THE LIFE-CYCLE OF THE INTENSE IOP-14 STORM DURING CASP-II .2. SENSITIVITY EXPERIMENTS

In this paper, two consecutive (48- and 54-h) predictions of the IOP-1 4 storm during the Canadian Atlantic Storm Program II (CASP II) with a mesoscale version of the Canadian Regional Finite-Element (RFE) model are used as control runs to investigate the sensitivity of the cyclon e development to different initial conditions and various physical and dynamical processes. It is shown that obtaining appropriate initial c onditions is crucial in helping to improve the operational predictions of initial cyclogenesis and its subsequent amplification, particularl y for travelling disturbances that propagate from upstream data-sparse regions. It is found that at the cyclone's mature stage i) dry dynami cs account for more than 70% of the cyclone's total depth; ii) the RFE model predicts the weakest (+32 hPa) and deepest (-12 hPa) storms (wi th respect to the control-predicted) in the absence of the oceanic sur face characteristics and Greenland topography, respectively; iii) conc urrent surface sensible and latent heat fluxes have relatively weak po sitive impacts (-5 hPa) on the explosive deepening of the storm; iv) a ll experimental storms except for the no-ocean-surface run satisfy the ''oceanic bomb'' criterion; and v) the observed and predicted non-cla ssical frontal structures, such as the cold frontal ''fracture'', the ''bent-back'' warm front, the ''T-bone'' thermal pattern and warm core structure, fail to develop in the absence of the oceanic surface char acteristics. The results reveal that i) the IOP-14 storm is baroclinic ally driven in nature and it is only modulated by other physical proce sses and ii) the explosively deepening nature and the previously docum ented non-classical frontal structures appear to result from weak surf ace drag over the ocean.