source: palm/trunk/DOC/app/chapter_4.2.html @ 61

Last change on this file since 61 was 61, checked in by raasch, 15 years ago

further preliminary changes for revision 3.2

  • Property svn:keywords set to Id
File size: 226.2 KB
Line 
1<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
2<html><head>
3<meta content="text/html; charset=ISO-8859-1" http-equiv="content-type"><title>PALM chapter 4.2</title></head>
4<body><h3 style="line-height: 100%;"><a name="Kapitel4.2"></a>4.2 <a href="#Laufparameter">Runtime
5parameters</a> and <a href="#Paketparameter">package
6parameters</a></h3>
7<h3 style="margin-bottom: 0cm; line-height: 100%;"><a name="Laufparameter"></a>Runtime parameters:</h3>
8<br><br><table style="text-align: left; width: 100%;" border="1" cellpadding="2" cellspacing="2"> <tbody>
9<tr>
10<td style="vertical-align: top;"><font size="4"><b>Parameter
11name</b></font></td> <td style="vertical-align: top;"><font size="4"><b>Type</b></font></td>
12<td style="vertical-align: top;"> <p><b><font size="4">Default</font></b> <br> <b><font size="4">value</font></b></p> </td>
13<td style="vertical-align: top;"><font size="4"><b>Explanation</b></font></td>
14</tr> <tr> <td style="vertical-align: top;"><a name="averaging_interval"></a><span style="font-weight: bold;">averaging_interval</span><br>
15</td> <td style="vertical-align: top;">R<br> </td>
16<td style="vertical-align: top;"><span style="font-style: italic;">0.0</span><br> </td>
17<td style="vertical-align: top;">Averaging interval for
18all&nbsp;output of temporally averaged data (in s).<br><br>This
19parameter defines the time interval length for temporally averaged data
20(vertical profiles, spectra, 2d cross-sections, 3d volume data). By
21default,&nbsp;data are not subject to temporal averaging. The
22interval
23length is limited by the parameter <a href="#dt_data_output_av">dt_data_output_av</a>.
24In any case, <span style="font-weight: bold;">averaging_interval</span>
25&lt;= <span style="font-weight: bold;">dt_data_output_av</span>
26must hold.<br><br>If
27an interval is defined, then by default the average is calculated from
28the data values of all timesteps lying within this interval. The number
29of time levels entering into the average can be reduced with the
30parameter <a href="#dt_averaging_input">dt_averaging_input</a>.<br><br>If
31an averaging interval can not be completed at the end of a run, it
32will be finished at the beginning of the next restart run. Thus for
33restart runs, averaging intervals do not
34necessarily begin at the beginning of the run.<br><br>Parameters
35<a href="#averaging_interval_pr">averaging_interval_pr</a>
36and <a href="#averaging_interval_sp">averaging_interval_sp</a>
37can be used to define different averaging intervals for vertical
38profile data and spectra, respectively.<br> </td> </tr>
39<tr> <td style="vertical-align: top;"> <p><a name="averaging_interval_pr"></a><b>averaging_interval_pr</b></p>
40</td> <td style="vertical-align: top;">R<br> </td>
41<td style="vertical-align: top;"><span style="font-style: italic;">value of <a href="#averaging_interval">averaging_<br>
42interval</a><br>
43</span> </td> <td style="vertical-align: top;"><p>Averaging
44interval for output of vertical profiles&nbsp;to
45local
46file <font color="#000000"><font color="#000000"><a href="chapter_3.4.html#DATA_1D_PR_NETCDF">DATA_1D_PR_NETCDF</a>
47</font></font>and/or&nbsp; <a href="chapter_3.4.html#PLOT1D_DATA">PLOT1D_DATA</a>
48(in s).&nbsp; </p> <p>If
49this parameter is given a non-zero value, temporally
50averaged vertical profile data are output. By default, profile data
51data are not subject to temporal averaging. The interval length is
52limited by the parameter <a href="#dt_dopr">dt_dopr</a>.
