Discussion: Contour Intervals & Form Lines

Mapping question here...

We're completing a new map here in Washington state, and I've got some questions about contour intervals and form lines.

According to IOF mapping standards, the accepted contour intervals are either 5m or 2.5m.

The area is a complex of low sand dunes and small ponds. The 2.5m interval doesn't really do it justice, so should we either:

a) map it at 2.5m and put a helluva lot of form lines on it
b) map it with a smaller interval, say 1.25 or 1.5m, no form lines

I am certainly not a mapping expert, and I am sure some of them will contribute.

My question is what is the smallest contour interval that is actually usable, or put differently, such that you can discern variations in the terrain? Do you have a mapper and/or base map sufficiently precise to place 1.25 or 1.5 meter contours? As a runner, I am not sure I would be able to make use of those. And my fear is that unless you have a very highly skilled mapper, the output may not be much good.

Well I think j-man has basically hit on the key issue. I think the most overlooked part of the IOF mapping specs is the section 2 - General Requirements, in particular 2.4 Generalization and Legibility. Remember the ultimate purpose is to represent the terrain for the orienteer passing through the area at competition speed, not to produce the most technically accurate/detailed map.

Out of curiosity, where is the terrain? Doesn't sound like typical WA stuff.

2.5 is regularly used for sanddune terrain in NZ, still with form-lines, point features etc used for detail that falls in between. Sometimes this in-between detail is necessary to show shape that is there, sometimes it can be showing something so subtle you can't really see it in the forest.

ye we have a detailed sand dune map in Ireland using 2.5m contours with form lines... and that can be really confusing as it is... i would say dont use a smaller scale it would be far too confusing.. but again i am no expert either

We trained on some pretty detailed sand dune terrain in Denmark that was 2.5 meter contour. You can see samples of the maps somewhere on the jwoc2010.dk web site. Look for maps Blokhus and Vester Troup. One of the maps was 1:15000 and it was very hard with naked eyes though with my 3.5 magnifier glasses I was okay. The other map was 1:10,000 which made reading the detail much easier.
I'm not a mapper either but off hand I would say keep the contour interval at 2.5 meter but adjust the scale until its legible.

I, too, am by no means an expert mapper, but my personal opinion is that the IOF Spec requirement for 2.5 or 5m is an example of the Scandinavian bias of the rules of orienteering. But I think j-man has nailed the point. If you use a smaller interval, do it in a way that makes the map usable to the runner.

Especially for sprint maps, the rule against intervals smaller than 2.5m takes away some clarity of the terrain. I ran on a sprint map last week that had a few really interesting small reentrants in otherwise large expanses of lawn, and they weren't mapped. I like putting controls in shapes that are out-of-view until you get close. If they're not mapped, you can't use them as control locations.

I probably wouldn't attempt a less-than-2.5m map unless the basemap was good. I've seen some Lidar basemaps that were good enough (as basemaps).

Blegg, I suspect Patrick is talking about the area around the Potholes Reservoir, where one of the US champs events will (hopefully) be held next year.

What scale? In general I think lots of form line dashes with ends would be far harder to interpret than solid contour lines which don't end. So terrain that needs lots of form lines is really asking for a smaller contour interval. If that's the best way to do justice tro the terrain then that's what you should use. The IOF standard is only a guideline other than for World Champs etc.

Here in New Zealand we use many sand dune maps in various parts of the country, like birdman has stated we always use 2.5m contour interval for these terrain types. As a mapper consistancy is important so runners get used to what is and isn't put on the map. Of course it is always subjective, some will be over mapped, some under, depending on you point of view. I have found very few occasions when you need more than one formline between 2.5m contours, it's not supposed to be done - but possible to show something complex. The moment you start picking even smaller intervals could be the beginnig of really over mapped areas which is not ideal.

Looks fun!

