Skip to content

CADWRDeltaModeling/oceantracker

 
 

Repository files navigation

OceanTracker

Fast particle tracking in structure and unstructured grids

OceanTracker is a fast extendable code for offline particle tracking in structured and unstructured grids [1]. OceanTracker currently supports structured grid ROMs and GLORYS (NEMO/COPERNICUS) for fixed z level vertical grids and unstructured grids of SCHISM and DELFT3D-FM.

Fast computational approaches to numerical interpolation makes OceanTracker extremely fast even on a single computer core [2]. In addition, computations are split to run in parallel across available physical computer cores.

OceanTracker’s speed enables millions of particles to be simulated. This significantly increases the range of particle behaviours that can be modeled and the quality of statistics derived from the particles. To eliminate the need to store and wade through the analysis of vast volumes of recorded particle tracks, the code has the ability to calculate statistics on the fly, such as heat maps and connectivity between regions.

OceanTracker code is highly flexible and extendable by the user, whether run by a new user with a text file of parameters, or by an expert adding their specialised code for novel particle behaviours or statistics, to the computational pipe line

More about Ocean tracker and gallery

Installing Oceantracker

Oceantracker User Guide

Running Oceantracker

Howto tutorials

[1] Vennell, R.,Steidle. L., Scheel, M.,Chaput. R., Knight, B. and Smeaton, M, OceanTracker 0.5: Fast Adaptable Lagrangian Particle Tracking in Structured and Unstructured Grid, preprint, https://eartharxiv.org/repository/view/8387/

[2] Vennell, R., Scheel, M., Weppe, S., Knight, B. and Smeaton, M., 2021. Fast lagrangian particle tracking in unstructured ocean model grids , Ocean Dynamics, 71(4), pp.423-437.

About

Fast Lagrangian Ocean Particle Tracking-Unstructured Grids

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 71.4%
  • HTML 23.1%
  • Python 5.4%
  • Other 0.1%