Radar sensors work in the microwave part of the spectrum and so can make images even when there is cloud cover. They are active sensors that can operate both day and night.
Sentinel 1 is a European radar imaging system and provides global data every 12 days.
PALSAR is a Japanese radar imaging system. Earth Blox uses its annual mosaic which is a continuous global map for an entire year.
The data are all georectified and analysis ready.
Using Radar Data
TABLE OF CONTENTS
When should you use radar imagery?
- Radar imagers can see through clouds, so every image is guaranteed cloud-free.
- Sentinel 1 works well for picking up objects on the ocean, including ships, wind turbines and icebergs, as well as artificial structures on the land surface (buildings, bridges, etc).
- Radar is also good for tracking changes in agricultural crops.
- Radar is particularly good at picking up areas of flooding.
- PALSAR, with its longer wavelengths, is good at mapping forests and forest change.
Sentinel 1 (S1)
- Sentinel 1 is a C-band radar imaging system that provides global data every 12 days. When both of the satellites are operational, there is a 6-day exact repeat cycle. However, the nature of the orbit allows overlapping images as frequently as every 3 days at the equator, less than 1 day at the Arctic, and 1-3 days over Europe, Canada and the main shipping routes.
- The first Sentinel 1 data is from 17th October 2014 until the present day.
- C-band means it has wavelengths of around 5cm long. This means it is good for tracking changes in agricultural crops, or other short vegetation.
- Since radars are active sensors, Sentinel 1 can collect data on both the day (ascending) and night (descending) sides of the orbit.
- All the S1 data in Earth Blox are derived from Interferometric Wide Swath mode. This means the spatial resolution is 20x20m. (Note: the grid spacing of the pixels is 10x10m, but the resolution is still 20x20m.)
- PALSAR is a Japanese L-band radar imaging system.
- L-band has wavelengths of around 23cm and so is good for discriminating larger vegetation (i.e. trees, rather than crops). It will provide a higher forest/non-forest contrast than C-band.
- Earth Blox uses its annual mosaic which is a continuous global map for an entire year.
- Data is available from 2007-2010, and then annually from 2015.
- PALSAR Forest Change products are also visible under the FOREST->Other forest change datasets block.
How to use the radar block
- Select the Use This Dataset block from the toolbox and from Imagery choose Landsat or Sentinel 2.
- By default it will include the cloud filtering and visualisation blocks with defaults (if you leave Add Recommended Blocks checked), and so will look like this:
- By default the bands that are made available are in linear scaling. The bands are also available in logarithmic scaling (deciBels) and those bands will have a "_dB" after their name. dB scaling is a logarithmic scaling of the data and often makes it easier to see the features in the image.
- By default the block will select both the ascending "ascending" (when the satellite is travelling northwards in its orbit, and looking East), and the "descending" ( going South and looking West). You can use a Filter->Select Orbit block to restrict to just one of these.
- The HH, VV, VH and HV options in the RGB output refer to the polarisation channel. Radar signals can be transmitted or received as either horizontal (H) or vertical (V) waves. For instance, the HH channel is both transmitting and receiving in H, whereas the HV channel is transmitting H and receiving in V. (the "cross-polarisation").
- Sentinel 1 is a little more complicated. In the Polar regions, the satellite collects HH and HV. Everywhere else it collects VV and VH. This is because HH gives better contrast between sea ice and open water.
- When using Sentinel 1, for most use cases either VV or VH will be the best option to select.
- PALSAR has HH and HV only. HV provides the best contrast between forests and non-forest.
- Finally, the Composite option allows you to combine multiple dates together to create a new image. If you select a time period that covers multiple image acquisitions, then using the Mean option create an image layer that is the average of those scenes. The other options would allocate to the final pixel value the minimum or maximum value, or the median, over all the images that have data over that pixel.
- Remember to add a Layer name.
- You are now ready to go. Click RUN WORKFLOW and view the results on the map.
We always track the feedback you give us so that we can improve how we help you.
Please let us know if you found this one helpful.