[en] A method to reconstruct missing data in sea surface temperature data using a neural network is presented. Satellite observations working in the optical and infrared bands are affected by clouds, which obscure part of the ocean underneath. In this paper, a neural network with the structure of a convolutional auto-encoder is developed to reconstruct the missing data based on the available cloud-free pixels in satellite images. Contrary to standard image reconstruction with neural networks, this application requires a method to handle missing data (or data with variable accuracy) in the training phase. The present work shows a consistent approach which uses the satellite data and its expected error variance as input and provides the reconstructed field along with its expected error variance as output. The neural network is trained by maximizing the likelihood of the observed value. The approach, called DINCAE (Data INterpolating Convolutional Auto-Encoder), is applied to a 25-year time series of Advanced Very High Resolution Radiometer (AVHRR) sea surface temperature data and compared to DINEOF (Data INterpolating Empirical Orthogonal Functions), a commonly used method to reconstruct missing data based on an EOF (empirical orthogonal function) decomposition. The reconstruction error of both approaches is computed using cross-validation and in situ observations from the World Ocean Database. DINCAE results have lower error while showing higher variability than the DINEOF reconstruction.
Disciplines :
Earth sciences & physical geography
Author, co-author :
Barth, Alexander ; Université de Liège - ULiège > Département d'astrophys., géophysique et océanographie (AGO) > GeoHydrodynamics and Environment Research (GHER)
Alvera Azcarate, Aida ; Université de Liège - ULiège > Département d'astrophys., géophysique et océanographie (AGO) > GeoHydrodynamics and Environment Research (GHER)
Licer, Matjaz
Beckers, Jean-Marie ; Université de Liège - ULiège > Département d'astrophys., géophysique et océanographie (AGO) > GeoHydrodynamics and Environment Research (GHER)
Language :
English
Title :
DINCAE 1.0: a convolutional neural network with error estimates to reconstruct sea surface temperature satellite observations
MULTI-SYNC project (contract SR/00/359), Consortium des Équipements de Calcul Intensif (CÉCI), funded by the F.R.S.-FNRS under grant no. 2.5020.11, COST action ES1402 – “Evaluation of Ocean Syntheses”
Funders :
F.R.S.-FNRS - Fonds de la Recherche Scientifique [BE] BELSPO - SPP Politique scientifique - Service Public Fédéral de Programmation Politique scientifique
Abadi, M., Agarwal, A., Barham, P., Brevdo, E., Chen, Z., Citro, C., Corrado, G. S., Davis, A., Dean, J., Devin, M., Ghemawat, S., Goodfellow, I., Harp, A., Irving, G., Isard, M., Jia, Y., Jozefowicz, R., Kaiser, L., Kudlur, M., Levenberg, J., Mané, D., Monga, R., Moore, S., Murray, D., Olah, C., Schuster, M., Shlens, J., Steiner, B., Sutskever, I., Talwar, K., Tucker, P., Vanhoucke, V., Vasudevan, V., Viégas, F., Vinyals, O., Warden, P., Wattenberg, M., Wicke, M., Yu, Y., and Zheng, X.: TensorFlow: Large-Scale Machine Learning on Heterogeneous Systems, available at: https://www.tensorflow.org/ (last access: 13 November 2018), 2015.
Alvera-Azcárate, A., Barth, A., Rixen, M., and Beckers, J.-M.: Reconstruction of incomplete oceanographic data sets using Empirical Orthogonal Functions. Application to the Adriatic Sea Surface Temperature, Ocean Model., 9, 325-346, https://doi.org/10.1016/j.ocemod.2004.08.001, 2005.
Alvera-Azcárate, A., Barth, A., Beckers, J.-M., and Weisberg, R. H.: Multivariate reconstruction of missing data in sea surface temperature, chlorophyll and wind satellite field, J. Geophys. Res., 112, C03008, https://doi.org/10.1029/2006JC003660, 2007.
