Discussion: Special flights for getting Lidar data

I'm just wondering how common it is for clubs to pay for special flights to get Lidar data for areas they want to map.

I know that EMPO did for the map at Moreau (for the 2010 classic champs), also that UNO and EMPO are in the process of scheduling flights, each covering multiple areas that they are interested in mapping, or remapping. At a cost of maybe 5K or so per flight?

I also know that there is existing Lidar data available for some places, and not for others, and that the quality can vary a lot.

Having recently done field work on a small map using a basemap from Lidar on which the contours were just fabulous (we just got lucky, the available and free Lidar was real good), it seems to me that getting good Lidar should be a high priority. And if that means paying for a flight, than that's what should be done, especially if they can do several areas in the same flight.

So is anyone doing this? Or is the attitude to just go with whatever may be available in the public domain?

Obviously this also ties in with the thread about what clubs should do with their money, but this seems worth a separate thread.

And if anyone who has arranged flights wants to share the pros and cons from their experience, cool.

I think its more than a flight. The data has to be processed into useful elevation data, perhaps using ground control. And I have noticed from the LIDAR I have acquired that there's a big difference in the "noisiness" of the data. Guess they can be filtered out but it suggests there are consultant's fees as well as aeroplane costs involved.

Greg Lennon is the expert and will likely weigh in but QOC is, depending on someone's plane having a bit of free time about to spend a chunk of our cash on new LiDAR data covering a set in Virginia of four of our existing maps that are either decades old or, in one case, recently made on a basemap from rather low quality LiDAR, plus at least one new area. $5K is apparently the minimum to make it worth their while putting a plane in the air so one reason we're asking for so many areas at once is to get the cost per unit area lower. It'll no doubt take us at least a few years to get the subsequent fieldwork done.

How far away are we from having publicly available LIDAR data of the whole country? (How much coverage do we have now?)

Wikipedia lists where lidar data is complete--
http://en.wikipedia.org/wiki/National_LIDAR_Datase...

Don't know if it's all publicly available.

Mark Dominie is rather experienced with this.

getting good Lidar should be a high priority

Putting on quality, well attended, well publicized, well sponsored events on maps close to major population centers should be the highest priority.

I think CNYO got a private flight done for all of our stuff north of Syracuse, such as happy valley, Salmon river, and others. Mark D. Would Know more about it. right now we are searching to find out if there is Lidar data for the Syracuse area. Based on some of my work with Mark, finding free or cheep Lidar data of large areas is pretty worthwhile because you can produce good maps for local events training and B meet type stuff for low costs. For A meet quality stuff if you have the money its probably nice to get very specific data to what you are looking for. It seems when CNYO gets a private flight they combine a few different ares so they have data for later even if they don't make maps right away. So if your getting a flight for a certain area look around to see if there is other stuff that can be combine in the flight that won't cost to much more. This is just what I've picked up from my work with Mark and CNYO. Don't know all of the details.

One time I was talking with Mark Dominie about the lidar base maps and one comment I seem to recall is that he worked with one (several?) lidar bases that had the boulders on them while usually the boulders are processed out of the data. Guess it depends on the processing. Maybe Eddie can tweak the processing to show boulders (or likely boulders) but it would be nice to have them on if possible though I doubt any of the publicly available data would show boulders. (Sorry, I'm getting off topic but it's sort of related).

It would be much appreciated if someone who has done this before could weigh in on the specific tasks needed to arrange such a flight. What companies or entities do Lidar surveys? How much terrain can an aircraft cover in one sortie? Is there any equipment that must be procured, or do the contracted parties have all of that?

"you can produce good maps for local events training and B meet type stuff for low costs."

This point is continually getting exaggerated.
The end product of a lidar flight is only a base map, which is only a minor cost compared to fieldwork and drafting. The current top of the line lidar base map is still only marginnally better than a good traditional O-photogrammetry base map, and most lidar products are worse. In general, lidar base maps have better contours, but worse point and line features. This is changing as the lidar data processing continues to progress, but the end result is still a base map, Yes, a good base map will reduce field checking time, but not come close to eliminating it.

A priority, yes, but but I think there numerous other issues tthat are more important to O growth and sustainability.

