Intersect vs Union vs Erase in GIS: A Visual Step-by-Step Guide to Vector Overlay Analysis

Spatial overlay analysis is one of the core techniques in vector GIS because it allows you to combine or compare geographic features based on their location. Understanding intersect vs union vs erase in GIS is especially important when completing university vector analysis assignments because each tool applies a different spatial logic and produces different geometry and attribute results. Choosing the wrong overlay operation can create misleading outputs, unexpected features, or apparent topological errors.

This guide explains the three operations using practical workflows in ArcGIS Pro and QGIS, with emphasis on what happens to both geometry and attribute data.

---

What Is Spatial Overlay Analysis?

Vector overlay analysis combines two or more spatial datasets to create a new dataset based on their geographic relationships.

For example, suppose you have:

• A land-use polygon layer
• A flood-risk polygon layer

You might want to determine:

Which parts of each land-use category fall inside the flood-risk zones?

An overlay tool can split the polygons where they overlap and transfer relevant attributes into the resulting layer.

The most important operations are:

• Intersect → keep only overlapping areas.
• Union → keep the complete extent of all input polygons.
• Erase/Difference → remove one layer’s footprint from another.

---

Intersect vs Union vs Erase: Comparison Matrix

When comparing intersect vs union vs erase in gis, Intersect is the primary tool used when you only care about shared spatial boundaries.

Tool Name	Geometry Inputs Required	Output Feature Type	Real-World Use Case
Intersect	Two or more overlapping feature layers	Features representing only areas common to the inputs	Finding buildings located inside flood zones
Union	Two or more polygon layers	Polygon features covering the full combined extent	Combining land-use and administrative planning zones
Erase / Difference	Input layer + erase layer	Input features with the erase area removed	Removing protected areas from a development suitability zone
Dissolve	One polygon/line layer with a common attribute	Merged features with selected boundaries removed	Combining several administrative polygons into one region

Important: Dissolve is not itself an overlay operation like Intersect, Union, or Erase. However, it is frequently used before or after overlay analysis to simplify spatial boundaries.

---

1. Intersect: Keep Only the Overlap

How Intersect works

Think of Intersect as:

A ∩ B = the area shared by A and B

If Layer A contains land-use polygons and Layer B contains flood zones, Intersect keeps only the portions where the two datasets overlap.

Example

Imagine:

• Layer A = agricultural land
• Layer B = flood-risk zones

After Intersect:

Output = agricultural land located within flood-risk zones.

This is useful for questions such as:

• Which roads cross flood-risk areas?
• Which parcels are inside a protected area?
• Which land-use classes occur within a watershed?

---

Step 1: Prepare your input layers

Make sure both datasets:

1. Use an appropriate coordinate reference system.
2. Have valid geometries.
3. Cover the geographic area being analyzed.
4. Are loaded into your GIS project.

For example:

Land_Use.shp
Flood_Zones.shp

---

Step 2: Open the Intersect tool in ArcGIS Pro

In ArcGIS Pro, open:

Analysis → Tools → Intersect

Add your input layers.

[Insert Screenshot of Intersect Tool Settings in ArcGIS Pro]

Typical configuration:

Input Features:
    Land_Use
    Flood_Zones

Output Feature Class:
    LandUse_Flood_Intersect

Output Type:
    Polygon

Click Run.

---

Step 3: Examine the output

The output contains only the portions of the input features that overlap.

For example:

Land Use
████████████

Flood Zone
      ███████████

Intersect
      ██████

The resulting polygons represent the common geographic area.

Attribute behavior

Intersect also combines relevant attributes from the input layers.

For example:

Land_ID	Land_Type	Flood_ID	Risk_Level
15	Agriculture	F03	High
21	Residential	F03	High

This allows you to perform further analysis using both datasets.

---

QGIS equivalent

In QGIS, go to:

Vector → Geoprocessing Tools → Intersection

Depending on the QGIS version and interface, the tool can also be found through the Processing Toolbox.

[Insert Screenshot of QGIS Intersection Tool Settings]

Set:

Input layer: Land_Use
Overlay layer: Flood_Zones

Then choose an output location and run the operation.

---

2. Union: Keep the Full Combined Extent

How Union works

Union is different from Intersect because it does not discard areas that belong to only one input layer.Unlike other operations in intersect vs union vs erase in gis, Union preserves all geographic features regardless of overlap.

Think of Union as:

Combine the complete spatial extent of the input polygon layers and split them wherever their boundaries intersect.

Suppose you have:

• Layer A = land-use zones
• Layer B = zoning boundaries

Union creates polygons representing:

1. Areas belonging to A only
2. Areas belonging to B only
3. Areas where A and B overlap

---

Step 1: Prepare polygon layers

Union generally requires polygon feature layers.