53In any case <b>averaging_interval_pr</b> &lt;= <b>dt_dopr
54</b>must
55hold.</p>If an interval is defined, then by default the average
56is calculated
57from the data values of all timesteps lying within this interval. The
58number of time levels entering into the average can be reduced with the
59parameter <a href="#dt_averaging_input_pr">dt_averaging_input_pr</a>.
60<p>If
61an averaging interval can not be completed at the end of a run, it will
62be finished at the beginning of the next restart run. Thus for restart
63runs, averaging intervals do not
64necessarily begin at the beginning of the run.</p> </td> </tr>
65<tr> <td style="vertical-align: top;"><a name="call_psolver_at_all_substeps"></a><span style="font-weight: bold;">call_psolver_at_all_<br>
66substeps</span></td> <td style="vertical-align: top;">L<br>
67</td> <td style="vertical-align: top;"><span style="font-style: italic;">.T.</span><br> </td>
68<td style="vertical-align: top;">Switch
69to steer the call of the pressure solver.<br> <br>
70In order to speed-up performance, the Poisson equation for perturbation
71pressure (see <a href="#psolver">psolver</a>) can
72be called only at the last substep of multistep Runge-Kutta
73timestep schemes (see <a href="chapter_4.1.html#timestep_scheme">timestep_scheme</a>)
74by setting <span style="font-weight: bold;">call_psolver_at_all_substeps</span>
75= <span style="font-style: italic;">.F.</span>.
76In many cases, this sufficiently reduces the divergence of the velocity
77field. Nevertheless, small-scale ripples (2-delta-x) may occur. In this
78case and in case
79of non-cyclic lateral boundary conditions, <span style="font-weight: bold;">call_psolver_at_all_timesteps</span>
80= <span style="font-style: italic;">.T.</span>
81should be used.&nbsp;<span style="font-weight: bold;"></span></td>
82</tr> <tr> <td style="vertical-align: top;"><p><a name="fcl_factor"></a><b>cfl_factor</b></p>
83</td> <td style="vertical-align: top;">R<br> </td>
84<td style="vertical-align: top;"> <p><i>0.1,
850.8 or 0.9</i> <br> <i>(see right)</i></p>
86</td> <td style="vertical-align: top;"> <p lang="en-GB">Time step limiting factor.&nbsp; </p>
87<p><span lang="en-GB">In the model, the <span lang="en-GB">maximum
88allowed </span>time step according to CFL and
89diffusion-criterion
90dt_max is reduced by </span><a href="chapter_4.1.html#dt"><span lang="en-GB">dt</span></a> <span lang="en-GB">=
91dt_max * <b>cfl_factor</b>
92in order to avoid stability problems which may arise in the vicinity of
93the maximum allowed timestep. The condition <i>0.0</i>
94&lt; <b>cfl_factor</b>
95&lt; <i>1.0 </i>applies.<br> </span></p>
96<p><span lang="en-GB">The default value of
97cfl_factor depends on
98the </span><a href="chapter_4.1.html#timestep_scheme"><span lang="en-GB">timestep_scheme</span></a><span lang="en-GB"> used:<br> </span></p> <p><span lang="en-GB">For the third order Runge-Kutta scheme it
99is <b>cfl_factor</b> = </span><span style="font-style: italic;">0.9</span><span lang="en-GB">.<br> </span></p> <p><span lang="en-GB">In case of the leapfrog scheme a quite
100restrictive value of <span style="font-weight: bold;">cfl_factor</span>
101= <span style="font-style: italic;">0.1 </span></span><span lang="en-GB">is used because for larger values the velocity
102divergence
103significantly effects the accuracy of the model results.</span><a href="chapter_4.1.html#scalar_advec"><span lang="en-GB"></span></a><span lang="en-GB"> Possibly larger values may
104be used with the leapfrog scheme but these are to be determined by
105appropriate test runs.<span style="font-family: times new roman;"><br>
106</span></span></p> <span lang="en-GB"><span style="font-family: times new roman;"></span><font face="Times New Roman">The default value for the Euler
107scheme is <span style="font-weight: bold;">cfl_factor</span>
108= <span style="font-style: italic;">0.8</span> .</font></span></td>
109</tr><tr> <td style="vertical-align: top;"> <p><a name="create_disturbances"></a><b>create_disturbances</b></p>
110</td> <td style="vertical-align: top;">L<br> </td>
111<td style="vertical-align: top;"><span style="font-style: italic;">.T.</span><br> </td>
112<td style="vertical-align: top;"> <p>Switch to
113impose random perturbations to the horizontal
