Skip to article frontmatterSkip to article content
Site not loading correctly?

This may be due to an incorrect BASE_URL configuration. See the MyST Documentation for reference.

Preamble

Learning Objectives

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

Course Material

Data and tools required for this lab session are available below 👇

📂 Labworks data:
Download the dataset containing the Sentinel-2 bands:

📥 session_1_data.zip

After downloading:

  1. Unzip the archive

  2. Place the extracted folder in a known location on your computer

🛰️ The Sentinel-2 mission:

Sentinel-2 is part of the Copernicus Earth Observation programme of the European Union, developed and operated by the European Space Agency (ESA).

The mission consists of a constellation of satellites equipped with a Multispectral Instrument (MSI) designed to monitor land surfaces, vegetation, soils, water bodies and coastal areas. :contentReference[oaicite:0]{index=0}

Sentinel-2 provides high-resolution optical imagery with 13 spectral bands ranging from the visible to the shortwave infrared region of the electromagnetic spectrum. :contentReference[oaicite:1]{index=1}

These bands are acquired at three spatial resolutions:

Spatial resolutionNumber of bands
10 m4 bands
20 m6 bands
60 m3 bands (mainly designed for atmospheric correction purposes)

This combination of spectral richness, spatial resolution and frequent revisit time (~5 days) makes Sentinel-2 particularly well suited for applications such as:

Sentinel-2 infographic Source: ESA Copernicus Sentinel-2 mission overview.

Sentinel-2 spectral bands

The Sentinel-2 sensor records reflectance in multiple regions of the electromagnetic spectrum, including:

Sentinel-2 spectral bands

These spectral bands allow the analysis of different properties of the Earth’s surface, such as vegetation health, soil moisture, and water content.

🛠 Labworks tools:

If QGIS is not installed on your computer, you can download the LTR (Long Term Release) from the official website: https://qgis.org

For this session, a portable version of QGIS is also provided.

Steps to run QGIS portable:

  1. Download the archive below

  2. Unzip the folder

  3. Launch QGIS by double-clicking on: launch_qgis_portable.bat

📦 Portable QGIS: 📥 QGIS

Tasks

Task 1 – Data download

  1. Download the dataset from the link provided above.

  2. Unzip the archive.

  3. Place the extracted folder in a known directory on your computer.

Inside the folder you should find several Sentinel-2 spectral bands stored as raster files.

Task 2 – Opening the data in QGIS

  1. Open QGIS

  2. Create a new project

  3. Load the raster files

Menu: Layer → Add Layer → Add Raster Layer

Once loaded, the layers will appear in the Layers panel.

Questions

Task 3 – Inspecting image properties

In QGIS, explore the properties of the satellite image.

Step 1 – Raster properties

  1. Right-click on one of the raster layers

  2. Open Properties

  3. Go to the Information tab and observe the different characteristics

Questions

  1. What are the image dimensions (rows and columns)?

  2. What is the pixel size?

  3. Which CRS is used?

  4. How many bands does the dataset contain?

  5. What is the data type of the raster values?

  6. Is a NoData value defined?

Step 2 – Pixel values

Use the Identify Feature tool in the QGIS toolbar alt text and click on different pixels corresponding to various land cover types:

Questions

Task 4 – Exploring individual spectral bands

Each band represents the reflectance of the Earth’s surface measured at a specific wavelength of the electromagnetic spectrum.

The Sentinel-2 dataset provided in this lab includes the following bands:

BandDescription
B2Blue
B3Green
B4Red
B8Near Infrared (NIR)
B11Shortwave Infrared (SWIR)

Questions

Task 5 – Creating a multiband stack (VRT)

The Sentinel-2 bands are provided as separate raster files, each corresponding to a different wavelength. To work with them together (for visualization or analysis), we first create a multiband stack. Instead of creating a new raster file on disk, we will create a Virtual Raster (VRT).

A VRT is a lightweight file that references existing rasters and virtually combines them into a single multiband dataset, without duplicating the data on disk.

Steps

  1. Open the tool:

Raster → Miscellaneous → Build Virtual Raster (VRT)

  1. Add the 5 spectral bands in the following order: B2 B3 B4 B8 B11

  2. Check Place each input file into a separate band

  3. Set Resolution → Highest resolution

Some Sentinel-2 bands have different spatial resolutions (10 m or 20 m). Choosing the highest resolution will resample the 20 m bands to 10 m so that all bands share the same grid.

  1. For the resampling method, select: Nearest neighbour

  2. Save the file in a known directory as: S2_band_stack.vrt

  3. Run the tool.

The result is a multiband virtual raster that combines the five Sentinel-2 bands into a single layer.

Task 6 – Creating a Natural Colour Composite

Satellite images are often visualized by combining three bands.

To create a natural colour composite:

  1. Right-click the VRT layer

  2. Open Layer Properties

  3. Go to Symbology

  4. Select Multiband color

Assign the following bands:

ChannelBand
RedB4
GreenB3
BlueB2

This combination approximates what the human eye sees.

Questions

Task 7 – Creating a False Colour Composite

False colour composites highlight specific land surface features.

Use the following band combination:

ChannelBand
RedB8
GreenB4
BlueB3

In this representation:

Questions

Task 8 – Exploring Spatial Resolution

  1. Zoom in on Band B2

  2. Then zoom in on Band B11

Observe the differences in pixel size.

You can also consult the metadata to verify the spatial resolution.

Questions

Task 9 – Interpreting land cover

Using the different visualizations, try to identify:

Using the identify features tool, record approximate pixel values for several land cover types in the following bands:

Questions

Further Reading and Resources

Troubleshooting – QGIS portable version

Some students may experience issues where raster processing tools do not appear or do not work correctly in the portable version of QGIS.

If this happens, you can reload the core plugins manually using the Python console.

Steps

  1. Open the Python Console in QGIS
    Menu: Plugins → Python Console

  2. Paste and run the following code:

from qgis.utils import unloadPlugin, loadPlugin, startPlugin

core_plugins = ["processing", "db_manager", "MetaSearch", "grassprovider"]

for plugin in core_plugins:
    try:
        print(f"Reloading plugin: {plugin}")
        unloadPlugin(plugin)
        loadPlugin(plugin)
        startPlugin(plugin)
        print(f"{plugin} successfully reloaded\n")
    except Exception as e:
        print(f"Error while reloading {plugin}: {e}\n")

print("Done — plugins should now be active.")

This will reload the core processing plugins and usually fixes issues with the Raster Calculator or Processing tools.
Computer Vision & Remote Sensing
Session 1 — Introduction to Computer Vision & Earth Observation
Computer Vision & Remote Sensing
Session 3 — Vegetation indices and simple NDVI analysis