Building a crop map for South Africa using Sen2-Agri

Description

What is Sen2-Agri ?

The Sen2-Agri system is an operational standalone processing system generating agricultural products from Sentinel-2 (A&B) and Landsat 8 time series along the growing season. These different products consist of:

• monthly cloud-free composites of surface reflectance at 10 – 20 m resolution;
• monthly dynamic cropland masks, delivered from the agricultural mid-season onwards;
• cultivated crop type maps at 10 m resolution for main crop groups, delivered twice along agricultural seasons;
• periodic vegetation status maps, NDVI and LAI, describing the vegetative development of crops each time a cloud-free observation is recorded.

The Sen2-Agri system is free and open source, allowing any user generating near real time products tailored to his needs at its own premises or on cloud computing infrastructure

What can I do with this EO Solution? Description of the solution and what is it meant for.

1. monthly cloud-free composites of surface reflectance at 10 – 20 m resolution;
2. monthly dynamic cropland masks, delivered from the agricultural mid-season onwards;
3. cultivated crop type maps at 10 m resolution for main crop groups, delivered twice along agricultural seasons;
4. periodic vegetation status maps, NDVI and LAI, describing the vegetative development of crops each time a cloud-free observation is recorded.

Sen2-Agri allows to perform near real time agriculture monitoring at national scale and at parcel resolution, all for a determined region and a predefined growing season. This Knowledge Package (KP) explains how to use the Sen2-Agri software with an example over South Africa.

Essential elements of the Sen2Agri Knowledge Package.

• Research publications
• Software code
• Documentation: Software user manual, Video
• Example of area of interest: Polygon Shapefile of the AOI
• In-situ data for a region of South Africa: Polygons shapefile
• Links to cloud environments (AWS – ESA Food Security TEP) where Sen2-Agri is ready to run.
• Results for South Africa

Any user can use Ready to run Instances?

For both AWS – ESA Food Security TEP the user needs to have an account in one of the platforms.

Advantages of using the Sen2-Agri?

Sen2-Agri is an end-to-end solution. Once the system is running, the user can define an area of interest and the growing season date, set the sensors (Sentinel-2 or Landsat 8) and uploads the in-situ data. The rest of the processing is done automatically. There is no need to manually pre-process the remote sensing imagery because the system downloads them, performs the corrections and applies the cloud mask before creating the composite. Furthermore, this KP contains ready-to-run instances so users may access the software without having to perform the complete installation.

Step-by-step guide using the "ready to use package" with links to specific resources.

• Read the Publication
• Access the Instance
• Run the code, uploading the test area and in situ data e.g. South Africa
• Compare your output map with the output map from the paper
• Use the EO Solution for your area of interest

If you want to test, contact the GEO Secretariat to have access to the ready to run instances If you want you can download it and run it in your local computer.

Reproducibility

Reproducibility in other areas (Sentinel 2/ Landsat 8)

Once the system is running, the user can define an area of interest and the growing season date, set the sensors (Sentinel-2 or Landsat 8) and uploads the in-situ data. The rest of the processing is done automatically, the most important point is to have in situ data compliant with the required fields (See below)

Where the Sen2Agri solution gets the images?

The importance of in situ data for validation and calibration

To create accurate crop-type maps, the system relies on ground samples in polygon shapes (this is what allows parcel resolution mapping) with an area of minimum 20 m2 .

Characteristics of the in-situ data (Metadata) to collect.

Metadata of the ground samples (in situ data) format are described in the Software user manual from page 99 to 102. An example over South Africa may be found in the related resources for testing purposes.

JECAM guidelines:

In order to be used by the system, the in-situ samples need to follow the JECAM guidelines for cropland and crop type definition and field data collection.

What type of Hardware shall the User have to run the EO Solution in the local machine?

If you intend to use the Sen2-Agri system on a local site (~90 000km²), the minimum system requirements to install and run this system are as follows:

• CPU: 8 Cores
• RAM: 32GB
• HDD Storage: 4 TB
• SSD Storage: 250 GB (optional – for temporary files)

If you intend to use the Sen2-Agri system on a national site (~500 000km²), the minimum system requirements to install and run this system are as follows:

• CPU: 16 to 32 Cores
• RAM: 64GB to 128 GB
• HDD Storage: 10 TB is enough even if 15TB is recommended
• SSD Storage: 1 TB (optional – for temporary files)

Ancillary links and documents:

• Sentinel-2 for Agriculture (esa-sen2agri.org)

Who to contact for support.

• GEO Secretariat: secretariat@geosec.org specifying is regarding support for the use of the Sen2Agri Solution in the GEO Knowledge HUB
• Forum Sen2Agri: https://forum.esa-sen2agri.org/

Knowledge Resources