I've been looking into this quite a bit and it would appear Nexus have in fact got enough funding for everything they wanted.
The Merseyrail example you mention is going to cost them £460 million, in return they will be receiving 52 trains at 65m in length, with dual voltage capabilities amongst other things.
If we work on each trainset costing £8.5 million, and we order 40 trainsets that would be a total of £340 million, a new depot would probably cost around £30 million. So a £370 million total. It's only £8 million more than what they have to spend.
I have a feeling that part of the contract for building the trains will include a new depot to be built. So whoever wins will also find themselves building a depot, possibly with Nexus paying them yearly for upkeep and possibly for them (the train builder) to have their own staff at the depot. Much like Siemens has been doing.
40 trains is even less than Nexus wanted - the plan was for 84/42, depending on the 4/2 car per unit configuration. Either way, that's less than we have now. Yes, we do have a few that are more-or-less retired, but Nexus has only been able to maintain the timetable without regular cancellations by reducing the service a little.
Admittedly, the newer trains will have less faults-per-km than the existing fleet, but that's purely due to age, and when this new fleet reaches the same point, we'll be in a similar place; not accounting for expansion/increase in service/other factors.
The original intention was that the signaling system would be replaced along with the rolling stock at this part of the modernisation scheme, yet we've seen no mention of this.
My issue is that we've been given less than was asked for in monetary terms (from my memory of the numbers), based on a request for less than we have now, and it seems the design spec is not actually looking to take advantage of what is a once-in-a-generation opportunity to not only shape the service Metro provides, but what Metro actually is (and how it will compete in the market).
Capacity is mostly fine on evenings and in the middle of the day, but rush-hour and event periods are becoming noticeably worse. Capacity can be added by either or a combination of less seats, longer trains, more trains or faster speeds. Some of these options are mutually exclusive without other changes, for example you can't have faster speeds AND more trains; unless you have an advanced dynamic-block signalling system.
Sadly, it seems that rather than exploring any of these other options, Nexus and the NECA are happy to simply remove seats. As everyone will know, this has been a contentious issue, most notably from the Sunderland side of the region.
On the matter of the depot, I'd be surprised to see Gosforth remaining the primary fleet depot. The new trains will have requirements that simply can't be met cost effectively in the existing location. Any new manufacturer will be likely to insist on a new or heavily refitted facility, as appropriate servicing facilities are a core part of their ability to meet availability obligations. One such example could be the requirement for equipment capable of lifting two or more entire 4 car trains for work; if that design is chosen and the manufacturer doesn't propose and get agreement from Nexus on a non-articulated design.
Refitting Gosforth would be very expensive, and operationally complicated due to the requirement to service the existing fleet during any conversion.
My prediction would be the insistence of a new depot, with funding either being found from operational expenditure under a lease agreement, or from financing based on the realisation of Gosforth's land-value when the new depot comes online and the old asset can be disposed of. Hopefully if this option is taken though, we'll still see a line through the existing depot site plus some stabling roads maintained, as the location is operationally beneficial.