Coordinates System

[Anonymous]

Hi All,

I think my question is for people who are living in WA, or someone know much about US coordinate systems:

What is World Geodetic System of 1984 and how's difference between it and NAD83 Washtington state planes?

Thank you,

[ccoles]

WGS-84 is a coordinate system designed for world-wide navigation. It takes into account the spherical nature of the planet. The coordinates are given in deg-min-sec format, common with polar coordinates, for latitude and longitude.

The Washington state plane system is a square coordinate system specific for the state of Washington. It supposes a flat plane across the face of the state. Northing and Easting are given in feet, based on straight line-of-sight distance.

Why do you ask?

[Anonymous]

WGS84 and NAD83 are for most intents and purposes, one and the same unless you are doing some very high precision geodetic work. Academics will shout me down for saying that...

[antoniovinci]

---

ccoles wrote:

WGS-84 is a coordinate system... 

---

Well, I would say that WGS-84 is definetely NOT a coordinate system.

It's a datum (like Nad83) i.e. a 3D model of Earth, to which several coords' systems refer, e.g. LL84.

[ccoles]

okay, how about this nice explaination?

http://www.hypack.com/new/portals/1/PDF/sb/10_08/WGS84-NAD83.pdf

[Anonymous]

Yup, good non-technical explanation. State plane coordinate system is the projection of the curved (spherical) datum onto a flat surface for mapping. You can have state plane coordinates in NAD83 or WGS84, or a myriad of other different datums.

Therefore the need for metadata when providing coordinates that one should specify the datum, projection, and unit (intl ft, survey ft, meters).

[ccoles]

The original question was based on coordinate systems, not datum references. I've always associated WGS-84 datum to reference Lat and Long coordinates, and NAD83 to State Plane Coordinates. BTW, WGS-84 datum is not a selection within the state plane coordinate systems within Civil 3D. LL can select  NAD83 or WGS-84 however.

[Anonymous]

Just to be clear, datums and projections are independent. Any projection can be associated with any datum.

Typically, WGS84 is only used as a datum with no map projection and is expressed in either lat/long or ECEF Cartesian system.

NAD83 can also be expressed in MGRS, UTM, or any number of projections, as well as lat/long as a pure datum. There are also three or four different flavors of NAD83 depending which province or state you are in.

See http://www.maptoaster.com/maptoaster-topo-nz/articles/projection/datum-projection.html for a general primer - I quickly glanced through and it seems to be fairly accurate.

Also, for us Canadians, see ftp://ftp.geod.nrcan.gc.ca/pub/GSD/craymer/pubs/nad83_hydroscan2006.pdf

For Americans, you can refer to http://www.ngs.noaa.gov/faq.shtml

[Sinc]

There's still a lot of amiguities in terminology in all this stuff.  It's still a relatively new area of study, and things in this area are changing rapidly, so that's understandable.  As time goes on, hopefully we'll get more consistent in naming.

As of right now, there are two different things people consider to be a "coordinate system".  One of them is better-called a "geodetic coordinate system", and operates off of radial angles and distances from the origin (like a Polar Coordinate system in 2D).  The typical example of this is the Lat-Long coordinate systems, where you specify a Lat-Long-Elevation triplet of values.  However, even a coordinate system like this (like LL83) is dependent on a datum.

The "Cliff's Notes" version:  We start with an Ellipsoid  This is a sphere that is flattened by a specific amount.  You can also think of this as an oval that is rotated in 3D, the same way you can rotate a circle in 3D to create a sphere.  Then we hook that Ellipsoid to the Earth in some pre-defined way, in order to create a Datum.

Some of the Datums and Coordinate Systems are usually mentioned by the same name.  For example, there's a "WGS84" datum.  There's also also something that people often call a "WGS84 Coordinate system", more-appropriately called an "LL84" Coordinate System, which is basically Lat-Long-Elev on the WGS84 datum.  This essentially happens because we start with the WGS84 Ellipsoid, and then hook it to the center of the Earth, with the Y-axis pointed toward the Greenwich Meridian and the Z-axis pointed toward the rotational North Pole.  WGS84 Ellipsoid + centered on the Earth = WGS84 Coodinate System (aka LL84).

Usually, though, when we say "Coordinate System", we're referring to a grid projection onto a datum.  In other words, we usually use "Coordinate System" to refer to something that has Northing,Easting,Elev instead of Lat,Long,Elev.  You can argue a "geodetic coordinate system" like LL84 is still a "coordinate system", but very-common usage will have "coordinate system" used only to refer to systems that use Northing/Easting, i.e. more like (the 3D equivalent of) a Cartesian Coordinate system as opposed to a Polar Coordinate System.

All of this terminology is currently being used in rather contradictory fashion, which is not a good thing for the industry as a whole.  Eventually, the industry will probably settle on a "standard convention".  But that hasn't happened yet, which is why so many problems hinge on terminology.  In the 2D world, we can use "Cartesian Coordinate System" or "Polar Coordinate System", and there is absolutely no confusion.  We still do not have the same level of consensus in terminology, industry-wide and global-wide, when it comes to 3D work.