If you are planning to use it for an IOF sanctioned event (WRE race for example) then you better either follow the mapping spec precisely or apply to the IOF mapping commission now for any deviations (rather than be dissappointingly turned down for the sanctioning later ;-)

I dont think you should go below 2.5m contours. Lots of form lines is fine

I am somewhat surprised this question is being raised for a US champs area. Who is making the map - not professionals?

I'm surprised that this question is coming up while they're "completing" the map!

From the beginning, I thought this was an awkward, "messy" topic for an AP discussion thread, and for that reason I have refrained from commenting.

But that was before picking up on the "completing" status, the time urgency, the importance of the map, and the great potential of this terrain, assuming all the above comments are to be taken at face value.

PLEASE get the best, most experienced, mapper or map consultant available, ON SITE, as soon as possible. This AP discussion makes for good academia, but it is no place to solve this very real and urgent mapping issue.

It looks like you have something very special here. Please get it done right the first time.

To be fair, this could be a local map that has nothing to do with the US champs: we only have inglis' word that it is anything important. Also, 'completing' could just be a misleading word choice. Maybe kupackman might like to clarify.

From my own experience (I've run on sanddunes in seven different countries), I never saw an area that couldn't be mapped at 2.5m/1:10000 + formlines for those with decent eyesight.

And getting it done right first time always pays off in the long run, even if it costs a bit more and you kid yourself you make a loss on the first event.

For general US mapping issues, I recommend the policy which has never let me down i.e. "agree with EricW".

I'll admit to being completely ignorant to pretty much every point raised in this thread.

All I can request is please don't put more than one form line between each line. IMHO, that mis-represents the land more than leaving form lines completely off the map, especially at 2.5m. (I also believe that that practice violates IOF standards (or at least did at one time), but I have seen it more times than I care to).

Ok, I didn't intend to stir up this much with this topic, so I'll try to answer everyone's questions.

Terrain:
The terrain is in between Potholes Reservoir and Moses Lake, Washington. As I mentioned before, it's sand dune terrain, and it will be used for a middle distance race on June 26th, 2010. Map scale will be 1:10,000.

Basemap:
The basemap is excellent, and is essentially done. We have a copy of it with 2.5m contours and another with 1m contours. The catalyst for asking this question was from a popular personal log from about a month ago regarding form lines on O-Ringen maps and IOF mapping standards (a thread that several of the contributors here also participated in).

Basically, my original question should have been, "Hey, we're completing a new basemap here, and we have excellent contour resolution. What are people's thoughts on the IOF standard of 2.5m contours and form lines versus going with a smaller contour interval?"

I apologize for starting an "awkward, messy thread" without consulting our mapper first. I just saw the two versions of basemaps, thought about the previous discussion on AP, and decided to ask the question.

Based on responses here, and closer inspection of the basemaps, 2.5m with form lines is definitely the way to go. It's the current IOF standard, and it's consistent with what's mapped in Ireland, New Zealand, and Denmark, among other places. Thanks to those that contributed!

Timeline:
The basemap is essentially done, and the mapper will begin dedicated, on-site, field-checking on October 1st, the earliest we can safely start on-site mapping activities, due to seasonal off-road vehicle closures on certain regions of the map. The mapper has already been on site, and is the same mapper who created the excellent Salmon La Sac map. For what it's worth, the course designer has already been on site three times, and is working with two different course/event consultants.

Who produced the basemap? When were the photos taken?

just to add some thoughts from the netherlands. If it's very detailed you could consider to use a 1:7500 scale instead of 1:10000. We've done that over here for some dune-areas. Last year in France and year before that in Czech republic I ran also on maps with these scales in areas with lots of details (contours and also rocks/cliffs etc.).
In general the terrain in Holland is flat. If we would exactly stick to the 2.5m line we probably would have a lot of maps without much contourlines. This doesn't mean we officialy us a smaller interval but it's left to mapper to make a good impression of the terrain. And we've quite some maps made by excellent foreign mappers in this way.

