DocumentCode :
2535675
Title :
Artificial Neural Network for Predicting Extreme Sea Level Variation Associated with Severe Storms
Author :
de Oliveira, M.M.F. ; Ebecken, Nelson F F ; de Oliveira, J.L.F. ; Nunes, Luis Manoel Paiva
Author_Institution :
Centro de Tecnol., UFRJ, Rio de Janeiro, Brazil
fYear :
2010
fDate :
23-28 Oct. 2010
Firstpage :
121
Lastpage :
126
Abstract :
This paper presents an Artificial Neural Network (ANN) model developed to predict extreme sea level variation in Santos basin on the Southeast region of Brazil, related to the passage of frontal systems associated with cyclones. A methodology was developed and applied to Petrobras water deep data set. Hourly time series of water level were used in a deep point of 415 meters. 6-hourly series of atmospheric pressure and wind components from NCEP/NCAR reanalysis data set were also used from ten points over the oceanic area. Correlations and spectral analyse were verified to define the time lag between the meteorological variables and the coastal sea level response to the occurrences of the extreme atmospheric systems. These correlations and time lags were used as input variables of the ANN model. This model was compared with multiple linear regression (MLR) and presented the best performance, generalizing the effect of the atmospheric interactions on extreme sea level variations.
Keywords :
atmospheric movements; atmospheric pressure; data analysis; delays; environmental science computing; neural nets; regression analysis; sea level; spectral analysis; storms; weather forecasting; Brazil; NCEP/NCAR reanalysis data set; Petrobras water deep data set; Santos basin; Southeast region; artificial neural network; atmospheric pressure; atmospheric system; frontal system; multiple linear regression; oceanic area; sea level variation prediction; severe storm; spectral analysis; time lag; time series; wind component; Artificial neural networks; Atmospheric modeling; Correlation; Predictive models; Sea level; Surges; Tides; ANN model; Harmonic model; sea level; tide; time serie;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Neural Networks (SBRN), 2010 Eleventh Brazilian Symposium on
Conference_Location :
Sao Paulo
ISSN :
1522-4899
Print_ISBN :
978-1-4244-8391-4
Electronic_ISBN :
1522-4899
Type :
conf
DOI :
10.1109/SBRN.2010.29
Filename :
5715224
Link To Document :
بازگشت