MODELING THE L BAND ACTIVE AND PASSIVE MICROWAVE SIGNATURES OF PERENNIAL FIRN AQUIFER
Marzo 25, 2026CROSS-LOCATION NEURAL NETWORK TRANSFER LEARNING FOR RAIN ATTENUATION ESTIMATION AT MILLIMETER-WAVELENGTHS
Marzo 25, 2026S. Sebastianelli1, L. P. D’Adderio1, D. D’Armiento1, P. Sanò1, D. Casella1, D. Herndon2, G. Panegrossi1
1CNR-ISAC, 2University of Wisconsin-Madison – CIMSS
Mediterraean Hurricanes (hereafter medicanes) belong to a particular subcategory of Mediterranean cyclones that often cause devastating floods, storm surges and windstorms, leading to significant disruption and casualties (despite their intensity rarely exceeds category 1 on the Saffir-Simpson hurricane scale). During their mature phase, they acquire a tropical cyclones (TCs) typical features such as a warm core (WC), a cloud free region (i.e. the eye) of quasi-calm winds around the center surrounded by spiraling rain bands, and a closed cyclonic circulation. Generally, they originate as extra-tropical systems showing a cut-off from the main flow allowing the intrusion of relatively warm stratospheric air leading to a top-bottom WC development. In certain instances, during the mature phase medicanes show a tropical-like transition acquiring a deep axi-symmetric WC of diabatic origin in proximity of the center. In the tropical-like phase the WC is attributable to the deep moist convection (DC) that is produced as a consequence of latent heat exchange with the sea surface. The diabatic process provides the energy for self-sustaining the medicane or potential intensification. It has been observed that the wind field shows different characteristics depending on the phase of the medicane life cycle. In particular, during the mature phase, a closed ring-shaped band of strong near-surface winds appears around the eye. In addition, within the closed-ring of strong winds an area of maximum winds (AMW) can be always recognized. As for TCs, the radius of maximum wind (RMW) is derived as the distance between the medicane’s rotational center (RC) and the AMW. The RMW can provide information on both the intensification of medicanes and reaching the mature phase in analogy with what happens for TCs. During the mature phase, a significant decrease in RMW, often down to a few tens of kilometers, is associated with the cyclone intensification. To properly calculate the RMW, it is first necessary to identify the RC. To this end, the Medicane Rotational Center Automated Detection (MeRCAD) algorithm has been recently developed. However, it has been demonstrated that MeRCAD can be used to ascertain the RC’s position of several types of cyclones (namely, extratropical cyclones, medicanes or TCs). MeRCAD algorithm needs the ingestion of the wind speed and direction data collected over the sea surface. Data provided by the Advanced SCATterometer (ASCAT) on board MetOP satellites are used for this purpose. The ASCAT Coastal Winds at 12.5 km Swath Grid data (wind speed and direction) are provided in NetCDF format by the European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT) SAF on Ocean and Sea Ice (OSI SAF) with global coverage in 1.5 days. In addition, according to their availability, also data collected by the WindRAD instrument on board the Feng-Yun 3E satellite have been processed only for Medicane Daniel at this time. Similarly, the Automated Rotational Center Hurricane Eye Retrieval (ARCHER) algorithm is widely used to objectively identify the center of TCs. ARCHER is developed by the TC group of the Cooperative Institute for Meteorological Satellite Studies (CIMSS) of the University of Wisconsin-Madison. In this work, the ARCHER algorithm considers only brightness temperature data collected by means the 183.3 GHz channels observations of the Microwave Humidity Sensor (MHS) on board MetOp satellites (ARCHER can work with different high frequency channels from different radiometers). The objective of this study is to provide a comparative analysis between MeRCAD and ARCHER in order to establish whether ARCHER can be applied to medicane case studies, or whether further developments are necessary. All the medicanes that occurred in the Mediterranean basin from 2011 until 2023 are taken into account for the analysis. Preliminary results show that there is not always full agreement between the two algorithms depending on both the strength of the cyclone and the phase of its lifecycle. Usually, during the mature phase, when the cyclone is more intense and the ring shaped is more defined, the locations of the RCs provided by the two methodologies are closer to each other. In addition, MeRCAD appears to be more effective in identifying the medicanes’ RC, even during their development phase.
