐scale physical processes.