12 replies to this topic

[raynorfan]

I'm looking at obtaining 1M contour data for a plot of land fairly hilly, 40M elevation change, approx 200 acres.

What would be the lowest cost method?

What would be the easiest method?

Thanks for helping a newbie.

[BEAVER]

I'm looking at obtaining 1M contour data for a plot of land fairly hilly, 40M elevation change, approx 200 acres.

What would be the lowest cost method?

What would be the easiest method?

Thanks for helping a newbie.

It would take just few seconds to generate contours like that from DEM files. Where is this land?

[raynorfan]

It's a hypothetical question based upon an idea.

If I wanted to generate a set of contour lines at 1M for any plot of land does that base data exist or does it have to be obtained (via survey, LIDAR, etxc...)

It was my assumption the base data available is no where near that type of resolution.

[Hans van der Maarel]

It's a hypothetical question based upon an idea.

If I wanted to generate a set of contour lines at 1M for any plot of land does that base data exist or does it have to be obtained (via survey, LIDAR, etxc...)

It was my assumption the base data available is no where near that type of resolution.

It very much depends on the country/region, as well as the kind of quality you want to get out of it.

For example:
I'm working on a map for the Mount St Helens area. Did some searching and found 10 meter NED data and 3 meter LIDAR data for free download (yay!). From both of those, a 1 meter contour line dataset could be interpolated, in a rather short amount of time.

A brand new survey or LIDAR dataset of the same region would be more accuracte, but also a lot more expensive and time consuming.

Is that additional accuracy worth all the extra expenses and efforts? Will you actually see it in your end result?

[Rob]

It's a hypothetical question based upon an idea.

If I wanted to generate a set of contour lines at 1M for any plot of land does that base data exist or does it have to be obtained (via survey, LIDAR, etxc...)

It was my assumption the base data available is no where near that type of resolution.

It depends on how you are going to use them. Any reasonable amount of accuracy would require either a field survey or LIDAR air survery for 1m contours imo. I've worked on survey crews that created 1m contour datasets and we'd shoot points about every 5-6' in flat areas and 2-3' where there was more relief.

[raynorfan]

Idea is to get 3M contour lines for an existing golf course.

At 3M I was hoping the data existed or could be obtained easily.

My assumption is that 1M data would have to be field / aeirial surveyed and would be expensive / time consuming.

[Hans van der Maarel]

My assumption is that 1M data would have to be field / aeirial surveyed and would be expensive / time consuming.

Not really... A golf course isn't a terribly large area, so that changes a lot of things... (in fact, disregard my previous post). If you can round up some friends and strap some GPS receivers to golf carts, you can cover quite a bit of ground in a day. Set the receivers to log one point per x seconds (depending a bit on your speed). Measure the distance from the GPS receiver to the ground, automatically substract that from all readings and you've got yourself a pretty good model. Grid it and interpolate contour lines at whatever interval you please.

It pays to spend some time finding out what the best value is for the log interval and to plan which parts you need to cover.

So yes, I think it can be done in a short amount of time, with little effort and cheap material. Heck, I'd volunteer to drive the cart if it wasn't for the fact I'm on a different continent

[Charles Syrett]

I'm not sure about the golf cart / GPS method -- though it certainly sounds like fun! I've used a sub-metre GPS unit (MobileMapper) to measure points on a rural property and had pretty alarming results. Even letting the unit sit on the point for 20 minutes at a time, I've gone back to retake the point and seen elevation differences of up to 3 metres. I don't see any simple way of doing this and getting accurate results. A traditional measure of acceptable contour accuracy is half the contour interval, so you would need measurements that are within .5 metres. With a really good (=expensive) GPS unit, with a good antenna, and lots of time and knowhow, and post-processing, you may be able to achieve this!

Charles Syrett
Map Graphics
http://www.mapgraphics.com

My assumption is that 1M data would have to be field / aeirial surveyed and would be expensive / time consuming.

