1

I am using a very simple sample dataset to create a Euclidean distance surface from the source point. I used the Euclidean Distance tool and also the Accumulate Distance tool. To my surprise, these two tools created different surfaces! I can't figure out why I am getting these two different surfaces:

enter image description here

Here, Euc_Distance_2 is created by the simple Distance Accumulation tool in ArcGIS Pro. The values here are not correct; mathematically, the highest distance from the Source point must be 630.71 meter [cell sized 30x30]. Can anyone explain what is going on here?

Here is the difference surface: enter image description here

4
  • Euclidean Distance is depreciated, use Distance Allocation. Given your cell size is 30m a difference of less than half a cell could be explained by cell centre vs cell edge assuming you're not using a cost surface for Distance Allocation. Is your source data in a projected coordinate system or is it geographic units? Feb 13 at 3:17
  • @MichaelStimson The data is projected into a Meter based coordinate system. ArcGIS Pro recommended using Distance Accumulation instead of Euclidean Distance, which in both cases must create the same values from the center of one cell to the center of another cell.
    – sMartines
    Feb 13 at 17:14
  • 1
    I'd suggest to minus one from other to check if there IS any difference
    – FelixIP
    Feb 13 at 18:17
  • @FelixIP I have done that, I edited the question to show the difference layer.
    – sMartines
    Feb 13 at 19:27

1 Answer 1

2

Your experiment shows that Euclidean Distance and Accumulate Distance (former Cost Distance) are 2 different things, no matter what is recommended. To test it I created 2500 points (projected system, meters) and run:

  1. Near on upper left point
  2. Eucledian distance to the same point
  3. Cost Distance - raster of 1 (I don't have SA in my PRO) with target being the same point. No barriers, vertical factors etc.

I sampled both output rasters and compared to true straight line length, computed by Near tool. Result of following query on points dataset:

abs( "NEAR_DIST"- "EUC_DIST")<0.001

enter image description here

Proves that Euclidean Distance tool computes true straight distance from cell centers to target, all 2500 pairs are within 1mm difference. In reality they are as identical as DOUBLE (from NEAR) can be when compared to FLOAT (from rasters).

This query:

abs( "NEAR_DIST"- "COST_DIST" )<0.001

Produces very different output:

enter image description here enter image description here

Numbers are the same for 148 points only.

In fact distance from every cell that is not on one of eight cardinal direction from target will differ from straight line length:

enter image description here

2
  • ArcGIS Pro has marked the "Euclidean Distance" tool as "deprecated"; and instead explained that Distance Accumulation can calculate "Straight Line" distances. That is based on my question; I have no idea how the straight-line distances are calculated in the Distance Accumulation tool!
    – sMartines
    Feb 14 at 1:44
  • 1
    Last picture shows how. It is least cost path going from cell to cell to reach destination, I.e. almost never straight
    – FelixIP
    Feb 14 at 4:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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