Category 1-B Texas Standard Survey
More fun.
The client is expecting something to meet a published standard required for that type of work then that is what will be done. Before you put more words in my mouth, I will not do work for a client that asks to ignore a standard. If they ask to meet the standard, but not to incur extra costs greatly exceeding that standard, that its their call, it is their dime.
Proper QC will be employed to meet that standard (I advocate doing better than the standard if at all possible within the budget).
Now I am eager to see how you try to twist my words, or make up more for me.
Category 1-Beep Survey
> Before you put more words in my mouth, I will not do work for a client that asks to ignore a standard. If they ask to meet the standard, but not to incur extra costs greatly exceeding that standard, that its their call, it is their dime.
Okay, you sign up to do, say, a pipeline job in Texas, which I assume means hiring a bunch of Texas licensees to do the actual work of surveying the cadastral elements of the pipeline project. Do you:
(a) insist that all work returned meet all applicable minimum technical standards and disqualify RTK for certain parts of the work where it is obvious that it won't meet those standards or do you
(b) explain to your contract surveyors that there isn't enough time or money to do the work correctly (i.e. to minimum technical standards) - and anyway it's just a *pipeline* for Pete's sake! - and proceed from there.
Category 1-Beep Survey
I am beginning to see a pattern. By all of these comments about RTK, and all the rest, are you really begging someone to sell you on it?
It appears to be the case.
While I am probably less qualified than you to argue math, I am a long time expert in being hard headed and crusty. So, to speak to what I am an expert in:
I often attempt to learn by arguing with a person. I usually end up making them mad because this is often disrespectful. If I were to ask them straight up, I would learn a lot, and stop looking unprofessional.
Stats For Dummies (And/Or Lazy People)
> s(pooled) = SQRT [(s(1)^2 + s(2)^2 + ... + s(n)^2)/n]
>
> In other words, the estimate of the standard error from n differences is just the SQRT of the mean of the variances for each of the differences.
Statistics being a field of study that makes my head hurt, how might one apply the above concept to a situation in which each point was visited more than twice? Say you have 16 individual points, and you have RTK shots on some of them twice, some three times, and some four times. Can the method be generalized to deal with the spread of the values at each point -- which is the same as the difference when only two values are available -- or does that muck up the math?
No Category 1-Beep Survey
Good questions and very easy to answer:
"Okay, you sign up to do, say, a pipeline job in Texas, which I assume means hiring a bunch of Texas licensees to do the actual work of surveying the cadastral elements of the pipeline project. Do you:"
The applicable standard will be met. Period, or the job will not be taken.
"(a) insist that all work returned meet all applicable minimum technical standards and disqualify RTK for certain parts of the work where it is obvious that it won't meet those standards or do you"
The applicable standard will be met and appropriate tools selected for any portions of the job where they may meet the standard. It is hard to elaborate on the tool selection process without the specific standard. There are different standards in all states and provinces including TX for different types of work. Your scenario is missing that so I will discuss do some generic tool selection steps:
i) Verify what the standard is. The client may not be aware that a higher standard is required. If they will not budget for enough to meet that standard then we will not do the job. No comprimise. A client may also mistakenly assume that they need to meeet a higher standard than required. If they want budget for that then we will gladly do the job. Bottom line on this step: The applicable standard will be met.
ii) Site considerations. If there is insufficient sky view where observations will be required, that may disqualify GPS for those portions. If there are terrain obslacles, natural hazards, restricted areas, and more, then static GPS will be considered. If the site conditions allow, and if the leg distances are appropriate for the type of survey and applicable standards, PPK/RTK/RTN may be considered. The bottom line in this step is: the applicable standard will be met.
iii) The right crew for the tools. If it is decided that total stations are to be used, the crew must be qualified both in theory and practice. Experience is neccesary. Same for static GPS and PPK/RTK/RTN. I would not have anyone, even if licensed, atempt use of a particular tool that does not have the experience. Bottom line on this is: the applicable standard will be met.
iv) Control of measurement. All of the above apply.
As Dr. R.B. Buckner, the respected and respctful published educator and expert on Surveying Measurements and Their Analysis (in the book of the same name) put it:
"Control means to be able to wisely select methods or equipment which either increase or decrease precision and accurachy according to specifications and/or costs" he goes on to say "It may simply mean that the surveyor notices that the results of a certain method repeatedly yeild much better accuracy than requried and decides to modify the system. The modification may mean substituting a less expensive instrument, resulting in a less costly job and an opportunity to use the more precise instrument elsewhere". Wise man that doctor. Bottom line on this is: the applicable standard will be met.
v) Consider a mix. Examples: Closing a traverse on a properly executed static GPS campaign. Adding EDM observations to a static GPS campaign and LSA for the short distances. Many options on a case by case basis. Bottom line on this is: the applicable standard will be met.
vi) Bid the job based on what it will take to meet the standard. If the client is unwilling to pay that, then we will not do the job. As for RTK, as it has not been demonstrated that RTK is categorically and irrefutably disqualified for absolutely all cadastral work everywhere, then it is not assumed to be completely off the table until the applicable standards and all site conditions are considered. Bottom line on this is: the applicable standard will be met.
vii) Be prepared to switch tools. If during the course of a job, a tool is not meeting standard, then the next tool will be deployed. If this increases cost, the client will be informed, but if it resulted from my own poor estimating then I may eat some costs. Bottom line on this is: the applicable standard will be met.
