Discussion: North Lines

The exciting thread on the EOC has been hijacked (as often happens) by something only slightly related. I apologise for my part in that, and herewith start up a new thread for something that interests me. Others may well start threads about formlines or all the other rules in the book that may or may not have been broken.

It seems that in Sweden the rule is to print maps aligned to the national grid, which is 3 degrees away from magnetic north. Why? Around here (where the discrepancy can be anywhere from 18 to 24 degrees) this would be unthinkable. But more to the point, it's a piece of cake(*) to set up the map to mag north. It's easy to relate it to the national grid with the "real-world" link in OCAD, and to switch from paper to national coordinates.

* Though the ease of doing it hasn't stopped some mappers who should know better from overlooking this, heh heh.

Did you want some discussion on this or is this just a general comment thread?

About 10 degrees in some parts in Sweden, like Tårnedalen.

Is it anything to do with Sweden's 'anything goes' GPS rule? I've competed in Sweden, a long time ago admittedly, and the maps then were aligned to magnetic North.

As a general rule though, if they (SOFT) want to do it, fine, but if thy are holding IOF events (WOC, JWOC, World Cup, etc.) then the maps should be to ISOM / ISSOM standards.

My interpretation of these standards is that technical issues like North lines, symbol sizes and colours, course markings etc. are an absolute given, while interpretation issues require some flexibility on the part of the mapper and controller/event advisor etc. This is not to say that there should be open slather to use multiple form lines, or overmap green, or whatever.

"It seems that in Sweden the rule is to print maps aligned to the national grid"

WTF are you talking about?

Utilizing magnetic N lines not parallel to map's edge is 'heretical'...somehow doing it this way hijacks the "organized" way the mind prefers to work (or has come to work) when reading an orienteering map. Of course, the mind can/will adjust but why ask that it do so?

Old New England map (orange lines for another purpose) w/ MN at about 45 degrees:

Nixon, are you sure you did not subconsciously use text "Brelja" or those undergrowth stripes under you thumb compass capsule to align map on your way to #6 at EOC long final. Would guide you off, a bit right, and make it easier to make a parallel error.

Simmo said it all.

In a prior discussion, some suggested advantages were national database integration, ease of combining maps, gps-tracking alignment, non-orienteering uses. but I would still be interested to know the official motivations.

No, 6 was just a parallel error, but 2 in the middle Q was almost certainly due to the offset north lines.

That's not what my point was, I know what the EOC maps were like, I was there and I ran on them. But to say it is now the rule in Sweden because of one set of races is a bit of an exaggeration.

And if you look at 6 on the long final I went too far right, if i used the green lines, text edge of map I would have ended up too far left...

:)

I cannot follow a compass bearing so it doesn't matter which way the north lines are pointing. I just use Simmo to guide me to the controls.

Now how about a local C meet, where you just want to fit the whole area on 11x17 paper at 1:15000? What would be the greater sin - a non ISOM scale, or north lines that aren't parallel to the paper?

Blessed are those who need to fit a local event on 11"×17".

Nixon, it is written in point 5.2 of the Swedish Mapping Standard. In italics to show it is a deviation from ISOM.

The primary feature is very flat and very wet, and there are canoe marathoners in my state.

Mr Wonderful, I's say smallest sin might be printing it at 1:15 000, north lines not parallel to the paper, but objects like texts, marsh & veg stripes and such aligned to north lines (not paper as Swedes do). And no nice frames drawn around the paper, so the only wrong aligned edge would be the the edge of the paper. But what do I know.

Jagge is right about the correct way to print a map that is not aligned to the paper, and the map posted above by bl did not do it properly. The key items are:
1) If magnetic north is not parallel to the edge of the paper, then it must not be anywhere close, probably 20 degrees difference minimum.
2) All map elements that would normally be aligned with north or with the edge of the paper must be aligned with magnetic north. This includes all text, point features, and pattern symbols.
3) No frames or any other ink lines aligned with the edge of the paper.
An example, the only way to fit this map on 11x17 paper at 1:10000:

(Note that the little corner marks are not to be printed, they're just there to indicate the limits of the available paper size.)

This is good to know. One of my upcoming sprint mapping projects is a 1:2000 area that fits perfectly (8.0" x 10.5") on a standard sheet. However, the street grid here is about 42 degrees off magnetic north.

However, the street grid here is about 42 degrees off magnetic north.

