OOI Profiler Nitrate Data Retriever

This package simplifies the retrieval of in situ nitrate data from Ocean Observatories Initiative Endurance Array profiling moorings. The retrieved data is then binned and quality controlled before being saved to the local machine.

Installation

Assuming you have Python and conda installed:

1. Clone the repository and switch to the directory:

   git clone https://github.com/andrew-s28/ooi-profiler-nitrate-retriever.git
   cd ooi-profiler-nitrate-retriever/
   

2. Create and activate a conda environment:

   conda env create -f environment.yml
   conda activate ooi
   

3. You can now retrieve data:

   python ./scripts/retrieve_profiler_data.py CE01ISSP -s 2017-08-15 -e 2020-12-16 -p ./data
   

Usage

The only stations currently implemented are the Oregon Inshore Surface Piercing Profiler (CE01ISSP) and the Oregon Shelf Surface Piercing Profiler Mooring (CE02SHSP). This script accesses the NUTNRJ dataset, which includes in situ nitrate concentration data as well as typical oceanographic data (e.g., salinity, temperature, and density). Both a binned dataset and the raw data without any modification are saved to the local machine.

⚠️ The binning process is resource intensive: The binning process requires using the groupby operation, which is expensive in xarray. Optimization using flox to speed up the groupby has been implemented, but the code can still take ten to twenty minutes to finish if downloading all available data onto a typical consumer machine.

Only a site name is required. The only currently implemented sites are CE01ISSP and CE02SHSP. Separate sites by a space to get data from more than one:

python ./scripts/retrieve_profiler_data.py CE01ISSP CE02SHSP

Find help using the --help flag:

python ./scripts/retrieve_profiler_data.py --help

usage: retrieve_profiler_data.py [-h] [-p [path]] [-f [file]] [-s [start]] [-e [end]] [-n [njobs]] site [site ...]

Download OOI Endurance Array profiler nitrate data, QC fits, bin, and save.

positional arguments:
  site                  Site to download from, either CE01ISSP or CE02SHSP

options:
  -h, --help            show this help message and exit
  -p [path], --path [path]
                        Path to folder for saving output files
  -s [start], --start [start]
                        Start date in YYYY-MM-DD format (default: 2010-01-01)
  -e [end], --end [end]
                        End date in YYYY-MM-DD format (default: 2024-06-16)
  -n [njobs], --njobs [njobs]
                        number of jobs to run in parallel when running quality control - see joblib.Parallel docs for more details

Example

You can view an interactive Python notebook in the examples directory that goes through each step used in the data retriever script. After cloning the repository and setting up your environment, simply open and run the notebook ooi_profiler_nitrate_analysis.ipynb to view the steps taken in the access, quality control, binning, and validation of OOI profiler nitrate datasets.

License

This package is licensed under the MIT License. See the LICENSE file for more information.