Mixing Epochs in Basis of Bearing

[Jedwards]

Hi all,

I'm reviewing a Record of Survey where the basis of bearing is established between two NGS control points, but the issue is this:

One point is listed in NAD83(2011), Epoch 2010.00

The other is still published in NAD83(1992), Epoch 1991.35

The map also applies a combination factor of 0.99991473 at one of the points (KS1811) to convert between grid and ground distances.

My Questions:
Is it acceptable to establish a basis of bearing using two points that are on different datum realizations and epochs? Or does this introduce significant positional error?

Would applying the combination factor in any way account for the epoch difference? (I assume not, but I want to confirm.)

If transformation is needed, would using NGS’s NCAT tool be appropriate to bring the older point forward to NAD83(2011), Epoch 2010.00?

Would it be a better practice to only use points from the same epoch/datum to avoid any ambiguity on a record map?

Looking to make sure the methodology is defensible, especially here in California where tectonic movement is a factor. Appreciate any insight or recommended resources.

Thanks in advance!

[DWoolley]

I heard the news today oh boy, why did it have to be Tara Browne instead of a random epoch mixer?!

Locally, we realize 0.15' of movement per year. If it is the same in your area, the difference between the coordinate epochs is 2010-1991 = 19 years at 0.15' per year is 2.85 feet. Restated, 2.85' was introduced into the survey. Typically, the NGS points are held fixed and the "error" is transferred onto the surveyed points. Decide for yourself if this is significant.

As for the combination factor, it should be calculated for the site specific. The NGS or CSRC points are typically not representative the site due to distances from the site surveyed.

It came to my attention recently we have folks working in real-time epochs - which is actually against the law (California Public Resource Code). I recently published a video on topic here: https://www.youtube.com/watch?v=O4A_lbWgrxM&t=112s and the second half of the video here: https://www.youtube.com/watch?v=QEowP5QRhv4&t=193s

DWoolley

[Mike Mueller]

Is it acceptable to establish a basis of bearing using two points that are on different datum realizations and epochs? Or does this introduce significant positional error?

No. It is not correct and depending on the particulars of the two points it could introduce quite a bit of error.

If you check the vector map of the area in question you can get a rough idea of what the difference might be. For example here in Sonoma County there are many locations that have 2.5 cm a year vector pointed to the NW. So 1991.35 to 2010 is roughly 46 cm off. If the points have a azimuth that is NE/SW then the change in the azimuth between the 1991.35 and the 2010 position due to plate tectonics could be rather extreme. If the bearing between the two points is roughly in line with the area's vector then there is less issue.

Would it be a better practice to only use points from the same epoch/datum to avoid any ambiguity on a record map?

Yes. This is really the best option.

Another consideration is that they are mixing realizations as well. I would defer to some of the more expert folks here on the forum, but IIRC there was up to 10-30cm or so of change between those two realizations? Personally I would ask the submitter if they are ok with 10-50cm of error in their survey and to have a CS note explaining the technical issues related to mixing realizations and epochs like that with a statement that all positions may be off by 10-50cm. Does your area have a PPC that could help explain the situation in a way that might let the submitting surveyor save face?

If transformation is needed, would using NGS’s NCAT tool be appropriate to bring the older point forward to NAD83(2011), Epoch 2010.00?

I believe that is the correct methodology now. However it is a subpar process compared to utilizing two positions published on the same realization and epoch.

Would applying the combination factor in any way account for the epoch difference? (I assume not, but I want to confirm.)

No. In theory, the combination factor wouldn't really change much between realizations and epochs, since it is a function of the point's location in the plane and its elevation. But the combination factor won't account for the sort of error that may be introduced from the tectonic movement and the realization shifts.

One thing you might suggest they do is to provide a Lat and Long for each of those positions both in 1991.25 and in 2010. The Lat and Long are in theory a fixed grid that showcase the actual movement of the chunks of metal through time so by showing both, they will get a chance to see the change in SPC? However I am not sure how much you consider your role to be educator :)

Mikey Mueller, PLS 9076
Sonoma County

PS Take my post with a grain of salt :) I hope someone more expert chimes in and either corrects me or confirms what I said.

