The name ‘weathering’ has always been taken to indicate a direct relationship between the atmosphere and the breakdown of rocks, i.e. at or very close to the surface. This is so easy to test that it comes as a surprise to find that nobody has really tried until recently (Yokoyama, T. & Matsukura, Y. 2006. Field and laboratory experiments on weather rate of granodiorite: Separation of chemical and physical processes. Geology, v. 34, p. 809-812). Tadashi Yokoyama and Yukinori Matsukura of universities of Osaka and Tsukuba, Japan, placed small cut tablets of identical fresh granodiorite in three position: at the surface, buried above the water table and buried beneath the water table in one small catchment. These samples stayed there for 10 years. The only sample to show much sign of chemical breakdown of minerals was that buried below the water table. Does anyone claim that there is weather in groundwater? Just exposing fresh granodiorite in the laboratory to a constant flow of water chemically similar to the groundwater doesn’t accomplish the weathering (it is 50 times slower than when samples are buried). Chemical weathering needs to involve soaking, when grain boundaries break down so that individual grains can become detached and allow yet more penetration.
Most geoscientists who work on topics that involve chemical weathering, such as the changing release of tracer isotopes of strontium to estimate rates of weathering in the past, assume that it is all done by atmospheric carbon dioxide dissolved in rainwater or released by organisms in soil. It is accomplished by hydrogen ions that can be released by a great deal more processes than the formation of vary weak carbonic acid (e.g. organic acids and breakdown of sulfides). It now seems very clear that chemical weathering is a product of groundwater and burial, so should we call it weathering at all?