It often seems like people refer to any kind of ice stuck on something as frost. If one looks in books or the Internet, one can usually find a specific term for the many different and interesting ice formations, but a term used by one group of people generally differs from that used by another. Wouldn’t it be nice for scientists and naturalists (at least) to have some generally universal, agreed-upon way to describe any surface-ice formation? Snow crystals have such a system, so why not surface-ice forms?

We could try to establish word-names for each formation, like “frost flower”, but different people are used to using different names for the same thing, and they are unlikely to give up the habit, particularly if they are unfamiliar with the language of the term’s origin. For example, there are several well-used terms for the ice columns in dirt that I discuss in “Ice on the Rocks” including “frost heave”, “needle ice”, "ice columns", and “pipkrake”. But “needles” also names a type of snow. Another example: The white, curly ice hairs that can extrude from plant stems and logs is sometimes called “frost”, “frost flowers”, “Ice flowers”, "needle ice", and “sap crystals”. But the two flower terms are also used for several other, very different things. Though a few terms seem to be used consistently in English, like “hoarfrost” and “icicle”, most aren’t. And probably none work across all languages. So, instead of using word-names, a set of symbols may work best.

Wilson Bentley gave a good start to such a universal code for surface ice way back in 1907. His article “Studies of frost and ice crystals” gave an ingenious method for classifying several types of surface ice [1]. The article is a fantastic source of information about hoarfrost, window-pane frost, and ice that grows in bodies of water like lakes and rivers. Unfortunately, I know of no one except Bentley who ever used his system. This lack of use is probably due to issues unrelated to his system’s merits, so  his system might need only a little revision and some promotion to get it into use. My purpose here is to suggest such a revision.

Bentley’s system used three letters to classify many kinds of ice-forms:

Position 1: capital letter giving the kind of frost or place of deposition.

Position 2: capital letter giving the characteristic form of the ice.

Position 3: capital letter giving how common the ice form is.

Position one could be "W" for window frost, "H" for hoarfrost, "I" for window-ice, "M" for massive ice (e.g., ice in puddles, lakes, and rivers),  and "S" for hailstones. For short, we could call it the WHIMS system. As an example, the following ice formation is “IFA” in his system.

Although the “I” stands for window-ice, ice on smooth black metal, as in the above case, has similar growth patterns.

The second letter “F” stands for feather-like, and the “A” means that it is among the most common types of I-forms. The ‘striped tail’ part is mine; I called this the striped-tail formation before I learned of Bentley’s system. Though his system is a great start, I see some problems. One, the first letter seems to ignore the forms of frost that grow on other surfaces that do not produce patterns like those on glass (e.g., rough plastic). Two, his distinction between window-frost (W) and window-ice (I) doesn’t seem right; I think most examples he gave for W can be lumped with I, being examples of ice that formed in a film of liquid water. Three, his system doesn’t include many types of ice, such as ground ice and doesn’t distinguish ice that formed from a succession of sources with ice that formed from only one source. Moreover, his choice of second letters may cause some confusion because one letter may represent several different characteristic forms; for example, the “F” in “IFA” stands for feather-like, but F can also stand for fiber-like, filament-like, and flower-like. Another unclear point about his second letter is that ‘the characteristic form’ may refer to the shape of the individual crystals or the shape of the overall pattern. I think these two cases should be separated.

The third letter in the WHIMS system, being about how common a type is, seems biased to observations Bentley made at his farm. However, we could just as easily view the third letter as representing the order in which the form was named. For example, if the last letter is “D”, then we can understand this as meaning that the ice form was named after the “C” form. The nice thing about this is that we can keep adding new forms as they are discovered.

Later, Ukichiro Nakaya published a symbolic system for classifying falling snow [2]. His system appears similar to Bentley’s, but instead of using all capital letters, only the first position has a capital letter. The second position has a number, and the third a small letter. For example, most of the snow crystals shown on this blog by Mark are P1d, P1f, and P1g. The first letter gives the main characteristic of the crystal, with most being either “P” for plate-like (i.e., tabular) or “C” for columnar. A nice feature of Nakaya’s system is that the symbol shows when the main characteristic changes; for example, “CP1a” is a crystal that began as a columnar type (C), and then P1a crystals grew on the two ends of the column.

So, considering the good and bad points of these two systems, here is my idea for a revised system. I call the system “BEDFISH” for a reason mentioned below. In this system, a name for any ice type consists of four positions.

Position 1: a capital letter designating the main source of water of the ice.

Position 2: lower-case letter or letters designating the overall pattern or size of the ice formation.

Position 3: a capital letter to characterize the shape of the individual crystals.

Position 4: number, starting from 1, indicating either how common the form of this water source (or sources) or when it was named.

I came up with seven symbols for the first position or source of water:

B: from bulk liquid, meaning puddle-size or larger.

E: from extruded water; for example, soil moisture or plant sap.

D: from droplets, condensed from vapor or deposited as a droplet.

F: from a film, thick or thin, and including water in a narrow channel.

I: isolated crystals formed on a surface but from the vapor.

S: from snow crystals lightly deposited on a surface.

H: hoar frost, meaning crystals similar to I, but densely packed.

Hence the name BEDFISH. These symbols cover all the cases I thought of, and may be sufficiently general to cover other cases. Bentley’s system did not include forms that fit into E and D. For example, WHIMS did not include rime, large frozen droplets on a surface, icicles, ice stalagmites, and other ice forms that fit here in group D. Icicles don’t perfectly fit into any of the seven categories, but is most similar to other types in D.

An example:

HuT1 = hoarfrost (H), uniformly spread over a surface (u), consisting of tabular crystals (T) of type 1.

For this system to cover as many cases as possible, there should be a way to designate successive sources of water. A simple way combines symbols using a slash “/” as a separator, with the most recent source first. For example, HcC2/Fg3 could represent hoarfrost (H) with a “combed” look (c), made of columnar crystals (C) of type 1, but clearly growing from a frozen film (F) with a grainy (g) pattern type 3. For this case I assumed that the shape of the crystals in the frozen film is unknown, and this explains the lack of a second capital letter in the F part of the name. (This is just a possible case – I haven’t settled on these meanings for “c” and “g”. Besides, maybe this should be "u" for uniform, instead of "c" because the combed look comes from the Fg3.)

 

Let me know if you have suggestions.

- JN

References

[1] Bentley, W. A. (1907) Studies of Frost and Ice Crystals. Monthly Weather Review. This was published in five successive issues, from August through December. Each part is available freely online, with the last part (e, below) having 274 crystal images.

a) http://docs.lib.noaa.gov/rescue/mwr/035/mwr-035-08-0348.pdf

b) http://docs.lib.noaa.gov/rescue/mwr/035/mwr-035-09-0397b.pdf

c) http://docs.lib.noaa.gov/rescue/mwr/035/mwr-035-10-0439.pdf

d) http://docs.lib.noaa.gov/rescue/mwr/035/mwr-035-11-0512b.pdf

e) http://docs.lib.noaa.gov/rescue/mwr/035/mwr-035-12-0584.pdf

[2] Nakaya, U. (1954) Snow Crystals: Natural and Artificial (Harvard University Press).