Oregon's coordinate system is governed by Administrative Rule so updating them is much simpler.
Yes, I wish we had that option so our DOT could just take care of the updates.
Were you involved in the 2014 GSVS work done in Iowa by NGS?
I think you would have to say we were involved as a facilitator. They seemed to understandably want to use their own people for the work. But they certainly kept us in the loop. We recovered a few marks they tied to.?ÿ
The main point of my question must have been misunderstood. I'm trying to explain the 'shift' to a user who pretty much only knows how to add points, stake out points, and use basic collector tools. I think it is important that the users start collecting the Metadata as soon as possible. This will help convert back and forth easier in the future and develop better habits for note taking. I also have been telling them the predicted release of 2025 and direct them to the NGS website for more information.?ÿ?ÿ
I used the reference to a total station setup as an example for the county engineer using an occupied point and back sight on a datum like 5000.00 / 5000.00. I told him if your control points were all adjusted to 5001.00 / 5001.00 he will see a shift in the rest of the points he is collecting or staking out. This will happen if they are using the RTN for a project before the release of '2022' and continue with the project after the RTN is adjusted to the new datum. I thought this was a simple explanation for a county engineer with little GNSS knowledge.?ÿ
Do you think this is a good basic explanation, or should a few more details be added??ÿ
If you are using SPC or a regional coordinate system the shift won't be like going from 5000 to 5001. It will be more like 5000 to 9000. The false N and E will make the difference easily noticed.
Vertically, it will be more subtle and the metadata will need to be strictly watched.
The OP appears to be referring specifically to the use of a VRS network to a localization. He is asking if his local coordinates will change from a localization created from a VRS in say 2020. Then re-using the same localization with his same VRS in 2025. This is a very good question. The VRS provides Geographic coordinates (Lat/Lon/Ellip). The Cartesian coordinates N,E,Z are computed via the projections or a localization. It will be up to each VRS network administrator to determine what Geographic coordinates they will publish of the VRS. If they change then you will need to adjust your parameters in your DC. It is a parameter within your data collection software that will be easy to identify and need to be actively acknowledged by the end-user. In Magnet Field for example if the end-user wants SPC you select a Projection, a Datum, and a Geoid model. For localization, you still need to select the Datum and Geoid model, on a VRS this is very important to calculate from Geographic to Cartesian coordinates.?ÿ
Today we use Datum NAD83_NO_TRANS, with Geoid 2018. These parameters may change to something like NSPS2022 and Geoid2022.?ÿ
The OP is correct in that proper Metadata will be required.?ÿ
I believe we misunderstood your question because of your terminology. When using a VRS or RTN most assume you would utilize a known projection and not a localization or a calibration. Also the terms datum and projection can be confusing.?ÿ
Unfortunately using a localization with a VRS or RTN will be very dangerous without proper Metadata.?ÿ
Our RTN does not broadcast cartesian coordinates or elevation. It broadcasts the current NGS CORS realization of datum and?ÿ ellipsoidal height- whatever that may be and adjusts when CORS updates. The projection and/or calibration and elevation calculation happens at the rover. Use of the RTN after any adjustment requires a new calibration to align with cartesian coordinates assigned to a point before the adjustment just like when you align a new survey to an old one. The transformation from one adjustment to the next is not a constant xyz value across the RTN any more or less than NGS CORS is a constant across their network.?ÿ I wish I had a dollar for every phone call we got asking if we changed the Geoid again. The answer was we don't use one- silence.?ÿ
From a geodetic perspective (either thinking in latitude/longitude/ellipsoid height or Earth-Centered, Earth Fixed XYZ) the components of the shift will vary depending on location. In 3D, the shift is about 2 meters, mostly in ellipsoid height.?ÿ
Horizontal change map:
HorizontalNorthAmericanPlate.jpg (600??449) (noaa.gov)
Ellipsoid change map:
EllipsoidHeights.jpg (600??449) (noaa.gov)
Note that the ellipsoid height will change by about 1.5 meters. Orthometric heights won't change as much, but will still be changing. The geoid model will account for a lot of the difference in ellipsoid heights from NAD83 to NATRF2022, but the differences won't be trivial at about 0.5 meter.