[David Kendall]

I have no dog in this fight but to play devil's advocate here:

For the reasons described, I expect that the state plane coordinate values, if published on the survey, are out to lunch.

I recall a couple of knuckleheads on this here forum, cannot recall which ones, asserting that the basis of bearing is irrelevant, so long as it is stated clearly.

Since I see very little chance of someone using NGS mons to directly retrace parcel boundaries, this local movement may be irrelevant. Another thing to consider is that if the distance between the two points is small, the monuments could be moving in the same direction over time, thus rendering the tectonic vectors irrelevant. Essentially, I submit that if the (short geodetic) record distance between the two points matches measured, it may be a good tie, and enough data to allow the retracing surveyor to follow in the footsteps.

I would need more information to understand the situation and how the basis of bearings is being used. I agree fully with the assertions made and the suggestion to place a CS note to highlight the potential for error. If there is other identifying information on the map sufficient to facilitate retracement then maybe it would not disqualify the survey.

Trying to establish a geodetic BOB in this way seems awkward. With a very long line and using NCAT to match up the coordinate value metadata, it could possibly be done with sufficient accuracy but there are probably better ways

[Jedwards]

I heard the news today oh boy, why did it have to be Tara Browne instead of a random epoch mixer?!

Locally, we realize 0.15' of movement per year. If it is the same in your area, the difference between the coordinate epochs is 2010-1991 = 19 years at 0.15' per year is 2.85 feet. Restated, 2.85' was introduced into the survey. Typically, the NGS points are held fixed and the "error" is transferred onto the surveyed points. Decide for yourself if this is significant.

As for the combination factor, it should be calculated for the site specific. The NGS or CSRC points are typically not representative the site due to distances from the site surveyed.

It came to my attention recently we have folks working in real-time epochs - which is actually against the law (California Public Resource Code). I recently published a video on topic here: https://www.youtube.com/watch?v=O4A_lbWgrxM&t=112s and the second half of the video here: https://www.youtube.com/watch?v=QEowP5QRhv4&t=193s

DWoolley

Thanks for the information DWoolley. I also watched the linked videos and determined that requirements per PRC 8815.5 were triggered per this language in the Basis of Bearings statement.

"...WITHIN THE CALIFORNIA STATE PLANE COORDINATE SYSTEM, ZONE 2 AND ORIGINATE FROM MONUMENTS ESTABLISHED BY THE NATIONAL GEODETIC SURVEY (NGS)..."

-Jacob Edwards

[Jedwards]

Would it be a better practice to only use points from the same epoch/datum to avoid any ambiguity on a record map?

Yes. This is really the best option.

Another consideration is that they are mixing realizations as well. I would defer to some of the more expert folks here on the forum, but IIRC there was up to 10-30cm or so of change between those two realizations? Personally I would ask the submitter if they are ok with 10-50cm of error in their survey and to have a CS note explaining the technical issues related to mixing realizations and epochs like that with a statement that all positions may be off by 10-50cm. Does your area have a PPC that could help explain the situation in a way that might let the submitting surveyor save face?

If transformation is needed, would using NGS’s NCAT tool be appropriate to bring the older point forward to NAD83(2011), Epoch 2010.00?

I believe that is the correct methodology now. However it is a subpar process compared to utilizing two positions published on the same realization and epoch.

Mike,

I think I will push back and require that if the Basis of Bearing is going to be established between NGS monuments, they must be from the same datum realization and epoch. Otherwise, we’re introducing unnecessary error — and I’m not comfortable with that, especially on a Record of Survey that may be referenced in future maps.

At the very least a note like you suggest explaining the inconsistencies to satisfy 8764(7) of the PLSA.

