This study presents a revised surface layer formulation for the Weather Research and Forecasting (WRF) model, aiming to improve the simulation of atmospheric boundary layer (PBL) processes under various stability conditions. The revised formulation replaces the Kansas-type similarity functions with those proposed by Fairall et al. (1996) and Cheng and Brutsaert (2005), which provide a more accurate representation of the Monin–Obukhov similarity theory across a broader range of atmospheric stabilities. Additionally, the revised formulation removes or reduces limits on certain variables, such as the friction velocity $ u_* $, to enhance consistency and reduce undesirable effects. The revised scheme improves the simulation of turbulent fluxes, which are more efficient during the day and less efficient at night. This leads to a sharper afternoon transition in the PBL, with significant impacts on meteorological variables. The revised formulation also allows for a more accurate representation of near-surface diagnostic variables, such as wind speed, temperature, and specific humidity, when compared to observations from a mesoscale network in the northeast of the Iberian Peninsula. The revised surface layer formulation addresses several limitations of the original scheme, including the use of Kansas-type similarity functions and the imposition of artificial limits on variables. These limitations were found to affect the self-consistency of surface layer variables and the accuracy of stability definitions. By removing or reducing these limits, the revised formulation improves the relationship between the bulk Richardson number $ R_i_b $ and the Obukhov length $ z/L $, leading to a more accurate representation of atmospheric stability. The revised formulation also introduces changes to the bulk transfer coefficients, which are calculated using the integrated similarity functions. These changes allow for a more accurate representation of the transfer coefficients for momentum, heat, and moisture, leading to improved simulations of turbulent fluxes and near-surface variables. The revised surface layer formulation was tested through a series of numerical simulations, showing significant improvements in the simulation of atmospheric boundary layer processes. The results indicate that the revised formulation provides a more accurate representation of the surface layer under various stability conditions, leading to better agreement with observations. The revised formulation also reduces the restrictions on variables such as $ u_* $, leading to a more consistent and accurate representation of the surface layer.This study presents a revised surface layer formulation for the Weather Research and Forecasting (WRF) model, aiming to improve the simulation of atmospheric boundary layer (PBL) processes under various stability conditions. The revised formulation replaces the Kansas-type similarity functions with those proposed by Fairall et al. (1996) and Cheng and Brutsaert (2005), which provide a more accurate representation of the Monin–Obukhov similarity theory across a broader range of atmospheric stabilities. Additionally, the revised formulation removes or reduces limits on certain variables, such as the friction velocity $ u_* $, to enhance consistency and reduce undesirable effects. The revised scheme improves the simulation of turbulent fluxes, which are more efficient during the day and less efficient at night. This leads to a sharper afternoon transition in the PBL, with significant impacts on meteorological variables. The revised formulation also allows for a more accurate representation of near-surface diagnostic variables, such as wind speed, temperature, and specific humidity, when compared to observations from a mesoscale network in the northeast of the Iberian Peninsula. The revised surface layer formulation addresses several limitations of the original scheme, including the use of Kansas-type similarity functions and the imposition of artificial limits on variables. These limitations were found to affect the self-consistency of surface layer variables and the accuracy of stability definitions. By removing or reducing these limits, the revised formulation improves the relationship between the bulk Richardson number $ R_i_b $ and the Obukhov length $ z/L $, leading to a more accurate representation of atmospheric stability. The revised formulation also introduces changes to the bulk transfer coefficients, which are calculated using the integrated similarity functions. These changes allow for a more accurate representation of the transfer coefficients for momentum, heat, and moisture, leading to improved simulations of turbulent fluxes and near-surface variables. The revised surface layer formulation was tested through a series of numerical simulations, showing significant improvements in the simulation of atmospheric boundary layer processes. The results indicate that the revised formulation provides a more accurate representation of the surface layer under various stability conditions, leading to better agreement with observations. The revised formulation also reduces the restrictions on variables such as $ u_* $, leading to a more consistent and accurate representation of the surface layer.