Good morning everyone.
Please, I would like to get from the Sentinel-1 products an image like this:
Which Sentinel-1’s product should I use?
Should I take a picture of my interest only use the zoom/subset or I should set some parameters after performing Terrain Correction?
Thanks in advance.
In short - you unfortunately can’t. At least not with freely available Sentinel-1 data. IW Mode has a spatial resolution of 10 meters and SM mode one of 5 meters. As shown here, SM data is currently only collected at few locations.
The image you are showing is a SPOT image. SPOT is a high-resolution optical satellite, that means it uses wavelengths in the nanometer spectrum (the light that our eyes perceive)
Radar satellites operate at a microwave spectrum, thus cannot achieve high resolutions as easily as optical sensors (or only at the cost of reduced coverage). Additionally, they are affected by speckle effects which make the image noisy. Third, they operate as a side-looking system which leads to geometrical distortions in the image which cause obscured areas (especially in urban environments).
Even if you would have very high resolution imagery (TerraSAR-X Staring Spotlight, for example) you will never get as sharp and complete images as the one demonstrated above. It’s a matter of physics.
You could, theoretically, use a stack of maybe 12 TerraSAR-X images with spatial resolutions of 0.25 to 3.5 meters and apply a multi-temporal filter. But each of these scenes roughly costs 2000 Euro.
Have a look at those tutorials which explain what SAR sensors can achieve and where are the limitations.
Introduction to SAR remote sensing
A tutorial on synthetic aperture radar
Thank you very much, your explanation is fantastic, you are helping me a lot.
Please could you answer these my further questions?
This product S1A_IW_GRDH_1SDV_20170210T051855_20170210T051920_015217_018EB5_1497I
has a resolution of 20x22 m, with or without multilook used in SNAP, is it correct?
I understand this from here:
https://sentinel.esa.int/web/sentinel/user-guides/sentinel-1-sar/resolutions/level-1-ground-range-detected
How do I decide which zoom/subset I have to stop in order to obtain a good photo with a resolution of 20x22 m?
Excuse my current ignorance in this matter.
you already asked that in this topic yesterday:
I haven’t answered because I’m not sure by myself. So you would have to be patient for others to answer this, I’m sorry. Please don’t start multiple topics for the same questions in oder to catch more readers.
Excuse me, you are always very helpful.
My employer wants an image of Sentinel-1 at full resolution close to the ground.
I still don’t understand how to get this picture from Sentinel-1’s products, therefore I’m worried.
Thanks again.
Yes, I understand, I’ve been many years in your same situation so now it’s important for me to use SNAP to get these blessed photos.
You should study the SNAP and S1TBX tutorials that are available online. Do terrain correction to 10m pixel size into the projection you need. The resolution should be close to the original ~20m of the GRD products.
Thanks, I’m reading tutorials from twenty days.
I have performed Terrain correction to 10m pixel.
Please, how do I decide which zoom/subset I have to stop in order to obtain a good photo with a resolution of ~20m? This is my problem.
Press the button (looks like magnifying glass) marked “P” in the navigation tool-window. This will show individual pixels in full display resolution. Then you can export your view to for example a .jpg with right-clicking on the image and selecting from the menus that pop up.
Excuse me, is it correct to say that these two photos have the same resolution ~20m but only changes the zoom?
Now I think that it is an individual choice to stop zoom in order to obtain a good photo with a resolution of ~20m.
Thanks.
Yes, it is the same product, just different levels of zoom.
your level of zoom is completely independent from the spatial resolution or the pixel size. If you make a subset, the pixel size of the image stays the same.
Unfortunately I mistakenly thought that SNAP decide even better photo but instead it is an individual choice.
Now I don’t know to choose the best answer so thank you to everyone. 
Hello,
I have a similiar question as mast. I am being asked about the resolution of a processed image. Original product was Sentinel-1 Level 1 GRD product, IW swath. Problem is, that I used Multilooking proccess with 5 look in both range and azimuth. But does it change the resolution, if it is GRD product? And what number should I say? You know, if you are processing data from Sentinel-2, some bands have 10 m resolution, some 20 m and some 60 m. You have just 1 number there.
Thank you for the answers.
S1 GRD data is already of square pixel geometry (10x10m), so there is no ultimate need for multi looking unless you really need lower resolution.
Thank you.
But I already used the process of Multilooking and I can not change it anymore. Is there a way to tell how much was the resolution changed? And is the value of spatial resolution 10 meters originally?
It worth to take a look firstly at Acquisition resolution Level-1 SLC,
Then
The spatial resolution of GRD products ,
Thank you. But what does this mean?
Is there any option to get 1 number like you have with Sentinel-2 bands? That you simply have a band with resolution of 60 m?when you enter a number of looks in the module, you see the resulting spatial resolution. As GRD is 10x10, any multiple of looks is simply the corresponding tenfold resolution in meter.
2 looks = 20x20 m
3 looks = 30x30 m
4 looks = 40x40 m
and so on
Thank you. I really appriciate the help from both of you. This is what I needed to know.
One can multilook in range, azimuth or both:
2 looks = 10x20m
4 looks = 20x20m
40x40m pixel size should correspond to ~16 looks, roughly speaking.