I have been able to find LiDAR data by calling the local county tax assessor's office, or the county GIS office. Sometimes a few phone calls to the right person is all you need.

@Iansmith--I can't help answer any of your questions but the Northeast Lidar project is supposed to produce very good lidar for the entire eastern 25%+/- of Massachusetts and I think that may be complete by now. The same project also included the costal areas of Maine and NH but at reduced resolution (or whatever it's called). I have seen maps showing the extent of coverage but can't seem to find one right now but you could google Northeast Lidar Project.

From my brief experience, it seems that the big difference between a photogrammetric base map, and a lidar base map is cost (IF lidar data is already available in the public domain) and time. (read on...)

(I think EricW knows this story)
An HVO member wanted to make an O-map of a county park near his home. He was able to find public lidar data and, using free/shareware, produced a serviceable base map in about 8 hours, including the learning curve. Once he got the hang of the process he could bang out a base map in about 30 minutes. Fieldchecking, on the other hand...

The map of Silas Condict Park, in Kinnelon, NJ, will debut at a local event (date tbd) this Spring. The area includes Harriman-like terrain.

The Wikipedia page for US LiDAR (which I started and do most of the updating for) lists data that is publicly available at nominal (or no) cost, for counties with near complete coverage. Collections that are minor portions of a county aren't tabulated. More sources of ~free LiDAR are listed on www.lidarbasemaps.org.

For the upcoming LiDAR collection being paid for by QOC, I've put together a one-page spec sheet, which I'm happy to email anyone considering a similar project. It's primarily based on the USGS National Geospatial Program ‘Lidar Guidelines and Base Specification’, Version 13 (2010).

Quick Google search of North East Lidar Project gave:
http://megug.org/wp-content/uploads/2010/10/LiDAR-...

Stating:
----------------------------------------------------------------------------------------------------------------
July 1, 2011 All remaining project deliverables due*
*Statewide data sets will be organized and delivered once acceptance of each state’s dataset has occurred and will include boundary tiles shared between states
----------------------------------------------------------------------------------------------------------------

Looks like I have to get mapping here in Camden, ME. Excellent! We have a great State Park here.

The other big difference between a photogrammetric base map and a lidar base map is that soon (if not already) you will not be able to find anybody still making photogrammetric base maps. There are no new orienteering photogrammetrists, and the old ones are retiring. I think Ivar Helgesen may have already burned his stereoplotter.

EMPO got into LIDAR for the Moreau map precisely because the photogrammetry we had already paid for from Sterling Surveys was deemed worthless by Mark Dominie when he (finally) began doing the field work. The hemlock forest obscured too much of the surface for the aerial stereo-photo/photogrammetry process to work at all reliably. So, we hustled and hired Kucera International (in Ohio) to do a LIDAR run. The cost was $6300, and that was a reduction from the original quote as we were able to get another club interested in joining in with surveys of their own areas. Mark reduced his initially quoted charge quite a bit once he had the LIDAR data to work with, and he was able to get the mapping done in very short order over the summer, to allow us to host the event on schedule.

I believe Mark used a lot of the rock data from the LIDAR, as well as the contour data. I have certainly received sufficient feedback to know that the results he produced from that data were quite good, and he did it in a relatively short timeframe.

Going forward, EMPO now has a source of LIDAR data for all of Albany County (the county hitched onto some flights being done for flood issues on the Hudson/Mohawk rivers), which is our most central area. We have engaged Eddie Bergeron to produce new basemaps, in OCAD, for three of our locations using that data. I personally am trying to use the data to produce new basemaps for some other (smaller) areas. Eddie can do much more than I can with the data (which I can see from the drafts he is providing me), but just doing what I can do will greatly improve some of our local event maps.

Meanwhile, we have again hired Kucera to perform another LIDAR flight, looking for A Meet level maps in Rensselear and Washington counties. We are getting the data for three maps (two new, one revised) for $9200, and that's down a bit from the original quote because they offered us a chance to delay and hitch a ride with some other stuff they were doing for New York State.

We will never try photogrammetry again, having blown several thousand dollars on what was deemed useless. LIDAR is the way to go, at least in our neck of the woods.