-Jacob Edwards

[Jedwards]

Since I see very little chance of someone using NGS mons to directly retrace parcel boundaries, this local movement may be irrelevant. Another thing to consider is that if the distance between the two points is small, the monuments could be moving in the same direction over time, thus rendering the tectonic vectors irrelevant. Essentially, I submit that if the (short geodetic) record distance between the two points matches measured, it may be a good tie, and enough data to allow the retracing surveyor to follow in the footsteps.

The NGS monuments are relatively close together (about 3,600 feet apart), and several record maps are also referenced. My main concern is whether using different datum realizations and epochs would introduce error that affects retracement or future referencing. Given the short distance between the NGS points, are you saying that error might be considered mitigated?

I would need more information to understand the situation and how the basis of bearings is being used. I agree fully with the assertions made and the suggestion to place a CS note to highlight the potential for error. If there is other identifying information on the map sufficient to facilitate retracement then maybe it would not disqualify the survey.

For reference, here is the BoB statement, identifying information removed for privacy. Anything you note would be much appreciated.

"THE BEARINGS SHOWN HEREON REFERENCE THE NORTH AMERICAN DATUM OF 1983 (NAD83) 2011 EPOCH WITHIN THE CALIFORNIA STATE PLANE COORDINATE SYSTEM, ZONE 2 AND ORIGINATE FROM MONUMENTS ESTABLISHED BY THE NATIONAL GEODETIC SURVEY (NGS) AND DOCUMENTED AS “xxxxxxx” PID: xxxxx AND “xxxx” PID: xxxxx. ALL DISTANCES ARE IN U.S. SURVEY FEET AND GROUND MEASUREMENT. TO EQUATE WITH GRID MEASUREMENT MULTIPLY BY THE COMBINATION FACTOR 0.99991473 APPLIED AT NGS POINT xxxxxx."

-Jacob Edwards

[Mike Mueller]

Based on that BoB statement I am assuming this is some sort of attempt to provide a map that is "on state plane coordinates" sufficient for some downstream user to import other data sets. Since the distances are relatively small leaving the map on ground vs grid distances doesn't really impact the accuracy of the downstream data sets since most of those that I have seen are not concerned with a few tenths. One position with a SPC or Lat Long and a rotated CAD file is enough for the average downstream user to get something that is "good enough".

The concern I would express to that map's submitter is that they are potentially having a rotational issue upwards of the 10-50 cm range at the far limits of the survey. If the map's submitter were to inform their client that there was this much potential error, would that client be OK with it? What about that client's client?

As David mentioned, I personally don't think what is used as the basis of bearings for a record of survey matters much. Interior angle relationships are what matter, so what north your CAD file thinks is F8 up doesn't really change the relative constellation of a map's monuments. However providing a CAD file to a client that claims to be on a grid basis of bearings when its not is another matter. That is like saying 2+2=4 out loud but writing 3.9 as the answer. Its close enough that most folks won't care, but its still wrong.

Good luck with your conversation!

Mikey Mueller, PLS 9076
Sonoma County

[DWoolley]

Jacob Edwards:

Another thought on combination factors (CF) and “ground” distances: ground distances typically introduce distortion, whereas grid distances represent measurements in their best form.

Here’s a simple way to illustrate the issue. Take any four CORS/CGPS coordinate values. To determine the ground distance between two stations, average/mean the combination factors of the respective stations. For example, for stations 1-2-3-4, compute the ground distance from 1-2 using the average/mean CF of stations 1 and 2, then do the same for 2-3, 3-4, and 4-1 - each line has its own CF. Now traverse those ground distances—what happens? The traverse won’t close; it will mis-close by several feet. That’s the problem: using a single mean CF distorts otherwise accurate measurements.

The solution? Work entirely in grid from the control ties to the project and keep the project in grid whenever possible.

One caution: the underlying record is often in ground. Don’t mix record (ground) values with grid indiscriminately. For example, some work in grid (correctly) but then use a record radius of 1500' to reestablish a corner—this creates conflict/error. Another problem, folks create "ground" maps and do not provide the scale point or factor applied. Don't get me started on Low Distortion Projections.

