HYSPLIT Forum: hysplitbbs.arl.noaa.gov

Hi
As we know the HYSPLIT is a Lagrangian Model and calculate Trajectories and concentrations through Lagrangian equations. puff(lagrangian-Gaussian and Lagrangian-top hat) and particle(Lagrangian-trajectory) model are the models which HYSPLIT use. but the question is that in the Draxler and Hess 1997 mentioned that "The model calculation method is a hybrid between Eulerian and Lagrangian approaches, advection, and diffusion(include dispersion and diffusion) calculations are made in Lagrangian framework while concentrations are calculated on a fixed grid(refer to Eulerian model)." I have tried to understand the definition of latter phrase but it has still remained unanswered for me, I conclude that all the equations used in model calculations are based on Lagrangian equations but there is an assimilation with a Eulerian approach to calculate the concentration value in a grid. Am I right? If not can you elaborate on this, please?

Hi, yes, I think that you are "right" in the way you are thinking about this, but I will try to elaborate briefly.

The quoted statement you mention is contained in the following reference: Draxler, R. R. and G. D. Hess (1998). "An overview of the HYSPLIT_4 modeling system of trajectories, dispersion, and deposition." Aust. Meteor. Mag. 47: 295-308.
{I found this reference online today at: http://citeseerx.ist.psu.edu/viewdoc/do ... 1&type=pdf}

As you have surmised, in this context, the Eulerian descriptor simply refers to the fact that the Lagrangian dispersion results -- calculated via Lagrangian particles or puffs -- are aggregated onto a user-defined concentration grid. This grid (or grids) is defined in the CONTROL file input for a given HYSPLIT simulation. Note that these concentration grids are used in a simple way. For particles or puffs, the model simply adds up the mass in each grid cell in each sampling interval. Remember, the grid cell dimensions (i.e., lat-long spacing, extent, vertical levels) and sampling frequency and type of sampling are defined by the user in the CONTROL file. The use of the Eulerian descriptor here does not mean that the dispersion itself is being simulated via an Eulerian approach, in this context.

However, it should also be pointed out that in recent years, an actual Eulerian dispersion simulation modeling capability has been added to the HYSPLIT model as an option. This "Global Eulerian Model" or "GEM" is now a form of the model that can be run. In the Users Guide http://www.arl.noaa.gov/documents/repor ... _guide.pdf, this is currently found in Concentration / Special Runs / Global. When using this new functionality of the model, the simulation can be done in "pure" Eulerian fashion, in which dispersion and everything is done in a gridded fashion, or it can be done in a new hybrid manner in which dispersion is initially done with a Lagrangian approach and then at some user-defined age after emission, the mass of particles or puffs are numerically transferred to an Eulerian grid and the dispersion of that mass is continued within the simulation using an actual Eulerian dispersion methodology.