Tom Taylor
12-09-2003, 08:59 AM
I'm seeking any/all feedback on this idea.
The FGDC published reporting standards that superceded the FGCS standards and specifications (FGDC-STD-007.2-1998
http://www.fgdc.gov/standards/status/textstatus.html). However, they did not associate any procedural and processing specifications. This could be a real problem. For example, You could report a positional tolerance of 2 cm on monuments that are only 20 meters away from each other. This has the potential of degrading any coordinate system associated with those values.
So, Ive come up with the following:
The main question with the FGDC reporting specification is: what are the procedural and processing specifications (guidelines) associated with the reporting standard? The feds have not provided any connection between the two. So, can we in a clear and cogent way associate FGCS procedural and processing specifications (guidelines) to the FGDC reporting standard until anything comes from the feds? I believe the answer is yes.
Here is my idea:
* The FGDC is requiring a reporting standard that shows a positional tolerance (band width) for the values on a station. Namely, the error ellipse for a station's value at the 95% confidence level will be equal to or below a certain value. Typically, 2 cm, 5 cm, 1 dm, 1 m, etc.
* This reporting standard is a better way of communicating survey accuracy as it allows for a single standard for evaluating results from the use of different surveying instrumentation.
* There are no procedural and processing specifications (guidelines) associated with the FGDC reporting standard.
* The FGCS specifications (guidelines) and standards were based on a precisional ratio.
* The FGCS procedural and processing specifications (guidelines) have withstood the test of time for about 15 years.
In my opinion, associating the reporting standard to the tried and true procedural and processing specs is the way to go.
How do we associate the two?
* The main objective is to establish a minimum station spacing distance that you will be reporting positional tolerances on. The FGDC specs could allow for small distances between stations while reporting a tight positional tolerance. This could lead to all kinds of problems and the degradation of the CSRN.
* What is the minimum station spacing? I would suggest that we take any particular positional tolerance for two stations and calculate what that positional tolerance is at the 95% confidence level for the two at the same time. This is a typical error analysis approach. Then determine what the minimum station spacing would be based on the procedural and processing specs that you want to correspond with your desired results. For example, 2 cm positional tolerances surveyed to first order FGCS specs. With this approach you can mix any FGDC band width with any FGCS order procedural and processing spec.
* Ok, Say at the 95% confidence level station A has a desired positional tolerance of X1 and station B has a desired positional tolerance of X2. The error ellipse at the 95% confidence level of the 2 combined is: Y = sqrt(X1**2 + X2**2). And say you want to use the FGCS first order specs. This means that to figure your minimum station spacing distance (Z) you would use the 1:100,000 precisional ratio at the same time as Y. ie: Z = (100,000)(Y). Note: all values are in meters. You would then place your stations a minimum of Z meters apart and survey them using the first order FGCS procedural and processing specifications.
* Confused? (I hope not).
How about an example? Suppose you want both stations to have a positional tolerance of 2 cm and you want the confidence of the first order specs. So, Y = sqrt(2**2 +2**2) = sqrt(8) = 2.828 cm = .02828 meters. And, Z= (100,000)(.02828) = 2828.xxxx meters. So, you would place your stations a minimum of 2.828 Kilometers from each other and use the first order specs.
This could be done with any FGDC positional tolerance and any FGCS order spec.
A chart could be readily made.
Additional thoughts:
* This approach may be an "interim" approach; however, it will increase the confidence of geodetic professionals in the quality of the positional tolerance reported results while helping to prevent the degradation of the CSRN.
The FGDC published reporting standards that superceded the FGCS standards and specifications (FGDC-STD-007.2-1998
http://www.fgdc.gov/standards/status/textstatus.html). However, they did not associate any procedural and processing specifications. This could be a real problem. For example, You could report a positional tolerance of 2 cm on monuments that are only 20 meters away from each other. This has the potential of degrading any coordinate system associated with those values.
So, Ive come up with the following:
The main question with the FGDC reporting specification is: what are the procedural and processing specifications (guidelines) associated with the reporting standard? The feds have not provided any connection between the two. So, can we in a clear and cogent way associate FGCS procedural and processing specifications (guidelines) to the FGDC reporting standard until anything comes from the feds? I believe the answer is yes.
Here is my idea:
* The FGDC is requiring a reporting standard that shows a positional tolerance (band width) for the values on a station. Namely, the error ellipse for a station's value at the 95% confidence level will be equal to or below a certain value. Typically, 2 cm, 5 cm, 1 dm, 1 m, etc.
* This reporting standard is a better way of communicating survey accuracy as it allows for a single standard for evaluating results from the use of different surveying instrumentation.
* There are no procedural and processing specifications (guidelines) associated with the FGDC reporting standard.
* The FGCS specifications (guidelines) and standards were based on a precisional ratio.
* The FGCS procedural and processing specifications (guidelines) have withstood the test of time for about 15 years.
In my opinion, associating the reporting standard to the tried and true procedural and processing specs is the way to go.
How do we associate the two?
* The main objective is to establish a minimum station spacing distance that you will be reporting positional tolerances on. The FGDC specs could allow for small distances between stations while reporting a tight positional tolerance. This could lead to all kinds of problems and the degradation of the CSRN.
* What is the minimum station spacing? I would suggest that we take any particular positional tolerance for two stations and calculate what that positional tolerance is at the 95% confidence level for the two at the same time. This is a typical error analysis approach. Then determine what the minimum station spacing would be based on the procedural and processing specs that you want to correspond with your desired results. For example, 2 cm positional tolerances surveyed to first order FGCS specs. With this approach you can mix any FGDC band width with any FGCS order procedural and processing spec.
* Ok, Say at the 95% confidence level station A has a desired positional tolerance of X1 and station B has a desired positional tolerance of X2. The error ellipse at the 95% confidence level of the 2 combined is: Y = sqrt(X1**2 + X2**2). And say you want to use the FGCS first order specs. This means that to figure your minimum station spacing distance (Z) you would use the 1:100,000 precisional ratio at the same time as Y. ie: Z = (100,000)(Y). Note: all values are in meters. You would then place your stations a minimum of Z meters apart and survey them using the first order FGCS procedural and processing specifications.
* Confused? (I hope not).
How about an example? Suppose you want both stations to have a positional tolerance of 2 cm and you want the confidence of the first order specs. So, Y = sqrt(2**2 +2**2) = sqrt(8) = 2.828 cm = .02828 meters. And, Z= (100,000)(.02828) = 2828.xxxx meters. So, you would place your stations a minimum of 2.828 Kilometers from each other and use the first order specs.
This could be done with any FGDC positional tolerance and any FGCS order spec.
A chart could be readily made.
Additional thoughts:
* This approach may be an "interim" approach; however, it will increase the confidence of geodetic professionals in the quality of the positional tolerance reported results while helping to prevent the degradation of the CSRN.