Absurd GIS Inaccuracies

There is only one thing you can do......drag to site and give them a tape measure. Once they have proven to themselves that their data is wrong, ask how you can trust any of the data.

A colleague once brought up an internet mapping service with photographic overlay. The image resolution was so good it clearly showed the colleague's rather distinctive car both outside the office and outside the home at the same time.

Does this sort of thing count?

Only if it was a classic Ford.

it's a good idea to drag them over with a tape (or might they prefer laser or GPS?) and try to explain it to them, but I've noticed that it's difficult to explain anything to an expert....

As to the wayward car, I can think of a lot of places to park that would have proven a lot worse than simply bending the space-time continuum.

This is a quantum phenomenon. When he drives it, it will collapse the wave function.

Are we talking about GIS or GPS? Some people use these terms interchangeably and they really are two different things that can be used together

With a geographic information system (GIS), you can link information (attributes) to location data, such as people to addresses, buildings to parcels, or streets within a network. You can then layer that information to give you a better understanding of how it all works together. You choose what layers to combine based on what questions you need to answer.

The Global Positioning System (GPS) is a satellite-based navigation system made up of a network of 24 satellites placed into orbit by the U.S. Department of Defense. GPS was originally intended for military applications, but in the 1980s, the government made the system available for civilian use. GPS works in any weather conditions, anywhere in the world, 24 hours a day. There are no subscription fees or setup charges to use GPS.

If we are talking GPS 30.2-29.12=1.08 isn't to bad. There are many factors that would give an explanation.

What kind of tape was used for the measurement you mention? Steel or cloth?

How carefully was the tape read & pulled?

How many satellites were locked in when the reading was taken?

How careful were the field crews when they took the shots?

Were these actual experts or salesmen that were trying to sell a product and just trying to impress people that don't know any better. Why the accuracy of so many decimal places? MOST of the time The real world doesn't need it this accuracy level.

The only time we've need this type of accuracy is when we were working for an air plane manufacturer that built fighter jets and needed the machinery to build them precisely placed. And we sure as H*** didn't use a GPS unit to do it. The civilian versions just are NOT designed to be that accurate.

It would be nice to have survey grade data ALL of the time, but since there are only 24 hrs in a day...

I just want to reitereate, the GIS profession true, DOESN'T always have the most accurate data, however, I did notice you said they were using 30m pixesl. I, for one, use LANDSAT for (at minimum) the largest scale projects, say a medium sized city. I noticed you said this was used for a farm? I assume this is probably larger than 30m? So why not use aerials that have 1 ft pixels or 1 m? I betcha that pivot gets ALOT closer to the actual point.

Although I am by no means a GIS expert. I do have some theories to explain the difference. GIS is primarily based on satellite and aerial survey photos, as has already been noted. On the other hand traditional surveying relies on multiple readings of a chain or tape. In the USA the standard surveying chain is 66 feet long, a little of 20 meters. Surveyors also use an optical instruments called a "transit" in conjunction with a magnetic compass and level to determine direction and elevation angles. Although, the better equipped surveyors are now using GPS units to measure distance and direction (as well as altitude).

Therefore, the survey team is measuring line segments along a varying terrain, while the satellite and aerial photos show a continuous distance over a projected flat (or corrected to account for the average curvature of the earth) surface, which can be accurately divided. There could easily be some error introduced by this difference, unless the terrain were exceptionally flat, with no major hills or valleys..

Another possible cause of error would be the digitalization (pixellizing) process itself, which assumes that if the majority of the pixel is of one color, then that value is assigned to the entire pixel, even though this is not actually the case. This sounds like it could very well be your case, since they were looking at a specific "green" area.

The error could be exaggerated substantially if the plat line is parallel or nearly parallel to the pixels. This is because the true edge of the plat would lie within a row (or column) of pixels and the entire row would be added to the digital area, even though, in reality only 50% or more of the pixel is within the plat. An angled line could have an almost equal error on both sides, cancelling most of the error out.

You may also have a combination of these errors, and as luck might have it, the errors may be additive instead of cancelling. Sounds like Murphy was up to his old tricks!

Like you, I was going to suggest it might be an effect of topography.

Also possibly of definition: is the area of a 100-m x 100m field that is flat all round the edges but has a peak in the middle and a 30-degree slope everywhere except at the very edge 1-hectare or 1.15-hectares? (I suspect it depends whether you are buying or selling)

Fyz

Ha! Very good. I think it would depend on whether or not you sub-divided as well.

Buy a 1km x 1km tract of land (100ha) as you described, with a nice big hill in the center, roughly 30 degree slope everywhere but the edges where it was originally surveyed. That land sells for $1000 per hectare so your cost is $100,000. Now subdivide that land into 1ha plats by resurveying each 1ha. If your math is correct (its not the geometry I really am examining here) then you now have turned 100ha of land into 115ha, creating 15 new hectares or making yourself a nice profit of $15,000 without even raising the price per hectare!

They were using GIS to establish maximum lawful entitlements to water (a scarce commodity in our parts) The errors are causing expropriation without compensation.

A GPS is not all that accurate in our part of the world (because of fewer satellites). Up to 30m error has been observed. It is still useful because the error will be almost constant for all the points.

Will take few GIS people on a field trip one off these days (some of them hasn't seen a irrigation system before). What should I use to measure? a certified tape or matches?

I notice another funny error (why do they use all the bad examples in my pres sens).

I watershed of a catchment area was placed on the centre line of a river. The name of the river should be changed to clever waters.