For example:

Land_Use
+
Planning_Zones

---

Step 2: Open Union

In ArcGIS Pro:

Analysis → Tools → Union

[Insert Screenshot of Union Tool Settings in ArcGIS Pro]

Configure:

Input Features:
    Land_Use
    Planning_Zones

Output Feature Class:
    LandUse_Planning_Union

Run the tool.

---

Step 3: Understand the output

Suppose Layer A covers one side of your study area and Layer B covers another side, with some overlap.

Union preserves:

A only
A + B
B only

Unlike Intersect, the areas that do not overlap are retained.

---

Handling NULL Attributes

This is one of the most important concepts for students.

Suppose a polygon exists only in Layer A. There is no corresponding feature from Layer B.

The output may therefore contain:

Land_Type	Planning_Zone
Agriculture	NULL
Residential	Urban
Forest	NULL

NULL does not necessarily mean an error.

It usually means:

There was no corresponding attribute value from that input layer for that particular output feature.

Example

If a polygon belongs to Land_Use but does not overlap any Planning_Zone, its planning-zone field may contain NULL.

When analyzing the output, distinguish between:

• NULL because no spatial overlap existed
• NULL because the source data was incomplete
• NULL caused by an attribute-processing problem

---

QGIS Union

In QGIS, use:

Processing Toolbox → Vector overlay → Union

[Insert Screenshot of QGIS Union Tool Settings]

Select:

Input layer: Land_Use
Overlay layer: Planning_Zones

Run the tool and inspect the resulting attribute table.

---

3. Erase / Difference: Remove a Spatial Footprint

Erase performs almost the opposite conceptual operation to Intersect.

Think of it as:

Start with Layer A and cut away the area occupied by Layer B.

In set notation:

A − B = areas of A that are not inside B

Example

Suppose:

• Input = proposed development areas
• Erase layer = protected wetlands

The resulting layer shows:

Proposed development areas excluding wetlands.

This can be useful in suitability analysis, environmental planning, and land-management workflows.

---

Step 1: Identify the input layer

The input layer is the dataset you want to preserve.

For example:

Input:
Development_Areas

---

Step 2: Identify the erase layer

The erase layer contains the geographic footprint that should be removed.

Erase:
Protected_Wetlands

---

Step 3: Run Erase in ArcGIS Pro

Open:

Analysis → Tools → Erase

[Insert Screenshot of ArcGIS Pro Erase Tool Settings]

Configure:

Input Features:
    Development_Areas

Erase Features:
    Protected_Wetlands

Output Feature Class:
    Development_Excluding_Wetlands

Click Run.

---

Step 4: Interpret the result

Imagine the development polygon looks like this:

Before:

████████████████
████████████████
██████░░░░██████
██████░░░░██████
████████████████

The ░ area represents the protected wetland.

After Erase:

████████████████
████████████████
████████████████
████████████████
████████████████

The wetland footprint has been removed from the development area.

The output retains the remaining portion of the original input features.

---

Erase vs Difference in QGIS

QGIS commonly uses Difference for this type of operation.

Use:

Processing Toolbox → Vector overlay → Difference

[Insert Screenshot of QGIS Difference Tool Settings]

Configure:

Input layer:
Development_Areas

Overlay layer:
Protected_Wetlands

The resulting layer represents:

Development_Areas − Protected_Wetlands

---

Intersect vs Union vs Erase: Visual Logic

A useful way to remember the three operations is:

INTERSECT
A + B
→ Keep overlap only

UNION
A + B
→ Keep everything

ERASE
A − B
→ Keep A, remove B

Or conceptually:

Operation	Question it answers
Intersect	“Where do these datasets overlap?”
Union	“What does the complete combined area look like?”
Erase/Difference	“What remains after removing this area?”

---

Common Errors and Troubleshooting

Spatial overlay tools can produce unexpected results when the input data has geometry, projection, or attribute problems.

1. Invalid Geometries

A polygon may contain:

• Self-intersections
• Duplicate vertices
• Unclosed rings
• Overlapping polygon parts
• Other geometry inconsistencies

These problems can cause overlay operations to fail or produce unexpected features.

Solution

Run a geometry validation or repair process before performing the overlay.

In ArcGIS Pro, look for:

Repair Geometry

[Insert Screenshot of ArcGIS Pro Repair Geometry Tool]

In QGIS, use:

Processing Toolbox → Fix Geometries

[Insert Screenshot of QGIS Fix Geometries Tool]

After repairing the data, rerun the overlay operation.

