Latest Contributions by MelitaKennedy

Michael, You can check the parameter values in the geographic_transformations.pdf file that's in the Documentation folder of your ArcGIS installation. The combination you listed actually works out to a null transformation. Both use the same parameter values. I often put in "equivalents." The parameters were originally for ITRF00 / WGS84 to NAD83 HARN or CORS96, for instance, but knowing that people have labeled their data with NAD83 instead, I used the same parameters. Sometimes it's also unclear whether the transformation (usually from the NGS CORS website) should be used for Alaska as well as the lower 48. Melita ... View more

The projection portion isn't more accurate, but the geographic coordinate reference system (GCS, GeoCRS) is. The implication is that the latitude-longitude values that were projected to zone 4 (or if you unprojected the data back to lat-lon) are on the 2011 re-adjustment of NAD 1983. A lot of data and its metadata/coordinate system information is confused. Many people use "NAD 1983" which should technically be used only for the 1986 realization for data that's really on a later re-adjustment like NSRS2007 or 2011 or a HARN re-adjustment. In ArcGIS 10.4.0 (out soon), we're adding support for Geocon v1, which does have some transformations between HARN and NSRS2007 and NSRS2007 and 2011. We're hoping to get v2 into a later release. Melita ... View more

Hi Jack, The first step is to get some test points into "HARN"/HPGN. You could use NGS's NADCON program. I think you have to download it, as the online calculator doesn't seem to support NAD83 to HARN. You could also extract a few points and reproject them using the Project Tool and just set the appropriate HARN grid for the area. For the HARN to NSRS2007 to 2011 transformations, you'll have to use NGS's GEOCON v2.0 program. It's...a bit complicated. So first follow the "show all realizations" link on the left and look up your state. Then look through the user guide to find out the what's allowed in the "free format" (or use Blue Book if you know it). For NC, it looks like you'll do an HARN to FBN, then FBN to 2011. I think the biggest change will be from NAD83 to HARN, and the others will be smaller. ... View more

I have only a tiny bit of knowledge about CityEngine, but some general ideas about coordinate systems. Does the DEM have a defined coordinate system? There are ways that it could overlay in ArcMap without a coordinate system, but not work in CityEngine. What is the DEM's coordinate system? Melita ... View more

Hi Jack, Maybe nothing, depending on the accuracy of your data. Is the other data (using NAD 1983 like the data frame/map) really on NAD 1983 (1986) or could it be a mix of HARN, NSRS2007, 2011, CORS96, etc.? In 10.4.0, we added support for GEOCON v1 grids/transformations, which convert between HARN and NSRS2007 and between NSRS2007 and 2011. That also means a lot of concatenation of transformations you have NAD27 or NAD83 in the mix. Right now, ArcMap only supports concatenating 2 transformations, you could convert from HARN to 2011, but not the original NAD83 to 2011. Prior to 10.4.0, ArcGIS will piece together two transformations to convert from NAD83 to 2011, usually via ITRF00 or ITRF08. Most of those transformations are already approximations (where I used the same parameters with different GCS) that I put in, to remove most of the offsets between the two systems. For instance, once suggestion in ArcMap is WGS_1984_(ITRF00)_To_NAD_1983 + WGS_1984_(ITRF00)_To_NAD_1983_2011 Those are both the same transformation, just with different NAD 1983 realizations. It was originally designed for approximately CORS96/HARN data, but if you've a NAD 1983-defined data that's really mixed... ... View more

To expand on Chris' answer a little bit, a world file describes the cell size, any rotations, and the coordinates of the lower left corner of the raster. Here's an example from Wikipedia: 32.0 0.0 0.0 -32.0 691200.0 4576000.0 There's no information on the coordinate system, and it could be any UTM zone, a bunch of Gauss-Kruger zones, and probably a few other things. There's no way to know, unless the coordinate system is also defined in the TIFF, you have knowledge of the coordinate system, or can pick a likely coordinate system based on your knowledge of the location that the data represents. Melita ... View more

The difference between the false easting in your listed CRS that's using US survey feet, and the meter version of the zone is 456 km. Could the LIDAR have the wrong coordinate system defined? Does it look like it could be sized wrong as well as offset? Melita ... View more

You might want to call up your regional office (if in the US), otherwise the distributor and ask for an account rep. They may be able to connect you to the right person. Melita ... View more

If you use a PCS, it should be densifying relative to the plane. A GCS should densify the line based on the geodesic line between the two points. ... View more

Make sure the input data has an ellipsoid-based VCS defined, Check on the vertical option. Make sure the output VCS is also ellipsoidal. ... View more

Er, what? If you performed an equation-based transformation (3 or 7 parameter) and had ellipsoidal heights for the input data, then you would see small differences in the lat/lon values depending on whether you converted the ellipsoidal heights to the output gcs's ellipsoid. Otherwise, it would just be whatever the horizontal transformation does. ... View more

However, now that 10.4.0 is with the release team (AKA no more development changes), there are some vertical transformations possible in ArcGIS for Desktop 10.4.0. It's currently for vector data only. The Project tool has a check box to enable vertical transformations. The input data must have vertical data and a vertical coordinate system defined. When you set the output coordinate systems, you can now set both the horizontal (geographic/projected) and vertical coordinate systems. We look up a combination horizontal and vertical transformation using a table, hvtdefaults.json, which is in the pedata folder of the install. It's known to be incomplete. If it comes up with transformations that you don't agree with, you can either edit the json file or try to perform the projection/transformation in multiple steps. Vertical transformation data tends to be in grid files which are large. For 10.4.0, there's now a separate ArcGIS Coordinate Systems Data setup which can optionally install the US geoid2012 files and VERTCON files (NGVD29 / NAVD88), and the world EGM84 and EGM2008 1'x1'  or 2.5'x2.5' geoid models. There's some supporting functionality in ArcObjects too. Here's a picture: Melita ... View more

Okay, weirdly enough, for those transformation parameters, I added ITRF_2000_To_   NAD 1983 HARN | CORS96 | 2011 Why did I omit NSRS2007? I dunno. Trying to reconstruct my thinking, I believe I decided that I shouldn't add ITRF2000 to NAD 1983 because the transformation wasn't designed for that, but for CORS96 and, to a lesser degree, HARN. I did put it in with WGS84 as an approximation to ITRF2000 and converting to NAD 1983, which is another approximation. In your particular case, you're starting out with NAD 1983, so the "most accurate" transformation or pair of transformations is the one you have, rather than a direct one which as zero parameters, so even more of an approximation. Technically, I would move NAD 1983 (1986) to HARN (NADCON/HARN transformation), assume it's relatively close to CORS96 (and redefine it as NAD 1983 CORS96), then use ITRF_2000_To_NAD_1983_CORS96 to convert to ITRF00. Sorry, rather than getting simpler, geodetic transformations are just getting more complicated with time! Melita ... View more