114velocity field.&nbsp; </p> <p>With <b>create_disturbances</b>
115= <i>.T.,</i> random
116perturbations can be imposed to the horizontal velocity field at
117certain times e.g. in order to trigger off the onset of convection,
118etc..<br> </p> <p>The temporal interval between
119these times can be steered with <a href="#dt_disturb">dt_disturb</a>,
120the vertical range of the perturbations with <a href="#disturbance_level_b">disturbance_level_b</a>
121and <a href="#disturbance_level_t">disturbance_level_t</a>,
122and the perturbation amplitude with <a href="#disturbance_amplitude">disturbance_amplitude</a>.
123In case of non-cyclic lateral boundary conditions (see <a href="chapter_4.1.html#bc_lr">bc_lr</a>
124and <a href="chapter_4.1.html#bc_ns">bc_ns</a>),
125the horizontal range of the perturbations is determined by <a href="chapter_4.1.html#inflow_disturbance_begin">inflow_disturbance_begin</a>
126and <a href="chapter_4.1.html#inflow_disturbance_end">inflow_disturbance_end</a>.
127A perturbation is added to each grid point with its individual value
128determined by multiplying the disturbance amplitude with a uniformly
129distributed random number.
130After this, the arrays of u and v are smoothed by applying a
131Shuman-filter twice and made divergence-free by applying the pressure
132solver.<br> </p> <p>The random number generator to
133be used can be chosen with <a href="chapter_4.1.html#random_generator">random_generator</a>.<br>
134</p> <p>As soon as the desired flow features have
135developed
136(e.g.&nbsp; convection has started), further imposing of
137perturbations
138is not necessary and can be omitted (this does not hold for non-cyclic
139lateral boundaries!). This can be steered by assigning
140an upper limit value for the perturbation energy (the perturbation
141energy is defined by the deviation of the velocity from the mean flow)
142using the parameter <a href="#disturbance_energy_limit">disturbance_energy_limit</a>.
143As soon as the perturbation energy has exceeded this energy limit, no
144more random perturbations are assigned<br>
145.&nbsp; <br>
146Timesteps where a random perturbation has been imposed are marked in
147the local file <a href="chapter_3.4.html#RUN_CONTROL">RUN_CONTROL</a>
148by the character "D" appended to the values of the maximum horizontal
149velocities. </p> </td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="cross_normalized_x"></a><b>cross_normalized_x</b></p>
150</td> <td style="vertical-align: top;">C*10&nbsp;
151<br>&nbsp;&nbsp; (100)</td> <td style="vertical-align: top;"><i>100 * ' '</i></td>
152<td style="vertical-align: top;"> <p>Type of
153normalization applied to the x-coordinate of vertical
154profiles to be plotted with <span style="font-weight: bold;">profil</span>.</p>
155<p>This parameter only applies for &nbsp;<a href="chapter_4.2.html#data_output_format">data_output_format</a>
156= <span style="font-style: italic;">'profil'</span>.</p><p>If
157vertical profiles are plotted with the plot software <span style="font-weight: bold;">profil</span> (data on
158local file <a href="chapter_3.4.html#PLOT1D_DATA">PLOT1D_DATA</a>,
159parameters on local file <a href="http://www.muk.uni-hannover.de/%7Eraasch/PALM_group/PLOT1D_PAR">PLOT1D_PAR</a>)
160the x-values of the data points can be normalized with respect to
161certain quantities (e.g. the near-surface heat flux) in order to ensure
162a better comparability. This type of normalization then applies to all
163profiles of one coordinate system (panel). The normalization quantities
164are re-calculated for each output time of each individual profile. If
165temporally averaged profiles are output (see <a href="#averaging_interval_pr">averaging_interval_pr</a>),
166then the normalization quantities are also temporally averaged
167accordingly. If the value of a normalization quantity becomes zero,
168then normalization for the total respective coordinate system (panel)
169is switched off automatically (also for the y-axis).<br> </p>
170<p>By default, the normalization quantities are calculated as the
171horizontal mean of the total model domain and and these values are also
172used for the normalization of profiles from subdomains (see <a href="chapter_4.1.html#statistic_regions">statistic_regions</a>).