Not really... A golf course isn't a terribly large area, so that changes a lot of things... (in fact, disregard my previous post). If you can round up some friends and strap some GPS receivers to golf carts, you can cover quite a bit of ground in a day. Set the receivers to log one point per x seconds (depending a bit on your speed). Measure the distance from the GPS receiver to the ground, automatically substract that from all readings and you've got yourself a pretty good model. Grid it and interpolate contour lines at whatever interval you please.

It pays to spend some time finding out what the best value is for the log interval and to plan which parts you need to cover.

So yes, I think it can be done in a short amount of time, with little effort and cheap material. Heck, I'd volunteer to drive the cart if it wasn't for the fact I'm on a different continent

[Adam Wilbert]

I have to agree with Charles. GPS is designed to measure horizontal position on a plane that is perpendicular to the satellite signals. Trying to measure elevation using a signal that is essentially coming straight down will include a lot of error. Survey grade GPS units include an altimiter to help mitigate this, but even they have a margin of perhaps +/- 10 feet.

[Hans van der Maarel]

I have to agree with Charles. GPS is designed to measure horizontal position on a plane that is perpendicular to the satellite signals. Trying to measure elevation using a signal that is essentially coming straight down will include a lot of error. Survey grade GPS units include an altimiter to help mitigate this, but even they have a margin of perhaps +/- 10 feet.

Bummer... I knew there was a catch somewhere.

Would the errors be random or sort of regular?

[Paul H]

I have to agree with Charles. GPS is designed to measure horizontal position on a plane that is perpendicular to the satellite signals. Trying to measure elevation using a signal that is essentially coming straight down will include a lot of error. Survey grade GPS units include an altimiter to help mitigate this, but even they have a margin of perhaps +/- 10 feet.

I disagree with this statement. GPS signals are coming from all angles, and the angles are always changing. There is rarely a signal that is coming straight down, perpendicular to a horizontal plane, and when it is, it is only for an instant. In fact, GPS is only measuring the time it takes for a signal to go from the satelite to the receiver. In that regard it is measuring distance parallel to a line between the satelite and the receiver. By using several signals, triangulation can occur, and a position can be plotted in three dimensions.

[Charles Syrett]

Whatever the geometry, even the folks who sell these units will caution that vertical accuracy is about half the horizontal accuracy. And I know this from experience anyway. If I stand under a big sky, picking up WAAS signals (but without external antenna), I can watch the altitude readings roaming up and down over about 10-20 feet. I don't think I'd want to try using such readings for 1 metre contours.

Charles Syrett
Map Graphics
http://www.mapgraphics.com

I disagree with this statement. GPS signals are coming from all angles, and the angles are always changing. There is rarely a signal that is coming straight down, perpendicular to a horizontal plane, and when it is, it is only for an instant. In fact, GPS is only measuring the time it takes for a signal to go from the satelite to the receiver. In that regard it is measuring distance parallel to a line between the satelite and the receiver. By using several signals, triangulation can occur, and a position can be plotted in three dimensions.

[Paul H]

Whatever the geometry, even the folks who sell these units will caution that vertical accuracy is about half the horizontal accuracy. And I know this from experience anyway. If I stand under a big sky, picking up WAAS signals (but without external antenna), I can watch the altitude readings roaming up and down over about 10-20 feet. I don't think I'd want to try using such readings for 1 metre contours.

Charles Syrett
Map Graphics
http://www.mapgraphics.com

I disagree with this statement. GPS signals are coming from all angles, and the angles are always changing. There is rarely a signal that is coming straight down, perpendicular to a horizontal plane, and when it is, it is only for an instant. In fact, GPS is only measuring the time it takes for a signal to go from the satelite to the receiver. In that regard it is measuring distance parallel to a line between the satelite and the receiver. By using several signals, triangulation can occur, and a position can be plotted in three dimensions.

Vertical accuracy is about half that of horizontal accuracy because you can only pick up satelites above the horizon. So, in essence, there are potentially less satelites to make accurate elevation measurments. (The most precise elevation measurement would come from a satelite directly overhead (rare), or far below the horizon (impossible)). I agree that elevation measurements are not as accurate as measurements on the plane, but I disagreed with the explanation.