5

I have an older map that I didn't create. I'm just trying to get everything to line up on it, and having a hard time.

The data frame is in NAD_1927_UTM_Zone_12N, as are the point shapefiles (derived from a .dbf table that I had to repair the data source to) and one line shapefile. All the other data is in StatePlane (either 1927 or 1983) and that's what shows up in the layout view (aerial, roads, buildings, polygons, etc.)

I've changed the points (via right click, data, repair data source, browsing for the data and then editing "Coordinate System of Input Coordinates" to StatePlane, but that doesn't work. The point shapefiles aren't visible anywhere, even if I zoom to the layer or go to Full Extent.

When the points are in the UTM projection, they're at a ridiculous scale (1:1,000,000) and they're all in a straight line instead of spread out like the old PDF version of the map I have.

I'm trying to make some simple edits, but I can't fix the bigger issue of shapefiles not lining up.

Although I don't know if this is the problem, when I open the map I get a notice that says, "This document may contain VBA code but you don't have a VBA license." I would just assume that scripts don't run, not that that would affect projections...

Any insight would be greatly appreciated. Thanks.

edit: No, the pdf is uneditable. I did realize the XY were being mapped to the same field, which was wrong (the XY data is in northings and eastings), but correcting that didn't fix the problem. Unfortunately, whenever I try to add data, browse for projections and the like, ArcMap crashes which is an unrelated server problem. I could copy the data to my own folder but there are a lot of files and some of them are linked and from other folders and ...still of course in different projections. Although I fixed the VBA license problem, so that has nothing to do with it.

I tried doing Data Management > Projections & Transformations > Features > Project and each time (regardless of which projection I choose) the new points project exactly where they currently are. I'm currently trying suggestion #1 at the moment.

3
  • is the pdf a 'geopdf' that you could convert to .shp or .gdb?
    – Mapperz
    May 3, 2011 at 21:50
  • Are you sure you're mapping the x & y fields correctly? A straight line would indicate you're mapping both to the same field, regardless of what else is happening. May 4, 2011 at 4:56
  • 1
    It's very useful to post a sample point from the problematic dataset. Happily, you mentioned that the area of interest is somewhere in NAD27 UTM zone 12N. Would you narrow that down (what state plane zone?)? Could the data be in decimal degrees?
    – mkennedy
    May 4, 2011 at 21:32

4 Answers 4

6

I will disregard all of the facts from your question.
Because often when I KNOW something is in a projection and I still have problems it ends up being the known projections that needed to be changed.

Here is my first thought. It would be fairly rare (not impossible) to find data in a dbf with UTM coordinates.
Most often the xy in a dbf are there because nobody has converted from lat lon. (could be WGS84 but more likely nad27).

Next: the method I use is to leave all that is confusing me and get something I know. Get one layer that I have no question about (preferably not associated with the current problem) and put it in it's own new mxd.
Right click the layer to check the crs.
Go and get one of the other layers that overlays with the new known data.
If it overlays you just verified that it is using the correct coord system.

If it doesn't then you have to use deductive reasoning looking at scale, juxtpostion, and generally "are they using the same numbering system"? -1,-1 +1,+1 +1,-1 -1,+1.

If the second is the case what you can do is to temp rename the prj file so that it is not associated with the shape file any longer.

Add that data into the mxd again and start guessing which system it actually is in. To do this right click on the layers at the top of TOC (table of contents). go to the properties (this is the document properties) and select the coordinate system tab.

select some of the "alleged" crs and when or if you find the correct one all your data will line up. (this is because you are using "on-the-fly projection" and the unknown data "is not" being transformed [because there is no prj associated with it]).

It is a difficult task that can be very confusing and is at least a little dangerous to the data. Take your time and walk through it carefully.

2

First make sure that on the fly reprojection is enabled (assuming that is still an option in Arc10). If so, it may be beneficial to first reproject your point shapefile. There are many ways to accomplish this. I would recommend ogr2ogr available for free through FWTools. The command would be something like this:
ogr2ogr -s_srs EPSG:26712 -t_srs EPSG:2229 outfile.shp infile.shp

You'll have to change those EPSG codes as appropriate.

In ArcGIS Desktop you can reproject features using the Project tool under Data Management> Projections and Transformation> Feature in ArcToolbox.

1
  • 1
    @Paul- actually the "Define Projection" tool should be used only if your shapefile does not already have an existing (or all-together wrong) projection file associated with it. If you want to actually reproject the data from one coorinate system to another, you should use the "Project" tool. May 3, 2011 at 22:13
2

I've changed the points (via right click, data, repair data source, browsing for the data and then editing "Coordinate System of Input Coordinates" to StatePlane, but that doesn't work. The point shapefiles aren't visible anywhere, even if I zoom to the layer or go to Full Extent.

Rather than changing the Coordinate System, I believe you need to reproject the data so that they are both using the same system. Have a look at the help documentation about projections, and the tool you need Project is under the toolbox Data Management.

3
  • If the data doesn't overlay with its current coordinate system definition, projecting it to a new coordinate system is not going help. ArcMap will project data on-the-fly to the data frame's coordinate system. If data doesn't show up, there's something wrong with the layer's coordinate system.
    – mkennedy
    May 4, 2011 at 21:29
  • My point is that changing the projection within the shapefile, does not re-project the data.
    – djq
    May 5, 2011 at 14:42
  • Ah, I see what you're saying. It seems to me that the underlaying problem is that the data's coordinate system is incorrect. I was trying to point out that if GIS_User2736 doesn't get that corrected, Project won't work.
    – mkennedy
    May 5, 2011 at 20:29
1

I've dealt with a lot of this do to hole-in-the-wall survey companies. Try every projection you can think of..including changing the unit of measure on the projection..including the unit of measure on the projection(repeated intentionally..=) ). I've had to change the unit of measure on State Plane data to be in meters and UTM data unit of measure to be in one of the feet..there's like 10 different type of feet.

Write each variation of projection/unit of measure down that you try on a post-it until you figure it out. Hope this helps.

To get started.. open a new mxd, add some imagery(or some other layer that helps you know you're in the right spot) and then add your data, then just change your data frame using the above tip and this may help. Good luck!

P.S. I've had a .dbf with xy(utm/state plane coords) instead of lat/longs.

Not the answer you're looking for? Browse other questions tagged or ask your own question.