SAR in QGIS

While most data processing happens in SNAP, you can use those processed images to do some analysis and visualization of SAR data in QGIS as well. The ASF website has a number of good “recipes” for using SAR data in various platforms, including QGIS. Let’s go through a similar classification exercise to look at our processed images in the QGIS platform.

Exercise 4.1 Open and view SAR imagery in QGIS.

1. Open QGIS.
2. Create a new project by selecting the New Empty Project template.
3. Save your project as intro-radar.
4. Click on Layer > Add Layer > Add Raster Layer….
5. Select the ... next to the Raster dataset(s) field.
6. Navigate to the intro-radar-data > subset_0_of_S1A_IW_GRDH_1SDV_20230120T090602_20230120T090627_046865_059EA0_7322_Cal_ML_TC.data folder. We know this is the folder for the fully corrected image, since the file ends in Cal_ML_TC.
7. Select the Sigma0_VH and the Sigma0_VV bands.
8. Click Add and then close the window.
9. Convert the bands into decibels. The bands we converted into dB in our SNAP analysis did not save in the file, so we have to do that in QGIS to properly view the image.
   1. Go to Raster > Raster Calculator…
   2. In the Raster Calculator Expression field, enter the following: 10*log10(VV).
   3. Now highlight the portion of the equation where it says VV and input the actual band by double-clicking on the Sigma0_VV band in the Raster Bands field. 4. Your equation should now read: 10*log10(“Sigma0_VV”).
   4. Click on the ... next to the Output layer field. Save the file in the intro-radar-data folder and name it Sigma0_VV_db.
   5. Click OK.
   6. Repeat steps a-e for the Sigma0_VH band.

You should see two new bands added to the map canvas. These look brighter and more similar to the db bands we viewed in the SNAP software. Now let’s manipulate the bands to create another water classification and an RGB image.

Exercise 4.2 Create an RGB image in QGIS.

It is important to note that there are no standardized rules on how to represent microwave data visually, nor specific rules on how to compose a 3-channel color image from only two input channels. The HH, HV, HH/HV and VV, VH, VV/VH combinations for R, G, B are both commonly used in remote sensing for generating an RGB image from Sentinel-1 C-band SAR imagery.

1. Since we are using a DV type image, we will be using the VV band for R, VH for G, and a ratio band VV/VH for B. First, create the ratio band.
   1. Select Raster > Raster Calculator….
   2. In the Raster Calculator Expression field, enter the following: "Sigma0_VV_db@1"/"Sigma0_VH_db@1"
   3. Click on the ... next to the Output layer field. Save the file in the intro-radar-data folder and name it Sigma0_db_ratio.
   4. Click OK.
2. Merge the bands into a single image to generate an RGB image.
   1. Select Raster > Miscellaneous > Merge…
   2. Select ... next to the Input Layers field and check the boxes next to Sigma0_VV_db, Sigma0_VH_db, and Sigma0_db_ratio. Drag the bands to ensure they are listed in that exact order.
   3. Click OK.
   4. Check the box Place each input file into a separate band.
   5. Select ... > Save to File… next to the Merged field. Save the file in the intro-radar-data and name it sentinel-1-rgb. Click Save.
   6. Click Run.
   7. Click Close to close the window.

Great! Now we have an RGB image generated in QGIS. It should look fairly similar to the one we generated in SNAP. Next, let’s replicate the water binary classification methodology that we did in SNAP.

Exercise 4.3 Classify water/non-water in QGIS.

1. Confirm the threshold for our classification scheme.
   1. Right-click on the Sigma0_VH_db@1 layer name and select Properties.
   2. Click on the Histogram tab.
   3. Click Compute Histogram.
   4. A histogram with two peaks should appear. Confirm the value that separates the water values (minimum peak) from the non-water values (maximum peak).
2. It appears that the value is about the same, so we can use the same threshold that we did in the SNAP software.
3. Create a binary image.
   1. Select Raster > Raster Calculator….
   2. In the Raster Calculator Expression field, enter the following: 255*(Sigma0_VH_db@1<-18.38)
   3. Click on the ... next to the Output layer field. Save the file in the intro-radar-data folder and name it water_nowater.
4. Color the image. QGIS gives you a lot more flexibility with the color options, so get creative and make the image look nice!
   1. Right-click on the water_nowater image name in the Layers panel and click Properties.
   2. Select the Symbology tab.
   3. Next to the Render type field, choose Singleband psuedocolor from the dropdown menu.
   4. Now, select a color ramp or make your own. Make sure to follow the golden rules of data visualization.

Terrific! Now you have a beautiful image that illustrates the water bodies in the region in QGIS.

Congratulations! You completed the Introduction to Synthetic Aperture Radar (SAR) workshop. Don’t forget to complete the feedback form, and get ready to build on this material to do even more complex analyses in future workshops. Way to go!