Just some comments on the noticeability of cheating.
First of all, an aerial photograph of a cruiser (which you are going to get when and how?) will not tell you much. I would estimate the deck-area coefficient to be at minimum 0.6, even for ships with a block-coefficient of 0.4. A deck-area coefficient of 0.5 is achieved by the edges of the deck (as seen from above) being straight from the bow to the beamiest point of the ship, and then straight from that point to the stern. Once the sides of the deck start curving out (being convex) that coefficient goes up. I have yet to see a design where the sides as seen from above curve inwards (concavely). Also, take note that there is A LOT of cutaway under the stern. Submerged bulges or other appendages are also unlikely to be discovered by aerial photography - remember that you are unlikely to get good, clear, accurate aerial photographs of any power's vessels unless they are in your home area. To be properly usable, the angle from the axis of photography must also be known, and one also has to take into account any differences between length+beam of deck and length+beam at waterline.
Also, the combination of draught and loading-submergence of the hull is a significant factor - a 10408ton (standard) design I have done earlier has a submergence of 699tons/foot. This means that floating one foot deeper, it would displace 11107 tons standard - and how are you going to spot the difference of submergence of one foot, particularly when you most likely will never have the chance to get an accurate measurement of the ship's actual draught, or the freeboard from the surface to the lowest row of scuttles? Also, keep in mind that when the ship is observed, it will almost always displace more than its standard tonnage, since it will always carry fuel and reserve feedwater - and you will not necessarily know how many tons of that my ship can or does carry. The one exception is if it has been in combat and worked itself low on ammunition and fuel.
With the Japanese cruisers, the one thing that lead the Western powers to suspect that they were severely overweight (despite the Japanese attempts to build the ships so that they would have the proper standard tonnage) were the capabilities of the ships - I believe British naval architects are on record as saying that the Japanese cruisers were either within tonnage-limits but built of cardboard, or else severely overweight, when asked why they themselves couldn't build equally good cruisers. As such, increased protection going into cheating tonnage will make suspicion less likely, while cheating to allow for instance a 30knot 16x21cm/8.2inch cruiser on ostensibly 13000tons standard is likely to cause much suspicion abroad.
With regards to larger vessels, such as capital ships, discrepancies even as large as several thousand tons are unlikely to be much noticed - if my typical