Abstract This study investigates the complex dynamics of burst evolution in the Australian summer monsoon under present‐day and warmer climate conditions. Using reanalysis data, historical atmosphere‐only simulations (AMIP), and simulations with uniformly increased sea surface temperatures (+4K; AMIP+4K), we examine how seasonal‐mean precipitation and burst behavior respond to warming across eight climate models. A consistent link emerges between higher seasonal‐mean precipitation and a greater number of bursts. A moist static energy budget analysis reveals that horizontal moisture advection governs burst evolution, while model‐specific parameterizations add variability in burst characteristics. Some models show gradual transitions governed by horizontal moisture advection, others display abrupt shifts absent in reanalysis. Among the three models with AMIP+4K simulations, MRI‐ESM2‐0 and BCC‐CSM2‐MR reproduce seasonal precipitation patterns close to their AMIP counterparts, with burst frequencies nearly unchanged in BCC‐CSM2‐MR and slightly reduced in MRI‐ESM2‐0. Only MRI‐ESM2‐0 captures a gradual, physically consistent burst evolution resembling reanalysis, whereas BCC‐CSM2‐MR still shows abrupt transitions. IPSL‐CM6A‐LR, the third model with AMIP+4K simulations, shows a gradual evolution but produces stronger mean precipitation and fewer bursts—contrary to the typical positive precipitation–burst relationship. Cluster‐based composites show that precipitation changes arise from differing contributions of thermodynamic and dynamic components of vertical moisture advection. In MRI‐ESM2‐0, enhanced precipitation in warming scenarios is mainly driven by stronger thermodynamic moistening, whereas in IPSL‐CM6A‐LR, dynamically induced drying offsets thermodynamic gains, particularly in specific burst clusters. These results emphasize model‐dependent monsoon burst responses and the need to separate dynamic and thermodynamic influences to strengthen confidence in future hydro‐climate projections.
Mohanty et al. (Thu,) studied this question.