Submesoscale (1-10 km) satellite observations like sea surface temperature and ocean color data show biogeochemical tracers structured in filaments, elongated and strongly contrasted structures. Filaments have been found to organize both physical and biological processes, like mixing, vertical exchanges, and the structuring of ecosystem communities. Due to the pres-ence of clouds, these observations however provide a sparse and unpredictable coverage. Here I will show how global filament information can be retrieved by currently available altimetry data. Even if altimetry data does not resolve the submesoscale, it contains information on the mesoscale turbulence, that is able to generate tracer filaments through the stirring pro-cess. Therefore, by combining non-local spatial and temporal variability of sea surface heights, Lagrangian methods like the Lyapunov exponent are able to quantify the stirring process and to predict the position of submesoscale tracer filaments. The reproduced filaments are well in agreement with in situ and remote sensed submesoscale data. The possibility of constructing synoptic maps of filaments allows to explore the submesoscale structure of the biogeochemical and ecological processes of the ocean surface. The calculation of filaments from real time altime-try allows to conceive novel satellite assisted in situ experiments (like the recent LOHAFEX campaign), by which filaments are tracked and sampled in the course of their evolution.