4.1 The Trajectory Calculation




In the previous trajectory calculation to test the model installation, after selecting the source location, the sequence of simulation configuration, model execution, and trajectory display, was entirely automatic, with each step hidden within the GUI script. In all subsequent menus, each step will be addressed in sequence.

  1. To start a new trajectory calculation, from the top bar of the main menu select Trajectory / Setup Run which will open a menu with multiple entry options.

  2. The Trajectory setup menu is used to enter information, such as the starting time on the first line. Enter 83 09 25 17 to correspond with the time of the start of the tracer release (1700 UTC 25 September 1983).

  3. On the second line change the number of starting locations to 1 and then press the Setup starting locations button to open a new window where you should now enter 39.90 -84.22 10.0, which corresponds to the latitude and longitude of the tracer release location at 39.90N, 84.22W, at a starting height of 10 meters above ground level.

  4. The other options on the setup menu will be discussed in more detail later, but for now enter total run time of 68 hours (the trajectory duration). The vertical motion method should be using the data field from the meteorological input file, the name of the output file with the trajectory end point latitude-longitudes should be called tdump, and after pressing the clear button add the meteorological data file captex2_narr.bin in the captex tutorial directory.

  5. After all the changes have been made to the setup menu, press the Save button which copies all the information to a special file called default_traj and then closes the menu. Now the model simulation can be started from the Trajectory / Run model menu tab. This button causes two actions. First, the trajectory setup information is copied from default_traj to a file called CONTROL, which is the input name required by the HYSPLIT trajectory model executable (hyts_std.exe). Second, the GUI then launches the model.

  6. As the model is running a progress message percent complete is displayed. Do not try to Exit this window until the calculation is complete or a beep is heard indicating a premature stop.

  7. By default all calculations take place within the \hysplit4\working directory. The calculation created the trajectory endpoints file tdump. To turn this ASCII file into a graphic, open the Trajectory / Display / Trajectory menu.

  8. This multi-entry Trajectory Display menu provides for several options, which will be discussed in more detail later as needed. For now, no changes are required and the tdump input file will be converted to a Postscript output file called trajplot.ps using the default map background file.

  9. Pressing the Execute Display menu button will run the trajplot program and then open the Ghostscript viewer to display the trajectory graphic.

  10. If your simulation result matched the previous picture, then you should save the information contained in the Trajectory setup menu by selecting the Save as button from the trajectory setup menu and entering a file name such as traj_base_control.txt. The default directory for these Save as actions is in the working directory. In the future, this information can be loaded using the Retrieve menu to reload the simulation information into the GUI. An example of the CONTROL file can also be found in the tutorial files directory.

In summary, the basic trajectory calculation consists of three steps: setup, run, and display. Note how the low-level trajectory (started at 10 m AGL) follows the southern shoreline of Lake Ontario and Lake Erie. From the sampling data we already know the tracer from this short 3 hour release was spread all over New York and Pennsylvania, and even into southern Canada, suggesting that a single trajectory is insufficient to represent this pollutant dispersion episode. In subsequent sections, we will explore the different applications of the trajectory calculation as well has how multiple particle trajectories can be used to more accurately simulate pollutant dispersion.