Coverage of shapefile for strata/multi tile processing


Is it ok to use the same shp file uploaded when creating a site in the website interface for multi tile processing?
Or do I need to create another shp file that fully covers the landsat+sentinel tiles and change the insitu folder ownership to sen2agri-service?

I assumed since my site already intersects my landsat and sentinel tiles (four tiles for each sensor), it is suffice to use it in the multi tile processing in situ data.


I’m not sure I understand the question. The system should download both L8 and Sentinel-2 tiles that intersect your site definition. After you create the site and the products are downloaded, you don’t have to do anything else with that shapefile.

Are you trying to use the same shapefile for both the site definition and the crop type in-situ data?

I’m sorry for my confusing question.

The ishapefile I’m referring to is the one found in the “/mnt/archive/insitu/{site_name}/strata/” file path (for multi-tile processing).

What I did is to use the shp file in creating a site in the web interface to be placed in the “/mnt/archive/insitu/{site_name}/strata/”. Or should I use a similar crop type in situ data to that file path?

No, the strata shapefile must have polygons with a numeric ID field.

But you probably don’t need to use one. The file is used not for multi-tile processing, but for larger sites where there are regions with different climate, either because of distance, features of the terrain, or other factors. The effect is that different classification models are trained for each region separately. Even so, we’ve had mixed results with this approach, and using a single stratum often gave better results.

If you don’t need to use this feature, you can leave that directory empty.

Ok, now I understand.

Thank you very much.

Hi, Inicola!

Can you please clarify one more time, it is a little confusing for me. If I already have a product, how can I check, was it single- or multi-tile processed? What does the “Single stratum: True” in the output mean? If the product was single-tile processed, should I get different training/validation polygons for each of the tiles in the AUX_DATA/***ISD***.zip archive?

Thank you in advance!

Hi mosin,

For L4* products, the default mode of the system is “Single-stratum” processing. That means that the entire site (one or several tiles) will be processed as one stratum -> the classifier will use all the provided in situ data to create one single classification model -> the classification model will then be applied to each tile of the site. Therefore of course you don’t need in situ data covering each of the tiles. But please, be aware that the spatial distribution and the diversity of your dataset will be a direct driver of the quality of the output maps.

In case you are providing the system with a strata shapefile (possible when editing your site in the GUI from v1.8), the system will use this shapefile to stratify your site and build one classification model for each of the strata. In this situation, each stratum have to be covered by in situ data samples. The resulted output map will be a mosaick of the individual stratum classification.

Finally, if the output was processed with multiple strata, you should get different training/valdiation polygons split in the output products allowing you to check which samples have been used for the training over each stratum.

Let me know if it helps,

Yes, it helps a lot! Thank you!

I was confused by the following sentence in the Sen2-Agri-Software-User-Manual-2.7 p. 88: “This mode (multi-tile mode) requires a stratification which divides the region to monitor in different regions.”. So, I thought that if I do not provide a stratification file than the single-tile mode will be used. Now, it’s clear for me that the default mode is multi-tile even if not using stratification file, and the all tiles together will be used as one single stratum by default.

1 Like