173Instead of this, they can be calculated from the data of a certain
174subdomain by using the parameter <a href="#normalizing_region">normalizing_region</a>
175(however, they are used again for all subdomains and even for the total
176domain).&nbsp; </p> <p>The user can choose between
177the following normalization
178quantities: <br> </p> <table style="text-align: left; width: 100%;" cellpadding="2" cellspacing="2"> <tbody> <tr> <td style="vertical-align: top;"><i>'wpt0'</i></td>
179<td style="vertical-align: top;">Normalization with
180respect
181to the total surface sensible heat
182flux (k=0 ).</td> </tr> <tr> <td style="vertical-align: middle;"><i>'ws2'</i></td>
183<td style="vertical-align: top;">Normalization with
184respect
185to w<sub>*</sub> <sup>2</sup>
186(square of the characteristic vertical wind speed of the CBL)</td>
187</tr> <tr> <td style="vertical-align: top;"><i>'tsw2'</i></td>
188<td style="vertical-align: top;">Normalization with
189respect
190to the square of the characteristic
191temperature of the CBL theta<sub>*</sub> (this is defined
192as ratio of
193the surface heat flux and w<sub>*</sub>).</td> </tr>
194<tr> <td style="vertical-align: middle;"><i>'ws3'</i></td>
195<td style="vertical-align: top;">Normalization with
196respect
197to w<sub>*</sub> <sup>3</sup>.</td> </tr>
198<tr> <td style="vertical-align: middle;"><i>'ws2tsw'</i></td>
199<td style="vertical-align: top;">Normalization with
200respect
201to w<sub>*</sub><sup>2</sup>theta<sub>*</sub>
202(for definition of theta<sub>*</sub> see <span style="font-style: italic;">'tsw2'</span>).</td>
203</tr> <tr> <td style="vertical-align: middle;"><i>'wstsw2'</i></td>
204<td style="vertical-align: top;">Normalization with
205respect
206to w<sub>*</sub><sup>2 </sup>theta<sub>*</sub>
207(for definition of theta<sub>*</sub> see <span style="font-style: italic;">'tsw2'</span>).</td>
208</tr> </tbody> </table> <p>For each
209coordinate system (panel) to be drawn (see <a href="#cross_profiles">cross_profiles</a>)
210an individual normalization quantity can be assigned. For example: if <span style="font-weight: bold;">cross_normalized_x</span> =
211<span style="font-style: italic;">'ws2'</span><i>,'ws3'</i>,
212then the
213x-values in the 1st coordinate system are normalized with respect to w<sub>*</sub><sup>2</sup>
214and in the 2nd system with respect to w<sub>*</sub><sup>3</sup>.
