1 of 63

Proximity Analysis

​

Jeff Jenness

jeffj@jennessent.com (928) 607-4638

2 of 63

Main Topics

2

3 of 63

Main Topics

  • Methods to determine cell values at points

​

​

​

3

4 of 63

Main Topics

  • Methods to determine cell values at points

​

  • A few common ways to analyze data by proximity

​

​

​

​

4

5 of 63

Main Topics

  • Methods to determine cell values at points

​

  • A few common ways to analyze data by proximity
    • The Spatial Join function

​

​

​

5

6 of 63

Main Topics

  • Methods to determine cell values at points

​

  • A few common ways to analyze data by proximity
    • The Spatial Join function
    • The Near tool

​

​

​

6

7 of 63

Main Topics

  • Methods to determine cell values at points

​

  • A few common ways to analyze data by proximity
    • The Spatial Join function
    • The Near tool
    • Raster distance functions

​

​

​

​

7

8 of 63

Zonal Statistics as Table

  • Can extract the value of a raster cell at a point.

​

8

9 of 63

Zonal Statistics as Table

  • Can extract the value of a raster cell at a point.
  • Good when grouping points and calculating statistics on the group.

9

10 of 63

Zonal Statistics as Table

  • Can extract the value of a raster cell at a point.
  • Good when grouping points and calculating statistics on the group.
  • Not so good when just getting individual cell values for each point.

​

10

11 of 63

A problem with the Point Zonal Statistics analysis…

11

12 of 63

Extract Raster Cell Values to Points

  • Can do Bilinear Interpolation or Exact Values.

​

12

13 of 63

Extract Raster Cell Values to Points

  • Can do Bilinear Interpolation or Exact Values.
  • Can do multiple rasters at once!

​

13

14 of 63

Extract Raster Cell Values to Points

  • Can do Bilinear Interpolation or Exact Values.
  • Can do multiple rasters at once!
  • Can bring in all raster attribute values, if there are any.

​

14

15 of 63

15

Extract Raster Cell Values to Points

16 of 63

What are Exact Values?

16

A way to calculate the point value as the exact value of the cell it intersects with…

17 of 63

What is Bilinear Interpolation?

17

A way to calculate the point value as the weighted average of the 4 closest raster cell values…

18 of 63

What is Bilinear Interpolation?

18

Each cell value is weighted by the area of the opposite quadrant.

19 of 63

What is Bilinear Interpolation?

19

Called “Bilinear” because it is interpolated along two axes.

In this example, start by interpolating two values along Y-axis.

20 of 63

What is Bilinear Interpolation?

20

Next, interpolate final value between these two values along X-axis.

21 of 63

21

Questions:

  • Under what circumstances would Bilinear Interpolation be a more appropriate method to estimate point values?
  • When would Exact Values be more appropriate?

This is also a homework question!

22 of 63

22

Questions:

  • Under what circumstances would Bilinear Interpolation be a more appropriate method to estimate point values?
  • When would Exact Values be more appropriate?

This is also a homework question!

Hint:

  • Consider Categorical vs. Continuous data types

​

  • Categorical rasters would include
    • Land Ownership
    • Vegetation Type

​

  • Continuous rasters would include
    • Elevation
    • Slope
    • Canopy Cover

​

23 of 63

Vector Distance Methods

  • These functions calculate the straight-line distance between vector objects.

23

24 of 63

Vector Distance Methods

  • These functions calculate the straight-line distance between vector objects.
  • Say we’re interested in distance from survey points to the nearest road:

24

25 of 63

The “Spatial Join” tool

  • Vector-based Proximity Analysis Tool

25

26 of 63

The “Spatial Join” tool

  • Vector-based Proximity Analysis Tool
  • Can do lots of cool things.

26

27 of 63

The “Spatial Join” tool

  • Vector-based Proximity Analysis Tool
  • Can do lots of cool things.
    • Identify features that intersect other features.

27

28 of 63

The “Spatial Join” tool

  • Vector-based Proximity Analysis Tool
  • Can do lots of cool things.
    • Identify features that intersect other features.
    • Identify features within a distance of other features.

28

29 of 63

The “Spatial Join” tool

  • Vector-based Proximity Analysis Tool
  • Can do lots of cool things.
    • Identify features that intersect other features.
    • Identify features within a distance of other features.
    • Give you the distance to the nearest feature from another dataset.

29

30 of 63

The “Spatial Join” tool

  • Vector-based Proximity Analysis Tool
  • Can do lots of cool things.
    • Identify features that intersect other features.
    • Identify features within a distance of other features
    • Give you the distance to the nearest feature from another dataset.
    • Get all attributes of nearest feature.

30

31 of 63

The “Spatial Join” tool

  • You have joined tables before. You use the same function to do Spatial Joins.

31

32 of 63

The “Spatial Join” tool

  • You have joined tables before. You use the same function to do Spatial Joins.
  • Two ways to do it: the Add Spatial Join tool and the Spatial Join tool

32

33 of 63

The “Spatial Join” tool

  • You have joined tables before. You use the same function to do Spatial Joins.
  • Two ways to do it: the Add Spatial Join tool and the Spatial Join tool
    1. The Add Spatial Join tool creates a Join the way you would normally join tables.
      • Join is created in-memory in your ArcGIS Pro Session
      • Does not support a 1:Many relationship; can only join single nearest feature.

