1

When buffering a point, the result is inaccurate to the basemap.

I specify a 200 foot buffer around a point however the resulting buffer has a distance from the point of about 260 feet. Both the measure distance tool and the scale bar on the layout view agree that the 200 foot buffer is indeed greater than 200 feet.

The .MXD uses a coordinate system of WGS84 Web Mercator Auxiliary Sphere as I am using some of ESRIs available online datasets, however I have added other data in different projections.

These data include:

  • State-provided data (multiple point shapefiles) in State Plane UTM Zone 13N
  • Field collected data in NAD27
  • Federally provided data in GCS North America 1983 (Datum NAD83)

I am using the buffer tool in the Analysis toolbox to buffer the point collected in the field (NAD27)

I have tried re-projecting both my location and the resulting buffer however this only results in moving the shapefiles relative to my basemap, with no resultant change is size.

How can I correct this issue?

Note: The displayed NAD27 Point is positioned accurately to the WGS84 Web Mercator basemap. (projected "on the fly" as it was added to the .MXD)

Steps:

  • Open .MXD

  • Add Base-layer from ESRI Online database (Aerial Imagery WGS84)

  • Add Data in the form of .csv; this is the field gathered location of interest in NAD27 Decimal Degrees

  • Export xy data to shapefile keeping it in NAD27

  • add exported data to map and remove .csv event

  • Analysis Tools -> Buffer

  • Buffer location .shp

  • Measure distance from point to buffer boundary = inaccurate

This happens because the buffer is in NAD27 as it is applied to the point. the basemap layer is in WGS Web Mercator Auxiliary Sphere and the measure distance tool is measuring the distance on the basemap. If I change the map properties to display in NAD27 I encounter distortion in the effect of oblation of my once round buffers. I also encounter a "stretched" appearance to the basemap and the shapefiles (excluding points of course). This does however plot the buffer boundaries in the correct location with correct distances (verified with on-the-ground references in the basemap) - but they are ovals.

Is there a way to preserve the roundness of my buffers yet still have them displayed accurately without changing the basemap projection (and distorting the image)

  • 1
    What tool are you using to measure the buffer? Web Mercator is the least accurate. Movement, not change in size would be expected with incorrect transform. You may be expecting incorrect results. – Vince Jan 14 '16 at 21:56
  • 1
    Would you be able to list the precise steps, from starting ArcMap, that lead you to concluding that your buffer distances are incorrect, please? – PolyGeo Jan 14 '16 at 22:27
  • 1
    Mercator distance measurements are not accurate. Just look at the relative sizes of Greenland and South America on a globe. – Vince Jan 15 '16 at 0:28
  • 2
    The buffer IS accurate. Your measurement is inaccurate. – Vince Jan 15 '16 at 0:34
  • 1
    Please use the edit button beneath your question to revise it with your precise steps. Potential answerers may or may not read comments but they usually read and re-read the question body. – PolyGeo Jan 15 '16 at 1:15
2

Try changing the buffer method from Planar to Geodesic Arguably, in such small scale features this shouldn't change much, but stranger things have happened before.

Here's a case where I buffered some points in a geographic CRS with a planar buffer of 250 meters and displayed on a projected CRS. Using a geodesic method solved this issue for me.

enter image description here

  • I appreciate the response however this has not solved the problem for me. Stranger things have indeed happened though. – c0ba1t Jan 15 '16 at 0:33
1

Answer:

When adding XY data in NAD27 Decimal Degree Coordinates (from a .csv file

  • The buffer distances will be inaccurate to the basemap if the basemap is in a different GCS (Geographic Coordinate System) or PCS (Projected Coordinate System)

Solution:

  • display XY data from .csv file
  • export data by right-click on "filename.csv Events"
  • select data then export data
  • instead of exporting with the source data coordinate system, use the same coordinate system as the Data Frame

Now, 100 foot buffers will measure to be 100 feet in the measure tool, and the ground truthing distances will line up in the aerial imagery.

statement of problem:

Buffer tool used on NAD27 data point generates buffer in NAD27 GCS. This GCS has different distortion than the basemap data layer presented in a different coordinate system, therefor, yielding a buffer that appears to be inaccurate. In fact, the buffer is accurate but will only accurately correlate with NAD27 basemap imagery.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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