Consider doing some additional investigation of the smoothed DYNBAL fluxes introduced in ctsm5.4.055

[ctsm5.4.055](https://github.com/ESCOMP/CTSM/pull/4189) changed the dynbal conservation fluxes from being dribbled out in the same year to instead being released from pools with (by default) 20-year residence times.

In talking this through with @slevis-lmwg , he suggested that we flag a couple of things for possible long-term follow-up, via investigations and/or discussions:

(1) In various configurations, how far out of conservation are we due to these new 20-year-residence-time pools?

(2) Note that this new delay applies to all water and energy states; notably, this includes canopy liquid / snow. Are we okay with imbalances in those canopy stores also being released gradually (with a 20-year residence time)? That feels particularly weird, though could *maybe* be justified by thinking of the canopy water being held somewhere (non-explicit) in the soil and gradually released from there, rather than going immediately to runoff. Note that this arises from daily FATES changes, as well as from annual landcover change.

I'm not sure how to determine what's acceptable for these questions, so I'm not sure if this is actually worth further investigation, but I thought it made sense to open this issue so others can also weigh in on whether this is worth some follow-up at some point.