Sign in

Parcels

Parcels (Probably A Really Computationally Efficient Lagrangian Simulator) is a set of Python classes and methods to create customisable particle tracking simulations using output from Ocean Circulation models. Parcels can be used to track passive and active particulates such as plastic and fish.

91
mentions
24
contributors

Cite this software

DOI:

10.5281/zenodo.823561

What Parcels can do for you

Parcels (which is a ‘backronym’ for “Probably A Really Computationally Efficient Lagrangian Simulator”) is a set of Python classes and methods to create customisable particle tracking simulations using output from Ocean Circulation models. Parcels can be used to track passive water parcels in what is often referred to as Lagrangian Ocean Analysis (Van Sebille et al 2018), as well as active particulates such as plastic and fish.

Parcels has been used in more than 80 peer-reviewed articles. It has been used in applications as diverse as global oceanic transport of microplastic particles (e.g., Wichmann et al 2019), planning of routes of Underwater Autonomous Vehicles (Boulares and Barnawi 2021), mapping of the genetic connectivity and of fish (Sefc et al 2020, Schilling et al 2020, Lindo‐Atichati et al 2020) and assessment of the role of swimming in turtle hatchling survival (Le Gouvello et al 2020).

Parcels supports a multitude of Ocean General Circulation Model outputs, including from NEMO, MITgcm, HYCOM, POP, MOM and ROMS/CROCO. In general, it supports any vertical grid and (curvi)linear grid with the only limitation that it does not yet support unstructured meshes.

The unique feature of Parcels, compared to other community codes such as Ariane (Blanke and Raynaud 1997), TRACMASS (Döös et al 2017), OpenDrift (Dagestad et al 2018), CMS (Paris et al 2013), LADIM (Ådlandsvik and Sundby 1994) or TrackMPD (Jalón-Rojas et al 2019), is that in Parcels it is extremely simple to encode the ‘behavior’ of particles.

Using what are called ‘kernels’ in Parcels, users can write simple functions that act on the virtual particles and make them e.g., sink, beach, fragment or die. These actions can depend on the local conditions – such as sea water temperature, current speed or any other field that comes out of an Ocean General Circulation Model – as well as on properties of the particles themselves such as their age, density, or size. These kernels can then be combined with each other and with pre-loaded kernels such as for advection and diffusion. This kernel-driven computation makes Parcels highly versatile and adaptable to different types of virtual particles, including plastic and plankton as well as turtles and tuna (Scutt Phillips et al 2016).

Keywords
Programming language
  • Jupyter Notebook 93%
  • Python 7%
License
  • MIT
</>Source code

Participating organisations

Utrecht University

Mentions

Getting started with Parcels: general structure

Contributors

AG
Angus Gibson
DR
Daan Reijnders
DW
EvS
Erik van Sebille
JP
James Pringle
JB
Julius Busecke
Christian Albrechts Universität zu Kiel
MS
Miriam Sterl
Royal Netherlands Institute for Sea Research
NV
Noam Vogt-Vincent
RB
Roman Bezhenar
Institute of Mathematical Machines & Systems Problems of National Academy of Sciences of Ukraine (IMMSP)
SY
Stefanie Ypma