Time series with saturation and back

Hi guys,

I have a question about the modelling of a time series with a clear rising trend and then, as consequence of a shock in a correlated variable, a second descendent trend. something like this:


After a short period the shock in the correleted variable returns to normal values an I expect the solution to come back to the normal rising trend like the upper bound instead to fall down to zero.

do you have any suggestion reguarding the correct model approach in Prophet?

Thank you!