Installation

Requirements

To be able to install and use the s3lst-ds package in your project, you would need:

An Unix-like environment.
uv project manager.
A CDSE account.

Installation Steps

Install package from PyPI

Install the latest stable release from PyPI in the activated virtual environment using uv:
```
uv add s3lst-ds
```

Set CDSE credentials

Safely set your CDSE credentials as environment variables of the system:
```
uv run s3lst-ds-set-cdse
```

Install `esa-snappy` (optional)

Install esa-snappy Python package using uv:
```
uv add s3lst-ds[snap]
```
Install the backend SNAP Java package (if not already installed) and subsequently configure it using uv:
```
uv run s3lst-ds-install-snap
```

Note

CDSE credentials

CDSE credentials are required to download Sentinel-3 data. Script s3lst-ds-set-cdse will prompt you to provide their CDSE mail and password. The script will subsequently write them to file ~/.config/cdse_credentials.sh with user-only read and write permissions and source it in ~/.bashrc file. You may check the created credentials file using the command:

nano ~/.config/cdse_credentials.sh

Note that if you would like to remove the credentials from the file at a later time, you may run:

uv run s3lst-ds-unset-cdse

Better downscaling results may be obtained with `esa-snappy`

The default installation considers rioxarray for georeferencing the downloaded Sentinel-3 products. However, esa-snappy has been found to produce better results, and, because of that, it is herein availed as an optional tool. With its installation, both tools can be used for georeferencing.

Uninstall `SNAP`

If the you would to like to uninstall SNAP at a later time, you may run:

uv run s3lst-ds-uninstall-snap

`SNAP`'s memory limit

It is important to note that SNAP is configured with a limited amount of memory. To increase SNAP's maximum heap memory to for instance 64 GB, you would need to open file esa_snappy.ini nested in the installation directory of the virtual environment (herein assumed to be VENV) by doing

nano VENV/lib/python3.12/site-packages/esa_snappy/esa_snappy.ini

and writing in it:

[DEFAULT]
java_max_mem: 64G