Possible behind-the-scenes (ILM) design approach
I think it’s fair to say that ILM is almost automatically associated with
Star Wars and therefore it must have been quite a delicate matter (not without controversy back then) that the same guys who designed the spaceships for SW (e.g. Steve Gawley) were given the opportunity to design new spaceships for Star Trek.
At first glance the Earth Spacedock looks like a Star Trek version of the Death Star and the chasm of Excelsior like the hangar bay of an imperial Star Destroyer.
But the romantic in me says that this is just a deceiving appearance while instead the ILM designers and model makers possibly approached their Star Trek III creations rather with the intention to feature these as “counterparts” to their “cousins” in the SW Universe.
I.e. while the Death Star obliterated planets with antimatter, the Earth Spacedock was designed to
protect Earth from antimatter-loading accidents and lethal fallout (I wrote a yet unpublished treatise on the “orbital mushroom” last August), and while the hangar bay of Lord Vader’s
Devastator had a facility to “vaporize” Princess Leia’s ship (deleted piece of dialogue), one design purpose of the
Excelsior’s hangar bay could have been to enable the “repair” of smaller vessels since it appears to have essential characteristics of a miniature onboard spacedock to substitute the lack of such facilities at the outermost borders of Federation space (“Star Destroyer” vs. “Star Explorer”).
In-unverse rationalization approach
According to the context of ST III the
Excelsior was still “the great experiment” “ready for trial runs” and hadn’t been tested under normal space flight conditions, yet (“Looking forward to breaking some of Enterprise's speed records
tomorrow.”), and shrouded in classified mystery (“She’s
supposed to have transwarp drive”).
(This is how I imagine an early session of the Starfleet admiralty board under its chairman and veteran Robert Comsol, conjectural but inspirational, I hope)
“Mr. Thorndike, while you know that Starfleet and the Federation Council approved your orignal new starship class design, you’ve fallen far behind schedule because of this overambitious desire to reinvent the wheel [transwarp drive]. You’ve stated, I quote from your latest progress report, whose number escapes me because there have been way too many, that it is impossible to evaluate the performance and potential of transwarp based on computer simulations.
You’ve further stated that it is therefore inevitable to test transwarp drive under real space flight conditions but cannot assure that the outcome will meet the expectations.
Looking at your construction design plans, correct me if I’m wrong, you intend to integrate the transwarp drive components irrevocably
into the ship.
But should transwarp turn out to be failures, Starfleet will, in essence, end up with a new starship carrying redundant dead weight.
This is unacceptable.
It is the opinion of this board that the transwarp components should be implemented in a modular fashion in the engineering hull of the 20th starship design. Should the transwarp experiment fail, the space allocated to such components can and will be put to other uses.”
I’m definitely not suggesting that transwarp was a failure (just because Scotty had sabotaged the T-drive before it even got a chance to be tested

) – and in the early TNG phase Andrew Probert considered the
Enterprise-D to use transwarp drive where merely the “trans” had been dropped – but the modular nature of the components inside the chasm, from ST VI on, is rather evident, IMHO.
These modules look like they might perfectly fit into other Federation vessels, but stick out like a sore thumb inside the chasms of the Excelsior Class: A module designed exclusively for the Excelsior Class would fill the entire available chasm space, but what we see leaves plenty of unused (and wasted) space port and starboard of these modules which begs for explanation.
Bob
P.S. In ST VI the
Excelsior had been studying "gaseous anomalies" and I wonder if these studies required physical contact. I won't exclude the possibility that the
Excelsior literally "vacuum sucked" gaseous materials (here we go again, not too dissimilar what the chasm of Bespin's Cloud City in SW had been designed for...) for in-depth analysis into the chasm. The blue light visible in the chasm could be a part f it (though probably not a Cerenkov glowing effect).
Another hypothetical question would be how to resupply Constitution Class starships like NCC-1701-A operating in deep space with antimatter pod assemblies as illustrated in the Kimble cross-section drawing. Looks like the chasm has the right proportions to transport such antimatter pod assemblies for other starships (and in case of confinement failure could just "spit" these out in a manner of speaking).
In general I feel that the cargo transport capabilities of other starships next to the
Enterprise (i.e. Miranda and Excelsior Class) still merit further evaluation.