Discussion: Mapping rates

SLOC is currently evaluating a full remap of Cuivre River State Park. This is a very large map (25 KM^2), most of which is open woods. About 20% of the map is thicker vegetation. A small percentage is nasty. The base map is a 1987 map which was excellent in it's day and is still not terrible. The main adjustments will be trails and vegetation type/boundaries.

As it's been 7 years since we've tackled a project this size (Hawn State Park), we'd like to get some idea of what the going rate is for remapping a large area. We believe the current map is good enough that a new base map from aerial photos will not be necessary. The old map is not in OCAD, so cartography will be from scratch.

Any estimates would be appreciated, particularly from clubs that have recently undertaken a major remap or from mappers with similar experience. If you are a mapper and are interested in the job, feel free to indicate that as well, but we are not yet to the point of evaluating specific bids.

Replies can be made to this list, or directly to me (see my profile for email address).

Thanks,
Eric Buckley
SLOC mapping chair

We hired mappers from the Czech Republic in the spring/summer of '08. We paied them $20 CDN/hour for fieldwork and cartography.

Your overall costs will largely depend on the quality of the basemap and how fast they can get through the terrain. I recall 0.10 sq km/hr to be very slow, and 0.5 sq km/hr to be very fast but don't quote me on that. For the catrography add 25% of the fieldwork time roughly.

Any mapper who can do 0.5 km^2 per hour is a force of nature to be reckoned with; that is inconceivably fast.

Also, there are well-known issues with hiring foreign mappers in the US due to immigration issues. So, I'm not sure the above post is very helpful to SLOC either for the hourly rate or for the expected number of hours.

How do people make maps in the US given those issues? What are the domestic options?

The last big map we made (Hawn), we hired Zoran Krivokapic on an H1-B visa. A bit of a PITA, but certainly worth it given the final product. Frankly, given the complete lack of interest I've encountered over the past 2 years trying to find a mapper for Cuivre River, I may have to do this one myself. I'm certainly a lot closer to the 0.1Km^2/hour speed, even with a base as good as this one. 200 hours in the field does sound a bit daunting (and I'm not sure I'd even get it done that fast).

For A-meet quality, typical northeast terrain, using a good photogrammetric base, a rule of thumb we've used was 40 hours/sqkm. Various factors can make things faster or slower. For the section of Surebridge that I did, I think it was 111 hours/sqkm (I was not very fast)

I'm pretty sure Gouger is talking exclusively about the mapping of terrain that, while topographically complex, was largely open with patches of mostly near impenetrable forest. Given a high quality contour base and a skilled photogrammetrist's stab at the vegetation, I suspect field checking was about as fast as is possible for anything other than completely open land or uniformly white forest with excellent visibility and relatively few point features.

Personally, given my limited experience mapping in Mid-Atlantic forested terrain, even 0.1 km^2 per hour sounds like something to aspire to after a lot more practice.

For Eric's immediate problem, I'd say the crucial information not provided is approximately how many small pieces the thicker vegetation is divided into and how complex their shapes are, since it sounds as though most of the mapping effort will go into updating vegetation boundaries.

That's my take on it - vegetation will be the issue. I rather thought CR was a popular enough map that most who read this forum would already be familiar with it. Here's a typical section.

We don't have anything like the complexity typical of the NE and visibility is generally quite good. The contours are great already and there's no rock to speak of.

It seems to me it should go slightly faster than when I did S-F for Team Trials (that was under .1/hr, but not by too much) and much slower than the update of Hawn I just finished (very fast because the base map was only 7 years old). I'm sure there are mappers who could do it in under 200 hours, but I don't think I'm one of them.

Boy, I dislike AP as a forum for this type of map talk, but I'll try to go with the flow. I'll suggest that the Yahoo OMAP group or private emails are more appropriate.

Gouger's time numbers are off by about one decimal place.
His $/hr number may be realistic, but the terms that I am accustomed to are $/km2 of fieldwork, which includes the OCAD drafting, but different people may work differently.

Anything mapped faster than say 15 hrs /km2 is not worth doing. Its either uselessly simple terrain, or poor mapping.

For Cuivre River, my fastest estimate for one of the world's best mappers (efficient and good quality) would be ~20hrs/km2, but I would expect more like 25-30 hrs/km2, and that is still exceptionally fast for expected A meet standard mapping, by a professional, in 2010. I recognize that this terrain is relatively simple.

Jon is right to emphasize the vegetation. This will be the biggest job, but I wouldn't dismiss the contours as trivial.

My impression (at race speed) is that these contours could use some work, and this was by 1980 standards. The reentrants seemed to be sharper, and extended further into the ridges, with the ridge tops more narrow than shown, pinched tight by opposite side renntrants. I suspect this is a remnant of the original base map.

Note, almost no form lines, which might be appropriate for the generally steep hillsides, but I'll bet that the "shoulders" and flatter, complex sections could use some help.

My best OCAD contacts (Vlad Zh and Alexey Z say that fieldwork to OCAD ratio is close to 1:1, for good quality drafting. In my opinion the pay rate for OCAD work should be less than the fieldwork rate, but "25%" more still sounds low.

