Bivariate QGIS Plugin

Bivariate choropleth mapping for QGIS with 3×3, 4×4 and 5×5 vector/raster classification, Staridas palette import, Leaflet export, and native Print Layout legends.

Made by Hemed Lungo · Version 0.0.5 · QGIS ≥ 3.16

Version 0.0.5

This release adds two Print Layout legend styles derived automatically from the selected or detected bivariate box colours:

• Bivariate Box + Variable Ramps Legend — the complete matrix with two low-to-high component strips above it.
• Bivariate Variable Ramps Legend — the two compact component strips without the matrix.

Both styles work with 3×3, 4×4 and 5×5 palettes. Palette selectors in vector, raster, Leaflet and Print Layout workflows now display a small colour-grid preview to the left of every palette name. The Processing previews use QGIS native enum icon metadata, avoiding the abstract widget-wrapper errors reported on QGIS 3.40.

It retains the automatic Print Layout layer sensing and single axis convention introduced in 0.0.4:

• Variable 1 is always Y / vertical.
• Variable 2 is always X / horizontal.
• Vector Bi_Class codes store X first + Y second: A1, B2, C3, etc.
• Raster bivariate codes also store X first + Y second: 11, 22, 33, etc.
• Therefore the map colours/classes and the Print Layout legend matrix correspond directly.
• Print Layout Box and Diamond legends can auto-detect the bivariate layer actually rendered by a layout map.
• Supports both vector and raster bivariate layers.
• Detects 3×3, 4×4 and 5×5 bivariate class grids.
• Vector sensing reads categorized renderer classes such as A1 … C3, A1 … D4, or A1 … E5.
• Raster sensing reads paletted raster values such as 11 … 33, 11 … 44, or 11 … 55.
• The plugin prefers the live layer renderer, so symbology changes can be reflected by the layout legend.
• The legend properties panel includes Auto, Manual, explicit Source layer selection, and Rescan layout.

Main features

• 29 built-in bivariate colour palettes.
• Colour-grid previews beside palette names throughout the plugin.
• 3×3, 4×4 and 5×5 bivariate classification.
• Staridas labelled HEX, CSS and JSON palette import.
• Vector bivariate classification and styling.
• Raster bivariate generation and QML styling.
• Native QGIS Print Layout Box Legend and Diamond Legend items.
• Native Box + Variable Ramps and compact Variable Ramps legend items.
• Automatic raster/vector sensing in Print Layout.
• Transpose axes option for manual legend/palette workflows.
• Standalone Leaflet HTML export.

Installation

1. Download bivariate_plugin-0.0.5.zip from the 0.0.5 GitHub release.
2. Open QGIS → Plugins → Manage and Install Plugins.
3. Choose Install from ZIP.
4. Select the ZIP and install it.
5. Enable Bivariate QGIS Plugin.

For manual installation, place the plugin folder in your QGIS profile's python/plugins directory and restart QGIS.

One axis convention everywhere

The plugin now deliberately uses the same orientation in processing, map symbology, raster codes, and Print Layout legends:

Variable	Axis	Low → high classes
Variable 1	Y / vertical	1 → 3, 1 → 4, or 1 → 5
Variable 2	X / horizontal	Vector: A → C/D/E; Raster: 1 → 3/4/5

The important rule is:

Bivariate class code = X class first, Y class second.

For vector data, C3 means high Variable 2 / X and high Variable 1 / Y. For raster data, 33 means exactly the same thing.

Processing tools

The tools appear under Processing Toolbox → Bivariate QGIS Plugin → Cartography.

Bivariate Choropleth Classification

Classifies two numeric vector attributes into a 3×3, 4×4 or 5×5 bivariate class field.

The vector classifier uses:

• Variable 1 → Y axis / vertical
• Variable 2 → X axis / horizontal

The output Bi_Class is written as Variable 2 letter first + Variable 1 number second.

Example: Rainfall as Variable 1 and Temperature as Variable 2

Suppose you choose:

• Variable 1 = Rainfall
• Variable 2 = Temperature
• Grid size = 3×3

The map classification and the Print Layout legend both use this exact matrix:

                         RAINFALL — Variable 1 (Y)
                                  ↑
High rainfall (3)       A3        B3        C3
Mid rainfall  (2)       A2        B2        C2
Low rainfall  (1)       A1        B1        C1
                         └─────────┬─────────┘
                    Low temp    Mid temp    High temp   → TEMPERATURE
                       (A)         (B)         (C)        Variable 2 (X)

So:

Class	Meaning
A1	Low temperature + low rainfall
C1	High temperature + low rainfall
A3	Low temperature + high rainfall
C3	High temperature + high rainfall

If a polygon is styled as C3 on the QGIS map, the C3 / top-right legend cell in Print Layout uses that same class colour.

