Low Distortion Projections

[TTaylor]

Dave,

First off I want to say that Low Distortion Projections are not No Distortion Projections.

As you know I worked on many iconic projects in The SF Bay Area as well as hundreds of others. The iconic ones were The Cypress Reconstruction Project, The New Bay Bridge, and The Park Presidio Project in SF, the last two in a geodetic capacity and the first one as being a member of the main survey crew laying out the project.

All of the projects I worked on used a two parallel Lambert Projection, in the beginning that was NAD27 and then NAD83. At no point in time did we have any problems using a combined grid factors to move between grid and ground.

The Cypress project went up 880 to a new alignment then into The Maze at The Bay Bridge toll plaza and also out along 80 east for a ways. The control (NAD27) over the years was pushed west along 80 into the maze and north along 880 into the maze. Guess what??? The error between those at the meeting point in the maze was several feet. Go figure.

Well, we ended having two sets of coordinates: one for Right of Way and one for the project construction. We figured that as long as the project stayed inside the RW we can do whatever we want. The error for the construction coordinates was then spread thoughout the project making sure the maze area was precise and burying the error when the project was at grade.

The most critical time came when we layed out the bearing plates for the steel girders on top of the bent caps. Those had to be very precise to make sure the bolts on the bearing plates lines up with the holes in the steel girder.

We did this with no problem.

At the time we didn’t have the new fangled survey controllers of today. So, we put the CGF into the instrument in the form of a ppm along with atmospheric ppm info. The CGF is basically a ppm so we subtracted the CGF from one million and added that the atmospheric ppm to input into the instrument. Voila…. Grid converted to ground or verse visa.

OK, I get that this is a shaggy dog story so I’ll get to the point.

The new NSRS2022 projection for CA is a one parallel Lambert Projection where there is still CGFs regardless of what some at NGS might say. So, a tangent plane and not a secant plane. That means the ellipsoid (GRS 80 – the same as NAD83) pulls away from the tangent plane faster than it does with a secant plane. Hence one reason some may want Low Distortion Projections, which is a plane that approximates the general slope of the area covered. The other, of course, is because of elevation.

Again, it is NOT a no distortion projection which means for large projects you still have a correction factor for ground to grid. If your project is a rural road or a drainage ditch in the boondocks then it probably doesn’t matter. Same with making a Thomas Brothers map or a GIS.

The good thing about NSRS2022 is that one of the main reasons for creating it was to align with the International Terrestrial Reference Frame (ITRF) instead of assuming the North American Plate was fixed like they did with NAD83. Also, the representation of the NSRS2022 will be the GNSS satellites as opposed to the national CORS Network being the representation of NAD83 where the error at those stations was virtually zero.

Anywho…. I see a potential bigger problem with NGS “deprecating” the US Survey Foot in favor of the International Foot. In a previous webinar one of the slides showed that 42 states use the US Survey Foot mandated by law or some other formal notice. 7 mandated the International Foot and the rest (territories included) allowed for either or are silent on the issue.

So, at this recent webinar I asked the question ‘Why doesn’t NGS support the metric system along with a tool to convert to either the US Survey Foot or the International Foot depending on what the end user needs.” The response was, “NGS has worked in the Metric System for years and we are only supporting a conversion to feet at the request of the states, our customers.” I left it at that (I didn’t want to get into a chasing the tail type of exchange during the webinar). However, I couldn’t help wondering that if they are only providing a conversion to feet based on customer feedback, why not convert to the US Survey Foot instead of the International Foot as most of their customers work in that.

Well, when you have coordinates that are in the (2,000,000 500,000) meters range the difference between converting to the US Survey Foot and the International Foot is 2ppm or approximately 12-13 feet.

That means that people will either need to be very careful with their legacy documents such as as-builts and rw/property maps along with GIS systems based on the US Survey Foot or ignore NGS and continue using the US Survey Foot on future work.

2 cents

TT, PLS

[hellsangle]

Professor Taylor . . .

Excellent real-world example and excellent explanation!

(It is sad we can't embrace the metric system . . . I speculate the dumbest common denominator, the contractor, whined and took CalTrans off metrics.)

Thanks again, Professor!

Appreciatively yours. Happy Holidays!

Crazy Phil

[Steve Martin]

To throw 2 cents in:

SPCS83 as a legacy system will continue to be in terms of US survey foot, if you choose to use feet instead of meters.

SPCS2022 on the other hand will use the official definition of the foot, commonly called the International foot, if you choose to use feet units instead of meters.

See my short article on the Deprecation of the US Survey Foot in the California Surveyor when it comes out.

[Steve Martin]

The new NSRS2022 projection for CA is a one parallel Lambert Projection where there is still CGFs regardless of what some at NGS might say. So, a tangent plane and not a secant plane. That means the ellipsoid (GRS 80 – the same as NAD83) pulls away from the tangent plane faster than it does with a secant plane.

According to NGS, any 2 parallel LCC projection can be recast as a 1 parallel projection. They are mathematically the same. The reason NGS is going to 1 parallel LCC projections is so the number of parameters is the same as Mercator projections. It is a programming simplification.

NGS is not necessarily going to tangent plane LCC projections. They are moving the projection surface up or down and changing the projection axis to minimize the overall combined scale in the zone. Some projections will be non-intersecting. Also, they have algorithms to prioritize minimizing the overall combined scale factor in areas where the most people live. Computational abilities have come a long way since they original State Plane Coordinate Zones were defined in 1934.

Michael Dennis has some good presentations illustrating how these zone changes for SPCS2022 reduce overall combined scale factor.

On the NGS Presentations Library https://geodesy.noaa.gov/web/science_ed ... s_library/
look for "Grids for the Future: A New Approach for Designing State Plane Coordinate System Zones" dated 10-3-2022

[DWoolley]

TTaylor:

I am sorry to see your last two posts disappear. I read them over the holiday, watched a two more videos, and was preparing a response i.e. notes on my phone app during a lull in the holiday gatherings. I appreciated the technical aspects of the posts.

Agreed, the world is not flat.

DWoolley

[TTaylor]

Mr. Wolley,

I took them down as I subsequently realized the new projections are not a Single parallel or double parallel Lambert projection as originally touted but a somewhat modified projection based on a single parallel.

Ha.. My old school understanding of state plane coordinates needed updating to the new approach.

I'm still wary of this new approaches implementation.

[TTaylor]

Mr. Wolley,

I took them down as I subsequently realized the new projections are not a Single parallel or double parallel Lambert projection as originally touted but a somewhat modified projection based on a single parallel.

Ha.. My old school understanding of state plane coordinates needed updating to the new approach.

I'm still wary of this new approaches implementation.