Maps

Geographic grid-aggregation map

Aggregates spatial observations within a regular geographic grid.

  • Map
  • Geography
  • Aggregation

What it answers

How do regular spatial areas compare?

Shape
Regular spatial cells
What it shows
Local patterns and aggregated hotspots.

Example

The question

How are store visits spread across the country in even squares?

Store visits per 10 km square, September (grid map)

Store visits per 10 km square, September (grid map)The country is split into equal squares coloured by store visits, darkest navy for fewest and bright cyan for most. Visits cluster in the south-east.Visits per squareFewerMore
Store visits per 10 km square, September (grid map)The country is split into equal squares coloured by store visits, darkest navy for fewest and bright cyan for most. Visits cluster in the south-east.
The country is split into equal squares coloured by store visits, darkest navy for fewest and bright cyan for most. Visits cluster in the south-east.

Illustrative data. It belongs to no company.

When to use it

Use it when

Spatial aggregation on a declared regular grid.

Avoid it when

Incompatible cell areas or confusing counts with density.

How to read it

Read each cell's extent and aggregate value.

Note

Specify coordinate system, cell size and aggregation.

The data you need

Layout
Long table
Fields
LocationID: text; Longitude,Latitude: numbers; Value: nonnegative number
One row is
One point location and its value or weight. What is the grain?
Rules
State CRS. For WGS84, longitude is [-180,180] and latitude [-90,90]. Validate coordinate order.
Optional
For unweighted density, each point may have weight 1. Project coordinates appropriately for distance calculations.
Preparation
Validate the input grain and missing values. State CRS. For WGS84, longitude is [-180,180] and latitude [-90,90]. Validate coordinate order.
Units
State cell area and aggregate. Counts and per-area density have different denominators.
Example of the input layout: Long tableA long table: one row per observation, few columns. New data adds rows.LocationIDLongitudeLatitudeValueAABBC
A long table: one row per observation, few columns. New data adds rows.

Worked example

Equal-area cells contain 12 and 6 events. Counts differ twofold; if cell areas differ, compute density before comparing.

LocationIDLongitudeLatitudeValue
A18.0759.3330
B11.9757.7145