There are two kinds of unspent sMerits: those originated in the day to day received merit halvings, and the large remnant still unspent, originated in the initial sMerit airdrop. The former I estimated a week ago to be around the 34k area, and the latter is above the 400k area.
The >400 k (from the initial 600K airdrop) is not likely to suddenly be placed into circulation at short notice, and likely belongs to accounts that do not really want to be engaged in the merit system, or even accounts that have turned inactive over the past months. The other 34K sMerits have likely a greater chance of being awarded in the short run.

Moving some of these sMerits to current Merit Sources, however active they may be, would not help with one of the most important factors in my opinion, which is the number eyes on the field. Merit Sources, as we know, on aggregate do not finish off their monthly allowance, so adding more sMerits to their pool would likely not have a visible impact on the number of Receivers (it would on the amount of sMerit per TX though).

It would be better to appoint more Merit Sources (more eyes on the field), and in any case let some of them be the receivers of some of those unspent sMerits, although it would not really be necessary if they are dimensioned well enough from the start since it is system generated.