215Data
216of the further coordinate systems (if any are to be drawn) are not
217normalized.&nbsp; </p> <p>Using a normalization
218leaves all vertical profile data on
219local file <a href="chapter_3.4.html#PLOT1D_DATA">PLOT1D_DATA</a>
220unaffected, it only affects the visualization. Within <span style="font-weight: bold;">profil</span>, the
221normalization is steered
222by parameter <a href="http://www.muk.uni-hannover.de/institut/software/profil_beschreibung.html#NORMX">normx</a>
223which may be changed subsequently by the user in the parameter file
224(local file <a href="chapter_3.4.html#PLOT1D_PAR">PLOT1D_PAR</a>).<br>
225&nbsp;<br>
226The assigned normalization quantity is noted in the axes labels of the
227respective coordinate systems (see <a href="#cross_xtext">cross_xtext</a>).</p>
228</td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="cross_normalized_y"></a><b>cross_normalized_y</b></p>
229</td> <td style="vertical-align: top;">C*10&nbsp;
230<br>&nbsp;&nbsp; (100)</td> <td style="vertical-align: top;"><i>100 * ' '</i></td>
231<td style="vertical-align: top;"> <p>Type of
232normalization applied to the y-coordinate of vertical
233profiles to be plotted with <span style="font-weight: bold;">profil</span>.&nbsp;</p>
234<p>This parameter only applies for &nbsp;<a href="chapter_4.2.html#data_output_format">data_output_format</a>
235= <span style="font-style: italic;">'profil'</span>.</p><p>If
236vertical profiles are plotted with the plot software <span style="font-weight: bold;">profil</span> (data on
237local file <a href="chapter_3.4.html#PLOT1D_DATA">PLOT1D_DATA</a>,
238parameter on local file <a href="http://www.muk.uni-hannover.de/%7Eraasch/PALM_group/PLOT1D_PAR">PLOT1D_PAR</a>)
239the y-values of the data points can be normalized with respect to
240certain quantities (at present only the normalization with respect to
241the boundary layer height is possible) in order to ensure a better
242comparability. </p> <p>The user can choose between the
243following normalization
244quantities: <br> </p> <table style="text-align: left; width: 100%;" cellpadding="2" cellspacing="2"> <tbody> <tr> <td style="vertical-align: top;"><i>'z_i'</i></td>
245<td style="vertical-align: top;">Normalization with
246respect
247to the boundary layer height
248(determined from the height where the heat flux achieves its minimum
249value).</td> </tr> </tbody> </table> <p>For
250further explanations see <a href="#cross_normalized_x">cross_normalized_x.</a></p>
251</td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="cross_profiles"></a><b>cross_profiles</b></p>
252</td> <td style="vertical-align: top;">C*100&nbsp;
253<br>&nbsp;&nbsp; (100)</td> <td style="vertical-align: top;">see right<br> </td>
254<td style="vertical-align: top;"> <p>Determines
255which vertical profiles are to be presented in
256which coordinate system if the plot software <span style="font-weight: bold;">profil</span> is used.
257&nbsp; </p> <p>This parameter only applies for
258&nbsp;<a href="chapter_4.2.html#data_output_format">data_output_format</a>
259= <span style="font-style: italic;">'profil'</span>.</p><p>The
260default assignment is:&nbsp; </p> <p><b>cross_profiles</b>
261=&nbsp; </p> <ul> <p><span style="font-family: monospace; font-style: italic;">'
262u v ',</span><br> <span style="font-family: monospace; font-style: italic;">' pt
263',&nbsp; </span><br style="font-family: monospace; font-style: italic;"> <span style="font-family: monospace; font-style: italic;">'
264w"pt" w*pt* w*pt*BC wpt wptBC ',&nbsp; </span><br style="font-family: monospace; font-style: italic;"> <span style="font-family: monospace; font-style: italic;">'
265w"u" w*u* wu w"v"w*v* wv ',&nbsp; </span><br style="font-family: monospace; font-style: italic;"> <span style="font-family: monospace; font-style: italic;">' km
266kh ',</span><br style="font-family: monospace; font-style: italic;"> <span style="font-family: monospace; font-style: italic;">' l '
267,</span><br>
26814 * <span style="font-family: monospace; font-style: italic;">'
269'</span></p> </ul> <p>If plot output of
270vertical profiles is produced (see <a href="#data_output_pr">data_output_pr</a>),
271the appropriate data are written to the local file <a href="chapter_3.4.html#PLOT1D_DATA">PLOT1D_DATA</a>.