The IOF mapping standard was probably written pre-Lidar, so there was a practical reason for imposing some limit. With a Lidar basemap, you can choose your own contour interval but this won't necessarily improve the finished product - what it will allow you to do is experiment with alternatives without the expense of detailed photogrammetry that might be discarded. On a local detailed sand-dune map here in Tasmania, 1 metre contours at 1:10000 would be unreadable, but having the potential to produce a version at 1:5000 or bigger with 1 metre contours from a Lidar basemap offers interesting training and not-too-serious competition options.

Hawkeye touched on an important topic, perhaps relevant if you (kupackman or others) are working with a lidar data. The final map contour interval and base map contour interval are two very seperate issues. To make full advantage of the lidar's potential, I will encourage you to use a very small contour interval on the base map, limited only by what you can print legibly, and/or your mapper's preference.

For a 5m final map, I'll suggest trying 1.0m for the base, and for a 2.5m final map I'll suggest trying 0.5m. It helps the mapper's sanity if the line symbol for the core interval (5m or 2.5m) is highlighted by color or line width, like an index contour.

Mark Dominie (already doing this) introduced this concept to me. Field inspections were very convincing, with the caveat that my samples involved high quality (dense) data. Like photos and other base maps, all lidar is not equal. Poorer quality lidar might yeild more mess than help with these small intervals.

Interesting that mention of Pittwater by Hawkeye. ( I don't know of any other good map near Hobart). A Bendigo Club member has produced some contours for Pittwater based on the publicly available lidar data for the area. We decided it would be a good experiment for us prior to getting access to a bendigo lidar flight due later this month. Unfortunately, the data has been cleaned and simplified prior to publication so may not be much use for another Pittwater if you had one. Or are we missing something?

Production of a Lidar version of Pittwater is a work in progress, but only as a limited exercise to compare the two versions. There's a reasonable fit with the existing map using a 2m contour interval. As an experiment, I want to create a version with a 1 metre contour interval and try setting a course on both lidar and existing versions and see which is easier to navigate on. Peter Hoban has been doing similar work in Launceston, and his initial observation is that the Lidar contours are rather jagged and require some smoothing - he suggests using software called Quickgrid to generate a DXF file, rather than importing Lidar data directly into OCAD.

Re. 1:7500...
It has become standard here to use a blown up map scale for older competitors (45+), and this has been very well received. Also the ISOM doesn't really recognise the idea of anything other than 1:15000 maps - if you print the map at 1:10000, all the line widths and symbols are meant to be 33% bigger, if you print at 1:7500 they're meant to be twice the size. e.g. boulders should be 0.8-1.2 mm across.

Not that many people pay attention to this, including the last WRE in the UK.

(Yes, I am hoping to come to your event)

I did some 'playing' with the LiDAR data available from Tasmania's LISTmap (http://www.thelist.tas.gov.au/listmap/listmap.jsp?...). As TheInvisibleLog stated, the data available from that site has already been generalised, with the metadata staing vertical accuracy of +/-10m (although I suspect it should be +/-10cm, perhaps a typo?).
Generating contours from gridded data will always produce jaggered lines. This is why it is important to get access to the raw LiDAR data. For the information of other posters, the LiDAR scanners send out laser pulses and record the time taken to return to the sensor. Modern LiDAR scanners can detect up to 6 returns from a single pulse, the last one representing the ground (or impenetrable surface). Not every LiDAR pulse gets a return so you end up with a random set of points across the surface. LiDAR companies process the raw data and provide the gridded data (regularly spaced points across the surface) as it is easier to use.
I produced 2 quick maps I produced from the downloaded LiDAR data 1) 1m contours 2) 2.5m contours. On first inspection the 1m contour map I produced looked more similar to the existing Pittwater map (forgetting form lines) than the 2.5m version. You can draw your own conclusions from that.
I agree with EricW that contours generated from LiDAR (along witht he aerial photography) form a good basis for a mapper to 'interpret' the landscape as s/he sees appropriate for the area. The area I see most potential for LiDAR is assisting locate significant features ie rocks and gullies (re-entrants) along with vegetation density (runability).
Interested in chatting with others using/experimenting with LiDAR for orienteering maps.