If I wanted to project latitude & longitude data for the contiguous United States (the United States excluding Alaska and Hawaii), which projection would I use? I prefer more accurate distances followed by shapes
asked Apr 6, 2015 at 15:53
-
5different projections have different characteristics. What is more important: shapes, distances, or cardinal directions?– mr.adamCommented Apr 6, 2015 at 16:21
-
@mr.adam - Distance followed by shape– DenaliHardtailCommented Apr 6, 2015 at 17:21
-
4us48 otherwise known as alber usgs, USA_Contiguous_Albers_Equal_Area_Conic_USGS_version, EPSG:5070– Brad NesomCommented Apr 6, 2015 at 18:44
-
Related: gis.stackexchange.com/questions/104005– Chris WCommented Apr 6, 2015 at 21:59
Add a comment
|
4 Answers
The Albers equal area conic is the typical projection for historical USGS maps of the lower 48, it being a general-purpose low-distortion compromise for mid-latitude short and wide extents.
As a reference on map projections, I like the ESRI book Understanding Map Projections. Its first 30 pages are not unlike a short textbook, followed by ~70 pages of appendix on individual projections, their uses, strengths, weaknesses, etc.
- Albers, pg 37
- Lambert, pg 66
- equidistant conic, pg 53
-
1Good direction indeed! all the geographic coverage is illustrated visually.– LearnerCommented Aug 25, 2016 at 12:58
-
the one-sixth rule discussed in ESRIs book for albers equal area conic suggests standard parallels for the US of about 22 (26 - (49 - 26)*(1/6)) and 53 (49 + (49 - 26)*(1/6)). but I've found calculations over small areas appear to improve more from picking parameters close to the site than from picking a better projection (e.g. the more simple azimuth equidistant projection appears to work well for a single US state if the tangency point is chosen at the state centroid, better than an albers projection with these US-wide parameters) Commented Apr 29, 2019 at 6:20
ESRI has defined three projections especially for the contiguos United states. These are included in QGIS as well:
ESRI:102003 USA_Contiguous_Albers_Equal_Area_Conic
+proj=aea +lat_1=29.5 +lat_2=45.5 +lat_0=37.5 +lon_0=-96 +x_0=0 +y_0=0 +datum=NAD83 +units=m +no_defs
ESRI:102004 USA_Contiguous_Lambert_Conformal_Conic
+proj=lcc +lat_1=33 +lat_2=45 +lat_0=39 +lon_0=-96 +x_0=0 +y_0=0 +datum=NAD83 +units=m +no_defs
ESRI:102005 USA_Contiguous_Equidistant_Conic
+proj=eqdc +lat_0=39 +lon_0=-96 +lat_1=33 +lat_2=45 +x_0=0 +y_0=0 +datum=NAD83 +units=m +no_defs
So it is up to you which projection characteristics you need: equal area, equal distance or conformal.
Visit this page to see the differences: http://www.radicalcartography.net/?projectionref
-
It's not clear what you mean by "equal distance." Note that an "equidistant" projection typically gives accurate distances only to one (or sometimes two or three) fixed points on the map. Other projections may have relevant properties not shared by these, such as being cylindrical, minimizing the grid convergence, etc. The wide variety of possible choices is what led @mr.adam to ask for clarification concerning the objectives of the projection.– whuberCommented Apr 6, 2015 at 17:08
-
EPSG:102005 is referenced in two reliable sources as slightly different: +proj=eqdc +lat_0=0 +lon_0=0 +lat_1=33 +lat_2=45 +x_0=0 +y_0=0 +ellps=GRS80 +datum=NAD83 +units=m +no_defs Please see spatialreference.org/ref/esri/102005 and epsg.io/102005 Commented Feb 18, 2017 at 7:47
-
@BradHorn both links proof that those are not reliable sources. The center meridian and latitude of US projections should be inside the US, and not
lat_0=0 lon_0=0. The OGC WKT definitions of the same sites are correct.– AndreJCommented Feb 18, 2017 at 8:04 -
1I think there's a mistake. Given codes looks like ESRI codes, not EPSG. I think authority part should be fixed by stating
ESRI:instead ofEPSG:.– amaninCommented Nov 25, 2021 at 10:42
Snyder's GS48
PROJ has a Modified Stereographic of 48 U.S. (+proj=gs48) projection:
It is based on a modified-stereographic conformal projection detailed by Snyder (1987) (p.203–212). Here is the estimate of scale factors from the document:
The down-side of this projection is that it is seldom used and has limited support. For instance, to use with QGIS, a custom CRS needs to be created with a PROJ string +proj=gs48 +type=crs.
Registered projections
There are a few "CONUS" projections that are registered with EPSG. Here are some for specific datums:
If shape is important, consider a Lambert conic conformal projection, with two standard latitudes. Distances will be consistent in the vicinity of each of the standard parallels. See
- wikipedia.org/wiki/Lambert_conformal_conic_projection
- wired.com/2013/11/projection-lambert-conformal-conic
- georeference.org/doc/lambert_conformal_conic
- mathworld.wolfram.com/LambertConformalConicProjection
You could also consider some sort of "equidistant" projection. However, distance scale will never be true everywhere; only true from one or two points (in all directions) or from one line (in a single direction).
