Xianliang Chen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Xianliang Chen has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
- Jianping Gan
Center for Ocean Research in Hong Kong and Macau, Department of Ocean Science and Department of Mathematics, Hong Kong University of Science and Technology, Hong Kong, China
12 shared publications - Lin Fu
Department of Mechanical and Aerospace Engineering, The Hong Kong University of Science and Technology, Kowloon, Hong Kong
9 shared publications - James C. McWilliams
Department of Atmospheric and Oceanic Sciences University of California Los Angeles CA USA
3 shared publications - Song Fu
Tsinghua University
3 shared publications - Anjia Ying
The Hong Kong University of Science and Technology
2 shared publications - Cheng Cheng
Department of Mechanical and Aerospace Engineering, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong
2 shared publications
Latest Publications
- Open Access
- 17 August 2026
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Oceans
- Vol. 131 (8)
Dynamics of the Kuroshio Separation, Instability, and Associated Eddy Formation off Southern Japan
The Kuroshio is well‐known for its variability south of Japan, where its complex dynamics lead to the formation of meandering jet and eddies along the stream. In this study, we investigated the relevant processes and physical mechanism that remain ambiguous, using a well‐validated high‐resolution China Sea Multiscale Ocean Modeling System and a process‐oriented numerical modeling. We found that this variability along Kuroshio is chiefly governed by the dynamics of the boundary jet separation. Our analysis shows that the variability follows a sequential dynamical pathway. First, the Kuroshio separation is induced by an inverse along‐stream pressure‐gradient force associated with the highly nonlinear jet, which decelerates the nearshore flow and induces a coastal countercurrent. In wind‐favorable events, Ekman transport can further reinforce this countercurrent, thereby promoting separation. Near protruding coastal capes, bottom pressure torque and topographic vorticity adjustment help anchor the separation point. An accumulation of shear vorticity on the shoreside/seaward side of the separated Kuroshio forms a cyclonic/anticyclonic eddy dipole. Through active dynamic adjustment along the jet, the increasing barotropic and baroclinic instabilities strengthen the meandering jet and eddy, thereby maintaining the separation. This study provides new insight into how boundary‐current separation, topographic constraint, wind‐favorable countercurrent formation, and jet instability jointly regulate western boundary current variability.
- 17 August 2026
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Oceans
- Vol. 131 (8)
Open AccessDynamics of the Kuroshio Separation, Instability, and Associated Eddy Formation off Southern Japan
The Kuroshio is well‐known for its variability south of Japan, where its complex dynamics lead to the formation of meandering jet and eddies along the stream. In this study, we investigated the relevant processes and physical mechanism that remain ambiguous, using a well‐validated high‐resolution China Sea Multiscale Ocean Modeling System and a process‐oriented numerical modeling. We found that this variability along Kuroshio is chiefly governed by the dynamics of the boundary jet separation. Our analysis shows that the variability follows a sequential dynamical pathway. First, the Kuroshio separation is induced by an inverse along‐stream pressure‐gradient force associated with the highly nonlinear jet, which decelerates the nearshore flow and induces a coastal countercurrent. In wind‐favorable events, Ekman transport can further reinforce this countercurrent, thereby promoting separation. Near protruding coastal capes, bottom pressure torque and topographic vorticity adjustment help anchor the separation point. An accumulation of shear vorticity on the shoreside/seaward side of the separated Kuroshio forms a cyclonic/anticyclonic eddy dipole. Through active dynamic adjustment along the jet, the increasing barotropic and baroclinic instabilities strengthen the meandering jet and eddy, thereby maintaining the separation. This study provides new insight into how boundary‐current separation, topographic constraint, wind‐favorable countercurrent formation, and jet instability jointly regulate western boundary current variability.
