Abstract
The paper documents a concept of ocean forecasting system for ocean surface currents based on self-organizing map (SOM) trained by high-resolution numerical weather prediction (NWP) model and high-frequency (HF) radar data. Wind and surface currents data from the northern Adriatic coastal area were used in a 6-month long training phase to obtain SOM patterns. Very high correlation between current and joined current and wind SOM patterns indicated the strong relationship between winds and currents and allowed for creation of a prediction system. Increasing SOM dimensions did not increase reliability of the forecasting system, being limited by the amount of the data used for training and achieving the lowest errors for 4 × 4 SOM matrix. As the HF radars and high-resolution NWP models are strongly expanding in coastal oceans, providing reliable and long-term datasets, the applicability of the proposed SOM-based forecasting system is expected to be high.
Similar content being viewed by others
References
Abascal AJ, Castanedo S, Medina R, Losada IJ, Alvarez-Fanjul E (2009) Application of HF radar currents to oil spill modeling. Mar Pollut Bull 58:238–248
Abascal AJ, Castanedo S, Fernandez V, Medina R (2012) Backtracking drifting objects using surface currents from high-frequency (HF) radar technology. Ocean Dyn 62:1073–1089
Abbot J, Marohasy J (2012) Application of artificial neural networks to rainfall forecasting in Queensland. Aust Adv Atmos Sci 29(4):717–730
Abdul-Rahman A, Pilouk M (2008) Spatial data modelling for 3D GIS. Springer, New York
Aladin International Team (1997) The ALADIN project: mesoscale modelling seen as a basic tool for weather forecasting and atmospheric research. WMO Bull 46:317–324
Allard R, Rogers E, Martin P, Jensen T, Chu P, Campbell T, Dykes J, Smith T, Choi J, Gravois U (2014) The US Navy coupled ocean-wave prediction system. Oceanography 27:92–103
Barreto GA (2007) Time series prediction with the self-organizing map: a review. Perspect Neural-Symb Integr 77:135–158
Corchado JM, Lees B (2001) A hybrid case-based model for forecasting. Appl Artif Intell 15:105–127
Corchado JM, Aiken J (2002) Hybrid artificial intelligence methods in oceanographic forecast models. IEEE Trans Syst Man Cybern C 32(4):307–313
Cosoli S, Mazzoldi A, Gačić M (2010) Validation of surface current measurements in the northern Adriatic Sea from high-frequency radars. J Atmos Ocean Technol 27:908–919
Cosoli S, Bolzon G, Mazzoldi A (2012) A real-time and offline quality control methodology for SeaSonde high-frequency radar currents. J Atmos Ocean Technol 29:1313–1328
Cosoli S, Licer M, Vodopivec M, Malačič V (2013) Surface circulation in the Gulf of Trieste (northern Adriatic Sea) from radar, model, and ADCP comparisons. J Geophys Res 118:6183–6200
Crombie DD (1955) Doppler spectrum of sea echo at 13.56 Mc/s. Nature 175:681–682
Dorffner G (1996) Neural networks for time series processing. Neural Netw World 6:447–468
Duda RO, Hart PE, Stork DG (2001) Pattern classification, 2nd edn. Wiley, New York
Faraway J, Chatfield C (1998) Time series forecasting with neural networks: a comparative study using the airline data. J R Stat Soc Ser C (Appl Stat) 47:231–250
Fdez-Riverola F, Corchado JM (2002) A hybrid CBR model for forecasting in complex domains. Adv Artif Intell 2527:101–110
Geleyn JF (1988) Interpolation of wind, temperature and humidity values from model levels to the height of measurement. Tellus 40A:347–351
Harlan J, Terrill E, Hazard L, Keen C, Barrick D, Whelan C, Howden S, Kohut J (2010) The integrated ocean observing system high-frequency radar network: status and local, regional, and national applications. Mar Technol Soc J 44:122–132
Hsieh WW, Tang B (1998) Applying neural network models to prediction and data analysis in meteorology and oceanography. Bull Am Meteorol Soc 79:1855–1870
Ivatek-Šahdan S, Tudor M (2004) Use of high-resolution dynamical adaptation in operational suite and research impact studies. Meteorol Z 13:99–108
Kalinić H, Mihanović H, Cosoli S, Vilibić I (2015) Sensitivity of self-organizing map surface current patterns to the use of radial versus Cartesian input vectors measured by high-frequency radars. Comput Geosci 84:29–36
Kohonen T (2001) Self-organizing maps, 3rd edn. Springer, New York
Krasnopolsky VM (2007) Neural network emulations for complex multidimensional geophysical mappings: applications of neural network techniques to atmospheric and oceanic satellite retrievals and numerical modeling. Rev Geophys 45:3009