I agree that a new base map is probably not worthwhile, and that the geometry on this map is probably reliable. The drawback is that the mapper won't have many reliable point features to locate all the expected veg details and expected point features, and will have to do plenty of locational work to get the geometry correct.

25km2? Have you considered a smaller section?

Time for me to sleep.

Yes, I initially thought Gouger's progress numbers were per day, not per hour, and even per day they are spectacularly fast. The prevailing wage for surveyors in St. Louis metro area is here.

yes.....my initial numbers were off.....not sure what I was thinking....looking back at the fieldwork numbers submitted they were closer to 45hrs/km2.

the area we had mapped did have large open sections which did make it easier/faster than some. the final product mapped was slightly less than 10km2, and it took our mappers roughly 8 weeks(50 hrs/week) to do this.....25km2 is a huge area.....make sure you have lots of lead for your pencils:)

if you pay your mappers by the km2 rather than by the hour, you may just encourage them to go fast and get it done thus compromising quality. On the other hand if you pay someone by the hour then they can take their sweet time and maybe have too much detail.

biggest thing I learned from the whole process of selecting an area, looking for mappers, picking mappers, supervising mappers............mapping is definetly an art & a science....let a professional do it if you want professional results.

I also love it when people dismiss posts straight out without any hesitation...thanks feet.

Dismissal without hesitation? There is no one better! :)

Being a complete non-expert in mapping, maybe the answer is obvious to others, but here goes...

Wouldn't CR be a good candidate for a new LIDAR basemap?

Also, with all due respect to EricW -- in my mind, the foremost O-mapping expert in the USA -- I am glad this topic is being discussed on AP. It gives non-experts like myself insight into O-mapping we might have missed if it was on a dedicated forum.

Since no one else seems to want to try, I'll answer Guy with a definite maybe. Given access to the results of a high resolution LIDAR survey, many of the contour issues EricW believes are there might be substantially addressed. It should even be possible to extract changes to the trail network just from the LIDAR data if the data is good enough. However, the mapper would still have to check LIDAR-derived contours and edit them somewhat to match human perceptions so it's not clear that this would represent an improvement over asking the mapper to edit the contours on the current map, which all Erics seem to agree are pretty good. Also, high resolution LIDAR data doesn't always exist covering a particular area, nor is it necessarily freely or cheaply available when it does. Others are much more knowledgeable than I about the costs associated with commissioning a LIDAR survey.

I think that LIDAR is a great thing and would definitely want it as a starting point for a new map. However, we're talking about a REALLY BIG area, here and we already have decent contour data. A new basemap would be a significant and unecessary expense.

Back to the original question, cost is a real issue here. The whole point of mapping an area this big is to be able to use it for long events (ultra-long, goat, score-O). As with all orienteering, a good map is always better than a bad one, but certainly such events do not call for the same standard that one would apply to middle distance. Spending 50 grand to map it as such is simply a misallocation of club resources.

I'm not a particularly speedy mapper, but I'm not slow, either. I put in about 20 hours/Km2 remapping S-F for the 2006 Team Trials Middle and people seemed pretty satisfied with the results. My expectation from some preliminary fieldwork at CR is that I could do it a lot faster than that and still have a map usable for an A-meet long or ultra-long, but probably not something I'd want for a Championship Middle. But, I'd never use CR for that, anyway.

Bottom line is that we'd all love to have the best maps, but sometimes one has to go with "good enough". My question was really aimed at that: how much should we expect to spend to get a map that we can use for an quality long event? I realize it's hard to estimate these things, which is precisely why I posted it here. Many thanks to those who have kept the discussion reasonably focused and civil up to this point. AP threads often turn for the worse long before this.

I think that choosing the appropriate mapper is important for what you want.

Pro mappers often tend to do a detailed, WOC-quality (and thus expensive) job of every map, perhaps because that's their style, perhaps because the map doesn't seem right to them otherwise, perhaps because, for their reputation, they want any map with their name on it to be of a top standard.

Where I used to live, we hired a pro mapper that we had used and liked before, and asked him to just do some veg updates on a particular map, and made it very clear that we didn't want contour updates or major work...everyone felt the contours were fine; just some veg and trail changes over the years...the priority was to get veg and trails updated for the entire area, which meant quickly in the budget available. Well, needless to say, when we looked at the work in progress, nearly every single millimeter of every single contour was changed (and dozens of form lines added)...on a map that a dozen national champions said had just fine contours. Budget ran out quickly, with only a tiny area remapped...and thus most of the veg changes we wanted undone. That work was never used.

Now, I believe that mapper intended well...he just wanted to make a top notch map, and do it right. And nearly every mapper thinks their way is better. For a new map (like our previous use of him), he was great. For a limited touch-up on an existing map, as you're suggesting, he wasn't the right one to choose.

So, for the veg and trail update that you say you want (perhaps with the contour tweaks that Eric suggests), I suggest finding a mapper who is happy to do that kind of update...perhaps a more recreational mapper who does this as a retirement hobby, or such.

