Order - First, Second, Third, Etc.

[+Black Dog Trackers]

I have always wondered about Order in datasheets in regard to finding a location-Adjusteded mark with a GPS receiver. In the forum, we always say to people: 'well if it is a location-Scaled mark, it could be off by as much as 600 feet but if it is a location-Adjusted mark, then it's coordinates are way closer than you could see with your GPSr'.

How close is way closer?

 

In its NGS Documentation File, the section on ORDER AND CLASS: HORIZONTAL (about the middle of the document) gives relative accuracies for Orders A and B (but none of the other orders). These accuracy specifications are:

A: 5 mm +/- 1:10,000,000

B: 8 mm + - 1:1,000,000

The A specification is valid only for other stations of class A.

The B specificaiton is valid only for other stations of classes A or B.

 

For the other orders, the NGS Documentation File has a refercnce for the Standards and Specifications for Geodetic Control Networks. This standard gives the following table:

Classification	Minimum distance accuracy
--------------	-------------------------
First-order			  1:100,000 
Second-order, class I	1: 50,000 
Second-order, class II   1: 20,000 
Third-order, class I	 1: 10,000 
Third-order, class II	1:  5,000 

As with the A and B orders, these are relative specifications; relative to the coordinates of other survey stations of equal or higher order.

 

For whatever reason, datasheets don't have class I or II. It appears from another document that the exactl class is part of the job specifications of a survey and that exactness is not necessary to be shown in datasheets.

 

A newer (1989 versus 1984) FCGS standard gives this version here:

(I converted 1 cm to 10 mm, etc. for readability.)

Classification	Base error   Minimum distance accuracy
--------------	 ---------	 -------------------------
AA order				3 mm		   1:100,000,000
A order				 5 mm		   1:10,000,000
B order				 8 mm		   1:1,000,000
First-order			10 mm		   1:100,000 
Second-order, class I  20 mm		   1: 50,000 
Second-order, class II 30 mm		   1: 20,000 
Third-order, class I   50 mm		   1: 10,000 

It gives no Third-order class II specificaitons.)

 

Page 9 of the document gives a nice graphic view of this.

 

I couldn't figure out from reading the document what the "Base error" actually is. At first I thought, well hey, that is what I'm looking for! But, it isn't; it's something else, I think.

 

Page 27 of the same document has a somewhat disturbing table that says this:

Order		  Accuracy required
-----		  ---------------------
AA				+/- 0.5 meter
A				 +/- 0.5 meter
B				 +/- 0.5 meter
1 and lower	   +/- 25 meters

 

I don't know what this information means.

Further down on the same page there is this information (I changed it from paragraph to tabular form), saying what the standard deviation of the range residuals in the base line solution:

Order		  Standard deviation
-----		  ---------------------
A				 1 mm to 20 mm
B				 1 mm to 20 mm
1				 1 mm to 20 mm
2				10 mm to 40 mm
3				10 mm to 80 mm

 

I don't know if either of these tables answeres my question.

 

In this thread and this thread, DaveD has told us that the 5 decimal level of seconds is not meant to imply that level of accuracy. So, I want to know what is the level of accuracy for these orders?

 

When approaching a mark who's datasheet says that its coordinates are Second Order, does this mean that it is accurate to what (in terms of seconds or mm)? I believe I am looking for an absolute tolerance, not a relative tolerance. I realize from reading the documents that survey job specificaitons are given in terms of internal consistence (relative tolerance), however, I would think that the locations of all location-Adjusted marks in terms of absolute specifications would be of interest, and in a wy much more important. I can imagine some survey with a relative error of 1 part in a billion but all stations being a mile Northeast of where they should be.

 

A tolerance in millimeters can be more-or-less converted to a distance. It depends on your latitude of course, and whether you are talking E-W or N-S direction. I did some experimenting with my GPSr using the coordinates of the center of the contiguious U.S. and found the N-S difference was the larger when I varied the seconds part of the coordinates, so I used N-S. For one station being 0.1 seconds North of another, it is 10 feet North. If a waypoint is 1.0 seconds North of another waypoint, it is 100 feet North. Scaled coordinates are stated to be within +/- 6 seconds, which amounts to 581 feet according to my GPS looking in the center of the country. (These conversions to distance are not very different in my home area.)

 

Here's my question in tabular form:

Order	 +/-		 seconds  
------   ------	  ---------
AA	   A mm		   a
A		B mm		   b
B		C mm		   c
1		D mm		   d
2		E mm		   e
3		F mm		   f
Scaled  581 feet		6

What are the approximate values of A-F and a-f? For example, would Order 1 be about as close ad 0.01 seconds (1 foot)?

(Using a standard-deviation or 95%-confidence caveat for the values would seem reasonable and appropriate of course.)