Volcanic unrest and eruptions are typically characterized by transitions in intensity and style. These transitions are thought to be controlled by changing magma ascent rates and physicochemical processes occuring within the feeding plumbing systems and/or magma-hydrothermal interaction zones. Quantifying magma ascent rates from geochemical and geophysical records is therefore crucial to constrain and forecast volcanic activity, although understanding these records is a challenging task considering that volcanoes are highly non-linear dynamic systems. This session invites abstracts in which seismology, infrasound, remote sensing, gas geochemistry, and/or petrological observations; as well as physics-based, machine-learning, and/or analog modeling, have been used to investigate transitions in volcanic activity and quantify the effects of these dynamic processes. The goal of this session is to approach these issues from all angles, highlighting the gaps, limitations and advances within and between multiple disciplines.