How to list a species' range?

My votes are as follows, somewhat of a duplication of what I said before, but not quite:

First - Already written descriptions that make sense probably shouldn't be changed, unless the editor doing it is correcting it - just seems like a lot of unnecessary work. I often change ones that don't make sense to me, are incomplete, or are incorrect, as a matter of practice, but I don't mess with ones that already make sense.

Second - In this country most people "think" left to right and top to bottom, whether it is looking on a shelf, reading a map, or looking at a document or picture. That's the way we are tought, and that's the way most people will most easily comprehend a description of distribution. The one intuitive exception mentioned is when something crosses a border from a particular direction, or is found along a coast line, and is found in just a limited part of our area. Then something like "inland from the Atlantic Coast - - -" or "north from Mexico" make the most sense to me. So, weather it be listing places or land forms, I mostly favor right to left and top to bottom. Four corners often works well, and I think most people comprehend top left to top, then bottom left to bottom right best.
Second best to that (and sometimes more convenient) would be a clockwise approach, starting from the top left. When looking at something circular, most of us start near the top or upper left, and move clockwise.

Third - I definitely favor using geographic features (rivers, mountains, basins, deserts, coasts, whatever, or whatever) over politcal units, because they tend to provide natural and easy boundaries (or automatically define areas of occurence). Also something like a coast, river, or mountain front can give a single LONG easy boundary that doesn't require listing a bunch of political units.

Fourth - I mentioned international issues. I think it is always good to include the whole distribution, and I personally usually give that in very general terms before the more local distribution (big first then local), but it could be the other way-round. I do think it best to be very brief and general, especially when it's areas outside of the US and Canada being described. And, when both the US and Canada are included within one distribution area, I definitely prefer to treat it as one unit with one set of boundaries, and say something like "western U.S. and Canada" first, and then define the area more closely. Often I don't even think mentioning the country is necessary. Many people prefer to list the areas within each country separately, and will list the distribution in one country first, and then start over with the other country. This just wastes space, and is unnecessary. The international boundary is no different to the insects than is a state or provincial boundary, and both countries are in our covered area.

However, when Mexico is the country, since it is not part of our covered area, I prefer to just add something like "into Mexico", or "into northeastern Mexico".

Fifth - this is something I didn't see brought up (maybe it was), but it is similar to the physiographic province / geologic feature thing. Sometimes I will combine a basic habitat with the range when the habitat helps to define the range. It seems a natural union, but leaves the habitat field bare (or one can ellaborate there). So, I might say something like "Great Basin Sagebrush Desert", "Coastal Dunes" or "Grasslands", and then define the range in somewhat more detail.

This is overly simplified, but covers most of the basics. Obviously there are many individual cases that require individuality in describing their unique ranges.

1. Always start from SC, then counterclockwise :)
-or-
2. Come up with some simple way to generate a range map from points [states]- lot of different ways I can imagine doing this. I think, on balance, most folks would prefer visual data.

To simplify, you could select from 30 or so common distributions/regions - you could pick one or more and maybe some individual states.

It would be handy to overlay the range map with the BG data map - new state records might pop out.

One of the most common distributions is e. US. What is the best way to represent this in the most concise and understandable way?

TX-FL-NY-IA - This is my first choice. I like the fact that the first two states are listed as they occur, west to east.

e. US. - Very concise and understandable my all, but somewhat vague.

NY to IA, south to FL-TX - To me, this is an abomination. I don't like listing east to west, nor do I like zig-zag listings.

NY-FL, west to TX-IA - Ditto above comment.

IA-NY
TX-FL - Thinking outside the box. Four corners are listed as they occur, ie, NW-NE, above SW-SE. A somewhat interesting suggestion, but I doubt it would get wide acceptance.

e. US (TX-FL-NY-IA) - This may be the best compromise.

Texas to Florida to New York to Iowa - State names written out (regardless of order), to me, it too long.

Full disclosure, I reside in Texas.
.

I assume we are trying to standardize somewhat??

Anyway, here is basically the pattern that I think I tend follow the most. I don't really think about it while writing distributions, but I think this is pretty close to what I do.

It depends a lot on the nature of the distrubution and how large it is. However, I "try" to keep it simple, and when describing the distribution, I tend to start at the northwest and work east and then south as I list boundaries. I try not to list every political unit along a boundary, so instead of saying something like "British Columbia through Alberta, Saskatchewan, Manitoba, Ontario, and New York, and south through - - - ". I would probably say "from British Columbia to New York and south to - - - ".

If it is international involving Canada and the US, I usually list states and provinces without mentioning countries, but I might head it with "United States and Canada" or "Canada and United States" (I try to list the country with the bulk of the distribution first, but I have a habit of writing US first, so sometimes I forget that detail). If it ranges into Mexico, I usually just say "into Mexico"; or, perhaps "into northern Mexico" if the distribution doesn't go far into that country. I initially may have listed some Mexican States, but I tend not to now, since we don't cover that area, and since it takes a lot more space.

Of course sometimes the distribution takes in all of one country and part of the other, or even all of both countries, so I won't list states if it's the whole thing. If it's only a small area, I'll define the area within the state or county, or whatever. If it is a rare and endangered species, I try to state about where it is without actually giving details (such as "northwestern Colorado", instead of "only north of Rangeley, Colorado") [a made up example].

