Using Sentinel 1 in Arcmap

Hi all,
I am new in radar data and I really need to import the sentinel 1 data to Arcmap because I need many tools that I used to use in Arcmap (I know SNA does not have those tools). I use sentinel 1 data with GRDM and GRDH in thier names because I’ve heard they are gridded and ready to use. However, when I add the data in arcmap the image becomes distorted (means pixels are not square anymore). Any one may have any suggestion to safely import sentinel 1 in Arcmap? I appreciate your help.

SAR images are not geocoded by default. Before you can use them in a GIS you need at least to run the Range Doppler Terrain Correction. Here you can select an output coordinate system. In many cases calibrating to Sigma0 before this is also advisable. Please also see here: Radiometric & Geometric Correction Workflow

1 Like

Thanks. However should I do all these by their default? I mean for example for "Range Doppler Terrain Correction"I go from Radat tab> terrain correction > Range Doppler Terrain Correction and I put my input and press Run. So there is a message that says: INPUT should be a SAR product! I have to mention that in each extracted Sentinel 1 file (in the measurement folder) there are two tiff files one is vh, vv OR hh, hv which I opened one of them in SNAP. so can you give me a brief explanation that which of them should I use and why I receive this error message? Thanks a lot in advance

don’t just open the TIff files, import the SAR product by its metadata instead.

Please see this tutorial here: http://step.esa.int/docs/tutorials/S1TBX%20SAR%20Basics%20Tutorial.pdf
It explains the basic steps you need to perform to get a usable product.

There are more tutorials here: http://step.esa.int/main/doc/tutorials/sentinel-1-toolbox-tutorials/

Thanks ABraun, I have 3 more questions, I do appreciate your answers in advance:
1- So why you mentioned Sigma0 and in the discussion you put the link it has been mentioned “Calibration to Beta0”? I know there are many littretures about differences between beta, gama and sigma, but generally could you please give me a general indication that what is the difference and which is better? (my study area in in Northern Manitoba, Canada)
2- I was wondering: if all information for pre-processing are available in the metadata and it can be processed easily in SNAP, then why many of the data are not converted to GRD and they are still SLC in ESA website? I am asking because there is an option in SNAP for converting SLC to GRD, so I can do it quickly in SNAP. Am I right?
3- Because my supervisor asked me to find GRD data, I was wondering if there is any difference between the GRD result when I myself convert SLC to GRD and the ready GRDs on ESA website?
Thanks a lot for your response in advance

Some other things: I processed my GRD data with all steps you hve been mentioned in the link you sent in your reply (I mean: 1-thermal noise removal, 2- apply orbit file, 3- calibration to beta0, 4- speckle filtering, 5- radiometric terrian flattering, 6- tange doppler terrian correction). The final result is Gamma and it is uniformly gray! What I did wrong through this process?
And also
if I want to use SLC data can I just use RADAR > sentinel1- TOPS > S1_SLC to GRD and then should I use all pre-processing I mentioned above for the result. AM I correct?

You use Beta0 only if you don’t have information about the incidence angle (required for Sigma0) or when you want to perform Terrain Flattening afterwards which only takes B0. The general sequence (in terms of calibration quality is B0, S0 and G0.

2- I was wondering: if all information for pre-processing are available in the metadata and it can be processed easily in SNAP, then why many of the data are not converted to GRD and they are still SLC in ESA website? I am asking because there is an option in SNAP for converting SLC to GRD, so I can do it quickly in SNAP. Am I right?
3- Because my supervisor asked me to find GRD data, I was wondering if there is any difference between the GRD result when I myself convert SLC to GRD and the ready GRDs on ESA website?

Actually, most of the data is available as both SLC and GRD product. If you are not working on interferometry or polarimetry, GRD is sufficient. You can make a GRD out of every SLC, that is correct, but the quality is nearly the same. The only difference is that if you do it yourself you exactly know which steps were applied and under which parameters. But as long as a GRD is available, you should stick to that, because downloading and processing SLC takes much more time.

Some other things: I processed my GRD data with all steps you hve been mentioned in the link you sent in your reply (I mean: 1-thermal noise removal, 2- apply orbit file, 3- calibration to beta0, 4- speckle filtering, 5- radiometric terrian flattering, 6- tange doppler terrian correction). The final result is Gamma and it is uniformly gray! What I did wrong through this process?

You should check after each step if the result is still ok. Only then you can identify the step where something must have gone wrong. One error source is the DEM that is involved in terrain flattening and terrain correction. If it wasn’t downloaded correctly by SNAP no valid elevation information is available to process the image. But this is only a guess, first see at which step the valuse turn into uniformly gray.

if I want to use SLC data can I just use RADAR > sentinel1- TOPS > S1_SLC to GRD and then should I use all pre-processing I mentioned above for the result. AM I correct?

as stated above, this makes only sense if you don’t have the option to download the GRD directly.
Did you check all sources for GRD? Different archives sometimes do not store the exact same amout of data:

1 Like

did you by chance see this grey image in QGIS? If so, you need to check if contrast enhancement is selected in the layer properties (under style). Sometimges, QGIS doesn’t use contrast stretching, leaving a grey image while it actually covers real values. You can check the histogram of the data also in the properties or its min/max values in the raster metadata -

Thanks ABraun. I just tried to convert my first SLC to GRD and the error came up that “Cannot construct DataBuffer”. I don’t think I have RAM problem or Java problem! all my desktop computer and everything on it are brand new and I have lots of space! I read the same problem by others came up last year, but finally I didn’t get what to do and seems the conversation is not ended with a real solution. Any idea you might have?

the error on data buffer does indicate that you don’t have enough RAM. As I stated above, the process of generating GRDs from SLCs is time consuming because it involves a lot of processing steps. Sometimes the computer’s capacities reach their limits.
How much RAM do you have?

Did you check all mentioned data portals to see if there is really no GRD product for your desired images?

Yes, I checked all data portals and for my data they have the same GRD data. But I figured out why this error came up! My data was on the desktop and when I put it in the C drive and used it once more it works (actually it is working right now! don’t know what will happen at the end)! :slight_smile: Thanks by the way for all the helps