Preprints
https://doi.org/10.5194/egusphere-2022-539
https://doi.org/10.5194/egusphere-2022-539
 
29 Jun 2022
29 Jun 2022

Detecting anomalous sea-level states in North Sea tide gauge data using of autoassociative Neural Network

Kathrin Wahle1, B. Emil Vassilev Stanev1,2, and Joanna Staneva1 Kathrin Wahle et al.
  • 1Helmholtz Zentrum Hereon, Geesthacht, Germany
  • 2University, Sofia, Bulgary

Abstract. Sea level in the North Sea is densely monitored by tide gauges. These measurements are used for validation as well as for the detection of extreme events. Here, we focus on the detection of sea level states with anomalous spatial correlations.

An autoassociative neural network emulates the spatial correlations of gauges in a lower dimensional subspace. Anomalous sea-level states are defined by means of failure of the reconstruction model. Using spatially distributed data as input to the network thus reveals sea-level conditions with unusual spatial correlations. The corresponding atmospheric conditions indicate high wind tendencies and pressure anomalies. Quantitative analysis of such states might help assess and improve numerical model quality in the future as well as provide new insights into the nonlinear processes involved. The method has the advantage of being easily applicable to any tide gauge array without preprocessing the data or acquiring any additional information.

Kathrin Wahle 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 egusphere-2022-539', Anonymous Referee #1, 05 Aug 2022
  • RC2: 'Comment on egusphere-2022-539', Anonymous Referee #2, 10 Aug 2022

Kathrin Wahle et al.

Kathrin Wahle et al.

Viewed

Total article views: 176 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
127 40 9 176 3 3
  • HTML: 127
  • PDF: 40
  • XML: 9
  • Total: 176
  • BibTeX: 3
  • EndNote: 3
Views and downloads (calculated since 29 Jun 2022)
Cumulative views and downloads (calculated since 29 Jun 2022)

Viewed (geographical distribution)

Total article views: 146 (including HTML, PDF, and XML) Thereof 146 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 11 Aug 2022
Download
Short summary
Knowledge of what causes maximum water levels is often key in coastal management. Processes, such as strom surge and atmospheric forcing alter the predicted tide. Whilst most of these processes are modelled in present-day ocean forecasting, still there is a need for better understanding situations where modelled and observed water levels deviate from eachother. Here we will use machine learning to detect such anomalies within a network of sea level observations in the North Sea.