Photogrammetry is probably very dependent on the quality of the photos used. If the leaf cover is dense, the data will be poor. LiDAR requires some post processing that can effect the quality of the contours as well.

So, my question is:

If I was to commission special LIDAR flights to create orienteering basemaps, what specifications should I request from the company? I presume there is at least a minimum recommended resolution for data sampling, and probably one that is considered "ideal". Can someone who has experience in processing LIDAR data tell me what I should be asking for? I obviously don't want to pay for a flight that yields insufficient resolution, but I probably don't want to pay extra for unnecessary resolution that may simply create difficulties for post-processing.

In short: what should I be asking of my LIDAR provider?

60 cm post spacing is it, Eddie?

Wil, I've been in contact with this company that is based at the McMaster Innovation Park. http://www.geodigital.com/

There seem to be several ways mentioned of how LIDAR data is aquired.
1) Commission its production yourself to your specifications: Expensive
2) Buy it already produced by an agency probably for another purpose: Usually also expensive but occasionally you find a reasonable agency.
3) Use LIDAR data made 'freely available to the public' by a state, county or other body.
Option 3 is very attractive and I must admit to doing that but in one case I turned it down as the warning and disclaimer against commercial use of the data was too ominous. As I interpreted it the orienteering club could use the LIDAR data to produce a map but they then couldn't charge anyone for the use of that map.
Has anyone anywhere had to deal with this issue? What was the result?

In the U.S., I don't know of a case of data made available by a government agency that would require non-commercial use only. A county south of here was charging for its GIS data, more so for developers than for other uses. They were sued and lost, and required to make data available to everyone at no cost over the cost of copying and reproduction. [...] the [...] Court was [...] clear that California government entities do not have the right to use copyright law to restrict disclosure or impose limitations on the use of their data [...]

We pay for it with our taxes, presumably.

Tax money well spent in this case.

Unfortunately, your cousins north of the border have not spent as much tax money on LIDAR. So it'll have to come out of pocket instead. Oh well.

And is the 60cm post spacing correct/optimal?

You might want to email greg_l....see his post above

I suspect there is very little of orienteering significance that shows up at lower post spacings than 60cm but not at 60cm and/or not enough to justify paying for collecting and processing the increased amount of data but I lack direct experience of working with a lot of LiDAR data of varying quality - hopefully one or more of Eddie, Marc Dominie and Greg will weigh in. Eddie, in particular, must have made more basemaps from LiDAR of varying quality than anyone else in North America by now.

The post spacing you need will depend on the terrain and what you need to extract from the data. "Simpler" terrain or a contours-only basemap may be fine with (average) spacing of 1.0 to even 1.4m, whereas more complex terrain or basemaps including vegetative and other feature analysis will benefit from spacing from 0.5 to 1.0m.

But there are other factors beyond average spacing to take into account - just for starters, how homogenous over the area covered are the data points, or are there areas with much less coverage? Are the points all from the same flight line or from two or more flight lines? What was the angle of the beam and how did that affect both horizontal and vertical accuracy? Basically, don't just look at quantity (average spacing), think about quality too.

Understood. And I'm talking about getting special flights done, not using existing data. It'll need to be contracted out, and I was wondering what I should ask for in terms of specifications (so if there are other things I should specify besides post spacing, I'm all ears).

At the end of the day, the majority of the cost lies in getting the plane off the ground - so post spacing and other criteria don't affect the price so much. I just want to make sure that I am able to give the base-mapper a data set that they are happy with. So I will request whatever is needed to make that happen, even if it increases the cost incrementally.

Greg, if you (or anyone) are happier e-mailing me a specification sheet, that works too. But I figured that since the title of this thread was about special LiDAR flights, I thought the specifications might be useful to have embedded into the thread itself, archived for future use.

Thanks!

Sure, here they are. But keep in mind (1) you really should study the 2010 USGS v13 LiDAR guidelines, and (2) the specs below are somewhat tuned to the terrain we're having flown, but additionally, these specs are not perfectly self-consistent, nor intended to be, since it's important to have a discussion with each vendor over what they will (or can) actually do.

