This repository contains a python-based tool to extract lake subsets from the global ESA CCI Lakes dataset based on the code from the LakeCREST project.

This tool allows the user to specify the necessary lakes, variables and time-range to extract. The extraction is then executed using the CCI lakes maskfile and will create a merged netCDF file per lake. The extraction can be either run on a local copy of the Lakes CCI dataset or directly from the online source using OPeNDAP (slow!).

To run the script the following packages are needed:

• netCDF4
• numpy
• pandas

• Build a conda environment from the provided YAML file in setup/environment.yaml:
  conda create --name lakecrest_tools --file setup/environment.yaml

• (or) Install the necessary packages to an existing conda environment:
  conda install netCDF4 numpy pandas

• (or) Install the necessary packages listed in the dependencies using pip:
  pip install netCDF4 numpy pandas

To use the script efficiently it is necessary to move a local copy of the CCI Lakes dataset into the data/raw/ folder. For example, if using the CCI Lakes v2.0.2 dataset create a copy of the dataset with the yearly folders located within data/raw/v2.0.2/.

Before running the extraction from main.py the user can set the desired extraction parameters in the file. By changing the variable VERSION in scripts/constants.py it is possible to run the extraction for older CCI Lakes versions. Currently the necessary lakemask and data availability table are only provided for v2.0.2 and v2.0.1. If older CCI Lakes versions are needed the necessary files have to be provided by the user in data/auxiliary/.