Spinosaurus aegyptiacus

[DinoFan83]

Delving a little deeper into things, I've relatively recently come to notice something that's made me raise my eyebrows in regards to the FSAC-KK 11888 volumetric model of 3864.7 litres from Ibrahim et al. (2020), that has a marked effect on the mass estimates for both that specimen and the giant possible adult Spinosaurus.

More specifically, as can be seen here, the new volumetric estimate does not account for an appropriate amount of flesh on the spines. At all.
Making the correction for this would have quite a crucial effect on the mass of the animal - fortunately, there are 2 pre-existing parameters that can help us get the correction right: modern crested chameleons, and SpinoInWonderland's earlier GDI of Spinosaurus with segment masses.

In addition, I have word from SpinoInWonderland about this. He makes a suggestion that Spinosaurus would have proportionally more flesh than the oft-cited crested chameleon because its spines are much thicker and less spaced. That I could not agree more with - it's blatant in the hyperlinked images how well the spines match with his statements.

With these in mind, let's take a look at the flesh in the earlier GDI and the crested chameleon.

One thing is immediately clear - the GDI has far less spine flesh proportionally than the actual crested chameleon does, well under half as much. Considering how the spacing and robusticity of the spines compare, it's clear that using the segment mass of the GDI as-is would produce a major underestimate for the volume of a FSAC-KK 11888 with adequately fleshed spines.

However, no direct parameter exists for how much more fleshed Spinosaurus' spines would be, so bearing in mind that it's almost certainly still an underestimate, I'll simply assume a doubled thickness from the GDI segments to add to the new model. 

For example, the volume of FSAC-KK 11888 in that GDI is 4443.13 litres, with a ridge of 459.98 litres. Doubling the ridge, we get a total volume of 4903.11 litres with a ridge volume of 919.96 litres, meaning that doubling the ridge would make it account for about 18.8% the mass of the animal.

Applying this to FSAC-KK 11888 gives us 4591.26 litres, which gives us a mass of 4820 kg with a density 5% greater than that of water as proposed for spinosaurids by Greg Paul and Asier Larramendi (2020).

The really big specimens get an update in size as well. As of now, the only one of these that can be estimated from overlap with others is NHMUK R-16421, which SpinoInWonderland restores at 38.7% larger than FSAC-KK 11888.
That allows us to estimate it at 12250 litres in volume (4591.26x1.387^3), which gives a mass of 12860 kg with the recommended density - and it shouldn't be forgotten that the fleshing is still almost certainly too minimal, meaning the live animal has a very high chance of being larger than that.
This could change somewhat when more specimens are found that overlap with others and allow an estimate to be made, but for now, NHMUK R-16421 is the average, maximum, and minimum for possible adults that I am aware of.

Unsurprisingly, this size is quite consistent with what Nizar Ibrahim estimated in the 2020 'Ask Me Anything' on Reddit.

@GuateGojira 

Since you follow the information on Spinosaurus quite closely (as I do) and since we have had a number of discussions on it size, you may find this post to be worth a read. There's really no shortage of vindication for what Ibrahim said. Reply