Alvera-Azcárate, A., Barth, A., Sirjacobs, D., and Beckers, J.-M.: Enhancing temporal correlations in EOF expansions for the reconstruction of missing data using DINEOF, Ocean Sci., 5, 475-485, https://doi.org/10.5194/os-5-475-2009, 2009.
Alvera-Azcárate, A., Barth, A., Parard, G., and Beckers, J.-M.: Analysis of SMOS sea surface salinity data using DINEOF, Remote Sens. Environ., 180, 137-145, https://doi.org/10.1016/j.rse.2016.02.044, 2016.
AVHRR Data: NODC and Rosenstiel School of Marine and Atmospheric Science, AVHRR Pathfinder Level 3 Monthly Daytime SST Version 5. Ver. 5, pO.DAAC, CA, USA, https://doi.org/10.5067/PATHF-MOD50, 2003.
Beckers, J.-M. and Rixen, M.: EOF calculation and data filling from incomplete oceanographic datasets, J. Atmos. Ocean. Tech., 20, 1839-1856, https://doi.org/10.1175/1520-0426(2003)020<1839:ECADFF>2.0.CO;2, 2003.
Bolton, T. and Zanna, L.: Applications of Deep Learning to Ocean Data Inference and Subgrid Parameterization, J. Adv. Model. Earth Sy., 11, 376-399, https://doi.org/10.1029/2018MS001472, 2019.
Bretherton, F. P., Davis, R. E., and Fandry, C. B.: A technique for objective analysis and design of oceanographic experiment applied to MODE-73, Deep-Sea Res., 23, 559-582, https://doi.org/10.1016/0011-7471(76)90001-2, 1976.
Buongiorno Nardelli, B.: A novel approach for the highresolution interpolation of in situ sea surface salinity, J. Atmos. Ocean. Tech., 29, 867-879, https://doi.org/10.1175/JTECH-D-11-00099.1, 2012.
Chapman, C. and Charantonis, A. A.: Reconstruction of Subsurface Velocities From Satellite Observations Using Iterative Self-Organizing Maps, IEEE Geosci. Remote S., 14, 617-620, https://doi.org/10.1109/LGRS.2017.2665603, 2017.
Chicco, D., Sadowski, P., and Baldi, P.: Deep Autoencoder Neural Networks for Gene Ontology Annotation Predictions, in: Proceedings of the 5th ACM Conference on Bioinformatics, Computational Biology, and Health Informatics, BCB '14, ACM, New York, NY, USA, 533-540, https://doi.org/10.1145/2649387.2649442, 2014.
Cho, K., van Merrienboer, B., Gülçehre, Ç., Bougares, F., Schwenk, H., and Bengio, Y.: Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation, CoRR, available at: http://arxiv.org/abs/1406.1078 (last access: 10 February 2020), 2014.
Ebert, E., Wilson, L., Weigel, A., Mittermaier, M., Nurmi, P., Gill, P., Göber, M., Joslyn, S., Brown, B., Fowler, T., and Watkins, A.: Progress and challenges in forecast verification, Meteorol. Appl., 20, 130-139, https://doi.org/10.1002/met.1392, 2013.
Evans, B., Vasquez, J., and Casey, K. S.: 4 km Pathfinder Version 5 User Guide, NOAA, available at: https://www.nodc.noaa.gov/SatelliteData/pathfinder4km/userguide.html (last access: 16 March 2019), 2009.
Garcia-Gorriz, E. and Garcia-Sanchez, J.: Prediction of sea surface temperatures in the western Mediterranean Sea by neural networks using satellite observations, Geophys. Res. Lett., 34, L11603, https://doi.org/10.1029/2007GL029888, 2007.
Ge, R., Huang, F., Jin, C., and Yuan, Y.: Escaping From Saddle Points - Online Stochastic Gradient for Tensor Decomposition, CoRR, available at: http://arxiv.org/abs/1503.02101 (last access: 11 December 2019), 2015.