- 17 August 2026
- Published by American Physical Society (APS) in Physical Review Fluids
- Vol. 11 (8)
Total enthalpy transformation in compressible turbulence recovering the incompressible law of the wall
- 17 August 2026
- Published by American Physical Society (APS) in Physical Review Fluids
- Vol. 11 (8)
Total enthalpy transformation in compressible turbulence recovering the incompressible law of the wall
- Open Access
- 14 March 2026
- Published by Copernicus GmbH
On the nonlinear instability and submesoscale turbulence for western boundary currents and undercurrents
Western boundary undercurrents (WBUCs), present beneath almost all western boundary currents, are significant for transporting subsurface mass and energy and connecting regional circulations. Observations suggest that WBUCs can have strong ageostrophy and intra-seasonal variability, but many dynamic details remain unclear. Here, we analyze a representative model idealized from the Kuroshio Current and Luzon Undercurrent (KC and LUC); other instances like the Gulf Stream and the Deep Western Boundary Current are also feasible. A cross-shore mean flow section, extracted from realistic simulations, serves as the only input. We employ biglobal instability analysis (BIA) and high-resolution (up to 500m) regional simulations to reveal the nonlinear instability of the WBUC, its interaction with the upper-layer, and the induced mesoscale and submesoscale turbulence. First, we solidly verify BIA by showing that the predicted evolution of dominant eigenmodes closely agrees with the model results. Second, an upper-layer (depth < 500m) KC mode and a middle-layer (500 to 1500m) LUC mode are identified. The nonlinear instability of the KC mode leads to strong variability and periodic reversal of the LUC. The subthermocline-eddy-like LUC mode has stronger nonlinearity, but negligibly affects the upper layer. Qualitative and, in some cases, quantitative agreement with observations is obtained. The kinetic energy spectra for the subthermocline can exhibit the k scaling as the upper layer, jointly driven by the KC and LUC instability. Moderate centrifugal instability is identified for the LUC near the topography, leading to locally enhanced submesoscales and eddy fluxes. The present model has the potential to serve as a benchmark for global WBUCs, providing theoretical explanations to observational trends and helping improve the modelling for multi-layer circulations.
- 14 March 2026
- Published by Copernicus GmbH
Open AccessOn the nonlinear instability and submesoscale turbulence for western boundary currents and undercurrents
Western boundary undercurrents (WBUCs), present beneath almost all western boundary currents, are significant for transporting subsurface mass and energy and connecting regional circulations. Observations suggest that WBUCs can have strong ageostrophy and intra-seasonal variability, but many dynamic details remain unclear. Here, we analyze a representative model idealized from the Kuroshio Current and Luzon Undercurrent (KC and LUC); other instances like the Gulf Stream and the Deep Western Boundary Current are also feasible. A cross-shore mean flow section, extracted from realistic simulations, serves as the only input. We employ biglobal instability analysis (BIA) and high-resolution (up to 500m) regional simulations to reveal the nonlinear instability of the WBUC, its interaction with the upper-layer, and the induced mesoscale and submesoscale turbulence. First, we solidly verify BIA by showing that the predicted evolution of dominant eigenmodes closely agrees with the model results. Second, an upper-layer (depth < 500m) KC mode and a middle-layer (500 to 1500m) LUC mode are identified. The nonlinear instability of the KC mode leads to strong variability and periodic reversal of the LUC. The subthermocline-eddy-like LUC mode has stronger nonlinearity, but negligibly affects the upper layer. Qualitative and, in some cases, quantitative agreement with observations is obtained. The kinetic energy spectra for the subthermocline can exhibit the k scaling as the upper layer, jointly driven by the KC and LUC instability. Moderate centrifugal instability is identified for the LUC near the topography, leading to locally enhanced submesoscales and eddy fluxes. The present model has the potential to serve as a benchmark for global WBUCs, providing theoretical explanations to observational trends and helping improve the modelling for multi-layer circulations.
- Open Access
- 24 November 2025
- Published by Cambridge University Press (CUP) in Journal of Fluid Mechanics
- Vol. 1023, A40
Temperature–velocity relation for laminar adiabatic and diabatic hypersonic boundary layers
We study the temperature–velocity (TV) relation for laminar adiabatic and diabatic hypersonic boundary layers. By applying an asymptotic expansion to the compressible boundary-layer temperature equation, we derive a first-order equation for the TV relation, where the zeroth-order solution is found to be the classical Crocco–Busemann quadratic relation. The ensuing relation predicts accurately the temperature profile by using the velocity for hypersonic boundary layers with Chapman, power and Sutherland viscosity laws, arbitrary heat capacity ratios, variable Prandtl numbers close to unity and Mach number of up to 10. The Mach-number- and wall-temperature-independent quantities in our relation are also investigated. The present relation has the potential to function as a temperature wall model for laminar hypersonic boundary layers, especially for cold-wall cases.