Don't you just HATE that? It's like urban planners like to stick grid-pattern streets on 45° angles because otherwise the street directory looks boring.

Maybe, depending on latitude, it means that the sun won't ever be shining directly directly down the street when it's rising or setting, thus reducing glare problems for drivers. (I have no idea if that's a motivation, I'm just making this up.)

Not in Seattle. There are places that the sunset hits you directly in line with the street directions plus the beautiful reflection off Puget Sound - blinding.

And when you print a course on a map like this, you also need to align the control numbers with magnetic north.

Absolutely. They fall under the category of "text".

I woudl be hesitant to print the sprint map that Patrick mentions at any angle other than normal (i.e. with magnetic north parallel to the paper edges).

I tried to read the Swedish mapping standard jwolff lined, but I can't figure out with my language skills what it actually means. To me it looks like it says map frame should be rotated to grid, and "meridianer" (meaning magnetic north lines not actual meridians?) should not be parallel with frame/paper. But there is nothing about rotating objects to grid or frame, just rotating them to "north". Magnetic or something else? Go figure.

Semi-official response (since SOFT fired the person responsible for mapping issues I wouldn't expect an official one) about why maps in Sweden are printed mis-aligned to magnetic north:

"That the map should be in a (geographic) reference system is obvious for everyone who is thinking a little bit. If you are using GPS while fieldworking the map must be in a ref. system. If the map is rotated in OCAD you have forever lost the alignment to the reference, even if you try rotate it back it won't be correct again.

But historically Sweden has, long before GPS was common, prescribed that maps should be in RT90. The reason was then that SAR, police and military easily should be able to use the maps, that was also why the printing of coordinates in the frame was prescribed. If the maps were really used by these parties is more doubtful.

But nowadays it's a must for an effective handling of map databases that they have their place in "reality" and there is a great understanding in IOF about the swedish position in this matter."

Sorry about the bad translation. I tried to stick to the original text as much as I could.

And I don't really agree at all with any of these points myself and I'm not really sure if the "facts" are even true.

Yes, but orienteering is a sport practiced with compass (magnetic north) and not with a GPS.
The map is still on paper format and not a computer file.
There was nothing more easily and nobody stops you make a copy of that file, the declination of your choice and use it for any purpose you wish, and print the map with magnetic north,parallel edges of the paper

Thanks Erik, the best contribution so far to my question "why?"

I'm still struggling to understand though, is there a grand plan for a seamless orienteering map over all of Sweden? Certainly I have moved from thinking of "a map" to "a mapped area from which a map is extracted for an event". But I'm confident that OCAD has the tools to combine maps according to the real-world coordinate relationship established for each file.

What's happening at SOFT? Maybe I'll redouble my efforts to translate the article in Skogssport 1/2012 "Kartgruppen hoppade av i protest".

whoever wrote that doesn't seem to know how to use OCAD's real world and paper coordinates. If you set up the base map in real world coordinates (effectively aligned to grid north) there is no problem using GPS and no problem combining with other maps set up the same way. For O events you print in 'paper coordinates (ie: aligned to Magnetic north). and if the emergency services really are using the maps - they have to get them from somewhere - when they ask for maps it would be easy enough to run off a few copies in grid north or provide the map in an electronic format that is aligned to grid north - you could even have a square grid that is usually hidden but 'unhidden' for just such a purpose. The whole idea seems to based on a poor understanding of these things - or am I missing something?

Gruver - post a scanned copy of the article - I'll translate it for you.

Magnetic North is not a constant value. Sometimes it deviates 0,5 degree per year
http://magnetic-declination.com/
If map is used for few years it is more simple to update it: just move North lines to right position. You not need to change frame, all writings, logo ...
It's only question why do they this on EOC?

0.5 degrees per year would be extreme, so over the life of most orienteering maps, the change in declination is negligible.

• In Falun it's changing by 9.7 minutes per year: http://www.ngdc.noaa.gov/geomag-web/#declination - same in Stockholm.
  
• In New York by 1.6 minutes per year.
  
• In Sydney by 0.5 minutes per year.
  
• The largest change rate of declination I can find for any plausible orienteering area is for Whitehorse, Canada (21.7 minutes per year)
  

Not in Seattle.

Most of the city is on a true north grid, except for 1.2 square miles of downtown (which is where my mapping project is), where the streets run parallel/perpendicular to the Elliot Bay waterfront.

