Nonlinear Processes in Geophysics www.nonlin-processes-geophys.net 1023-5809 1607-7946 15 1 2008 10.5194/npg-15-61-2008 http://www.nonlin-processes-geophys.net/15/61/2008/ http://www.nonlin-processes-geophys.net/15/61/2008/npg-15-61-2008.html http://www.nonlin-processes-geophys.net/15/61/2008/npg-15-61-2008.pdf 61 70 2008-02-05 Artificial Neural Networks to reconstruct incomplete satellite data: application to the Mediterranean Sea Surface Temperature E. Pisoni enrico.pisoni@ing.unibs.it F. Pastor M. Volta Department of Electronic for Automation, University of Brescia, Via Branze 38, 25123 Brescia, Italy Centro Estudios Ambientales del Mediterraneo, C. Charles Darwin 14, Paterna, Valencia, Spain Satellite data can be very useful in applications where extensive spatial information is needed, but sometimes missing data due to presence of clouds can affect data quality. In this study a methodology for pre-processing sea surface temperature (SST) data is proposed. The methodology, that processes measures in the visible wavelength, is based on an Artificial Neural Network (ANN) system. The effectiveness of the procedure has been also evaluated comparing results obtained using an interpolation method. After the methodology has been identified, a validation is performed on 3 different episodes representative of SST variability in the Mediterranean sea. The proposed technique can process SST NOAA/AVHRR data to simulate severe storm episodes by means of prognostic meteorological models. Acock, M C.: Estimating Missing Weather Data for Agricultural Simulations Using Group Method of Data Handling, J. Appl. Meteorol., 39, 1176–1184, 2000. Beckers, J M. and Rixen, M.: EOF Calculations and Data Filling from Incomplete Oceanographic Datasets, J. Atmos. Ocean. Tech., 20, 1839–1856, 2003. Bishop, C M.: Neural networks for pattern recognition, Oxford University Press, 1995. Dirks, K N., Johns, M D., Hay, J E., and Sturman, A P.: A simple semi-empirical model for predicting missing carbon monoxide concentrations, Atmos. Environ., 36, 5953–5959, 2002. Gardner, M. and Dorling, S.: Artificial neural networks (the multilayer perceptron) – a review of applications in the atmospheric sciences, Atmos. Environ., 32, 2627–2636, 1998. Gardner, M. and Dorling, S.: Neural network modelling and prediction of hourly NO_x and NO₂ concentrations in urban air in London, Atmos. Environ., 33, 709–719, 1999. Gardner, M. and Dorling, S.: Statistical surface ozone models: an improved methodology to account for non-linear behaviour, Atmos. Environ., 34, 21–34, 2000. Houseago-Stokes, R. and Challenor, P.: Using PPCA to Estimate EOFs in the Presence of Missing Values, J. Atmos. Ocean. Tech., 21, 1471–1480, 2004. Iglesias, P., Jorquera, H., and Palma, W.: Data analysis using regression models with missing observations and long-memory: an application study, Comput. Stat. Data An., 2005. Junninen, H., Niska, H., Tuppurainen, K., Ruuskanen, J., and Kolehmainen, M.: Methods for imputation of missing values in air quality data sets, Atmos. Environ., 38, 2895–2907, 2004. Kilpatrick, K., Podesta, G., and Evans, R.: Overview of the NOAA/NASA Advanced Very High Resolution Radiometer Pathfinder Algorithm for Sea Surface Temperature and Associated Matchup Database., J. Geophys. Res., 106, 9179–9197, 2001. Kolehmainen, M., Martikainen, H., and Ruuskanen, J.: Neural networks and periodic components used in air quality forecasting, Atmos. Environ., 35, 815–825, 2001. Kondrashov, D. and Ghil, M.: Spatio-temporal filling of missing points in geophysical data sets, Nonlin. Processes Geophys., 13, 151–159, 2006. Kukkonen, J., Partanen, L., Karppinen, A., Ruuskanen, J., Junninen, H., Kolehmainen, M., Niska, H., Dorling, S., Chatterton, T., Foxall, R., and Cawley, G.: Extensive evaluation of neural network models for the prediction of NO2 and PM10 concentrations, compared with a deterministic modelling system and measurements in central Helsinki, Atmos. Environ., 37, 4539–4550, 2003. Legates, D. and McCabe, G.: Evaluating the use of "goodness-of-fit" measures in hydrologic and hydroclimatic model validation, Water Resour. Res., 35, 233–242, 1999. Millan, M., Estrela, M., and Caselles, V.: Torrential precipitations on th Spanish east coast: The role of the Mediterranean sea surface temperature, Atmos. Res., 36, 1–16, 1995. Nunnari, G., Dorling, S., Schlink, U., Cawley, G., Foxall, R., and Chatterton, T.: Modelling SO2 concentration at a point with statistical approaches, Environ. Modell. Softw., 19, 887–905, 2004. Paatero, P., Aalto, P., Picciotto, S., Bellander, T., Castaņo, G., Cattani, G., Cyrys, J., Kulmala, M., Lanki, T., and et~al., F N.: Estimating time series of aerosol particle number concentrations in the five HEAPSS cities on the basis of measured air pollution and meteorological variables, Atmos. Environ., 39, 2261–2273, 2005. Pastor, F., Estrela, M., Penarrocha, D., and Millan, M.: Torrential Rains on the Spanish Mediterranean Coast: Modeling the Effects of the Sea Surface Temperature, J. Appl. Meteorol., 40, 1180–1195, 2001. Perez, P., Trier, A., and Reyes, J.: Prediction of PM$_2.5$ concentrations several hours in advance using neural networks in Santiago, Chile, Atmos. Environ., 34, 1189–1196, 2000. Romanowicz, R., Young, P., Brown, P., and Diggle, P.: A recursive estimation approach to the spatio-temporal analysis and modelling of air quality data, Environ. Modell. Softw., 2005. Sanz, B M. and Marques, N M.: Total ozone time series analysis: a neural network model approach, Nonlin. Processes Geophys., 11, 683–689, 2004. Schneider, T.: Analysis of Incomplete Climate Data: Estimation of Mean Values and Covariance Matrices and Imputation of Missing Values, J. Climate, 14, 853–871, 2001. Seze, G. and Desbois, M.: Cloud Cover Analysis from Satellite Imagery Using Spatial and Temporal Characteristics of Data, J. Clim. Appl. Meteorol., 26, 287–303, 1987. Tangang, F., Hsieh, W., and Tang, B.: Forecasting the regional sea surface temperatures of the tropical Pacific by neural network models, with wind stress and sea level pressure as predictors., J. Geophys. Res., 103, 7511–7522, 1998. Wu, A., Hsieh, W W., and Tang, B.: Neural Network forecasts of the tropical Pacific sea surface temperatures, Neural Networks, 19, 145–154, 2006.