- 24 November 2025
- Published by Cambridge University Press (CUP) in Journal of Fluid Mechanics
- Vol. 1023, A40
Open AccessTemperature–velocity relation for laminar adiabatic and diabatic hypersonic boundary layers
We study the temperature–velocity (TV) relation for laminar adiabatic and diabatic hypersonic boundary layers. By applying an asymptotic expansion to the compressible boundary-layer temperature equation, we derive a first-order equation for the TV relation, where the zeroth-order solution is found to be the classical Crocco–Busemann quadratic relation. The ensuing relation predicts accurately the temperature profile by using the velocity for hypersonic boundary layers with Chapman, power and Sutherland viscosity laws, arbitrary heat capacity ratios, variable Prandtl numbers close to unity and Mach number of up to 10. The Mach-number- and wall-temperature-independent quantities in our relation are also investigated. The present relation has the potential to function as a temperature wall model for laminar hypersonic boundary layers, especially for cold-wall cases.
- 20 November 2025
- Published by Elsevier in Computers & Fluids
- Vol. 305
Development and assessment of an improved Baldwin-Lomax turbulence model for complex hypersonic flows in two and three dimensions
- 20 November 2025
- Published by Elsevier in Computers & Fluids
- Vol. 305
Development and assessment of an improved Baldwin-Lomax turbulence model for complex hypersonic flows in two and three dimensions
- 1 October 2025
- Published by American Meteorological Society in Journal of Physical Oceanography
- Vol. 55 (10) , 1715-1733
Nonlinear Instability and Subthermocline Turbulence for Western Boundary Undercurrents
Western boundary undercurrents (WBUCs), present beneath almost all western boundary currents, are significant for transporting subsurface mass and energy and connecting regional circulations. Observations suggest that WBUCs can have strong ageostrophy and intraseasonal variability, but many dynamic details remain unclear. Here, we analyze a representative model idealized from the Kuroshio (KC) and Luzon Undercurrent (LUC). A cross-shore mean flow section, extracted from realistic simulations, serves as the only input. We employ biglobal instability analysis (BIA) and high-resolution (up to 500 m) regional simulations to reveal the nonlinear instability of the WBUC, its interaction with the upper layer, and the induced mesoscale and submesoscale turbulence. First, we solidly verify BIA by showing that the predicted evolution of dominant eigenmodes closely agrees with the model results. Second, an upper-layer (depth < 500 m) KC mode and a midlayer (500–1500 m) LUC mode are identified. The nonlinear instability of the KC mode leads to strong variability and periodic reversal of the LUC. The subthermocline-eddy-like LUC mode has stronger nonlinearity, but negligibly affects the upper layer. Qualitative and, in some cases, quantitative agreement with observations is obtained. The kinetic energy spectra for the subthermocline can exhibit the k −3 scaling as the upper layer, jointly driven by the KC and LUC instabilities. Moderate centrifugal instability is identified for the LUC near the topography, leading to locally enhanced submesoscales and eddy fluxes. The present model has the potential to serve as a benchmark for global WBUCs, providing theoretical explanations to observational trends and helping improve the modeling for multilayer circulations.
- 1 October 2025
- Published by American Meteorological Society in Journal of Physical Oceanography
- Vol. 55 (10) , 1715-1733
Nonlinear Instability and Subthermocline Turbulence for Western Boundary Undercurrents
Western boundary undercurrents (WBUCs), present beneath almost all western boundary currents, are significant for transporting subsurface mass and energy and connecting regional circulations. Observations suggest that WBUCs can have strong ageostrophy and intraseasonal variability, but many dynamic details remain unclear. Here, we analyze a representative model idealized from the Kuroshio (KC) and Luzon Undercurrent (LUC). A cross-shore mean flow section, extracted from realistic simulations, serves as the only input. We employ biglobal instability analysis (BIA) and high-resolution (up to 500 m) regional simulations to reveal the nonlinear instability of the WBUC, its interaction with the upper layer, and the induced mesoscale and submesoscale turbulence. First, we solidly verify BIA by showing that the predicted evolution of dominant eigenmodes closely agrees with the model results. Second, an upper-layer (depth < 500 m) KC mode and a midlayer (500–1500 m) LUC mode are identified. The nonlinear instability of the KC mode leads to strong variability and periodic reversal of the LUC. The subthermocline-eddy-like LUC mode has stronger nonlinearity, but negligibly affects the upper layer. Qualitative and, in some cases, quantitative agreement with observations is obtained. The kinetic energy spectra for the subthermocline can exhibit the k −3 scaling as the upper layer, jointly driven by the KC and LUC instabilities. Moderate centrifugal instability is identified for the LUC near the topography, leading to locally enhanced submesoscales and eddy fluxes. The present model has the potential to serve as a benchmark for global WBUCs, providing theoretical explanations to observational trends and helping improve the modeling for multilayer circulations.
