Hi!

It should be a simple to fix your request in your request you shouldn’t specify the units of the resolution. Therefore, your request should be RESX=10 &RESY=10.

This is taken from the projection that your request the data in. For example, the resolution would be in degrees if requesting data in WGS84 but in metres if you were requesting pixels in EPSG 32633.

There are lots of examples of requesting data in our documentation like here:

Examples for S2L2A

Use these CURL and Postman examples to access atmospherically corrected Sentinel-2 L2A data with Sentinel Hub API.

I also find using requests builder useful as you can build up the request in an easy to understand user interface. You can also automatically set your resolutions in metres too.

Hope this helps you out

Will

Thank for your reply. Actually, I am using Request Builder to request such data. I defined the resolution as RESX=10 &RESY=10, but the selected area still seems too big.

Hi, that is because there is a limit on the number of pixels you can request with Process API. You can only request a maximum area of 2500x2500 pixels. You’ll need to reduce your selected area to be able to do this.

thank you so much. I actually did this. But the tiff file is quite smaller than Sentinel-2 geotiff file from other sources such as EarthExplorer. Is there any difference?

Hi, smaller in what sense? The area you have requested or the file size? Which data are you requesting; I’m assuming you are requesting Sentinel 2 L2A? If so, it will be the same as long as you are requesting the data in UINT16 data type.

Yes, I am requesting Sentinel 2 L2A, it is smaller in size in comparison to the same area that I downloaded from earthexplorer

Yes, so in area if you were requesting a full Granule of Sentinel 2 L2A it would be larger than 2500x2500 pixels. Therefore, using process API you will receive smaller areas. However, it is a more efficient way to request data though as you request the bands or products you need rather than every band via Earth Explorer.

In my case, -processing API-, I give resolution = 10.3 m for SENTINEL2_L1C. (Exact 10 m gives error) After this error I increased the value to resolution= 13.0 Resolved the issue

Hello, I wanted to circle back on this.

How would we get 10m L2A data that is larger than 2500x2500? It was not clear to me in this discussion…

Thanks!

Hi,

There are essentially three services that would allow you to do this (depending on how big of an area you want, each will have advantages or disadvantages):

• Large Area Utilities in Python: Large area utilities — Sentinel Hub 3.9.0 documentation (slower and less PU efficient, but simpler)

• The Asynchronous Processing API: https://docs.sentinel-hub.com/api/latest/api/async-process/ (requires an S3 bucket and works up to 10000*10000 pixels)

• Batch Processing API: https://docs.sentinel-hub.com/api/latest/api/batch/ (unlimited in terms of area, but requires an S3 bucket and an Enterprise account)