Overall: consistent with FEMA guidelines for production of 2 foot contours, including base specifications as described in USGS National Geospatial Program ‘Lidar Guidelines and Base Specification’, Version 13 (2010)

Timing: During leaf-off season, and when ground is snow-free

Flight plan: to include parallel flight lines with at least one cross flight line to assure positional accuracy; collection verified with either kinematic or static controls.
Preferable: scan angles of less than 30 degrees. Flight overlap 10% minimum.
Point spacing: 0.7 – 1.0m average
Returns: at least 3 or full waveform

Vertical Accuracy:
Bare-earth: 18.5cm RMSE
Vegetation: 30cm RMSE
Horizontal Accuracy: 0.5 – 1.0 meter RMSE

Coordinate Reference System: UTM, NAD83, meters

Data Products/Deliverables:
Mass/raw point cloud, including intensity, LAS v1.3 (ASCII optional)
Classified point cloud, LAS v1.3; possibly including vegetative and building
classifications
Bare earth
DEM

Optional:
Breaklines
DTM
Contours (1m)

Awesome Greg - thanks! I really appreciate it.

Wil,
Georgian nordic, Torrance barrens, Nobel, ?? (all of them?)

I wonder how soon LiDAR flights will be done with unmanned drones...
(over the USA)

You mean you haven't seen them yet?

@GuyO, it looks like your LIDAR drones could be roaming the skies soon pretty soon:
http://diydrones.com/profiles/blogs/found-cheapest...
maybe in a few years O-clubs will be doing their own LIDAR flights?

What about 3m USGS elevation data? How good is it compared to getting your own flights? I have been using it for some small mapping projects, it seemed okay at least for contours.

Do you mean NED 1/3" or NED 1/9"? I'm not sure I am aware of "3m USGS elevation" data.

NED 1/3" is basically useless. In most cases, it is retro-derived from 7.5' quads. NED 1/9" gets you usable rogaine maps, and is perhaps almost good for orienteering bases in areas of ample relief, such as most of California. However, NED 1/9" is in almost all cases derived from lidar data that is also available through USGS. You can massage that data to also derive things such as vegetation boundaries.

Not very good for fine details, or developing any of the other products you can get from lidar, or for making 2.5/5m contours as required for a standard map.
Great for small parks where having any map is better than nothing, so you can use them for a local meet or training.

Yes, I mean NED 1/9" which gives a grid with about 3m resolution, certainly not enough for boulders and such. It is free though.

I have not tried NED 1/3".

The lidar data that was used to create NED 1/9" will have better resolution. For Henry Coe, the post spacing was about 1.4 m.

You can of course process elevation datasets to derive contours with any interval above some reasonable minimum. You should have no problem contouring NED 1/9" into 2.5 m, 5 m, or 10 m contours.

I though he was referring to a premade set of 3m contours, ie http://www.mass.gov/mgis/hp.htm

That dataset covers all of MA, but is not USGS data. We used it as the base for many of our park-O maps before we had the lidar data.

@jman: No. But, then again, I'm not Iranian.

What made me think of using unmanned drones was greg_l's description of the flights -- which sound exceedingly repetitive/boring/tedious for the pilot.

They sound a lot more fun than flying from pt A to pt B, which is what most pilots do most of the time and is terribly boring if you can't see anything.

If you were doing LiDAR of the same piece of terrain over and over, that might be boring. But doing a new one every time is probably pretty interesting. However, drones may ultimately provide a more economical way to get the work done.

(Cristina provided the perspective of one kind of pilot, I can provide another. When I go hang gliding, I often end up just doing laps back and forth in front of a familiar ridge, sometimes for hours. There are hang glider pilots who consider this to be unbearably dull, but I still go do it, for entertainment. Flying a new place every time would be great!)

I have no idea if this is actually the case, but flying lidar passes sounds like it might be a bit like flying search patterns. While perhaps not 'exciting', it can be interesting and demanding. My meager experience dealing with pilots leads me to believe that any flight time is good flight time, but especially if it involves something other than autopilot at high altitude. Don't imagine there's much time for for girly mags/Angry Birds/action movies while flying lidar passes. ;-)

Might provide some of the interest of cropdusting without being as hazardous.

Thank you for the pilot POVs!

I should talk to my uncle in NC -- who builds his own planes.