Recommended citation: Pérez-Alarcón, A., Sori, R., Stojanovic, M., Trigo, R.M., Nieto, R., Gimeno, L. (2024). The origin of precipitation in Zorbas medicane (2018): Assessing different Lagrangian moisture tracking approaches. MedGU Annual Meeting. Barcelona, Spain, 25-28 November 2024.

Abstract

There are several moisture tracking approaches to establish the source-sink relation between evaporation and precipitation. In this line, Lagrangian methods have been widely used due to their ability to provide insights into long-term moisture sources. However, validating such analysis is challenging for the moisture tracking community due to the lack of observations. This study evaluates six combinations of threshold values (SOD08, FAS19, JK22, APA22, APA22_65, APA22_75) available in the Lagrangian atmospheric moisture and heat tracking (LATTIN) tool during the identification of the precipitating moisture sources along the trajectory of the medicane Zorbas, which occurred in the Mediterranean Sea in 2018. With this aim, the air parcel trajectories were extracted from the global outputs of the Lagrangian FLEXPART model fed by the ERA5 reanalysis. Results show an overall agreement in the origin of the precipitating moisture, predominantly coming from the Ionian, Aegean and Black Seas and the Sea of Crete. However, subtle differences among the approaches are observed in the intensity and spatial extent of the moisture sources pattern. Overall, tracking approaches using a critical value of relative humidity (RH) > 80% (SOD08, FAS19 and JK22) to filter precipitating parcels over the target region (the area limited by the outer radius of the medicane) underestimated the total Zorbas-related precipitation, while the most flexible method (APA22) overestimated it. Meanwhile, APA22_75 (RH>75%) almost matched the medicane precipitation from MSWEP and APA22_65 (RH>65%) from ERA5. In addition, by applying a bulk bias correction method using precipitation from the MSWEP and ERA5 datasets, all methods captured the Zorbas precipitation, but some differences were observed in the spatial location of moisture contribution maxima. It is worth noting that the intensity and extent of the bias-corrected pattern depends on the precipitation dataset. Based on our findings, sensitive analysis are required to select the most suitable approach according to the study goals.