Investigation of the Effect of Number of Time Steps on Ice Shapes Calculated by an Ice Accretion Code

摘要

A brief investigation was conducted of the effect of the number of time steps used by the ice accretion code LEWICE 2.0 in determining an ice shape. The investigation included a study of the LEWICE 2.0 validation database and approximately 30 additional LEWICE 2.0 rims. For more than 90 percent of LEWICE 2.0 cases in the validation database, the number of time steps used in calculating the ice shape was different from the 'default' number of time steps determined by LEWICE 2.0 with the automatic time-stepping feature on. In all these cases, the number of time steps was equal to the IFLO input value, which was determined in accordance with a 'minutes rule'. The authors recommend that the minutes rule be incorporated into LEWICE 2.050 that the default number of time steps determined by LEWICE 2.0 is consistent with the number of time steps used for the runs in the validation database. The accuracy of LEWICE 2.0 when using the current default number cannot be determined from the validation database. It was also found that the accuracy of LEWICE 2.0 ice shapes does not improve when the number of time steps is increased beyond a certain value. Specifically, when the number of time steps is increased beyond about 25 or 30, depending on the conditions, the ice shape predictions tend to diverge, and the upper horn (if present) tends to 'droop'. The authors recommend that the LEWICE 2.0 user be warned against inputting too large a number of time steps.