I thought this would be relevant
On Paul's estimate of ~2.5t that the OP derives the figure from:
"Paul (2010) estimated a mass of 2.5 tonne for a 5.5
metre long unidentified Euoplocephalus specimen,
and a few years later Paul (2016a) also estimated a
mass of 2 tonne for a 5.3 metre long unidentified
specimen. In his mass table Paul (2016b) identified
the Euoplocephalus tutus AMNH 5404 specimen with
a calculated mass of 1.95 tonne and a length of 5.47
metres. It should be noted that the Euoplocephalus tutus AMNH 5404 specimen appears to be relatively
less massive that the unidentified specimens. In general, the reconstructions show a light and agile looking animal. Paul was a student of Bakker, so perhaps
it is not surprising that his restorations show a fast
and agile Euoplocephalus."
"In general it would appear that many early estimates
of Euoplocephalus reconstructed it as a fast and agile
animal with a low mass in the 2.5 tonne range. But
why would such a fast and agile animal have devoted
so many resources to producing external armour?
Some of the more up-to-date reconstructions show
an animal that was a lot bulkier and sturdier. This
would certainly seem to correlate well with an animal
that had an impressive array of external amour.
Accordingly, a mass of 3.4 tonne was chosen as the
best estimate."
{1}And on
P.mirandai size
{2}References{1}Hurrell, Stephen W. "A Palaeogravity calculation based on weight and mass estimates of Euoplocephalus tutus."
{2}Moreno-Bernal, Jorge W. "Size and paleoecology of giant Miocene South American crocodiles (Archosauria: Crocodylia)." Journal of Vertebrate Paleontology 27.3 (2007).