Rob,
Sorry to disappoint you, but there is, in fact, no accurate of conversion of datums between NAD27 and NAD 83. This true especially from NGVD29 to NAVD88 vertical datums. Here is a quote directly from the NADCON readme file.
NADCON conversions between datums are approximate values based on models of real data. NADCON should be used only when data does not exist in the data base (NGSIDB) for one of the datums required. The accuracy of the transformations should be viewed with some caution. At the 67 percent confidence level, this method introduces approximately 0.15 meter uncertainty within the conterminous United States, 0.50 meter uncertainty within Alaska, 0.20 meter uncertainty within
Hawaii, and 0.05 meter uncertainty within Puerto Rico and the Virgin Islands. In areas of sparse geodetic data coverage NADCON may yield less accurate results, but seldom in excess of 1.0 meter. Transformations between NAD 83 and States/Regions with High Accuracy Reference Networks (HARNS) introduce approximately 0.05 meter uncertainty. Transformations between old datums (NAD 27, Old Hawaiian, Puerto Rico etc.) and HARN could combine uncertainties (e.g. NAD 27 to HARN equals 0.15 meter + 0.05 meter = 0.20 meter). In near offshore regions, results will be less accurate but seldom in excess of 5.0 meters. Farther offshore NAD 27 wasn't defined. Therefore, the NADCON computed transformations are extrapolations and no accuracy can be stated.
Oh, and remember, these accuracies are at the one sigma level ( 67% of the time) That means that the other 1/3rd of the time they could be off hundreds of feet (not likely but still statistically possible)
This may be accurate to you as a geocache hobbyist, but as a surveyor, you should be alarmed if you use this conversion in your work.
The same holds true, even moreso, with VERTCON. It's accuracies really stink in some places. I am in Florida, and I have personally seen this "conversion" used by surveyors, and checked their results. Not good enough for surveying. Now in areas where there are a lot of common datum points (ie having both NGVD, and NAVD dataums, the conversion is closer than out in the boonies where the common points are sparse. But even then, close ain't good enough for a surveyor. Actually, its scares me to learn of the number of surveyors that use VERTCON to transorm their vertical benchmarks to get a job done. I wouldn't do it, and what's worse is that they sign and seal this work. Not me, not gonna happen.
Since you are a surveyor, let me give you some more technical fodder. The reason that the transformations cannot be accurate is that the source references are not related, linearly, or otherwise. The horizontal datums generate from two totally related ellipsoids. NAD 27 is based on the Clarke Ellipsoid of 1866, with its initial point located at Meads Ranch, Kansas. NAD 83 is Earth centered and based on the Geodetic Reference System 1980 (GRS 80). Factor in the inherent survey "distortions" (an NGS derived term...heheheh, I read FUBARS) The two just can't be accurately related. Again, from the standpoint of geocahing, using small hand held code receivers, its close enough to use. For surveying....no cigar (where is Monica anyway) Now, even with high end dual frequency, dual P-code survey grade receivers, you can "localise" into a NAD 27 project and get very good results... but that is a topic for a whole new discussion, and has little to do with transformations.
Vertically, its even worse, NGVD assumes that "Mean Sea Level" all around the country is 0.00' (we learned that the Atlantic and Pacific Oceans are not at the same elevation when we cut the Panama Canal, hence the lock system [we may have known sooner.. but that is the popular tale]). Well when the huge least squares adjustment for NGVD occurred (by hand, mind you) it was constrained by having to hold 0.00 at all the tide station control points. That coupled with the survey "distortions" made it supposedly inaccuate. I suppose on a national scale it is, but in a local area, its only as inaccurate as the original leveling. (the adjustment errors are distributed accross the country). Now when NAVD was created, get this, it holds one (1) point as fixed, at Fathers Sound somewhere up in the Cold Country North of here, and adjusts a lot of the same runs, but they are minimally constrained (only one control point). So again, there is no direct corrolation between the systems. If NGVD had used only one fixed point, one could just apply the difference between the two systems and presto, an accurate transformation. ( a linear relationship)
Since we can see how the systems differ in nature, we can understand why they cannot be accurately related. However, because of pressure from surveyors and engineers, geologists, et al, the NGS came up with its best solution to translation between the datums. NADCON, and VERTCON. They basically find the closest common points to the desired shift position and "best fit" the solution between them. Obviously, the closer to a known dual point, the more accurate the shift value. Not a true least squares solution, but a good idea. Not including the surveying "distortions". ( heheheh I love that word)
As for NGS, well they have taken quite a beating over the years, going from over 6000 employees to around 600. They did at one time have "monument dusters" that drove around verifying marks everyday, but that died by the wayside as their personnel dwindled down. I know that they have used private sources to dust monuments for years and I was not trying to bash sport benchmarking, but I look at the posts here and see folks remark on how the Power Squadron missed locating a mark, and find pleasure in scoring a find over them ( I would and do too). My concern lies in reporting a loss. How many professionals read the find list and look at the list of recovery dates and see a reported destroyed or loss and don't look themselves? Most of them. Here is the rub, now they have to travel farther away to get a known point which equates to money. Increased fees to the clients for additional work. Should they scout ahead of time to know their control, yes in perfect world, but then again, if they pull their control from the net, and its marked gone.. they go looking for another... if that mark is still there, its a loss. Not trusting those notes myself, unless they were from a professional duster ( a DOT, County, or reputable surveyor) I have always gone to look myself. I have found some that the Squadron said were gone, and saved the taxpayer a whole buncha money. Your statement to only report the finds makes great sense to me... a note saying the monument is there helps me more than a note from that same someone that says its not. But that is just my two cents.
Thanks for you time, hope I offended no one.
Doc
PS
I didn't reread this, so if I have a spelling error, or a lost train of thought, I apologize.
