Data-driven Identification of Microphysical Processes by QVP-based Hierarchical Clustering of Polarised X-band Doppler Radar Observations

Correct, timely and meaningful interpretation of polarimetric weather radar observations requires an accurate understanding of hydrometeors and their associated microphysical processes as well as well-developed techniques that automatize their recognition in both the spatial and temporal dimensions of the data.

This study presents a novel data-driven technique for identifying different types of hydrometeors from Quasi—Vertical Profiles (QVPs) created from observations made by the NCAS X-band dual-polarization Doppler weather radar (NXPol). In this new technique, the hydrometeor types are identified in the data as clusters belonging to a hierarchical structure.

The number of different hydrometeor types in the data is not predefined and the method obtains the optimal number of clusters by implementing a recursive process. The obtained optimal clustering is then used to label the original data and to examine the temporal evolution of the observed microphysical processes.

The method can be used for different time series of vertical profiles as well as for the time series of radar volume scans. Further development includes making a step from the height versus time (2D data) to the volume observations in time (4D data) with application of other machine learning methods that extract data-driven parametrisations of microphysical processes as observed by the ground-based weather radar.