Session 3 — Vegetation indices and simple NDVI analysis

Preamble¶

• Tools: QGIS

• Data: Sentinel-2 spectral bands (same dataset as Session 2)

• Goal: compute and analyse a vegetation index (NDVI) and perform a simple land cover classification using thresholding.

Learning Objectives¶

By the end of this session, students will be able to:

• Compute a spectral index using the Raster Calculator

• Understand the NDVI formula and its interpretation

• Analyse the distribution of NDVI values using histograms

• Interpret vegetation patterns from NDVI maps

• Perform a simple land cover classification using NDVI thresholds

Course Material¶

We use the same Sentinel-2 dataset as Session 2.

📂 Labworks data
📥 session_1_data.zip

If needed:

1. Download the dataset

2. Unzip the archive

3. Place the folder in a known location on your computer

Tasks¶

Task 1 – Load the data¶

Open QGIS and load the spectral bands used previously.

Menu: Layer → Add Layer → Add Raster Layer

Task 2 – Computing NDVI¶

Vegetation indices exploit the spectral behaviour of vegetation.

Healthy vegetation:

• strongly reflects Near Infrared (NIR)

• absorbs red light

The Normalized Difference Vegetation Index (NDVI) is defined as:

NDVI = (NIR - Red) / (NIR + Red)

For Sentinel-2:

• NIR = B8

• Red = B4

Steps

Open the Raster Calculator: Raster → Raster Calculator

Enter the following expression:

(“B8@1” - “B4@1”) / (“B8@1” + “B4@1”)

Set:

• Output layer name: NDVI

• Output format: GeoTIFF

• Choose a destination folder

Click Run.

Questions

• What is the theoretical range of NDVI values?

• Which areas show the highest NDVI values?

Task 3 – Visualizing NDVI¶

Open Layer Properties → Symbology for the NDVI layer.

Select: Singleband pseudocolor

Choose a color ramp such as: Red → Yellow → Green

This visualization better highlights vegetation gradients.

Task 4 – Exploring NDVI values using the histogram¶

Open: Layer Properties → Histogram

Display the histogram of the NDVI layer.

Observe the distribution of NDVI values.

Questions

• What NDVI values correspond to vegetation?

• What NDVI values correspond to water or urban areas?

• Is the distribution unimodal or multimodal?

Task 5 – Pixel value inspection¶

Use the Identify Features tool and click on different land cover types:

• forest

• crops

• water

• urban areas

Record the NDVI values.

Questions

• Which land cover type has the highest NDVI values?

• Which land cover type has negative NDVI values?

Task 6 – NDVI classification using thresholds¶

NDVI values can be used to perform a simple land cover classification.

The idea is to assign a class number to each pixel depending on its NDVI value.

Each class corresponds to a type of land cover (water, soil, vegetation, etc.).

Step 1 – Open the Raster Calculator

Menu: Raster → Raster Calculator

Step 2 – Classification expression

Enter the following expression:

(“NDVI@1” < 0) * 1 + (“NDVI@1” >= 0 AND “NDVI@1” < 0.2 ) * 2 + (“NDVI@1” >= 0.2 AND “NDVI@1” < 0.5) * 3 + (“NDVI@1” >= 0.5) * 4

Save the output as: NDVI_classes.tif

The same logic is applied to all conditions, and the results are added together to produce the final classification raster. Since only one condition can be true for a pixel, the sum returns the corresponding class.

Output classes

The output raster contains integer class values:

Each pixel in the output image therefore represents a land cover class derived from NDVI thresholds.

Visualization

Open:

Layer Properties → Symbology

Select:

Paletted / Unique values

Assign colors to each class to visualize the classification map.

Questions

• How does this simple classification compare with the RGB image?

• Which NDVI threshold seems the most sensitive?

• What limitations do you observe with this simple thresholding approach?

Task 7 – Cartographic layout¶

In this task, you will create a cartographic layout to produce a map of the NDVI classification.

A map layout allows you to prepare a map for presentation or export by adding elements such as a title, legend, scale bar and north arrow.

Step 1 – Open the Layout Manager

In QGIS, open:

Project → New Print Layout

Give a name to your layout (for example: NDVI_map) and click OK.

A new layout window will open.

Step 2 – Add the map

1. Click Add Map in the toolbar.

2. Draw a rectangle on the page.

The map from your QGIS project will appear in the layout.

You can adjust the extent and zoom level if needed.

Step 3 – Add a title

1. Click Add Label

2. Click on the page

3. Enter a title, for example:

NDVI Classification – Casablanca study area

In the Item Properties panel you can adjust:

• font size

• alignment

• position

Step 4 – Add a legend

1. Click Add Legend

2. Click on the page

The legend will automatically display the NDVI classes.

You can edit the legend in the Item Properties panel:

• rename classes if needed

• remove unnecessary layers

Step 5 – Add a scale bar

1. Click Add Scale Bar

2. Click on the page

Select the map item associated with the scale bar.

You can adjust:

• units (meters or kilometers)

• number of segments

• style

Step 6 – Add a north arrow

1. Click Add Picture

2. Click on the page

3. In Item Properties, select a north arrow symbol

You can resize and reposition the arrow.

Step 7 – Add author and data source

Add a small text label with information about the map:

Example:

• Author: Your Name

• Data: Sentinel-2 imagery (ESA Copernicus programme)

Step 8 – Export the map

Export the final map:

Layout → Export as PDF

Save the file as:

NDVI_map_layout.pdf

Example of expected map layout