Convective initiation nowcasting using SEVIRI infrared and visible data

Infrared (IR) and visible (VIS) reflectance and texture data from the Meteosat Second Generation (MSG) Spinning Enhanced Visible and Infrared Imager (SEVIRI) are used to understand cloud-top signatures for growing cumulus clouds 30-60 min prior to known convective initiation (CI) events, or the first occurrence of a >35 dBZ echo from a new convective cloud. In the process, this study proposes how MSG IR and VIS fields may be used to infer three physical attributes of growing cumuli, namely cloud depth, cloud-top glaciation and updraft strength, with limited information redundancy. These three aspects are observed as unique signatures within SEVIRI and high-resolution visible (HRV) data, for which this study seeks to relate to previous research, as well as develop new understanding on what subset of IR information best describes these attributes. Data from 123 CI events observed during the 2007 Convection and Orograpically Induced Precipitation Study (COPS) field experiment conducted over Southern Germany and Northeastern France are processed, per convective cell, so to meet this study's objectives.

A total of 64 IR and 27 VIS interest fields are initially assessed for growing cumulus clouds, with correlation and principal component analyses used to highlight the top 20 IR and 8 most important VIS fields that contain the most unique information. For the IR channels, using multiple fields per physical attribute of growing cumulus clouds, a method is proposed on how developing cumuli may be quantified per 3 km sampling distance MSG pixel (or per cumulus cloud "object") towards inferring CI over 1 hour timeframes. For the VIS data, 8 fields are found to contain unique reflectance information towards estimating cloud-top glaciation, with the HRV data used to estimate various forms of texture and brightness variability within a scene.

The talk will present the details of the study, and focus more on how SEVIRI's and HRV data set provides us valuable new ways of estimating CI from geostationary data, especially, as we move toward to GOES-R era.