BIOLOGGING TOOLKIT

This package aims to handle biologging datasets, from raw data processing to specific applications on acoustic or inertial data. Most processing functions automatically access DTAG4 data, but classes also handle raw inputs.

It is divided into different sections :

• processing : Creates finalized dataset of processed data (sound pressure level data, animal posture and heading, jerk data, etc.) from raw data.
• applications : Uses the finalized dataset for specific use cases (wind estimation from acoustics, prey-catch attempts, drift dive detections, etc.).
• utils : Contains specific python functions used in the different modules.
• auxiliary : Python codes to download auxiliary data (sun position, all ERA5-reanalysis data)
• plot : Functions called in notebooks for interactive plotting

This package is meant to grow throughout the years, so please don't hesitate to contribute ! The main objective is to provide general codes with associated notebooks so that different users can have the same tools and parameters for their data analysis.

How to install the package

Download the package and unzip it or run the following command :

git clone https://github.com/gmanatole/SES_tags.git

Then install the necessary packages or run :

pip -r requirements.txt

How to run the code

Most code behaves in the same way.

Here is a simple example of how you can run the Acoustic class :

depid = 'ml18_185b'  # Enter name of deployement ID (Will be the name of the processed dataset created)
path = '/home/user/data/' #Enter path where dataset should be saved
raw_path = '/home/data/ml18_185b/'  #Enter path where data (raw (svw + xml) or sens5 for example) is saved | raw_path might also be name acoustic_path or inertial_path depending on the data handled
data_normalization = 'instrument'   # How to normalize data. If you don't know the instrument metadata, enter zscore
instrument = {'gain_dB':12, 'sensitivity':175, 'peak_voltage':1}
acoustic = Acoustic(depid, path=path, raw_path = raw_path, instrument = instrument)

acoustic.N = 60   # Modify timestep of dataset (in seconds). Defaults to 3.
acoustic()    # Get power spectral densities at desired timestep for all frequencies