All good points here.
The strings, anchored through holes behind the saddle, pull up and over the saddle and apply a constant force on that bridge, acting as a lever, that works to rotate the bridge forward. Engineering the bracing in front of the bridge, which includes the center of the “X”, is both a repeatable science and an artform. The smallest variation from one instrument to another in top thickness and stiffness, brace stiffness, bridgeplate placement, etc. can and will have an impact on that guitar soundboard’s performance, let alone its appearance.
If the bracing will permit it (aka “failure” to one degree or another), the string pull that seeks to rotate the bridge forward will cause the front of the bridge to plough downward (sinking the soundboard with it). In these cases you won’t typically notice severe bellying, as the stress is being relieved in front of the bridge. By contrast, should the X brace be a veritable unmovable rock, something you could seemingly stand on, the torque being applied to the bridge will NOT succeed in pushing the bracing and soundboard down in front of the bridge. If the structure of the lower bout will permit it (aka “failure” to one degree or another), the back of the bridge will tend to lift and, if the soundboard stays firmly attached to the bridge, the soundboard will lift (or “belly”) with it.
Most center soundhole, X-braced guitars achieve a balance, an equilibrium where there is both some depression (however slight) in front of the bridge and some raising (however slight) behind it. I am not suggesting that is a good thing, and some of us are working to try to change that, but it is how guitars have been built for a good long while now.
This knowledge is most applicable if a sunken soundboard is the result of bracing failure, either glue-related or structural. Such a failure could occur over time or, as kostask has suggested, could be the result of a sudden, catastrophic event.
jedzep has identified another issue that can explain a sunken soundboard, and that would be related to dryness. Most acoustic guitar soundboards are constructed to have a convex radius, or “arch”. It is a relative topic, but wood will shrink across its grain if moisture is removed from it, and will expand if moisture is absorbed. A lower bout of a jumbo guitar that left its properly humidified factory birthing center at 17” wide, if left to dry out, will absolutely measure less than 17” wide due to the soundboard shrinking in width and drawing the sides in. Additionally, that 28 foot radius arch (or whatever) will tend to flatten under such circumstances, as the bracing beneath the soundboard retracts across its grain as well. That loss of arch pulls the bridge down with it and can result in the appearance of a sunken soundboard.
I would remove the strings and attempt to humidify the guitar (don’t err on the other side and introduce a host of issues by getting it wet). If humidifying the guitar does not restore the arch in the soundboard, THEN you know you have damage to deal with.
