Sentinel-2 imagery in QGIS

Adding raster data

Now that we know the basics of QGIS, we can add our Sentinel-2 imagery. You can continue on with your “World Map” project or start a new one, it won’t matter.

• 1) On the Main Menu go to “Layer” > “Add Layer” > “Add Raster Layer”. If you prefer, you can also press Ctrl+Shift+R or use the “Open Date Source Manager” button immediately below the “New Project” button and click the “Raster” tab when it opens.

_{Remember, our satellite data is an image, so it is raster data rather than vector data. If you need a recap on the difference, check out Week 1.10 again.}

• 2) Click the “Browse” button (the button with three dots).

_{Click the “Browse” button.}

• 3) Navigate to the folder that contains your downloaded satellite imagery.
• 4) Keep double-clicking on the extracted folder until you get a choice of five different folders: “AUX_DATA”, “DATASTRIP”, “GRANULE”, “HTML” and “rep_info”.
• 5) Open the “GRANULE” folder.
• 6) Open the folder with your image ID, the example below is for image ‘L2A_T40QEM_A035724_20220425T064044’ – the numbers will be different for yours, but it will look similar.
• 7) Open the “IMG_DATA_R10m” folder.
• 8) Click the sixth image down, ending in ‘…_TCI_10m.jp2’ and click Open.

_{Select the TCI image.}

• 9) In the Data Source Manager click “Add” and then “Close”.

_{Adding our imagery from the Data Source Manager.}

_{Our imagery has been added to our project.}

_{What kinds of features can you see? What features do we miss at this resolution? What kinds of archaeological sites are likely to be visible in Sentinel-2 imagery? Let us know what you find in the comments!}

_{The remains of the Sumerian city of Girsu within modern farmland. The main part of the site was built almost five thousand years ago! Imagery courtesy of ESA.}

Exploring raster data

• 1) Click on the “Identify Features” button and click on a random part of the image.

_{“Identifying” a raster pixel.}

_{Wondering why these exact numbers are used? It’s down to binary – the number system that computers use to store and read data! If you include ‘0’, there are 256 possible values for Red, Green and Blue with this imagery. This is easy to represent with an 8-figure binary number – better known as an 8-bit number – from 00000000 (0) to 11111111 (255). Each of the eight digits has two options – it can either be a 1 or a 0 – and if you multiply 2 by itself eight times (2 to the power of 8 in mathematical terms) you get 256! This means that each individual pixel has 8-bits for red, 8-bits for green and 8-bits for blue – 24 bits in total. With 256 different options for red, green and blue, it is possible to make nearly 17,000,000 different colours!}

• 2) Find and identify a very bright pixel and see what sort of values it gives.

Red, green and blue should all be close to 255.

• 3) Find and identify a very dark pixel to compare.

This time all three are likely to be very low.

• 4) Find an area with dense vegetation such as a field, forest, park or garden and identify a pixel.

In this case, your green value should be significantly higher than red and blue.

Congratulations on reaching the end of Week 2! We have now covered the basics of true colour satellite imagery. Let us know if you have any questions in the comments. Come back next week to learn how we use multispectral imagery to extend our remote sensing toolkit beyond the visible light spectrum.