- Open Access
- 21 July 2025
- Published by Cambridge University Press (CUP) in Journal of Fluid Mechanics
- Vol. 1015, A40
Uncovering the forcing statistics in stochastic linear models for compressible wall-bounded turbulence
Modelling the nonlinear forcing is critical for linear models based on resolvent or input–output analyses. For compressible wall-bounded turbulence, little is known on what the real forcing looks like due to limited data, so the prediction agrees more qualitatively than quantitatively with direct numerical simulations (DNSs). Here, we present detailed forcing statistics of stochastic linear models, derived from elaborate DNS datasets for channel flows with bulk Mach number reaching 3. These statistics directly explain the success and failure of current models and provide guidance for further improvements. The benchmark linearised Navier–Stokes (LNS) and eLNS models are considered; the latter is assisted by eddy-viscosity-related terms. First, we prove the self-consistency of the models by using DNS-computed forcing as the input. Second, we present the spectral distributions of the forcing and its components. Third, we quantify the acoustic components, absent in incompressible cases, within the linear models. We reveal that the LNS forcing can exhibit relatively high coherence and low rank, very different from the modelled diagonal full-rank forcing. The eddy-viscosity-related term is not partial modelling of the LNS forcing; contrarily, the former is much larger than the latter, serving to disrupt the low-rank feature, enhance diagonal dominance and increase robustness across scales. The scales narrow in either horizontal direction are most susceptible to acoustic modes, while the others are little affected ( ${\lt}2\,\%$ in energy). Furthermore, the extended strong Reynolds analogy is assessed in predicting the density and temperature components.- 21 July 2025
- Published by Cambridge University Press (CUP) in Journal of Fluid Mechanics
- Vol. 1015, A40
Open AccessUncovering the forcing statistics in stochastic linear models for compressible wall-bounded turbulence
Modelling the nonlinear forcing is critical for linear models based on resolvent or input–output analyses. For compressible wall-bounded turbulence, little is known on what the real forcing looks like due to limited data, so the prediction agrees more qualitatively than quantitatively with direct numerical simulations (DNSs). Here, we present detailed forcing statistics of stochastic linear models, derived from elaborate DNS datasets for channel flows with bulk Mach number reaching 3. These statistics directly explain the success and failure of current models and provide guidance for further improvements. The benchmark linearised Navier–Stokes (LNS) and eLNS models are considered; the latter is assisted by eddy-viscosity-related terms. First, we prove the self-consistency of the models by using DNS-computed forcing as the input. Second, we present the spectral distributions of the forcing and its components. Third, we quantify the acoustic components, absent in incompressible cases, within the linear models. We reveal that the LNS forcing can exhibit relatively high coherence and low rank, very different from the modelled diagonal full-rank forcing. The eddy-viscosity-related term is not partial modelling of the LNS forcing; contrarily, the former is much larger than the latter, serving to disrupt the low-rank feature, enhance diagonal dominance and increase robustness across scales. The scales narrow in either horizontal direction are most susceptible to acoustic modes, while the others are little affected ( ${\lt}2\,\%$ in energy). Furthermore, the extended strong Reynolds analogy is assessed in predicting the density and temperature components. - Open Access
- 25 April 2025
- Published by Cambridge University Press (CUP) in Journal of Fluid Mechanics
- Vol. 1009, A39
Mean temperature–velocity relation and a new temperature wall model for compressible laminar and turbulent flows
The well-known quadratic temperature–velocity (TV) relation is significant for physical understanding and modelling of compressible wall-bounded turbulence. Meanwhile, there is an increasing interest in employing the TV relation for laminar modelling. In this work, we revisit the TV relation for both laminar and turbulent flows, aiming to explain the success of the TV relation where it works, improve its accuracy where it deviates and relax its limitation as a wall model for accurate temperature prediction. We show that the general recovery factor defined by Zhang et al. (J. Fluid. Mech., vol. 739, 2014, pp. 392–440) is not a wall-normal constant in most laminar and turbulent cases. The effective Prandtl number $Pr_e$ is more critical in determining the shape of temperature profiles. The quadratic TV relation systematically deviates for laminar boundary layers irrespective of Mach number and wall boundary conditions. We find a universal distribution of $Pr_e$ , based on which the TV relation can be notably improved, especially for cold-wall cases. For turbulent flows, the TV relation as the wall model can effectively improve the near-wall temperature prediction for cold-wall boundary layer cases, but it involves boundary-layer-edge quantities used in the Reynolds analogy scaling, which hinders the application of the wall model in complex flows. We propose a transformation-based temperature wall model by solving inversely the newly developed temperature transformation of Cheng and Fu (Phy. Rev. Fluids, vol. 9, 2024, no. 054610). The dependence on edge quantities is thus removed in the new model and the high accuracy in turbulent temperature prediction is maintained for boundary layer flows.- 25 April 2025
