In the precursor project DAROTA two new global empirical tides models, EOT08a and EOT10a, have been derived, the latter further improving the published and fully validated EOT08a. Moreover, a combination approach with GRACE K-Band range data has been developed and tested. The combination of altimetry and GRACE is essential for ocean tide modeling: on one hand we profit from the favorite sampling capability of long-term multi- mission satellite altimetry. At the same time we overcome the weaknesses of altimetry, e.g. over seasonally ice-covered oceans. The objectives of the present proposal are basically derived from the lessons learned from the precursor project. They are (i) to further strengthen the empirical tide analysis by additional altimeter data and a realistic error estimate (=> EOT12a) (ii) to seek the GRACE contribution to ocean tides not in terms of global spherical harmonics, but to use instead localizing base functions in particular over the polar oceans, (iii) to combine altimetry and GRACE using error information and variance-component estimates (=> EOT12ga), (iv) to validate EOT12ga by GRACE gravity field processing, ocean bottom pressure records, and GPS time series (promote EOT12ga as candidate for GRACE Release 06 processing), and finally (v) to investigate the proper treatment of admittance methods and nodal corrections. The ultimate goal of COTAGA is to provide with EOT12ga the most precise global description of the short-term tidal mass variations, a new reference model for GRACE gravity field modeling, satellite altimetry, Earth rotation, loading effects and other applications.