Sensitivities of the Asian Summer Monsoon Simulations to Physical Parameters for the Perturbed Parameter Ensemble of HadGEM3‐GC3.05

The simulation skill of the perturbed parameter ensemble (PPE) of HadGEM3‐GC3.05 on the mean climate pattern of the Asian summer monsoon (ASM) is evaluated in this study. The sensitivities of the model bias to the perturbed parameters are investigated based on metrics. The results show that the PPE mean (PPE‐20M) could effectively capture the general ASM precipitation and wind patterns, with a correlation coefficient of 0.78. PPE‐20M mainly shows positive precipitation biases over the tropical western Pacific, southern slope of the Tibetan Plateau, Indochina Peninsula, and South China and negative precipitation biases over the Indian continent and Bay of Bengal. The magnitude of the precipitation biases is more sensitive than its pattern to the variation of the perturbed parameters. Four parameters (ent_fac_dp, qlmin, ps_cloud‐ph, and psm) are found to be crucial for simulating the ASM precipitation intensity, and their combined effects are related to the simulated precipitation biases. Plain Language Summary The simulation of the Asian summer monsoon (ASM) has attracted much attention in recent decades. However, state‐of‐the‐art coupled models still show similar systematic biases when simulating monsoon precipitation from CMIP3 to CMIP6. Recently, a perturbed parameter ensemble (PPE) method has been applied in the development of HadGEM3‐GC3.05 with the aim of improving the overall simulation skill of the coupled model. The PPE can also be used to understand the related physical processes in the model simulations. Thus, in this study, we aim to evaluate the simulation skills of PPE on the mean climate distributions of ASM. Four parameters were found to be the most critical in the precipitation simulations, which could be potential tunable parameters to reduce the biases of precipitation intensity over the monsoon regions. This study provides possible pathways to reduce model bias that could benefit the model development community. Key Points The perturbed parameter ensemble mean captures the climate pattern of the Asian summer monsoon (ASM) but mainly overestimates precipitation over the western Pacific Four parameters (ent_fac_dp, qlmin, ps_cloud‐ph, and psm) are crucial for simulating ASM precipitation in coupled run The joint effect of these parameters, especially the deep entrainment amplitude, are dominant control for the bias in the western Pacific