Preprints
https://doi.org/10.5194/tc-2021-24
https://doi.org/10.5194/tc-2021-24

  01 Feb 2021

01 Feb 2021

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

Calibration of sea ice drift forecasts using random forest algorithms

Cyril Palerme and Malte Müller Cyril Palerme and Malte Müller
  • Development Centre for Weather Forecasting, Norwegian Meteorological Institute, Oslo, Norway

Abstract. Developing accurate sea-ice drift forecasts is essential to support decision making of maritime end-users operating in the Arctic. In this study, two calibration methods have been developed for improving 10-day sea-ice drift forecasts from an operational sea-ice prediction system (TOPAZ4). The methods are based on random forest algorithms (supervised machine learning models) and have been trained using either drifting buoy or synthetic-aperture radar observations for the target variables. Depending on the calibration method, the mean absolute error is reduced, on average, between 5.9 % and 8.1 % for the direction, and between 7.1 % and 9.6 % for the speed of sea-ice drift. Overall, the algorithms trained with buoy observations have the best performances when the forecasts are evaluated using drifting buoys as reference. However, there is a large spatial variability in these results, and the algorithms trained with buoy observations have particularly poor performances for predicting the speed of sea-ice drift in the Canadian Archipelago, along the east coast of Greenland, and north of Svalbard. In these areas, the algorithms trained with SAR observations have better performances for predicting the speed of sea-ice drift.

Cyril Palerme and Malte Müller

Status: final response (author comments only)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on tc-2021-24', Anonymous Referee #1, 09 Mar 2021
  • RC2: 'Comment on tc-2021-24', Anonymous Referee #2, 10 Mar 2021
  • RC3: 'Comment on tc-2021-24', Bruno Tremblay, 07 Apr 2021

Cyril Palerme and Malte Müller

Cyril Palerme and Malte Müller

Viewed

Total article views: 299 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
210 78 11 299 2 1
  • HTML: 210
  • PDF: 78
  • XML: 11
  • Total: 299
  • BibTeX: 2
  • EndNote: 1
Views and downloads (calculated since 01 Feb 2021)
Cumulative views and downloads (calculated since 01 Feb 2021)

Viewed (geographical distribution)

Total article views: 264 (including HTML, PDF, and XML) Thereof 264 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 19 Apr 2021
Download
Short summary
Two methods have been developed for calibrating sea-ice drift forecasts from an operational prediction system using machine learning algorithms. These algorithms use predictors from sea-ice concentration observations during the initialization of the forecasts, sea-ice and wind forecasts, and some geographical information. Depending on the calibration method, the mean absolute error is reduced between 5.9 % and 8.1 % for the direction, and between 7.1 % and 9.6 % for the speed of sea-ice drift.