- Published by Cambridge University Press (CUP) in Journal of Fluid Mechanics
- Vol. 1009, A39
Open AccessMean temperature–velocity relation and a new temperature wall model for compressible laminar and turbulent flows
The well-known quadratic temperature–velocity (TV) relation is significant for physical understanding and modelling of compressible wall-bounded turbulence. Meanwhile, there is an increasing interest in employing the TV relation for laminar modelling. In this work, we revisit the TV relation for both laminar and turbulent flows, aiming to explain the success of the TV relation where it works, improve its accuracy where it deviates and relax its limitation as a wall model for accurate temperature prediction. We show that the general recovery factor defined by Zhang et al. (J. Fluid. Mech., vol. 739, 2014, pp. 392–440) is not a wall-normal constant in most laminar and turbulent cases. The effective Prandtl number $Pr_e$ is more critical in determining the shape of temperature profiles. The quadratic TV relation systematically deviates for laminar boundary layers irrespective of Mach number and wall boundary conditions. We find a universal distribution of $Pr_e$ , based on which the TV relation can be notably improved, especially for cold-wall cases. For turbulent flows, the TV relation as the wall model can effectively improve the near-wall temperature prediction for cold-wall boundary layer cases, but it involves boundary-layer-edge quantities used in the Reynolds analogy scaling, which hinders the application of the wall model in complex flows. We propose a transformation-based temperature wall model by solving inversely the newly developed temperature transformation of Cheng and Fu (Phy. Rev. Fluids, vol. 9, 2024, no. 054610). The dependence on edge quantities is thus removed in the new model and the high accuracy in turbulent temperature prediction is maintained for boundary layer flows. - Open Access
- 23 December 2024
- Published by Cambridge University Press (CUP) in Journal of Fluid Mechanics
- Vol. 1001
Optimisation and modelling of eddy viscosity in the resolvent analysis of turbulent channel flows
The eddy-viscosity model, as initially used to model the mean Reynolds stress, has been widely used in the linear analysis of turbulence recently by direct extension. In this study, the mechanism of the eddy viscosity in improving the prediction of fluctuation structures with linear analysis is clarified by investigating the statistical properties of forcing, eddy-viscosity term and their correlations. From the direct numerical simulation (DNS) results of turbulent channel flows with $Re_{\tau }=186$ – $2003$ , the spatial correlation of forcing is partially cancelled due to its interaction with eddy-viscosity terms. The stochastic forcing after excluding the eddy-viscosity term is nearly uncorrelated spatially, which better matches the condition where the resolvent modes are consistent with the spectral proper orthogonal decomposition (SPOD) modes from DNS. With the above findings, an optimisation framework is developed by minimising the spatial correlations of the stochastic forcing. The optimised eddy-viscosity profiles nearly overlap with the mean-quantity-based model in the near-wall region, but have different maximum values. Compared with the mean-quantity-based model, the optimised results enhance the consistency between the resolvent and DNS results significantly. Based on the optimised results, a new modelling framework is further abstracted, leaving only one to-be-modelled parameter of the maximum value of the eddy-viscosity profile. This parameter follows distinctive rules with spanwise flow scales, based on which a simplified predictive model is constructed. The resolvent modes predicted by the new model exhibit high consistency with SPOD modes, which are essentially comparable to the optimised results for wide ranges of streamwise and spanwise scales.- 23 December 2024
- Published by Cambridge University Press (CUP) in Journal of Fluid Mechanics
- Vol. 1001
Open AccessOptimisation and modelling of eddy viscosity in the resolvent analysis of turbulent channel flows