And when you print a course on a map like this, you also need to align the control numbers with magnetic north.

This sounds like it may be tricky? I'm pretty sure that Purple Pen doesn't do this.

I would be hesitant to print the sprint map that Patrick mentions at any angle other than normal.

I'll see what happens if/when I actually want to use it. I want to map it because it would be fun to map, not necessarily because I want to have an official event there.

Seattle is changing at 10.4 minutes per year, so about the same as Falun.

Whatever the rules say now, I have a feeling that long term the Swedes on the right path. It's just better practice to make sure all your cartography is referenced to some well defined geodetic reference system

Before the computer age, maybe it didn't matter that much. But today, these coordinate systems are critical for enabling things like integrated GIS systems and GPS.

Magnetic north is a poorly defined moving target, and OCAD's so-called "real world coordinate" system is still pretty crude. (It's designed only for small regions, where the non-Cartesian nature of the coordinate system can be neglected).

People can, eventually, learn to deal with tilted north lines. But if you envision a future where GPS tracking is simple (and don't require annoying calibrations and affine transforms), or where you can pull up the local orienteering map database on your smart phone, anywhere in the world, you've got to start making your orienteering maps based on those geodetic systems.

Presumably, the people who are annoyed with Sweden for stubbornly opposing IOF can recognize how the IOF is stubbornly opposing the worldwide cartography community.

(Note - I realize that it's "possible" to be properly georeferenced, and still formatted to printing with alignment to magnetic north. This is not a completely trivial undertaking though. The real-world-coordinates option in OCAD is only a partial solution)

ok
we can change the rules, and AFTER we could run with a file.
But not before

I am currently mapping an area using the regions standard co-ordinate system. Come race day, the map will be rotated to magnetic north - about 8 degrees. Some symbols will also be rotated to align with magnetic north. Is there a consequent risk of decreased readability as a consequence? If so, is this an argument for any future standard to keep the symbology aligned to grid rather than magnetic? Discuss.

And just to highlight a new issue with regard to georeferenced maps.
I just received an email from a land manager granting tentative permission to use an area on the condition that we seek shape files for the location of an endangered species and overlay this on the orienteering map to guide the course setter. This particular orienteering map is not georeferenced or aligned to the local coordinate standard.

There are a number of symbols which are intended to be aligned to magnetic north (marshes, pits, x's, text, towers, etc...). It's best if you draft those in the rotated state (various ways to do that), because you never know when some little detail might get covered up, or you find that you need to cut a contour line, etc....

Marshes are a particularly nasty symbols to rotate. Undergrowth too. Although, it's odd that undergrowth hashing is not required to be aligned to north by ISOM, so young orienteers should definitely not get in the habit of aligning to that.

I would think a vector representation of a line or a symbol can be rotated without loss of sharpness, even if round-off error might shift its position by a very very tiny amount. But I suspect the answer to this will need an OCAD programmer and maybe depend on how the vector OCAD file is sent to the printer.

Like Blegg I have wondered about the paper-based view that OCAD has of the world (magnetic north up the page with real-world coordinates sort of an add-on) versus the GIS view of the world (objects have a position in a database and a map is just a view of the data).

For the moment the maps in my country are like postage stamps scattered on a large table. The fact that some are at a different angle to others (due to slightly varying grid-mag angle across the country) does not matter as far as I can see. Though I acknowledge the limits of my knowledge and hope for insight through this thread.

Were we to align our postage stamps with our table, would we then have a debate as to whether the rectangular table is properly sitting on the curved "floor" that is the earth?

Log's point about overlaying (national grid-based) data onto orienteering maps, brings me back to the here and now. We should be progressively geo-referencing all of our existing maps. Its easy, unless the process shows up significant distortion. In which case we need to be addressing the distortion.

Blegg, you suggest a number of symbols are intended to be aligned to north, but not undergrowth. I can find specific reference in the spec to some of the point symbols being oriented to north, but not the x's, not the marshes, and as you suggest not the undergrowth. However I have always believed (thru the examples I guess) that these should all be north-oriented. What's the basis for your interpretation that undergrowth is different?

In OCAD9 (not sure about O8) all that is required to rotate & realign all occurences of a symbol (point or area) is to right click on the symbol in the symbol box,[select by symbol], [all objects with selected symbol]. All of the occurences are boxed on the map, and the pattern can be rotated by drawing a line with the [direction of area patterns] on the tool bar - less than 10 seconds to realign the pattern or point symbol without moving its position.