Orthometricheights.jpg (600??450) (noaa.gov)
State Plane Coordinates will change too. RTN's don't operate in SPCS. RTN's send corrections related to ECEF. So the shift to your RTN and rover will be on the order of a couple of meters as shown above. The data collector takes the resulting LLH from your rover and projects it onto a grid, i.e. State Plane. All State Plane Coordinate System Zones will be changing with the new datum. They may have the same zone boundaries as NAD83 State Plane, but the parameters will be different. As a result, it isn't appropriate, in my opinion to view the change to the coordinates as a "shift". I suppose individually a particular point could be treated as a shift, but a geometry of points (like the outline of a property boundary) will have a transformation (translation, rotation, and scale). In some cases the rotation may be so small as to be negligible. For some Transverse Mercator based projections, there might not be any rotation, if the origin is the same. For Lambert based projections, the change from 2-parallel definitions to 1-parallel definitions, combined with the even-minute required origin for the new zones will mean there will be some rotational difference further from the origin, even if the central meridian is unchanged. All zones will be scaled to minimize linear distortion at the Earth surface (i.e. topographic surface across the zone, whereas the original SPCS designs were optimized for the ellipsoid. As a result there will be a scalar difference between SPCS83 and SPCS2022 in all cases, I would expect. The grid origin for all SPCS2022 zones will be different by at least 10,000 meters (as I recall) to ensure that the coordinates will be significantly different from one another so that users will immediately be able to distinguish between 83 and 2022 SPCS values. The same thing was done for the change from 27 to 83.?ÿ
Many States are employing LDP designs so that there will be more zones within a State than the 27 and 83 designs had. These will all have different geodetic origins and different grid origins so there will be different rotations and coordinate values for all of them compared to each other and compared to the SPCS83.?ÿ
Localizations will change too. A localization (or calibration for Trimble users) is a 7-parameter, best-fit, transformation between two coordinate systems. The 7-parameters are translation in N, E, and U, and rotation around N, E, and U-axes, and scale. When you perform a localization, either from an RTN or your own base, the data collector software performs a least squares best fit 7-parameter transformation from some projected grid system (i.e. State Plane Coordinates) to an unknown grid system (i.e. 5000, 5000). The software figures the translations in N, E, and if desired Up. It also figures rotation around three axes (N, E, and Up). As surveyors we are usually only interested in the rotation around the Up axis. It also determines the scale from the projection coordinates to the unknown grid coordinates. Note that scale is based only on a comparison of lengths and does not use any geodetic scale factor (such as grid factor and elevation factor), unless your specific software allows for it (some do, most do not).?ÿ
A note about the scale. The Grid Factor is based on the location of a point in the projection. For Lambert projections, changes in coordinates in a North-South direction will result in changes to the Grid Factor. For Transverse Mercator projections, changes in coordinates in an East-West direction will result in changes to the Grid Factor. So the scale of a geometry of points (i.e. the outline of a boundary) will change depending on the projection used. A perfect square in one projection may not be a perfect square in another projection because the linear distortion will be different for the different sides. For smallish projects, the difference is negligible, but it illustrates why a simple shift, rotate, and single scale, isn't exactly perfect and will be more apparent the larger the project and the bigger the difference in projection type and origin.
