Modeling the sensitivity of tropical cyclone Batsirai simulations to planetary boundary layer schemes

Tropical cyclones (TCs) in the Southwest Indian Ocean (SWIO) often cause severe socio-economic impacts, underscoring the need for accurate atmospheric modeling to improve forecasting and disaster preparedness. Despite advances in numerical modeling, persistent errors in simulating TC intensity, structure, and rapid intensification (RI) highlight uncertainties in physical parameterizations, particularly within the planetary boundary layer (PBL). This study examines the sensitivity of the Model for Prediction Across Scales–Atmosphere (MPAS-A) to PBL schemes and horizontal resolution in simulating TC Batsirai (2022). Six experiments were conducted using the Yonsei University (YSU) and Mellor–Yamada–Nakanishi–Niino (MYNN) PBL schemes at a uniform 60 km resolution and a variable-resolution mesh refined to 3 km over the SWIO. Additional tests explored MYNN closure levels (2.5 and 3.0) and two physics suites: mesoscale reference (MR) and convection-permitting (CP). Simulations were evaluated against IBTrACS, JTWC, RSMC, ERA5, and CFSR datasets, focusing on track evolution, landfall timing and location, intensity, wind–MSLP coupling, and key thermodynamic and dynamical structures. MPAS-A reproduced the large-scale track of TC Batsirai reasonably well but exhibited a northward drift and weaker peak intensity than observed. Higher resolution strengthened the simulated storm and improved wind–pressure coupling, particularly with MYNN. In contrast, YSU at higher resolution increased track errors. RI was underestimated across all configurations. Raising the MYNN closure level from 2.5 to 3.0 enhanced vertical mixing and storm intensity but shifted landfall southward. At 3 km resolution, both physics suites produced realistic tracks, with MR yielding stronger storms and tighter wind–pressure coupling than CP. High-resolution simulations combined with refined PBL parameterizations better captured Batsirai’s vertical structure, warm-core development, and updraft strength during peak intensification. These findings demonstrate that careful selection of PBL schemes and model resolution is critical for improving TC simulations over the SWIO.