Assessing the role of nudging, aerosols and run-up time in medicane simulations with WRF

Medicanes are small-scale cyclones with tropical characteristics that take place in the Mediterranean basin, showing hazardous features such as intense wind gusts and precipitation. Our ability to predict their consequences is of great importance for those cases of medicanes reaching coastal inhabited areas. Succeeding in a precise prediction of their characteristics is heavily subject to getting insight in the fundamental factors that are involved in their genesis, strengthening and maintenance. Given their small nature compared to the synoptic scale, RCMs are specially suitable for the simulation of these storms. However, when using RCMs, there are a number of configurations that must be controlled to specify the way the different physical and chemical mechanisms are solved during the simulation.


In this work, we evaluate the role of three different factors affecting the outcome of WRF, namely the run-up time, the inclusion or not of the on-line simulation of aerosols and the use of spectral nudging. To that end, six different medicanes have been simulated combining different possibilities for the aforementioned factors, resulting in a set of above 360 simulations. Although in principle the on-line simulation of aerosols is expected to have the strongest impact in the simulation of medicanes, it turns out that the run-up time -time delay from the simulation start to the medicane maximum intensity moment- is far more decisive in their successful development than the former. The results are also sensible to the use of spectral nudging, and the three considered factors end up having a considerable impact. Indeed, whereas the majority of their combinations lead to an erratic reproduction of the observed medicanes, there exist some combinations that allow reasonable results, showing that these configurations are in fact interdependent, i.e., the change in the simulation outcome due to a different configuration for one of the factors is dependent on the configuration of the others. This complicates the assessment on the influence of one factor alone, but facilitates gaining insight on the factors that control the genesis and maintenance of medicanes.