ArcGIS Pro

What's New in the Spatial Analyst Distance Toolset in Pro 2.5

The Spatial Analyst team has made some important improvements in the Distance toolset for ArcGIS Pro 2.5.
We have significantly improved the precision of the geoprocessing tools in the Distance toolset and the corresponding raster functions. We have also provided additional capabilities which let you answer some new “how much/how many/how far” questions.

These improvements allows us to remove the strong distinction between the “Euclidean” and “Cost” Distance tools. The number of tools has been reduced from 17 to 6. The new user experience is now simplified to:

  1. Enter Sources
  2. Do you have barriers?
  3. Do you have a surface?
  4. Do you have a cost surface?
  5. Do you need to consider direction of motion and other source characteristics?

The new toolset organization is shown in Figure 1 below. The new tools are:

Figure 1 - Organization of the Distance toolset at ArcGIS Pro 2.5

The original Distance tools that you are familiar with are available in the Legacy sub-toolset.
The Spatial Analyst team recommends that you move your Euclidean and Cost Distance analysis work to the new tools.

Why should I use the new tools?

At ArcGIS Pro 2.5, we’ve uniformly adopted a new algorithm [4] for all distance mapping tools. This algorithm removes the distortion in outputs caused by using a network model of cell connectivity, as shown in Figure 2 on the left below. This has the following benefits:

Figure 2- Output of legacy Cost Distance tool using a constant cost input with barriers encoded as ndoata, along with output of legacy Cost Path As Polyline tool.
Figure 3 – Output of new Distance Accumulation tool with barrier inputs, along with output of new Optimal Path As Line tool.

At ArcGIS Pro 2.4, we introduced this improvement in a limited context: Euclidean distance with barriers. ArcGIS Pro 2.5 generalizes and extends this work.

What else is new?

Along with this fundamental change to the behavior of the cost distance algorithm, several additional capabilities have been added.

Which geoprocessing tools and raster functions should I start using?

For each legacy distance tool that you are currently using, table 1 shows you which new tool to switch to.

Table 1 – Upgrade guide for distance mapping tools
If you used Then switch to
Cost Distance Distance Accumulation
Cost Backlink Distance Accumulation with back direction output
Cost Allocation Distance Allocation
Euclidean Distance Distance Accumulation
Euclidean Allocation Distance Allocation
Euclidean Direction Distance Accumulation with source direction output
Path Distance Distance Accumulation
Path Backlink Distance Accumulation with back direction output
Path Allocation Distance Allocation
Cost Path Optimal Path As Raster
Cost Path As Polyline Optimal Path As Line
Cost Connectivity Optimal Region Connections

For new workflows, the back direction raster replaces the back link raster. The Optimal Path as Raster/Line tools no longer accept a backlink raster as input (they will still accept a flow direction raster). If you were using the backlink raster in a workflow that involves something other than the Cost Path or Cost Path As Polyline tools, then the Spatial Analyst team would like to hear from you. The Euclidean distance blog mentioned above discusses the backdirection raster in more detail.

If you used the maximum_distance parameter in the legacy cost tools, you should now use the Maximum Accumulation parameter in the “Characteristics of the sources” parameter group.

These are the new raster functions.

The functions have the same parameters as the tools, with an important difference: the functions optionally emit multi-band outputs. This avoids having to run these (possibly expensive) global functions twice to get closely related outputs (like accumulated cost surface and back direction raster).

Also, the Least Cost Path raster function has been updated to use the Distance Accumulation tool behind the scenes.

Finally, although not directly related to these enhancements, your advanced use of the Distance toolset needs to consider if direction of motion between sources and destinations is important. If so, please check out this blog on Creating accumulative cost surfaces using directional data. .

References

1) Goodchild, M. F. (1977). An Evaluation of Lattice Solutions to the Problem of Corridor Location. Environment and Planning A. 9. 727-738. 10.1068/a090727.
2) Sethian, J. A. (1997). Tracking Interfaces with Level Sets: An “act of violence” helps solve evolving interface problems in geometry, fluid mechanics, robotic navigation and materials sciences. American Scientist Vol. 85, No. 3 (MAY-JUNE 1997), pp. 254-263
3) Sethian, J. A. (1999). Level set methods and fast marching methods: evolving interfaces in computational geometry, fluid mechanics, computer vision, and materials science (2nd edition). Cambridge University Press.
4) Zhao, H. (2005). A fast sweeping method for eikonal equations. Mathematics of computation, 74(250), 603-627.

About the author

Jim is a senior software engineer in the raster analysis group at esri. He has a Bachelor of Science degree in Computer Science from Hope College in Holland, Michigan, and a Master of Science degree in Computer Science from Rensselaer Polytechnic Institute in Troy, New York. He has been with esri 28 years and has had an office in almost every building, on almost every floor, of the esri campus.

Next Article

4 things to know about Esri's new Storyteller user type

Read this article