Geng, J., Fan, J., Wang, H., Ma, X., Li, B., and Chen, F.: High-Resolution SAR Image Classification via Deep Convolutional Autoencoders, IEEE Geosci. Remote S., 12, 2351-2355, https://doi.org/10.1109/LGRS.2015.2478256, 2015.
Gilleland, E., Ahijevych, D., Brown, B., Casati, B., and Ebert, E.: Intercomparison of spatial forecast verification methods, Weather Forecast., 24, 1416-1430, https://doi.org/10.1175/2009WAF2222269.1, 2009.
Hinton, G. E. and Salakhutdinov, R. R.: Reducing the Dimensionality of Data with Neural Networks, Science, 313, 504-507, https://doi.org/10.1126/science.1127647, 2006.
Hochreiter, S. and Schmidhuber, J.: Long Shortterm Memory, Neural Comput., 9, 1735-1780, https://doi.org/10.1162/neco.1997.9.8.1735, 1997.
Jo, Y.-H., Kim, D.-W., and Kim, H.: Chlorophyll concentration derived from microwave remote sensing measurements using artificial neural network algorithm, J. Mar. Sci. Tech., 26, 102-110, https://doi.org/10.6119/JMST.2018.02-(1).0004, 2018.
Jouini, M., Lévy, M., Crépon, M., and Thiria, S.: Reconstruction of satellite chlorophyll images under heavy cloud coverage using a neural classification method, Remote Sens. Environ., 131, 232-246, https://doi.org/10.1016/j.rse.2012.11.025, 2013.
Kilpatrick, K. A., Podestá, G. P., 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.-Oceans, 106, 9179-9197, https://doi.org/10.1029/1999JC000065, 2001.
Kingma, D. P. and Ba, J.: Adam: A Method for Stochastic Optimization, CoRR, available at: http://arxiv.org/abs/1412.6980 (last access: 18 April 2019), 2014.
Krasnopolsky, V., Nadiga, S., Mehra, A., Bayler, E., and Behringer, D.: Neural Networks Technique for Filling Gaps in Satellite Measurements: Application to Ocean Color Observations, Comput. Intel. Neurosc., 6156513, https://doi.org/10.1155/2016/6156513, 2016.
Krizhevsky, A., Sutskever, I., and Hinton, G. E.: ImageNet Classification with Deep Convolutional Neural Networks, in: Advances in Neural Information Processing Systems 25, edited by: Pereira, F., Burges, C. J. C., Bottou, L., and Weinberger, K. Q., Curran Associates, Inc., 1097-1105, available at: http://papers.nips.cc/paper/4824-imagenetclassification-with-deep-convolutional-neural-networks.pdf (last access: 12 September 2019), 2012.
LeCun, Y., Bottou, L., Bengio, Y., and Haffner, P.: Gradient-Based Learning Applied to Document Recognition, Proc. IEEE, 86, 2278-2324, 1998.
Locarnini, R. A., Mishonov, A. V., Baranova, O. K., Boyer, T. P., Zweng, M. M., Garcia, H. E., Reagan, J. R., Seidov, D., Weathers, K., Paver, C. R., and Smolyar, I.: World Ocean Atlas 2018, Volume 1: Temperature, edited by: Mishonov, A., NOAA Atlas NESDIS 81, 52 pp., available at: https://data.nodc.noaa.gov/woa/WOA18/DOC/woa18-vol1.pdf (last access: 5 April 2019), 2018.
Maas, A. L., Hannun, A. Y., and Ng, A. Y.: Rectifier nonlinearities improve neural network acoustic models, in: in ICMLWorkshop on Deep Learning for Audio, Speech and Language Processing, available at: https://ai.stanford.edu/~amaas/papers/relu-hybrid-icml2013-final.pdf (last access: 17 April 2019), 2013.
Masters, D. and Luschi, C.: Revisiting Small Batch Training for Deep Neural Networks, CoRR, available at: http://arxiv.org/abs/1804.07612 (last access: 12 November 2019), 2018.
