... d'Aérologie1
Corresponding author's address: 14 avenue E.Belin, F-31400 Toulouse, France (ghef@aero.obs-mip.fr). Comments that could help to improve this document are welcome!
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... documentations.2
available at http://www.aero.obs-mip.fr/mesonh/
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... discussion1.1
Suggestions are welcome!
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... origin)1.2
Remark: It is not necessary to calculate the evolution of a specific triplet of initial coordinates for each time origin, but actually sufficient to know the evolution of the single triplet $(x_{t_i},y_{t_i},z_{t_i})$ for each time interval $t_i <
t \leq t_{i+1}$, even if one wants to retrieve Lagrangian motions over longer periods. During the model run, computer memory and calculations thus remain required for never more than one triplet in the same time, long-range Lagrangian evolutions being obtained post-mortem as shown presently.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... grid1.3
see the MesoNH scientific documentation.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... transports)1.4
Note however a small difference with standard scalar variables. Since the latter are generally concentrations of atmospheric constituents, the numerical schemes are designed to ensure positivity. However the Lagrangian scalars $x_0$ and $y_0$ might have negative values (due to the inflow prescription at the lateral boundaries, see Table 1.1), so the minimum possible values is shifted down to -1.E+9 m (parameters XLG1MIN and XLG2MIN in moddlg.f90).
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... "father"1.5
i.e., the model in which the considered model is nested.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... list2.1
In reality CFILES(1) is never used by DIAG and therefore has not necessarily to be a segment end.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... files2.2
Except YINIFILE(1).
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... segments2.3
This is however not a theoretical constraint and DIAG will be improved in future versions.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... outside"2.4
Origins from outside the model domain may thus be given by the model fields LGXM, LGYM, LGZM or LGXT, LGYT, LGZT (bye virtue of the boundary conditions given in Table 1.1), but not by the post-processed fields X00i, Y00i, Z00i.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... frontogenesis2.5
...Advertisement... Schär and Wernli (1993) also recommend tracers of the geographical origin, such as $x_0$ or $y_0$, as excellent markers of fronts and frontogenesis. Try these, you won't be disappointed !
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.