?ÿ
More about scale. As has been stated, NIST has made the International Foot the official definition of the foot. This is within their Constitutional authority to do so, even if it presents some challenges for those of us in States that have adopted the US Survey Foot as the official foot. For the linear dimensions of smallish projects, the difference will be trivial, but for large coordinate values, the difference will be significant. Imagine a place with State Plane Coordinates of N 2,000,000 ft and E 1,000,000 ft. Using the wrong foot would result in an error of 4 feet in North and 2 feet in East, or a radial error of 4.5 feet.?ÿ
As surveyors, we will have to be familiar with all of these various components as any breakdown in properly selecting the correct datum, zone definition, geoid model, unit, could result in significant errors. Depending on the mistake, the error might be just small enough that it isn't easily apparent but still way outside of tolerance.
As surveyors, we will have to be familiar with all of these various components as any breakdown in properly selecting the correct datum, zone definition, geoid model, unit, could result in significant errors. Depending on the mistake, the error might be just small enough that it isn't easily apparent but still way outside of tolerance.
I wish I had a dollar for every phone call we got asking if we changed the Geoid again. The answer was we don't use one- silence.?ÿ
Considering we're supposed to be authorities (if not experts) on datums, projections and their relationships, the number of surveyors that do not understand basic geodesy is disturbing.
I'm torn between fear of the incredible number of screwups we are about to see and elation at the job security that will likely come from fixing said screwups.
"For Lambert based projections, the change from 2-parallel definitions to 1-parallel definitions,...."
We have to be careful here from a technical mapping viewpoint. Consider the the comparison of the NC preliminary 2022 to the current NCSPCS contained here:
https://www.ngs.noaa.gov/SPCS/images/SPCS2022_poster_left_panel.pdf
In the NAD83 version, two parallels, each with scale factor = 1.00000000 were defined at 34:20 and 36:10 (degrees:minutes north latitude.) From these two assumed parallels, the central parallel was calculated to be 35.2527586002 degrees and 0.999872591882 scale factor.
In the 2022 preliminary version, the central parallel is assumed to be 35 degrees 15 minutes north latitude, scale factor = 0.99995. Now two parallels, one north of the assumed central latitude and one south, with scale factors equal to 1.00000000 still exist and can be calculated.
The single parallel and two parallel Lambert projections are equivalent. The original NAD83 version can be exactly duplicated by assuming a central parallel of 35.2527586002 degrees and 0.999872591882 central scale factor.
Thus the choice of definition, one parallel or two, in itself, has no effect on relative distortions. However, the change in location of the central parallel does and it works as you described.
?ÿ
?ÿ
?ÿ
?ÿ
?ÿ
?ÿ
?ÿ
?ÿ
@mathteacher?ÿ I think they mention somehwere that they're not bound to the 'even minute paralel' convention anymore.
The difference between those two paralells would equate to .. 1/4 mile at that latiitude ??ÿ
No, it's much further. On the current NAD86 map, we can use NCAT to calculate state plane coordinates at 36-10, 79-00 and 34-20, 79-00, the two standard parallels on the central meridian. The eastings will be the same, the Northings, in USft, are 879,570.859 and 212,307.620.
The grid distance between the two is 667,263 feet. To get the ellipsoidal distance we divide this by the average scale factor and there's an aha moment here. The two scale factors are both 1, so their average is also 1. But we know from the geometry of the situation that the map is below the ellipsoid. We need to use the Simpson's Rule correction (SF1 + 4*SF(midpoint) + SF2)/6. Now the central parallel is not midway between the two standard parallels, but it's close enough for this. The average SF is then 0.99991506 and the ellipsoidal distance is 667,320 feet or 126 miles. One is north of Greensboro and the other actually passes partly through South Carolina.
If you look at the 2022 preliminary, it's not very different in either the central parallel or the scale factor for the central parallel, so the distance between the 2022 standard parallels should not be very different either.
I think that NGS decided that the location of the standard parallels is irrelevant. Doing so allows a central scale factor that is more representative of a state-wide scale factor. I did the calculation a couple of years ago using 0.99996 as the scale factor. It's on the hard drive of a dead computer, but I can access it. Otherwise, it's buried in a long-ago post here.?ÿ
We can build a .prj file that will work for these single parallel maps in GIS software. I'll try to do one for NC tomorrow using the 0.99995 central scale factor.
