This project involves the study of emerging methodologies, all of them physically and signal-processing based, devised to handle the ever increasing complexity of Earth Observation data with the goal of better characterization, monitoring, exploitation, classification and data fusion capabilities. Both partners have a strong complementarity in understanding the physics of sensors and the handling of complex datasets using advanced nonlinear methods. The objective is to take advantage of an existing partnership able to master the increasing spatial, temporal and spectral resolution capabilities of sensors in order to provide better answer to societal challenges, for example in agricultural monitoring. There is an important challenge in the handling of big data from Earth Observation, because the advent of high resolution data has brought the need of consistent processing along the scales of acquired signals, and the mass data provided by Earth Observation, most of it unused, has put forward the necessity of correct and precise fusion algorithms of different acquisitions in time, spectral components, are spatial resolution. However such a challenge needs new advances in signal processing because high resolution data has put forward some limitations of existing techniques; and it also needs the involvement in the building of sensors themselves, while the two partners of this proposition show an excellent complementarity w.r.t. these matters. Nonlinear methods have proven their adequacy in the handling of complex data, as they are able to match multiscale features which are usually non- attainable with linear approaches and stationary hypothesis. They provide optimal multi-resolution analysis that can be extremely useful for inferring properties along the scales of complex and turbulent signals, and are closely related with emerging techniques related to sparse representations and compressed sensing [4]. We will develop the foundations and explore the possibilities offered by nonlinear methods in the handling and monitoring of big data in Earth Observation and Universe Sciences. Nowadays, various satellite sensors are available and many of them are giving complementary information for earth observation but still it is challenging to use these big datasets in more analytical way so maximum information can be extracted. For example, Radar and Optical sensors are available for earth surface monitoring but due to limited availability of techniques they are still underused. Therefore, the proposed project will be focused to use more effectively these big datasets by doing research to analyse various methods like nonlinear methods for handling and monitoring big-data, suitable fusion techniques so the complementary information of different sensors can be obtained in one scale. The Indian partner has built an exceptional expertise in the acquisition and data fusion of Earth Observation data. Multisource data fusion is probably the most difficult aspect in the integration of satellite image data products. In fact, while fusion is relatively straightforward when using data from the same satellite, the integration of imagery originating from different satellites carrying different sensors, is quite complicated. Indian partner is developing adaptive fusion techniques [5-7]. The French partner is developing multiscale and nonlinear methods that suit particularly well the fusion of data acquired at different resolutions. In particular, the development of inference methods that optimize information transmission along the scales of complex acquired data has already proven extremely useful in Earth Observation and new data fusion answers brought by high resolution [1,2,3]. This proposed participation is centred on the following objectives: Development of Adaptive methods for multi-resolution data fusion. Development of techniques to extract complimentary data from different sensors in same scale. Monitoring the earth parameters with developed techniques and specifying the specification for monitoring system, Theoretical developments in nonlinear signal processing: reconstructible systems and multidimensional singularity exponents for multispectral high resolution data. Machine Learning methods for big datasets and high resolution data. Adaptive optics and turbulent data. Multiscale and nonlinear methods for data fusion. Nonlinear signal processing for multispectral and high-resolution astronomical datasets. Agricultural monitoring. High resolution dynamics in oceanography. High resolution mapping of atmosphere/ocean exchange flux, application to climate change. 4- Short bibliography [1] H. Yahia et al. "Motion analysis in oceanographic satellite images using multiscale methods and the energy cascade", Pattern Recognition, 43 (10), 3591-3604, 2010. [2] O. Pont, A. Turiel, H. Yahia, "Singularity analysis of digital signals through the evaluation of their unpredictable point manifold", Intern. Journal of Comp. Mathematics, Taylor and Francis, DOI: 10.1080/00207160.2012.748895, 2012. [3] A. Turiel, H. Yahia, C. Perez-Vicente. "Microcanonical mulifractal formalism -ageometrical approach to multifractal systems: part 1. Singularity analysis", J. Phys. A: Math. Theor. 41 (2008) 015501. [4] S. Mallat, "A wavelet tour of signal processing: the sparse way", Academic Press, 3rd edition, 2008. [5] G. R. Harish Kumar and D. Singh, "Quality Assessment of Fused Image of MODIS and PALSAR", Progress in Electromagnetics Research, , PIER B, 24, 191-221, 2010. [6] Chamundeeswari V V, Singh D, and Singh K, "An adaptive method with integration of multi-wavelet based features for unsupervised classification of SAR images," Journal of Geophysics and Engineering vol. 4 pp.384-393, 2008. [7] R. S. Gautam, D. Singh, A. Mittal, "Application of principal component analysis and information fusion technique to detect hotspots in NOAA/AVHRR images of Jharia coalfield", SPIE Journal of Applied Remote Sensing (JARS), vol. 1, 013523, [DOI: 10.1117/1.2771256], Regular Paper, 2007. [8] Triloki Pant, D. Singh and Tanuja Srivastava "Advanced Fractal Approach for Unsupervised Classification of SAR Images", Advances in Space Research, 45, 1338-1349, 2010.