The eddy-viscosity model, as initially used to model the mean Reynolds stress, has been widely used in the linear analysis of turbulence recently by direct extension. In this study, the mechanism of the eddy viscosity in improving the prediction of fluctuation structures with linear analysis is clarified by investigating the statistical properties of forcing, eddy-viscosity term and their correlations. From the direct numerical simulation (DNS) results of turbulent channel flows with $Re_{\tau }=186$ – $2003$ , the spatial correlation of forcing is partially cancelled due to its interaction with eddy-viscosity terms. The stochastic forcing after excluding the eddy-viscosity term is nearly uncorrelated spatially, which better matches the condition where the resolvent modes are consistent with the spectral proper orthogonal decomposition (SPOD) modes from DNS. With the above findings, an optimisation framework is developed by minimising the spatial correlations of the stochastic forcing. The optimised eddy-viscosity profiles nearly overlap with the mean-quantity-based model in the near-wall region, but have different maximum values. Compared with the mean-quantity-based model, the optimised results enhance the consistency between the resolvent and DNS results significantly. Based on the optimised results, a new modelling framework is further abstracted, leaving only one to-be-modelled parameter of the maximum value of the eddy-viscosity profile. This parameter follows distinctive rules with spanwise flow scales, based on which a simplified predictive model is constructed. The resolvent modes predicted by the new model exhibit high consistency with SPOD modes, which are essentially comparable to the optimised results for wide ranges of streamwise and spanwise scales. - 30 October 2024
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Oceans
- Vol. 129 (11)
Parameterization of the Vertical Mixing for the Luzon Undercurrent in the Northern Western Pacific Ocean
The Luzon Undercurrent (LUC) is one of the most significant western boundary undercurrents in the northern western Pacific Ocean (WPO), essential for subsurface water transport and connecting subtropical–equatorial circulations. Over the last three decades, abundant observational progress has been made in revealing the basic features of the LUC. However, the dynamics and thus successful modeling of the LUC remain unresolved. In this work, we conducted a high‐resolution (3 km, 60 levels) numerical investigation of the WPO circulation using the China sea multi‐scale ocean modeling system. We paid particular attention to the vertical mixing, aiming to reasonably resolve the LUC by modifying the vertical mixing parameterization. Based on physics reasoning and experiments of physically based modeling, we designed an adaptive mixing scheme (AMS), which used a Munk‐like function in the thermocline for enhanced vertical mixing in the areas of small Richardson numbers. Using the AMS, we reproduced the two‐layer WPO circulations well and captured the inshore component of the LUC consistent with observations. Furthermore, we studied the dynamics of the LUC by analyzing the momentum balance and found that the LUC is primarily maintained by baroclinic pressure gradient force due to strong thermocline tilting near the western boundary. Enhanced vertical mixing in this highly sheared region crucially provides sufficient geostrophic support to sustain the LUC. Also, nonlinearity and submesoscale motions contribute positively to the LUC. This work advances physical understanding of the current‐undercurrent system and improves numerical capability in capturing the LUC and WPO circulations.
- 30 October 2024
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Oceans
- Vol. 129 (11)
Parameterization of the Vertical Mixing for the Luzon Undercurrent in the Northern Western Pacific Ocean
The Luzon Undercurrent (LUC) is one of the most significant western boundary undercurrents in the northern western Pacific Ocean (WPO), essential for subsurface water transport and connecting subtropical–equatorial circulations. Over the last three decades, abundant observational progress has been made in revealing the basic features of the LUC. However, the dynamics and thus successful modeling of the LUC remain unresolved. In this work, we conducted a high‐resolution (3 km, 60 levels) numerical investigation of the WPO circulation using the China sea multi‐scale ocean modeling system. We paid particular attention to the vertical mixing, aiming to reasonably resolve the LUC by modifying the vertical mixing parameterization. Based on physics reasoning and experiments of physically based modeling, we designed an adaptive mixing scheme (AMS), which used a Munk‐like function in the thermocline for enhanced vertical mixing in the areas of small Richardson numbers. Using the AMS, we reproduced the two‐layer WPO circulations well and captured the inshore component of the LUC consistent with observations. Furthermore, we studied the dynamics of the LUC by analyzing the momentum balance and found that the LUC is primarily maintained by baroclinic pressure gradient force due to strong thermocline tilting near the western boundary. Enhanced vertical mixing in this highly sheared region crucially provides sufficient geostrophic support to sustain the LUC. Also, nonlinearity and submesoscale motions contribute positively to the LUC. This work advances physical understanding of the current‐undercurrent system and improves numerical capability in capturing the LUC and WPO circulations.