Lagrangian Particle Tracking
- Kester Scandrett
- Bui Bich Thuy (Deactivated)
- Duy Pham
- Nguyen Quang Trang (Deactivated)
EFDC+ Explorer incorporates the pre-and post-processing of Lagrangian particle track (LPT). The use of LPT's can be helpful when modeling oil spill tracks, emergency response, water quality applications, and plume tracking. When the Lagrangian Particle Tracking module is turned on, a LMC on the menu displays a report, as shown in Figure 1. Here the user can see the total number of particles and the number of groups that have been set, as well as the time for the release of the drifters and the end time for the observation of the drifters. Models with as many as 1.5 million drifters have been simulated, though loading time can increase with a large number of drifters.
Figure 1 Lagrangian Particle Tracking report.
The EE pre-processor provides full control for initial particle seeding, LPT computational option selection, and plotting. EE allows for a range of display options for the tracks, animations to the screen and or AVI files, and the ability to export any or all of the particle tracks to ASCII files. The LPT sub-model has been implemented with the following major options:
- Particles are free to move in full 3D,
- Particles can be fixed at a user-specified depth, and
- A random walk component can be added to the two options above.
To set the values related to the drifters, the user should RMC on the LPT tab in the Model Control form to open LPT Options form, as shown in Lagrangian Particle Tracking#Figure 2. This displays the various options for setting the drifters as described in detail in the following sections
Figure 2 Lagrangian Particle Tracking form.
Note that the numerical solution is separately divided into the advective transport and random components. This approach allows the user to enable (i.e., turn on random walk) or disable (advective transport only) the random components for either the horizontal and/or the vertical directions.
The method used to solve the differential equations is the Runge-Kutta 4 method: This method has the approximation of O(∆t4). It has been determined that the Runge-Kutta 4 method is preferred due to its higher level of numerical accuracy. It has been shown that the computational burden of the Runge-Kutta 4 method is not significant within the overall model run times compared to other methods tested.
The following links show how to configure and view LTP: