The timescale of the eyewall replacement cycle (ERC) is critical for the prediction of intensity and structure changes of tropical cyclones (TCs) with concentric eyewall (CE) structures. Previous studies have indicated that the moat width can regulate the interaction between the inner and outer eyewalls and has a salient relationship with the ERC timescale. In this study, a series of sensitivity experiments are carried out to investigate the essential mechanisms resulting in the diversity of the duration of CEs using both simple and full-physics models. Results reveal that a larger moat can induce stronger inflow under the same inner eyewall intensity by providing a longer distance for air parcels to accelerate in the boundary layer, thus there is greater inward absolute vorticity flux to sustain the inner eyewall. Besides, the equivalent potential temperature budget indicates that the vertical advection and surface flux of moist entropy can overbalance the negative contribution from the horizontal advection and lead to an increasing trend of equivalent potential temperature in the inner eyewall. This suggests that the thermodynamic process in the boundary layer is not indispensable to the inner eyewall weakening. It is also found that the contraction rate of the secondary eyewall, which directly influences the moat width, is subject to the activity of outer spiral rainbands. By directly introducing positive wind tendency outside the eyewall and indirectly promoting a vertically tilted eyewall structure, active convection in the outer region will impede or even suspend the contraction of the outer eyewall hence extending the ERC timescale.

Two numerical models are implemented in this study. One is the improved version of Ooyama-type three-layer model (OM3L) and the other is the full-physics Cloud Model 1 (CM1) of version 21.0. The original code of OM3L is developed by Prof. Yuqing Wang. If readers are interested in the improved OM3L model, please don't hesitate to contact us. The CM1 model is an open-source mesoscale atmospheric model that allows for idealized studies of atmospheric phenomena, whose source code can be downloaded from the wewbsite: https://www2.mmm.ucar.edu/people/bryan/cm1/. Model settings in two models can be refered to the detailed description in the paper and corresponding namelists presented here. 

Since raw datafiles of the simulation are very large, only the output that is most relavent to the research and is required to reproduce figures in the paper are elaborated in this repository. More information about this dataset can be found in the README section.