Preprints
https://doi.org/10.5194/amt-2021-137
https://doi.org/10.5194/amt-2021-137

  19 May 2021

19 May 2021

Review status: this preprint is currently under review for the journal AMT.

Using artificial neural networks to predict riming from Doppler cloud radar observations

Teresa Vogl1, Maximilian Maahn2, Stefan Kneifel2, Willi Schimmel1, Dmitri Moisseev3,4, and Heike Kalesse-Los1 Teresa Vogl et al.
  • 1Institute for Meteorology, University of Leipzig, Leipzig, Germany
  • 2Institute for Geophysics and Meteorology, University of Cologne, Cologne, Germany
  • 3Institute for Atmospheric and Earth System Research/ Physics, Faculty of Science, University of Helsinki, Finland
  • 4Finnish Meteorological Institute, Helsinki, Finland

Abstract. Riming, i.e. the accretion and freezing of SLW on ice particles in mixed-phase clouds, is an important pathway for precipitation formation. Detecting and quantifying riming using ground-based cloud radar observations is of great interest, however, approaches based on measurements of the mean Doppler velocity (MDV) are unfeasible in convective and orographically influenced cloud systems. Here, we show how artificial neural networks (ANNs) can be used to predict riming using ground-based zenith-pointing cloud radar variables as input features. ANNs are a versatile means to extract relations from labeled data sets, which contain input features along with the expected target values. Training data are extracted from a data set acquired during winter 2014 in Finland, containing both Ka-band cloud radar and in-situ observations of snowfall. We focus on two configurations of input variables: ANN #1 uses the equivalent radar reflectivity factor (Ze), MDV, the width from left to right edge of the spectrum above the noise floor (spectrum edge width; SEW), and the skewness as input features. ANN #2 only uses Ze, SEW and skewness. The application of these two ANN configurations to case studies from different data sets demonstrates that both are able to predict strong riming (riming index = 1) and yield low values (riming index ≤ 0.4) for unrimed snow. In general, the predictions of ANN #1 and ANN #2 are very similar, advocating the capability to predict riming without the use of MDV. It is demonstrated that both ANN setups are able to generalize to W-band radar data. The predictions of both ANNs for a wintertime convective cloud fit coinciding in-situ observations extremely well, suggesting the possibility to predict riming even within convective systems. Application of ANN #2 to an orographic case yields high riming index values coinciding with observations of solid graupel particles at the ground.

Teresa Vogl et al.

Status: final response (author comments only)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on amt-2021-137', Anonymous Referee #1, 28 Jun 2021
  • RC2: 'Comment on amt-2021-137', Anonymous Referee #2, 10 Jul 2021

Teresa Vogl et al.

Teresa Vogl et al.

Viewed

Total article views: 347 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
231 104 12 347 5 5
  • HTML: 231
  • PDF: 104
  • XML: 12
  • Total: 347
  • BibTeX: 5
  • EndNote: 5
Views and downloads (calculated since 19 May 2021)
Cumulative views and downloads (calculated since 19 May 2021)

Viewed (geographical distribution)

Total article views: 321 (including HTML, PDF, and XML) Thereof 321 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 20 Sep 2021
Download
Short summary
We are using machine learning techniques, a type of artificial intelligence, to detect graupel formation in clouds. It is important to better detect this process, because its dynamics are not sufficiently well understood. The measurements used as input to the machine learning framework were performed by cloud radars. Cloud radars are instruments located at the ground, emitting radiation with wavelenghts of a few millimeters vertically into the cloud and measuring the back-scattered signal.