PASSIVE MICROWAVE SNOWFALL RETRIEVAL AND COMPARISON WITH AIRBORNE MEASUREMENTS OVER THE ARCTIC REGION

A. Camplani¹, A. Pant², P. Sanò¹, D. Casella¹, M. Mech², S. Crewell², G. Panegrossi¹

¹CNR-ISAC, ²University of Cologne

The retrieval of snowfall over high latitude regions remains a challenging task in the field of meteorological sciences. The scarcity of ground-based instruments in these areas makes it very difficult to generate a long-term data set for climatological purposes. However, the analysis of snowfall regimes in polar regions is fundamental for understanding changes in the water cycle at global and regional scale.

Passive microwave (PMW) observations from polar-orbiting satellites can overcome the scarcity of data in polar regions, thanks to the scattering signal observed in the high frequency channels (> 90 GHz), related to the presence of snowfall, and to the excellent temporal coverage at the higher latitudes. In recent years, in the context of the EUMETSAT Satellite Application Facility for operational Hydrology (H SAF), two snowfall retrieval algorithms have been developed by CNR-ISAC for the NOAA Advanced Technology Microwave Sounder (ATMS): the Snow retrievaL ALgorithm for gpM – Cross Track (SLALOM-CT) algorithm, for global applications, and the High lAtitude sNow Detection and Estimation aLgorithm for ATMS (HANDEL-ATMS), tailored for the typical dry and cold conditions of the polar regions. These algorithms are based on machine learning models trained on the “coincidence datasets” built from nearly-coincident measurements from the CloudSat W-band Cloud Profiling Radar (CPR) and ATMS. However, several factors – the variability of the surface background emissivity, the extreme environmental conditions typical of the high latitudes, the weak scattering signal often damped by the presence of supercooled liquid water, the limitations of the CPR snowfall profiles used as reference – make the snowfall retrieval very difficult. Therefore, the validation of PMW-based snowfall retrieval with measurements available from dedicated field campaigns remains an element of great interest. In recent years, several airborne measurement campaigns have been undertaken in the Arctic region as part of the AC3 project to investigate the causes and feedbacks of the Arctic amplification. Various instruments were used to take these measurements, including radars, lidars, radiometers, photometers and nephelometers.

This presentation will show the results of the comparison between the snowfall measurements obtained from airborne observations – in particular, by using the 95 GHz radar and the 89 GHz radiometer that constitutes the Microwave Radar/radiometer for Arctic Clouds (MiRAC) – and the snowfall estimates obtained from SLALOM-CT and HANDEL-ATMS during AC3 flights. The analysis shows a good agreement in the detection of snowfall areas between the airborne radar and the satellite estimates, while larger differences are observed in the snowfall rate estimation. The analysis of larger datasets highlights a general difficulty in detecting very light snowfall events by using PMW observations and the fact that the environmental conditions have a significant and different impact on the performance of the two algorithms. Further developments in passive microwave snowfall retrieval based on the recently launched Arctic Weather Satellite and MetOp-SG-A1 Microwave Sounder (MWS) missions, and on the exploitation of the ESA/JAXA EarthCare mission, are foreseen in the future.