The Folk ternary diagram has been used in several large seafloor mapping programs and ocean plans. It has been selected as the foundation for the substrate classification in both the UK and US national classification standards.
The original Folk ternary diagrams were published in 1954 in a paper title "The distinction between grain size and mineral composition in sedimentary-rock nomenclature." (Journal of Geology, Vol. 62, pp. 344-359)
The figure above is the gravel-sand-mud ternary diagram. A sand-silt-clay diagram is also available in the original paper. The purpose of the paper was to introduce "a system of grain-size nomenclature of terrigenous sediments and sedimentary rocks." In it he states "The proportion of gravel is in part a function of the highest current velocity at the time of deposition, together with the maximum grain size of the detritus that is available; hence even a minute amount of gravel is highly significant. For this reason the gravel content is given major emphasis, and it is the first thing to determine in describing the specimen."
This seems to coincide well with the needs of marine scientists in more northern latitudes of the United States, as evidenced by the use of the diagram in Oregon and Massachusetts. It may be common that users alter the thresholds to accommodate local needs. There is evidence that coarser grain sizes correlate with higher biomass and biodiversity, so "hard bottom" is being preferentially protected in some areas. Therefore, the emphasis on even small amounts of coarser grains could be justified.
But why is it used? Is there more to the story than it being passed along?
The author of the Substrate Component of CMECS stated that Folk was used "due to the number of people that use Folk, particularly, those who wrote us comments about what system to use in the public review. About 3/4 of those who responded on that issue said "Use Wentworth and Folk". Among the 75% were many from NOAA and USGS, also saying "Wentworth and Folk". I like Folk for its simplicity, but Pre-CMECS I used Flemming for some work and I always liked the data that went into developing Flemming's system. In the end it came down to the public review comments though. It would have been hard to explain to FGDC why we didn't use Wentworth and Folk, if we had done otherwise."
A scientist associated with UKSeaMap stated, "The reason the Folk diagram was used was partially a result of what data were available." However, the UK scientists took the step to modify the thresholds and test them. I'm still not totally clear on how they initially chose the modified thresholds that were used, but it worked: "These thresholds were tested recently to assess their relevance to biological communities. The correlation is better for some habitats than others but it was decided that these were still the best thresholds to use for defining habitats according to the EUNIS classification system." (See UKSeaMap 2010 Technical Report 3.) Here is the UKSeaMap 2006 ternary diagram:
[This article would benefit from some citations. I might get around to it, but if you can recommend some citations, please leave a comment.]