Not sure if I have seen a rotated undergrowth pattern used for a narrow N-S aligned patch of bracken. This could otherwise be very confusing or even not drawn, depending on the relative position of OCAD's internal hatching grid

A narrow patch of undergrowth isn't worth putting on the map in the first place. I'm with j-man on this one. I'm pretty sure that the undergrowth symbol was originally intended only to denote an open area that looks appealing from a distance, but that is actually very slow when you get there, so you should avoid it. You can't map it as green, because you need to indicate its open appearance But over the years, the symbol has been abused, creeping into places where it doesn't belong. The control at the EOC is an example, the green lines make the technical contours virtually unreadable. 407 and 409 are abominations.

ISOM2000 has a marker (an arrow pointing up) indicating whether that symbol must be oriented to north. Strangely, undergrowth is not designated that way. I'm tempted to think it was just an oversight, because undergrowth is required to be aligned to north in ISSOM.

It does seems to me that symbols like 407, 409, not to mention 402 and 404 don't have much of a place on most modern hyper-detailed maps. But that's another matter.

Yes. I remember an IOF mapping conference while ISOM2000 was in preparation. The convenor said, "anyone got a better idea than stripes, we need something, but the stripes are not very satisfactory."

A small side issue, given that we do still have stripes. When you enlarge standard symbols for 1:10,000, they explicitly say to use the same dot screens, viz 60 "lines" per cm. They don't say the same about the line spacing, I wonder why not? They may (or may not) look a bit better at the 1:15,000 spacing.

Blegg: thanks. I was searching in the text and couldn't see the big red arrows! Yes surely the absence for undergrowth is an oversight.

Re the need for stripes at all, I could accept that there may be no need to distinguish good visibility forest; within what is normally mapped as 406/408 there is quite a range of visibility that is unable to be depicted, and we cope. But what would you do on open land with bracken, or recent logging with debris all over the place? I think the Swedes used to use diagonal yellow stripes for logged areas, but I think this may be as visually distracting as green stripes.

to be directed to north arrow is by the undergrowth symbol at:
IOF map comission web pages > ISOM 2000 > web version

http://lazarus.elte.hu/tajfutas/isom2000/webv/o44....

Thanks Jagge. I've been relying on a nice little printed booklet for many years, should have realised there might be some things that got fixed up. I hope that there aren't any significant differences between the booklet and the web:-))

We can find actual ISOM 2000 pdf version from the mapping comission's main page or from IOF page. It has IOF logo, there is listed comitee members and project group. This is Document.
And http://lazarus.elte.hu/tajfutas/isom2000/webv/o44.... is only web page, not document. Not signed, editor unknown, date unknown.
I think we should rely on Document, 407 and 409 shouldn't be aligned to north.

I'd ask before starting to rotate. Looks like mapping commission has controversial info on their web pages.

Those sure were supposed to be aligned to north in ISOM1990 Document and when ISOM2000 was released it was told those symbols were not changed. But who knows.

In the Swedish version 407 and 409 should be aligned to north, but then we all know that you cannot necessarily trust it to be a exact translation.

I cannot see anywhere if "north" is defined as magnetic north, or if you can get away with aligning it with true north too... ;-)

Yes there is a version control problem with these documents.

The IOF Mapping Commission maintains a SEPARATE website (the lazarus one) and there you can find for example under ISOM2000, some "Errata". This is a list of "tidying-up" things that presumably the commission would have liked to be changed in the published version. The Errata does NOT include the big red arrow on the undergrowth symbols, heh heh. Deliberate or an oversight? "Errata" was last changed 6 Aug 2000, at least that what it says.

On the other hand there are no "Errata" for the MTBO specs. Yet my dog-eared print with cover date 2010 valid from 15 May 2010 has been changed on the IOF and the Map Commission websites. The dimensions of connecting lines and marked routes have gone from 0.5 to 0.6. My question to the MTBO Commission as to what other subtle changes may have occurred has not received an answer. Only...

"I will investigate when the change of the 2010 May 15 map specification took place. I assume it was some mistakes that were corrected but nevertheless you are absolutely right that such changes must be communicated to the federations - I am not sure if that has happened." That was 6 months ago. No answer.

We must be somewhat understanding about version control though don't we. How many clubs have a watertight system for ensuring that the latest version of a map is always used?