Thank you both for your help! This has been rather enlightening to me. I do have two last questions for now out of curiosity, though: given the inaccuracy of the Smithsonian’s description, what actually causes the coral skeletons to dissolve at low pH (i.e. by what mechanism is the dissolution brought about), and what does the H+ actually do?
The increased H+ reduces the carbonate concentration in the water.
OK, here's a more realistic primer on solubility of anything.
Let's take a moderately soluble and simple material such as sodium chloride.
If you put a solid chunk of that salt into any water solution, sodium and chloride ions individually pop off off the surface and into the water. The rate of this process is largely independent of the concentration of sodium and chloride in the water. It always happens, and happens amazingly fast on a human scale. Likely millions per salt chunk per second.
If there is any sodium and chloride in the water, there is a simultaneous process of sodium and chloride ions each popping onto the surface from the water. The rate that this happens is directly proportional to the amount of sodium and chloride ions in the water.
There is some point in sodium and chloride concentration where these two processes exactly cancel each other out, and there is no net dissolution or precipitation, even though both are happening at a huge rate on a microscopic scale. The point is called the solubility limit, and no more of the solid chunk will disappear into the water, even if individual ions are flying back and forth.
Skipping now to calcium carbonate, all of the same processes exist, plus some others. The most important is that the concentration of carbonate changes with pH. Once below pH 8 in seawater, each drop of 1 pH unit drops the concentration of carbonate in the water by a factor of about 10 by converting more and more of it to bicarbonate:
CO3-- + H+ --> HCO3-
Bicarbonate plays no direct role in the solubility of calcium carbonate, so at lower pH, the carbonate concentration in the water declines, the rate at which carbonate pops back onto the solid surface declines, and thus the net solubility increases because the rate that calcium and carbonate pop off the surface was unchanged.
I hope that makes sense. :)