"The advent of ultrafast laser technology in the xuv and x-ray domains, by means of free-electron lasers and high harmonic generation, has opened the path to novel experiments at femto and attosecond time scales with direct applications on Chemistry and Atomic and Molecular Physics. These advances allow one to track in real time ultrafast electronic and nuclear dynamics induced by electromagnetic radiation. This largely unknown dynamics is at the focus of the proposal. Interpretation of ultrafast time-resolved experiments necessarily involves state-of-the-art time-dependent theoretical methods, together with advanced supercomputing resources. This remains a challenge due to the computational difficulty to include all electronic and vibrational degrees of freedom and the added complexity related to a multi-body Coulomb break up problem. ATTOTREND aims to develop a new theoretical tool based on time-dependent ab initio methods to account for the full dimensionality of the problem. Numerical methods combined with recently developed techniques to extract information from time-propagated wave-packets will be implemented. In particular, we seek to perform the first ever ab initio calculations on multiphoton double ionization of the hydrogen molecule including the dissociative channel. The second aim of this proposal is to carry out XUV attosecond pump-probe studies on molecules leading to single and double ionization. Double ionization processes are ideal to explore electron correlation and dynamics, and therefore to acquire a deeper knowledge of the basic mechanisms that governs the physics arising at the attosecond time scale. Finally, by combining these time-dependent methods with existing quantum chemistry tools, ATTOTREND also aims to explore these processes on more complex molecules of chemical interest."