How to set up the NUTS environment

The Alia-Space NUTS project provides an online tool to allow users obtaining merged and cropped data by selecting a data product, a Region of Interest (ROI), and a time period.

Set up the CONDA environment

To work with NUTS, start by creating a new conda environment. Let's call it nuts_env, but any valid name would do (change the following instructions accordingly):

conda create -n nuts_env python=3.8 geopandas owslib

Activate the environment:

conda activate nuts_env

Once the environment is activated, install the required libraries to work with Prefect and the Postgres DB:

conda install -y -c conda-forge prefect psycopg2

You are almost ready to execute your flow!

In the following paragraphs, we show you how to start a Prefect flow and an example of a request.

Prefect set up

Open a terminal and navigate to the repository folder. You should see the docker-compose.yaml, the create_db.sh, and the SENTINEL_workflow.py/DEM_workflow.py files in the current folder. If so, you are in the right place.

Once there, you have to execute the following commands:

docker-compose up

Open a new terminal (the docker output occupies the previous one) and go to the same folder. The first time you start up the system, execute the following commands:

./create_db.sh

prefect server create-tenant --name NUTS_tenant

Finally, go to the page localhost:8080 in a local browser and create a new project in the Prefect UI. Call the new project NUTS.

Now, you are ready to execute the last two commands in a terminal pointing to the repository path:

python SENTINEL_workflow.py

prefect agent local start

All you have to do now is connect to the DB to fill your request!

NOTE: if you want to run the DEM flow, use the following command instead of python NUTS_workflow.py:

python DEM_workflow.py

Connect to the DB

Open the Postgres IDE and connect to the NUTS DB.

name: nuts_db host: localhost port: 5432 username: nuts_user password: nuts_password

Once connected, you should see the following tables:

• user
• order_table
• job
• source_product

Case study

To enable a request in the Prefect flow, you must fill the user table and the order_table one in the DB. Specifically:

user:

user_id (a unique number is automatically assigned for each DB row inserting the name)

name

email address

order_table:

user_id (it should correspond to the unique user identifier)

shape_layer (i.e., NUTS_2)

shape_roi (i.e., Lazio)

source_mission (i.e., Sentinel-2)

source_data_type (i.e., S2MSI1C)

start_date (i.e., 2021-09-01)

stop_date (i.e., 2021-09-05)

This request will make you download 28 products of Sentinel-2 data.

To execute a run in which DEM data is requested, follow the example below:

user:

user_id (a unique number is automatically assigned for each DB row inserting the name)

name

email address

order_table:

user_id (it should correspond to the unique user identifier)

shape_layer (i.e., GADM_1)

shape_roi (i.e., Qina)

source_mission (i.e., DEM)

source_data_type (i.e., DS)

start_date (i.e., 2020-01-01)

stop_date (i.e., 2020-12-31)

Prefect is ready to execute your order: in the Prefect web interface, go to flow and select NUTS_workflow. Push the RUN button and insert the order_id associated to the DB request you want to perform.

Have fun!!