For larger grids the same rule is preserved. A 4×4 vector grid uses A–D on X and 1–4 on Y; a 5×5 vector grid uses A–E on X and 1–5 on Y.

Apply Bivariate Color Scheme (Vector)

Applies one of the built-in or imported palettes to a categorized vector layer. The styled layer stores its bivariate dimension and palette metadata so Print Layout legends can identify it reliably.

Bivariate Raster Generator

The raster tool follows the same convention as the vector tool and Print Layout:

• Raster A = Variable 1 = Y axis
• Raster B = Variable 2 = X axis
• Combined raster code = Raster B/X class first + Raster A/Y class second

For the Rainfall/Temperature example:

• Raster A = Rainfall
• Raster B = Temperature

Then raster code 31 means high temperature + low rainfall, while 13 means low temperature + high rainfall. Code 33 is high temperature + high rainfall.

This numeric order is intentional so the raster map and Print Layout legend use the same cell positions and colours.

Bivariate Style Generator (Raster)

Creates and optionally applies a paletted QML style. The style uses the same numeric code order expected by the Print Layout legend: X first, Y second.

Bivariate Leaflet Exporter

Exports a classified vector layer to a standalone interactive Leaflet HTML map.

Print Layout workflow

1. Add your styled bivariate raster or vector layer to the QGIS project.
2. Open a Print Layout and add a map item containing that layer.
3. Choose Add Item → Bivariate Box Legend or Bivariate Diamond Legend. You can also choose Bivariate Box + Variable Ramps Legend or Bivariate Variable Ramps Legend for the two new styles.
4. In Item Properties → Print Layout layer sensing, keep Auto — detect from Print Layout map or choose a specific source layer.
5. Use Rescan layout after changing the map's layer set or symbology.
6. The detected legend uses the same class colours and class positions as the map layer.
7. If no valid bivariate renderer is detected, switch to Manual and choose the palette/grid size yourself.

The detector first checks layers actually rendered by the Print Layout map through QgsLayoutItemMap.layersToRender(). If the layout has no resolvable map layers, it falls back to project raster/vector layers.

Print Layout labels

The Print Layout label fields now use the same variable numbering as the processing tools:

• Variable 2 (X) — horizontal axis
• Variable 1 (Y) — vertical axis

For Rainfall/Temperature enter:

• Variable 2 (X) = Temperature (°C)
• Variable 1 (Y) = Rainfall (mm)

This means there is no separate Print Layout axis convention to remember.

Print Layout properties

• Source layer: Auto, Manual, or an explicit raster/vector layer.
• Palette: built-in palette or custom Staridas import.
• Grid size: 3×3, 4×4 or 5×5.
• Transpose axes for manual workflows.
• Cell size and gap.
• Fit and center grid inside item.
• Variable 2 (X) label.
• Variable 1 (Y) label.
• Show axis labels.
• Show class codes.
• Outline colour and width.

Staridas palette support

The plugin supports labelled output from the Staridas Geography Bivariate Color Palette Builder, including formats such as:

A1 #FAEEC6
A2 #B8BD74
A3 #768D21

It also accepts CSS variables and JSON. Labels are used to preserve the correct bivariate class order even when entries are shuffled.

Repository structure

__init__.py
plugin_core.py
metadata.txt
palettes.py
palette_widgets.py
bivariate_provider.py
bivariate_choropleth.py
apply_bivariate_colors.py
bivariate_raster_generator.py
bivariate_style_generator.py
bivariate_export_leaflet.py
layout_items.py
bivariate_legend_box_generator.py
bivariate_legend_diamond_generator.py
icon.png

Requirements

• QGIS 3.16 or later.
• Python included with QGIS.
• GDAL/OGR and NumPy bundled with normal QGIS installations.

Credits

• Spiros Staridas — creator of the Staridas Geography Bivariate Color Palette Builder and its labelled palette export formats used by this plugin.
• Joshua Stevens — bivariate choropleth methodology reference.
• DataPlotly — inspiration for the QGIS custom Print Layout item registration pattern.
• Bivariate cartography community — palette inspiration.

License

GNU General Public License v2.0 or later.

---

Bivariate QGIS Plugin · v0.0.5 · Hemed Lungo