"(b) explain to your contract surveyors that there isn't enough time or money to do the work correctly (i.e. to minimum technical standards) - and anyway it's just a *pipeline* for Pete's sake! - and proceed from there."
No. I will bid the job on meeting the standard. If they are not willing to pay what it takes to meet the standard, then I do not take the job.
If it is not clear that meeting a stanadrd is actually important to me or anyone else, then there is nothing more we can do for you.
Category 1-Beep Survey
dmyhill.
I understand your frustration. People may argue the mertits of one tool or another or one interpretation of some surveying term or another until they are blue in the face. If the goal is to learn form it there may be less blue in the face.
If they have to rip into each other then it is no longer about learning from each other. I see subjects on this board that go on forever and never really change laws, standards, or rarely if ever change anyone's minds. In P&R that is often just fun, but there are board rules there. In surveying there are in many places professionalism and ethics rules from another kind of "Board". I see respected people besmirched (I saw that the other day and liked it), apparently just in the name of winning some argument (which is lost by default if a party has to besmirch). Not our place to try fix it though I am just as guilty as the next for getting caught up in it. Let's just face it, there are some subjects that will keep being "equine exhumation for flaggelation".
Category 1-Beep Survey
> I am beginning to see a pattern. By all of these comments about RTK, and all the rest, are you really begging someone to sell you on it?
Actually, the subject of the thread is found in the original post above. It was made many days ago and still stands. The fact that few RTK users were capable of defending the accuracy of their equipment was a bit surprising, I must admit; but that actually is consistent with expectation since RTK users tend as a group to be relatively unsophisticated.
Stats For Dummies (And/Or Lazy People)
> > s(pooled) = SQRT [(s(1)^2 + s(2)^2 + ... + s(n)^2)/n]
> >
> > In other words, the estimate of the standard error from n differences is just the SQRT of the mean of the variances for each of the differences.
>
> Statistics being a field of study that makes my head hurt, how might one apply the above concept to a situation in which each point was visited more than twice? Say you have 16 individual points, and you have RTK shots on some of them twice, some three times, and some four times. Can the method be generalized to deal with the spread of the values at each point -- which is the same as the difference when only two values are available -- or does that muck up the math?
Yes, the above method of forming a pooled estimate of the standard error can be easily generalized if you keep track of the degrees of freedom of each estimate of s from repeated positions on the same point. If there were n repeats on a point, then the standard error estimated from them will have (n-1) degrees of freedom.
When you form the pooled estimate take the SQRT of the weighted mean of the variances s^2. For example if you occupy three points, 1, 2, and 3 as follows:
So s(pooled) = SQRT [4(s(1)^2) + 1(s(2)^2) + 2(s(3)^2)/(4 + 1 + 2)]
And in that example, s(pooled) will have (4 + 1 + 2) degrees of freedom in case you want to pool it with further results.
Stats For Dummies (And/Or Lazy People)
I should add in case it isn't obvious that the standard errors of the N or E coordinates of a series of n repeated positions on a control point is formed as any ordinary estimate of standard deviation from a small sample :
s = SQRT [(sum of squares of mean deviations)/(n-1)]
RTK Could Meet The Standards ?
The whole point of the statistics arguement is to be able to show and prove one has met the standards. If you cannot do that then you have not met the standards.
Paul in PA
RTK Could Meet The Standards ?
Yes. That is correct in that the ussr must be able to demonstrate that it meets a standard. It depends what standard as to whether RTK could meet it.
It has not been shown that RTK or any other tool cannot meet EVERY standard for EVERY leg of EVERY type of cadastral work. There are urban and rural standrds. It may not meet certain standards for short legs but may meet standard on longer legs of the same, and whether it can or not should be analyzed by the user and demonstrated through statistical methods. Where did I or anyone else ever say or advocate otherwise? A scenario was posted yesterday and it was was answered respectfully and as complte as could be asnswered despite there being no standard sepcificed.
My official stance it is that a standard will be met by whatever combination of tools, methods and analysis can and do meet the standard, and if a change of plans is needed then so be it.