Patil, K. and Deo, M. C.: Prediction of daily sea surface temperature using efficient neural networks, Ocean Dynam., 67, 357-368, https://doi.org/10.1007/s10236-017-1032-9, 2017.
Pisano, A., Nardelli, B. B., Tronconi, C., and Santoleri, R.: The new Mediterranean optimally interpolated pathfinder AVHRR SST Dataset (1982-2012), Remote Sens. Environ., 176, 107-116, https://doi.org/10.1016/j.rse.2016.01.019, 2016.
Pisoni, E., Pastor, F., and Volta, M.: Artificial Neural Networks to reconstruct incomplete satellite data: application to the Mediterranean Sea Surface Temperature, Nonlin. Processes Geophys., 15, 61-70, https://doi.org/10.5194/npg-15-61-2008, 2008.
Rasp, S., Pritchard, M. S., and Gentine, P.: Deep learning to represent subgrid processes in climate models, P. Natl. Acad. Sci. USA, 115, 9684-9689, https://doi.org/10.1073/pnas.1810286115, 2018.
Renosh, P. R., Jourdin, F., Charantonis, A. A., Yala, K., Rivier, A., Badran, F., Thiria, S., Guillou, N., Leckler, F., Gohin, F., and Garlan, T.: Construction of Multi-Year Time-Series Profiles of Suspended Particulate Inorganic Matter Concentrations Using Machine Learning Approach, Remote Sensing, 9, 1320, https://doi.org/10.3390/rs9121320, 2017.
Ronneberger, O., Fischer, P., and Brox, T.: U-Net: Convolutional Networks for Biomedical Image Segmentation, in: Medical Image Computing and Computer-Assisted Intervention - MICCAI 2015, edited by: Navab, N., Hornegger, J., Wells, W. M., and Frangi, A. F., Springer International Publishing, Cham, 234-241, https://doi.org/10.1007/978-3-319-24574-4-28, 2015.
Rosenblatt, F.: The perceptron: A probabilistic model for information storage and organization in the brain, Psychol. Rev., 65, 386-408, https://doi.org/10.1037/h0042519, 1958.
Scherer, D., Müller, A., and Behnke, S.: Evaluation of Pooling Operations in Convolutional Architectures for Object Recognition, in: Artificial Neural Networks - ICANN 2010, edited by: Diamantaras, K., Duch, W., and Iliadis, L. S., Springer Berlin Heidelberg, Berlin, Heidelberg, 92-101, https://doi.org/10.1007/978-3-642-15825-4-10, 2010.
Schmidhuber, J.: Deep learning in neural networks: An overview, Neural Networks, 61, 85-117, https://doi.org/10.1016/j.neunet.2014.09.003, 2015.
Simonyan, K. and Zisserman, A.: Very Deep Convolutional Networks for Large-Scale Image Recognition, in: International Conference on Learning Representations, available at: https://arxiv.org/abs/1409.1556 (last access: 23 April 2019), 2015.
Widrow, B. and Hoff, M. E.: Associative Storage and Retrieval of Digital Information in Networks of Adaptive "Neurons", Springer US, Boston, MA, 160-160, https://doi.org/10.1007/978-1-4684-1716-6-25, 1962.
Wilks, D. S.: Statistical Methods in the Atmospheric Sciences, Academic Press, 1995.
Wylie, D., Jackson, D. L., Menzel, W. P., and Bates, J. J.: Trends in Global Cloud Cover in Two Decades of HIRS Observations, J. Climate, 18, 3021-3031, https://doi.org/10.1175/JCLI3461.1, 2005.
Zhou, Y., Wang, H., Xu, F., and Jin, Y.: Polarimetric SAR Image Classification Using Deep Convolutional Neural Networks, IEEE Geosci. Remote S., 13, 1935-1939, https://doi.org/10.1109/LGRS.2016.2618840, 2016.