?ÿ
?ÿ
Thought we were talking central parallel on the planar projection.?ÿ Interesting though.
the central parallel was calculated to be 35.2527586002 degrees
the central parallel is assumed to be 35 degrees 15 minutes north latitude
The 35.2527... is the central parallel on the current NAD 83 NC State Plane. The 2022 version is 35.25 degrees.
But look at how close to each other they are; just 0.0027... degrees apart.
Now look at the two scale factors. Current is 0.999872 and the 2022 is 0.99995. So the new one puts the map closer to the ellipsoidal surface which does make the two standard parallels closer together. The scale factor change is more than I thought last night, but I'll plug in that surviving hard drive and see how close the parallels really are.
Indeed, it is the scale factor that determines the distance between the calculated standard parallels. If it is 1, then the map plane is tangent to the ellipsoid and there are no standard parallels.
Trial and error coordinate and scale factor calculations put the Southern Standard Parallel at 34-40-30 and the Northern Standard Parallel at 35-49-25 for the 2022 version of the NC State Plane Coordinate system. In both cases, the scale factor is 1.0000001, so the precise parallels are a fraction of a second closer than the trial and error stopping point.
NGS hasn't given the false northing and easting, so we just take use those numbers from the NAD83 version. The northings on the central meridian that represent the latitudes above are 336,682.649 sft and 754754.634 sft.?ÿ
The distance between them is about 79 miles, which is less than half the distance between the standard parallels in the current NAD83 version.
The new one, then, is not a single parallel Lambert map. To be a single parallel map, the scale factor on the central parallel would have to be exactly 1. But it's a much better map and assigning the scale factor and central parallel is a superior development method.
Is it really part of the definition of a single-parallel Lambert that the s.f. has to be 1 there?
Can't the name be used while adjusting the projection to fit the terrain, as in the many LDPs that have come into use in recent times?
Increasing the grid scale factor is an attempt to get grid and ground closer. The Montana Coordinate system was switched from a two zone to a one zone Lambert with the change to NAD83 and the designers wanted all grid scales to be less than 1. So the north and south state lines are 1,,,,,,more or less. It created grid distances that are way less accurate compared to ground than sloppy chaining measurements. This I would argue is a hold over from pre GPS days when state plane was hand calculated with the state plane booklets. Increasing the grid factor nearer one will help push the grid distances up closer to ground, but in low elevation there will be areas with grid distances longer than ground. The old state plane designers didn't want that.
The single-parallel Lambert is tangent to the ellipsoid at the central parallel. The two-parallel Lambert is secant to the ellipsoid at the two standard parallels. In both cases, wherever the map touches the ellipsoid, the scale factor is 1. If the scale factor is less than 1 at the central parallel, then the two standard parallels exist, and the map is a two-parallel Lambert.
LDPs are an interesting case; they can be single or two-parallel design. What sets them apart is the combined factor, the product of the scale factor and the elevation factor. It will usually be the product of a scale factor that is less than (two parallel) or equal (single parallel) to 1 and an elevation factor that is greater than 1. This combined factor, then, can be 1, less than 1, or greater than 1, depending on the underlying design and the ellipsoid height selected.
It's in the LDPs that terminology becomes important. Loose descriptions of combined factors as scale factors lead to confusion. To me, it's always made sense to use NGS terminology: scale factor, elevation factor, and combined factor, but that's far from standard in the industry. As long as everyone is clear about meanings, it's unimportant, but that's often not the case.
This whole 2022 exercise is a great learning environment. So far, everything in the Lambert group has been duplicable from Stem's Manual 5.
As I think I understand Dr. Dennis, you can define a 1-parallel Lambert for any ellipsoid you want, not just for example the GRS ellipsoid currently used as our best overall model of the earth.?ÿ An LDP may have that scale factor=1 for a larger or smaller ellipsoid than GRS.