Kundu PK (1976) Ekman veering observed near the ocean bottom. J Phys Oceanogr 6(2):238–242
Liu Y, Weisberg RH, He R (2006) Sea surface temperature patterns on the West Florida shelf using growing hierarchical self-organizing maps. J Atmos Ocean Technol 23:325–338
Liu Y, MacCready P, Hickey BM, Dever EP, Kosro PM, Banas NS (2009) Evaluation of a coastal ocean circulation model for the Columbia River plume in summer 2004. J Geophys Res 114:C00B04. doi:10.1029/2008JC004929
Liu Y, Weisberg RH, Hu C, Zheng L (2011) Tracking the deepwater horizon oil spill: a modeling perspective. EOS Trans AGU 92(6):45
Liu Y, Kerkering H, Weisberg R (2015) Coastal ocean observing systems, 1st edn. Academic Press, London
Lermusiaux PFJ (2006) Uncertainty estimation and prediction for interdisciplinary ocean dynamics. J Comput Phys 217:176–199
Martinetz TM, Berkovitsch SG, Schulten KJ (1993) ‘Neural-Gas’ network for vector quantization and its application to time-series prediction. IEEE Trans Neural Netw 4:558–569
Mihanović H, Cosoli S, Vilibić I, Ivanković D, Dadić V, Gačić M (2011) Surface current patterns in the northern Adriatic extracted from high-frequency radar data using self-organizing map analysis. J Geophys Res 116:C08033
van der Merwe R, Leen TK, Lu Z, Frolov S, Baptista AM (2007) Fast neural network surrogates for very high dimensional physics-based models in computational oceanography. Neural Netw 20:462–478
Paduan JD, Washburn L (2013) High-frequency radar observations of ocean surface currents. Annu Rev Mar Sci 5:115–136
Prtenjak MT, Grisogono B (2007) Sea/land breeze climatological characteristics along the northern Croatian Adriatic coast. Theor Appl Climatol 90:201–215
Rako N, Vilibić I, Mihanović H (2013) Mapping underwater sound noise and assessing its sources by using a self-organizing maps method. J Acoust Soc Am 133:1368–1376
Roweis ST, Saul LK (2000) Nonlinear dimensionality reduction by locally linear embedding. Science 290:2323–2326
Rozier D, Birol F, Cosme E, Brasseur R, Brankart JM, Verron J (2007) A reduced-order Kalman filter for data assimilation in physical oceanography. SIAM Rev 49:449–465
Saylor JH, Miller GS (1988) Observation of Ekman veering at the bottom of Lake Michigan. J Great Lakes Res 14(1):94–100
Stanev EV, Ziemer F, Schulz-Stellenfleth J, Seemann J, Staneva J, Gurgel K-W (2014) Blending surface currents from HF radar observations and numerical modeling: tidal hindcasts and forecasts. J Atmos Ocean Technol 32(2):256–281
Tenenbaum JB, De Silva V, Langford JC (2000) A global geometric framework for nonlinear dimensionality reduction. Science 290:2319–2323
Thomson RE, Emery WJ (2014) Data analysis methods in physical oceanography. Elsevier, Amsterdam
Tudor M, Ivatek-Šahdan S (2010) The case study of bura of 1st and 3rd February 2007. Meteorol Z 19:453–466
Vilibić I, Mihanović H, Šepić J, Matijević S (2011) Using Self-Organising Maps to investigate long-term changes in deep Adriatic water patterns. Cont Shelf Res 31:695–711
Voronoi G (1908) Nouvelles applications des paramètres continus à la théorie des formes quadratiques. Journal für die Reine und Angewandte Mathematik 133:97–178
Wahle K, Stanev EV (2011) Consistency and complementarity of different coastal ocean observations: a neural network-based analysis for the German Bight. Geophys Res Lett 38(10):1–4. doi:10.1029/2011GL047070
Wehrens E, Buydens LMC (2007) Self- and super-organizing maps in R: the Kohonen package. J Stat Softw 21:1–19
Zhang GP (2003) Time series forecasting using a hybrid ARIMA and neural network model. Neurocomputing 50:159–175
Acknowledgments
The work on this paper was supported by the UKF project NEURAL (www.izor.hr/neural) and the IPA CBC project HAZADR (www.hazadr.eu). The SOM Toolbox version 2.0 for MATLAB was developed by E. Alhoniemi, J. Himberg, J. Parhankangas, and J. Vesanto at the Helsinki University of Technology, Finland, and is available at http://www.cis.hut.fi/projects/somtoolbox. Comments raised by anonymous reviewers are appreciated as greatly improving quality of the manuscript.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Kalinić, H., Mihanović, H., Cosoli, S. et al. Predicting ocean surface currents using numerical weather prediction model and Kohonen neural network: a northern Adriatic study. Neural Comput & Applic 28 (Suppl 1), 611–620 (2017). https://doi.org/10.1007/s00521-016-2395-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-016-2395-4