The tropical tuna treatment is the process created at the end of the 90s by EU scientists to estimate catches of the tropical tuna purse seine fishery, for which logbook declarations were known biased. The main purposes of this process are to compute the nominal catches and the catch and effort for RFMO. However, the evolution of fishing practices and the extension of the fishing grounds have challenged the T3 methodology in some parts of it processing. Thus, the used in the process of too large spatio-temporal sampling strata was specifically pointed out as the major cause of biases in the catch estimated. This paper so presents the new methodology developed to fix this issue and implications on the nominal catch estimates as well as catch and effort compared to the previous version. Finally, future improvements were discussed.