There is no central repository for sales statistics. There are many variations from publisher to publisher, depending for example on how many distributors carry your line, whether you are represented in the book trade, and so on. And, at the end of the day, adventures perform most poorly, as a general rule.
Having said that, the "whisper number" heard on the street is that for a lot of small D20 publishers it's now hard to sell more than 1000 copies of an adventure. (Even a few months ago, this figure would have been a lot higher.) With the number of new titles still coming on the market, this number is only likely to be driven lower.
To be honest, at 1000 copies, money is probably not being made, unless the retail price is high. Still, most publishers can hide from this fact by not accounting for all the expenses (e.g., if most of the labor involved in creating the book is "volunteer," or done by stakeholders who in theory would be getting a cut of profits...of which there are little or none), or by just not paying all the bills (leaving freelancers unpaid, etc.). My bet is that the market will drive the number down even lower, beyond the point where manufacturers don't even recover the cost of printing. Even then, there will be some willing to take a chance on publishing their own product, hoping or believing that it will catch fire and turn profitable (or accepting that publishing at a loss is a hobby activity in itself). At the risk of sounding Malthusian, we probably won't see a real bottom to the market until the typical newbie adventure publisher is only shipping 400 or fewer units in the first month of release.
Well before that point, publishers who have non-D20 games will find it more profitable to publish support for them instead of D20. Already we've seen a lot of the mid-sized publishers largely abandon adventures. Even with experience and quality on your side, it's just hard to get any sales traction fighting over the inherently limited market for adventures.