If the range does include other parts of the world, I try to always indicate that too, because I think it's important. So I might say "Holarctic", or more likely on BugGuide, because a lot of readers might not know what that means, I would probably write something like "boreal [or temperate, or ---- ] Eurasia and North America".

There may be better ways, but most other ways I find tend to take up a lot more space.

So, I hope that is the sort of thing you are looking for.

My concern is that your suggestion of something like "CA-GA-NB-WA" isn't intuitive. Will the average BugGuide user looking at the info page understand what is implied? It makes sense to me, but I just don't think all of our visitors and members will quickly grasp its meaning. It doesn't take much more time to write a more descriptive assessment of range, and it certainly doesn't take any longer to read "Transcontinental" or "Eastern United States and adjacent Canada" than "CA-GA-NB-WA" or "TX-FL-NB-ND". How we present information isn't just based on what we as editors/experts want to see, it should also be about what the community would want to see.

I think John Carr's suggestion below of "Eastern United States and adjacent Canada", "Eastern United States and adjacent Canada", "Southeast coastal plain", etc., is a good target for how we should describe distribution. It's a quick generalization of what we know about the insect's range, plus it is easy to read and understand. It can of course be followed by more specific information, such as published state/province records if desired and/or available.

In the 1979 Catalog of Hymenoptera of America north of Mexico, the editors recommended the use of a north to south and east to west bracketing distribution such as Maine south to Florida, west to Washington and California for species with abundant collecting records. Otherwise, it was just recommended to list the several states and provinces in the same logical order. For me, at least, the case of Pelecinus polyturator provides an interesting case.

On 9/24/12, a 9/9/12 photo from Prince Edward Island was submitted to BG, and the photographer wondered whether it was unusual to find that insect on that island and jokingly asked if it might just have blown in there, as the locals had never seen it before. I looked in the Hymenoptera catalog and found the distribution listed as Quebec and Ontario south to Argentina, but searching the web, I found a paper recording numerous records from Manitoba and a web site where a fellow posted a photo from Nova Scotia, saying that his 67 year-old father had never seen one of these before.

On 9/27/12, a fellow from New Brunswick uploaded an 8/27/12 photo to BG and said the insects were all over the place, sunning themselves in his garden and patrolling his yard, flying close to the ground and landing often.

Suggestion for distribution listing: PEI w. to MB, s. to FL and TX; also Mexico s. to Argentina.

Mike, thanks much for initiating this important discussion. You sure have many valid points. Consistency and some loose guidelines would be welcome indeed, but different organisms have very different distribution patterns (aquatic vs. land, flyers vs. non-flyers, etc.) and data on many obscure groups are very sketchy. This means that describing the ranges will take a lot of discretion and judgment in order to strike a balance between detail and brevity.

In general, "the four corners method" sounds like an acceptable compromise in cases where we cannot do without mentioning individual states – but let's feel free to add more ‘corners’ for odd-shaped ranges to better define the polygon. This method appears to be especially well-suited for ranges restricted to the continental interior, although here, too, describing them in terms of mountain ranges, river basins, etc. would be preferable.

In most cases, though, the format "xxx west/east to..." is much easier to visualize for me than any polygon.

The scheme should be rather flexible. I do not believe a single standard would work for all groups and all range types. It is fine to proceed from south to north when dealing with species whose range lies mostly south of the border (e.g., "[...ranging into] TX and AZ"), but otherwise I have a problem with the south-to-north directionality.

The northeastern part of the United States and adjacent Canada is arguably the best studied region of North America, and that’s why I always start there. (I can be biased: after all, it’s where I live – and my perspective is also affected by my Palaearctic background.)

For the ‘eastern’ spp. the Atlantic ocean is a clear eastern distribution limit, while the western limit is fuzzy and subject to change (advance); I would use the same logic for the western species and go from the Pacific Seaboard to the fuzzy eastern limit. So, as far as the ‘n-corners method’ is concerned, I would always start from the top of the map and from the oceans.

Ideally, though, range descriptions should be based on physiographic features rather than jurisdictions, which are purely artificial – and I would prefer seeing phrases like ‘boreal forests of NA’ or ‘Southeast coastal plain’ to anything mentioning states/provinces whenever possible. John Carr’s approach and wording look right to me. Too bad our handy sources often fail to provide the necessary range analysis and only list states/provinces instead.

The East is fairly uniform. Insects tend to be found anywhere east of the Great Plains in the United States, or in the Northeast United States, or in the southeast. Southern Ontario counts as United States. The southern Appalachian Mountains (south to Georgia) are Northeastern. The coastal plain may be distinct from the rest of the Southeast. Unless the distribution is very well known (e.g. butterflies and dragonflies) listing corners of records is less accurate than listing regions.

I have written many ranges like

"Eastern United States and adjacent Canada"
"Northeastern United States and adjacent Canada"
"Southeast coastal plain"
"United States and southern Canada East of Rocky Mountains"

changing these to lists of states would generally make the page less accurate and harder to understand.

We could add explanations of these standard ranges to the glossary, possibly under the corresponding ecozone names.

If we are going to list four corners, I would much rather move clockwise than counterclockwise (NW to NE, SE, SW). And then there are cases in which it is better to describe the area other ways: east of the Rockies, most of New England, perhaps?
I hope that the maps eventually become more informative than this.