Output of vertical profile data is enabled by providing values for the run parameters dt_dopr and data_output_pr which specify the temporal interval of the output times (in s) and the quantities for which output is to be made. Output times cannot be defined directly but only via the output time interval, starting from the beginning of the initial 3d run (t=0, but no profiles are written at the time t=0; exceptions see data_output_pr). By default, vertical profiles represent an instantaneous horizontal average of the total model domain at the respective output time. Temporal averaging can be initiated with the run parameter averaging_interval_pr. Beyond that, an output of profiles for user-defined subdomains is possible via the initialization parameter statistic_regions.
Output of desired profiles (temperature, velocity, etc.) is steered with the run parameter data_output_pr. If this parameter is not set by the user, then, independently of the value of dt_dopr, there will be no output! PALM creates the local file PLOT1D_DATA, which must be linked with a permanent file (see e.g. chapter 3.2) via a file connection statement in the mrun configuration file. At the end of the run the local file is copied to this file. Such a statement can look like this:
PLOT1D_DATA out:loc:tr pr ~/PLOT/$fname _pr_in.
If the respective mrun
mrun -d test -r “pr” …
then the local file PLOT1D_DATA is copied to the permanent file ~/PLOT/test/test_pr_in. However, the character string “pr” activating the file connection statement (see third column of the statement) must be given in the mrun call as argument of the option -r (and/or -o). If this is forgotten by mistake, the model outputs the data to the local file, but this is not copied to the permanent file and thus the data are not available for the user after the run has finished.
The created data file can be directly used as input file for the plot software profil. profil needs another file, the so-called parameter file, which specifies the plot layout. This file is created by PALM under the local name PLOT1D_PAR and also needs a file connection statement, e.g.:
PLOT1D_PAR out:loc:tr pr ~/PLOT/$fname _pr_par.
If you are in the directory ~/PLOT/test,the data can be plotted by entering
profil -d test_pr_in -p test_pr_par
The default plot layout as specified in the parameter file is as follows: the profiles of the different quantities are plottet into one or more coordinate systems (panels). Every profile is plotted only once, however several profiles of different quantities can be plottet into the same panel. If case of this, the different quantities are marked by different styles of lines. Which variables are drawn into which coordinate system is specified by the user via the run parameter cross_profiles. It is very important to note that only profiles are plottet, which are assigned to a coordinate system via cross_profiles. If a certain variable is selected by data_output_pr but not assigned to a coordinate system via cross_profiles, then the appropriate profile data are written to the file PLOT1D_DATA but they are not plotted! All panels for which data exist are plotted. For example, if only the profiles of the potential temperature are to be plottet in a panel but output of potential temperature is not set via data_output_pr, then the respective panel is not drawn (thus no 'empty' panels appear in the plot). The parameters profile_columns and profile_rows determine how the panels are arranged in columns and rows. The panels are plotted in the order given by data_output_pr starting in the top row from left to right. If the number of panels to be plotted is larger than the product profile_columns * profile_rows, then the plot is continued on a second page.
During a model run output of profiles to file PLOT1D_DATA is usually carried out for several times (as determined by dt_dopr). All of these profiles belonging to the same quantity are plotted in the same panel. The different output times are marked by different line colors. All profiles of a certain output time have the same color, even if they are plotted into different panels. Beyond that, a legend entry is made for each, containing the name of the respective quantity and the output time. This legend string is taken from file PLOT1D_DATA in which a comment line is placed ahead of each profile data.
For identification, each plot is provided with a title, which contains the model version number, the run identifier (base file name + number of the restart run), the name of the host where the run was executed, the date and time of the beginning of the run as well as the averaging domain (by default: 'total domain'). If the profiles are additionally temporally averaged, each panel gets an appropriate sub-label ('time averaged over… s'). The axes labels of the x axes of the individual panels are specified with the run parameter cross_xtext, the y axes always have the label 'height in m'. For better comparison, the profiles can be normalized with respect to different quantities (see cross_normalized_x and cross_normalized_y). The respective normalization is noted in the axes label.
As already mentioned, the profiles of a quantity for all output times are plotted into the same panel. This is not possible if a simulation needs restart runs, since the restart runs write the profile data to separate files (with appropriate cycle numbers) and create own parameter files. Thus by default profiles can only be drawn separately for each respective run. However there is a possibility to plot all profiles of a quantity in one plot, regardless of the number of restart runs. For this purpose the configuration file must be changed such that for restart runs the profile data are appended to already existing data (thus all data are output into one file). Then the configuration file has two entries for PLOT1D_DATA:
PLOT1D_DATA out:loc:tr pr# ~/PLOT/$fname _pr_in
PLOT1D_DATA out:loc:tra prf ~/PLOT/$fname _pr_in.
The first run of the job chain is now
mrun - D test - r “pr#…
For the restart runs, the "#" in the mrun call is automatically replaced by an "f". At the end of the job chain there will only be one data file containing the complete profile data. The file connection statement for the parameter file PLOT1D_PAR has not changed, thus a new cycle of the respective permanent file is created for each run. To plot the combined data with the parameter file created by the last run of the job chain, the run parameter use_prior_plot1d_parameters = .TRUE. must be additionally set for the restart runs. If this is forgotten, the parameter file is useless and the data cannot be plotted.
The default layout of the
vertical profiles can be changed by editing the
parameter file. For a detailed description of the profil parameters
manual (only in German).
Last change: $Id: chapter_4.5.2.html 62 2007-03-13 02:52:40Z raasch $