33

34 of 63

The “Spatial Join” tool

  • You have joined tables before. You use the same function to do Spatial Joins.
  • Two ways to do it: the Add Spatial Join tool and the Spatial Join tool
    1. The Add Spatial Join tool creates a Join the way you would normally join tables.
      • Join is created in-memory in your ArcGIS Pro Session
      • Does not support a 1:Many relationship; can only join single nearest feature.
    2. The Spatial Join tool is more powerful.
      • Creates a brand new dataset and leaves the originals unchanged.
      • Can extract information on multiple neighboring features

34

35 of 63

The “Spatial Join” tool

  • Purely vector-based, so you can’t analyze raster data with it.

35

36 of 63

The “Spatial Join” tool

  • A few things it doesn’t do…
    • It doesn’t have any option to give direction to the nearest object.

36

37 of 63

The “Spatial Join” tool

  • A few things it doesn’t do…
    • Only considers a single comparison feature class.
      • For example, you can’t find the nearest feature from multiple candidate feature classes.
      • If you want the distance to the nearest road, you can’t use this function to check multiple road feature classes simultaneously.

37

38 of 63

The “Spatial Join” tool

  • A few things it doesn’t do…
    • If you don’t specify Geodesic Distance, then distance calculated in the coordinate system and units of the target feature class.
      • This means that if your target features are in geographic coordinates, then the distance will be in “degrees” which is nonsensical!

38

39 of 63

The “Spatial Join” tool

  • Does give you full information on the nearest road.

​

39

40 of 63

The “Spatial Join” tool

  • Does give you full information on the nearest road.
  • Does give you distance to nearest road, IF:
    • Your “Match Option” is either Closest or Closest Geodesic
    • And you remember to specify a name for your new “Distance” attribute field

​

40

41 of 63

The “Near” tool

41

42 of 63

The “Near” tool

  • Better than Spatial Join in some ways, worse in others.

42

43 of 63

The “Near” tool

  • Better than Spatial Join in some ways, worse in others.
  • Gives very similar information, but with more options than Spatial Join.

43

44 of 63

The “Near” tool

  • Better than Spatial Join in some ways, worse in others.
  • Gives very similar information, but with more options than Spatial Join.
    • You can get the coordinates of the closest point on the nearby feature.

44

45 of 63

The “Near” tool

  • Better than Spatial Join in some ways, worse in others.
  • Gives very similar information, but with more options than Spatial Join.
    • You can get the coordinates of the closest point on the nearby feature.
    • You can get direction to that feature, although in an unusual manner.

​

45

46 of 63

The “Near” tool

  • Does not give you full information on the nearest road; only distance and ID value.

46

47 of 63

The “Near” tool

  • Does not give you full information on the nearest road; only distance and ID value.
  • Alters your source feature class by adding attribute fields to it.

47

48 of 63

The “Near” tool

  • Does not give you full information on the nearest road; only distance and ID value.
  • Alters your source feature class by adding attribute fields to it.
  • Can analyze multiple feature classes simultaneously.

​

48

49 of 63

The “Near” tool

    • Similar to the Spatial Join tool: If you don’t specify Geodesic Distance, then distance calculated in the coordinate system and units of the Input feature class.
      • This means that if your input features are in geographic coordinates, then the distance will be in “degrees” which is nonsensical!

49

50 of 63

Distance with Rasters

  • The “Euclidean Distance” tool
    • Every raster cell value is the distance from that cell to some specific object.

50

51 of 63

Euclidean Distance

  • Euclidean Distance in general means the straight-line distance between objects, as you would calculate from the Pythagorean Theorem.

​

​

​

​

​

​

51

52 of 63

Euclidean Distance

  • Euclidean Distance in general means the straight-line distance between objects, as you would calculate from the Pythagorean Theorem.

​

​

​

​

​

​

  • The Euclidean distance between Points A and B is the hypotenuse of the triangle and is calculated as:

​

​

52

53 of 63

Euclidean Distance

  • Euclidean Distance in general means the straight-line distance between objects, as you would calculate from the Pythagorean Theorem.

​

​

​

​

​

​

  • The Euclidean distance between Points A and B is the hypotenuse of the triangle and is calculated as:

​

​

  • Essentially it just means that this is the simplest way to calculate distance and does not include complications such as following routes (like roads or trails), accounting for difficult terrain or avoiding obstacles.

53

54 of 63

Alternatives to Euclidean Distance

  • There is not enough time to go into all of these in detail, but if you are curious I recommend looking into:
    • Network Analyst: Has a huge number of functions for determining things like “service areas” based on road networks and point locations (what areas would be most efficiently served by each point based on road networks).
    • Spatial Analyst Cost Distance tool
    • Spatial Analyst Path Distance tool
    • Corridor Designer tools (see www.corridordesign.org)
    • Circuitscape (see http://www.circuitscape.org/)

​

54

55 of 63

Network Analyst Example

55

56 of 63

Network Analyst Example

56

57 of 63

Network Analyst Example

57

58 of 63

Network Analyst Example

58

59 of 63

The “Cost Distance” tool

    • Every raster cell value is lowest cost it would take to reach that cell from some specific object.

59

60 of 63

Least-Cost Path Example

60

61 of 63

Least-Cost Path Corridor Example

61

62 of 63

Circuitscape Example

62

63 of 63

Circuitscape Example

  • Also useful for finding bottlenecks

63