Usually these days, when we're talking about a "georeferenced Coordinate System", we mean a Cartesian grid placed upon a projection from a specific datum.  In other words, we first need a datum (an Ellipsoid hooked to the Earth in a special way).  An example of a Datum would be either NAD83, WGS84, NAD29, etc.  Then we need a Projection, which converts that Datum to a 2D-representation.  Then we throw a 2D grid on that 2D representation, and we finally have a "Georeferenced Coordinate System".  I like to use the term "Georeferenced Coordinate System", to differentiate it from a "Geodetic Coordinate System", or lat/long/elev coordinate system like LL84.

Not sure if all that adds to clarity, or adds to confusion.  I have a paper that goes into some aspects of this in more detail here:

http://www.ejsurveying.com/Articles/Working_with_Grid_Coordinates.pdf

[Anonymous]

Sinc does it again.

[antoniovinci]

---

ccoles wrote:

WGS-84 is a coordinate system
...
I've always associated WGS-84 datum ... BTW, WGS-84 datum

---

I'm glad that finally you understood the difference between datum and coordinate systems, sir.

P.S.

Great clinic, Sinc, I appreciate it.

[Anonymous]

I used to teach this at the college level and now I'm even getting confused.

One needs to avoid talking about "coordinate systems" as a coordinate system is any reference system that describes 3-D space, either as a rectangular or polar system (xyz, N,E,ELEV, lat, long, h) and has an origin and definition to orient itself.

If curvature matters, an ellipse is chosen to fit the area of work that describes the shape of the earth . The ellipse is revolved on its axis and one derives an ellipsoid (and it can be local (like Clarke1866/GRS80) or geocentric (like WGS84)). That ellipsoid is given an origin (and the axis of revolution is defined) and now you have a datum (local datums like NAD27/NAD83 or world datums like WGS84) which Sinc refers to as geodetic coordinate system, although it could either be expressed with polar or rectangular coordinates as one has defined the latitude of origin, longitude of origin and which way is up.

Then one tries to project from the curved (datum) surface to a flat surface and that results in a "projection" as one is "projecting" to derive a flat plane on which to work (as plane trig math is much easier than spherical trig).

If curvature is not a concern and the work area is small, one arbritarily chooses an origin, assigns it coordinates, picks which way is north, chooses an arbritrary elevation, and away they go. This is a "local" coordinate system. If one must tie in a local system with other projects, one must define the relationship between their local coordinates and the other projects, typically via a horizontal and vertical transformation (translation, rotation, and scale typically (and sometimes time if one takes in tectonic effects)).

So one has coordinates N:6131735 E:539560 Elev 631.25

One must ask, which Datum? Which projection? Which unit?

The proper answer would be something like NAD83 (CSRS) UTM Zone 12 North in meters and decimals thereof, with elevations AMSL referenced to Canadian vertical datum 1928.

By the way, all this talk of datums - were about horizontal datums. Vertical datums are entirely another beast.

[Sinc]

---

@Anonymous wrote:

One needs to avoid talking about "coordinate systems" as a coordinate system is any reference system that describes 3-D space, either as a rectangular or polar system (xyz, N,E,ELEV, lat, long, h) and has an origin and definition to orient itself.

---

 Well, but that's what actually happens...  People use "coordinate system" to mean a Geodetic Coordinate System like LL84, or a projection-based Coordinate System like a UTM panel, or a local ground-based coordinate system.  They are three different beasts, so it's not helpful that we don't have any industry-wide clear terminology to differentiate between the three systems.

And I actually would not call a "Datum" a "Geodetic Coordinate System", although the difference between them is small.  A "datum" is simply a specific ellipsoid fixed to the Earth in a certain way.  WGS84 is a datum.  LL84 would be a coordinate system, based on the WGS84 datum.  But the definition of WGS84 includes a specific orientation, so a given point will have identical Lat/Long values in ANY coordinate system based on WGS84.  So as soon as we say "WGS84", we've defined the Lat and Long.  But strictly speaking, WGS84 is NOT a coordinate system, even though it's oriented and fixed in a way that defines its Lat and Long.  I think we'd start getting really confused if we said a "datum" is the same thing as a "geodetic coordinate system".

Fun stuff, eh? 

[Anonymous]

Yep, then try and get the concept of grid distances vs ground distances across...

Then get into heights above the ellipsoid versus geoid and watch the eyes start to cross.

[Sinc]

I taught a class last year at AU where I did exactly that...  I think wfb is still looking at me with eyes crossed... 

[wfberry]

Sinc-man:

Talking about me in public again?  Let's see, about November "something", I'll "comma" looking!

<G>

Bill

BTW, that document you posted earlier is definitely a remedy for about any sleeping disorder.