---

2. Unexpected Attribute Loss

Sometimes students focus on the map output and forget that overlay operations also affect the attribute table.

Before running an operation, inspect:

• Field names
• Data types
• Unique identifiers
• NULL values
• Join relationships

After running the operation, compare the input and output tables.

Good practice

Keep a unique identifier in every source layer:

Parcel_ID
Road_ID
Zone_ID

This makes it easier to trace an output feature back to its original dataset.

---

3. Unexpected Number of Features

An overlay operation can create more features than either original layer.

This is not automatically an error.

Why?

Because overlay operations can split polygons wherever boundaries cross.

For example:

Before:

┌──────────────┐
│              │
│      A       │
│              │
└──────────────┘

     ┌──────────────┐
     │      B       │
     └──────────────┘

The crossing boundaries can produce several smaller output polygons.

Always inspect the geometry and attribute table before assuming that additional features indicate a failed analysis.

---

4. Coordinate Reference System Problems

If datasets use incompatible or inappropriate coordinate reference systems, you may encounter:

• Misaligned layers
• Incorrect measurements
• Unexpected overlay locations
• Apparent gaps or shifts

For overlay analysis, confirm that your layers are correctly projected and spatially aligned.

For distance or area analysis, use an appropriate projected CRS rather than automatically relying on geographic coordinates.

---

5. Sliver Polygons

Tiny unwanted polygons can sometimes appear along boundaries because of:

• Slightly different digitizing boundaries
• Different data sources
• Coordinate precision
• Small geometry discrepancies

These are often called sliver polygons.

Before deleting them, determine why they were created. They may represent genuine small features rather than errors.

---

Dissolve Spatial Boundaries After Overlay

Overlay operations often create many small polygons because boundaries from multiple datasets are preserved.

If several neighboring output polygons share the same relevant attribute, you may be able to use Dissolve to merge them.

For example:

Before Dissolve:

┌───┬───┬───┐
│ A │ A │ A │
├───┼───┼───┤
│ A │ A │ A │
└───┴───┴───┘

After Dissolve:

┌───────────┐
│           │
│     A     │
│           │
└───────────┘

In ArcGIS Pro, use Dissolve.

In QGIS, use:

Processing Toolbox → Vector geometry → Dissolve

[Insert Screenshot of Dissolve Tool Settings]

However, do not dissolve automatically. Make sure the boundaries you remove are not analytically important.

---

A Practical University Lab Workflow

Suppose your assignment asks:

Identify agricultural land located within high flood-risk areas, then remove protected wetlands from the result.

A logical workflow would be:

Step 1 — Prepare data

Agricultural_Land
Flood_Risk
Protected_Wetlands

Step 2 — Validate geometries

Run:

Fix Geometries

on datasets where necessary.

Step 3 — Find agricultural land within flood zones

Run:

Agricultural_Land
        +
    Flood_Risk
        ↓
    INTERSECT

Output:

Agriculture_Flood_Risk

Step 4 — Remove protected wetlands

Run:

Agriculture_Flood_Risk
        −
Protected_Wetlands
        ↓
     ERASE

Output:

Available_Agricultural_Flood_Areas

Step 5 — Inspect the result

Check:

• Geometry
• Attribute table
• Area values
• Feature count
• NULL values
• Map topology

Step 6 — Dissolve if appropriate

If the assignment requires generalized regions rather than individual fragments, use Dissolve based on the appropriate attribute.

---

Quick Reference Box

️ Spatial Overlay Quick Reference

INTERSECT
Keep the overlap.
Use when you want to find features occurring inside or across another dataset.

UNION
Keep the full combined extent.
Use when you need information from all polygon areas, including overlapping and non-overlapping portions.

ERASE / DIFFERENCE
Remove an area from the input.
Use when you want to exclude a geographic footprint from another dataset.

DISSOLVE
Remove selected internal boundaries.
Use when neighboring features with the same attribute should be merged.

Before overlay: Check CRS + geometry validity.
After overlay: Check geometry + attributes + NULL values + feature count.

Final Takeaway

The easiest way to remember intersect vs union vs erase in GIS is to focus on the spatial question being asked:

Intersect asks “Where do they overlap?”
Union asks “What is the complete combined area?”
Erase asks “What remains after I remove this area?”

Understanding intersect vs union vs erase in gis ensures you select the correct overlay logic for your spatial assignments.  For university GIS assignments, don’t select an overlay tool simply because it produces a visually plausible map. First identify the spatial relationship you need, then choose the operation that expresses that relationship. Finally, inspect both the geometry and attribute table to verify that the output matches your analytical objective.