For most small projects (under 200–300 feet), the difference between grid and ground is negligible in my area—though higher elevations can affect that and mileage may vary. We lock our projects to grid and keep the project monuments in grid. Subdivision work or area computations eventually require transitioning to ground, but when using RTN GPS, positional errors often exceed grid/ground differences for distances under 600 feet. Again, elevation and working near the zone boundaries can amplify distortion.

Jacob, good job reviewing the record of survey and asking the right questions. As I mentioned, I’ve seen surveyors using RTN subscription corrections that broadcast real-time epochs instead of published epoch values. I am seeing work product without epochs stated, mapping angles, elevations etc. I’m also seeing maps with only single ties—like in the video. The troubling part is this isn’t new technology, and the laws governing it are decades old. Keep a close eye on the surveyor who submitted that record—what you found is likely just the tip of the iceberg in terms of negligence or incompetence. It is unclear to me how we got this far off the track.

Glad to hear the videos were helpful. Welcome to the rodeo, keep marking those donks out of the chute.

DWoolley

[Jim Roepel]

Apples are put in an apple basket and oranges are put in a basket for oranges. Elsewise you have fruit salad. Make them all equate by whatever transformation software they are using today and keep them separated. Why would you want to introduce that extra error into your system? I'm a traditionalist. Keep your project all on the same realization and epoch. Don't cross the streams. Don't create problems for yourself by a bad design to start.

[SPMPLS]

Section 8815.3 of the PRC addresses not using control that are not on the same epoch date. I interpret the last sentence of this to mean either HTDP or SCIP (CSRC product) may be used to align the epoch dates of control.

8815.3. When the published epochs of the controlling stations for a
survey using CCS83 coordinates are not the same, appropriate
adjustments shall be made to the horizontal positions of controlling
stations so that the coordinates of all the controlling stations are
consistent. These adjustments in the horizontal positions of
controlling stations shall be made in accordance with procedures and
values published by the NGS or CSRC.

[Steve Martin]

There was a question on the 2005 LS Exam (the one the Board released) that covered this exact scenario. Take a look at the CLSA Exam guide. I believe there is a solution there for it. Or you can PM me. I have a slide that covers it.

[Timothy J. Reilly]

If transformation is needed, would using NGS’s NCAT tool be appropriate to bring the older point forward to NAD83(2011), Epoch 2010.00?

It's actually the opposite. Instead of advancing the older epoch forward, the station with most current epoch needs to be turned back to match the older epoch. NGS data sheets include the superseded values of the station(s) from previous epochs. Those are the published values that satisfy the "known coordinate" requirement of PLS Act section 8771.5 and PRC section 8815.3.

[DWoolley]

...
Glad to hear the videos were helpful. Welcome to the rodeo, keep marking those donks out of the chute.

DWoolley

More ChatGPT experimenting:

what does this mean? marking those donks out of the chute.

The phrase "marking those donks out of the chute" is likely a mix of slang and idiomatic language, possibly with rodeo or gambling origins. Here's a breakdown:

Possible Interpretations:

1. Rodeo context:
“Marking out” is a rodeo term. In bronc riding, a rider must “mark out” their horse at the beginning — keeping their spurs above the horse’s shoulders when it leaves the chute. Failing to do so disqualifies the ride.

“Donks” could be slang for unskilled or weak animals (or metaphorically, people).

“Out of the chute” refers to the start — like an animal coming out of the gate in a rodeo.

So, in this context:

“Marking those donks out of the chute” could mean identifying or calling out the weak performers or rookies right from the start.

2. [Reference to poker, deleted]

TL;DR:

"Marking those donks out of the chute" means identifying the weak or inexperienced participants right at the beginning — whether in rodeo, poker, sports, or any competitive setting.

[AI surprises me to know to pick up on the reference. As a former rodeo guy, I occasionally use words not found in everyday speak - apparently ChatGPT does the same.]

DWoolley