If you hire a $20/hour pro mapper, I almost guarantee that they'll do the 40-80 hours/sq km type job, remapping everything. And the results will be excelllent...but an order of magnitude more work done than it sounds like you want or need.

My 2c.

I don't have any experience with LIDAR contours, but orienteering mappers draw contours how they look from ground level. They also have to make them 'fit' the rock or other landforms detail, eg cliffs, watercourses, earthbanks, knolls, depressions, even roads, etc. Remember that the inclusion of some of these features on the map will inevitably lead to a certain amount of 'displacement'.

No base map I've ever seen has not needed substantial 'tweaking' of contours, and I seriously doubt that LIDAR will change this.

True, Jim's story illustrates a) that every mapper's interpretation is different and b) some mappers will get carried away with contour changing. Some mappers and leading orienteers have become concerned about the inclusion of unnecessary form lines - there was a discussion on World of O recently.

The best mappers will literally 'run' each section of the map as they complete it (or employ a very competent orienteer to 'run' it for them), to ensure that they are not overmapping features and contours that competitors will not notice.

When employing a mapper, ask to see samples of their work, run them yourself if possible, and ask the mapper about their process and philosophy.

ebuckley raises some good points. He makes reference to a "a map that we can use for an quality long event" and contrasts that with one for a "Championship Middle." I wouldn't go there. But, I guess I will just ask rhetorically whether the last US Ultra Long Champs were a "quality long event?" I will reserve judgement, but your personal opinion should influence your mapping expectations.

Lidar-
Agreed, the best LIDAR application is for new mapping.

My main caution in this case is registration, getting the georeferenced LIDAR map to line up with the O map, for which the control is based on a human eye taking measurements off a 1:24 000 USGS paper map.

I've never done a matchup, but my guess is that the Omap will have noticeable distortions compared to the LIDAR, both across the map, as well as more localized twists that are more difficult to rectify. Not a concern for orienteering or O-mapping, but only when you try to match two different datum.

Worthwhile? Maybe, if you are willing and capable of dealing with this so that point features don't end up on the wrong side of linear features, and the LIDAR is cheap, and of very good quality. Not all LIDAR is the same.

If the planned standard of quality is less than Middle/Long A meet standard, this would change my estimations and comments.

I would like to hear more about an H1-B visa. This is a serious issue confronting DVOA right now. Separate thread?

Guy, point taken, I'm all for education, but there are also reasons why business is not conducted in public forums. I guess we're straddling the line on this one.

No disagreements with any other posts.

Again, thanks all. And I agree with EricW (as I typically do on such matters) that this forum is only appropriate for general discussion, not for specically committing to a bid. I made that point in the original post.

Since many who read this may someday run on the new map, I'd like to set the record straight on something: we are NOT looking for a quick and dirty local map. The old map already suffices for that. This needs to be mapped to a level where we can use it for an A-meet. My point was that in long events, the emphasis is on route choice as opposed to the fine navigation stressed in middle. As such, a SLIGHTLY greater level of generality is not only acceptable but, in my mind anyway, preferable since it's easier to evaluate a long leg without obscuring the big features.

I didn't attend ulta-long champs this year, so I can't comment on that specific case, but I would hope that anyone hosting an A-meet (much less a championship) would take map quality pretty seriously. Rest assured that I do.

EricW—there is some discussion going on with USOF and the H-1B, but nothing has been done yet. The allure of the H-1B is that with some recent administrative decisions that cut off rampant misuse of the H-1B by consulting companies, the visas are in ample supply and are, in my view, the most appropriate legal way to pursue for hiring non-U.S. mappers. (The recession added to the availability but it's a secondary effect on top of the government playing hardball with IT consulters.)

The cost overhead of the H-1B is not cheap (several thousand U.S. dollars per filing). There is also some paperwork involved. Payroll taxes for mappers' compensation are not cheap, either; and tax filings involve comparable paperwork.

I might be accused of doing too good a job of mapping - so be it. These remarks are aimed at club managers, not mappers for which I agree the o-map forum is more appropriate.

For any remap I insist on aligning the map to the national grid (using tools in OCAD). This enables information from various sources to be matched up - there are now aerial photos that have been corrected for distortion, LIDAR data, and consumer-level GPS's are good enough in the right conditions. These solve the problem of "A isn't right in relation to B, shall I move A or B?"

I do this because our early maps though good for their time are usually screwed up underneath. If we are lucky they only have the wrong magnetic north and scale, if the map has been added to it almost certainly has internal bends and twists and fudging around the join.

The immediate benefit - elimination of dodgy spots. With each "gut-feel" remap some of the A's and B's improve, others get further out. There are heaps of potential long-term benefits but they deserve a separate thread.

Totally agree on that point. Maps from old photo surveys typically have a fair bit of distorion. Flying a plane perfectly straight and level isn't as easy as it sounds. I always do a re-alignment before starting the cartography on a new map. This also alows me to set the declination precisely rather than taking my best shot from a compass bearing in the field.