Coupled ocean-biogeochemical models simulate the ocean circulation in combination with a biogeochemical model which simulates the transport of biogeochemical quantities as tracers that are then modified by a biogeochemical process model simulating, e.g. the growth of phytoplankton. The biogeochemical models are typically developed as empirical source-sink models given that no exact equations for the processes exist. To this end, it is also unknown how many species, or functional types, of phytoplankton or zooplankton are sufficient for a realistic simulation of the phytoplankton community in the ocean. The assimilation into coupled ocean-biogeochemical models involves several issues, which start from the different skill of the physical circulation model and the empirical biogeochemical mode. For the data assimilation into ocean-biogeochemical models both satellite data of physical quantities and optical data, so-called ocean color, which for example gives information on the total chlorophyll concentration, can be used. The biogeochemical quantities exhibit a non-Gaussian distribution, and at least for Chlorophyll it is known that it has an approximately log-normal distribution. Thus, the assimilation of biogeochemical models requires the consideration of non-Gaussian distributions. This presentation reviews different issues of the assimilation into ocean-biogeochemical models based on examples from global and coastal applications of ocean-biogeochemical modeling.