Spatial Joins¶

Goal: Master the spatial join — the operation that combines two layers based on where they are, not on a shared key.

Regular join vs spatial join¶

You'd use a spatial join when there's no shared ID — but the location says they belong together.

Classic example¶

"I have crime points and police precinct polygons. I want to count crimes per precinct."

flowchart LR
    Crimes[Crime points<br/>1,200 features] --> SJ((Spatial Join<br/>INTERSECT))
    Precincts[Precinct polygons<br/>22 features] --> SJ
    SJ --> Out[Precincts +<br/>crime count]

    classDef p fill:#fef3c7,stroke:#f59e0b,color:#92400e
    classDef pol fill:#dbeafe,stroke:#1e40af,color:#1e3a8a
    classDef o fill:#dcfce7,stroke:#10b981,color:#065f46
    class Crimes p
    class Precincts pol
    class Out o

Match options¶

In ArcGIS Pro, Spatial Join asks how the two layers should relate:

One-to-one vs one-to-many¶

This is the trickiest setting.

One-to-one (default)¶

If multiple features match a single target feature, the output summarizes them.

Precinct A overlaps 47 crimes  →  output row: Precinct A, COUNT = 47

You can choose how to aggregate each field: SUM, MEAN, MIN, MAX, FIRST, COUNT, etc.

One-to-many¶

Each match becomes its own row.

Precinct A overlaps 47 crimes  →  47 output rows, each with Precinct A copied in

Use one-to-many when you need each individual relationship preserved.

Field aggregation¶

When doing a one-to-one join with multiple matches, choose how each field merges:

Common spatial join patterns¶

Spatial join vs Summarize Within¶

Esri also has a Summarize Within tool that's a streamlined version of the points-in-polygon spatial join. It's often easier to use for "how many of X are in each Y, with stats" questions.

A common mistake¶

Practice¶

Next up¶

→ Geoprocessing Tools — buffer, clip, intersect, dissolve.