272Simultaneously, the model produces a parameter file (local name <a href="chapter_3.4.html#PLOT1D_PAR">PLOT1D_PAR</a>)
273which describes the layout for a plot to be generated with the plot
274program <span style="font-weight: bold;">profil</span>.
275The parameter <b>cross_profiles</b>
276determines how many coordinate systems (panels) the plot contains and
277which profiles are supposed to be drawn into which panel. <b>cross_profiles</b>
278expects a character string (up to 100 characters long) for each
279coordinate system, which consists of the names of the profiles to be
280drawn into this system (all available profiles and their respective
281names are described at parameter <a href="#data_output_pr">data_output_pr</a>).
282The single names have to be separated by one blank (' ') and a blank
283must be spent also at the beginning and at the end of the
284string.&nbsp; </p> <p>Example:&nbsp; </p> <ul>
285<p><b>cross_profiles</b> = <span style="font-family: monospace; font-style: italic;">' u v ',
286' pt '</span></p> </ul> <p>In this case the
287plot consists of two coordinate systems
288(panels) with the first panel containing the profiles of the horizontal
289velocity components (<span style="font-style: italic;">'u'</span>
290and <span style="font-style: italic;">'v'</span>)
291of all output times (see <a href="#dt_dopr">dt_dopr</a>)
292and the second one containing the profiles of the potential temperature
293(<span style="font-style: italic;">'pt'</span>).<br>
294</p> <p>Whether the coordinate systems are actually drawn,
295depends on
296whether data of the appropriate profiles were output during the run
297(profiles to be output have to be selected with the parameter <a href="#data_output_pr">data_output_pr</a>).
298For example if <b>data_output_pr</b> = <span style="font-style: italic;">'u'</span>, <span style="font-style: italic;">'v'</span> was assigned,
299then
300the plot only consists of one panel, since no profiles of the potential
301temperature were output. On the other hand, if profiles were assigned
302to <b>data_output_pr </b>whose names do not appear in <b>cross_profiles</b>,
303then the respective profile data are output (<a href="chapter_3.4.html#PLOT1D_DATA">PLOT1D_DATA</a>)
304but they are not drawn in the plot. <br> </p>
305The arrangement of the panels in the plot can be controlled
306with the parameters <a href="#profile_columns">profile_columns</a>
307and <a href="#profile_rows">profile_rows</a>.
308Up to 100 panels systems are allowed in a plot (however, they may be
309distributed on several pages).</td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="cross_xtext"></a><b>cross_xtext</b></p>
310</td> <td style="vertical-align: top;">C*40&nbsp;
311<br>&nbsp;&nbsp; (100)</td> <td style="vertical-align: top;">see right<br> </td>
312<td style="vertical-align: top;"> <p>x-axis labels
313of vertical profile coordinate systems to be
314plotted with <span style="font-weight: bold;">profil</span>.&nbsp;
315</p> <p>This parameter only applies for &nbsp;<a href="chapter_4.2.html#data_output_format">data_output_format</a>
316= <span style="font-style: italic;">'profil'</span>.</p><p>The
317default assignment is:&nbsp; </p> <p><b>cross_xtext</b>
318=&nbsp; </p> <ul> <p><span style="font-style: italic;">'wind speed in
319ms&gt;-&gt;1',&nbsp; </span><br style="font-style: italic;"> <span style="font-style: italic;">'pot. temperature in
320K',&nbsp; </span><br style="font-style: italic;">
321<span style="font-style: italic;">'heat flux in K
322ms&gt;-&gt;1',&nbsp; </span><br style="font-style: italic;"> <span style="font-style: italic;">'momentum flux in
323m&gt;2s&gt;2',&nbsp; </span><br style="font-style: italic;"> <span style="font-style: italic;">'eddy diffusivity in
324m&gt;2s&gt;-&gt;1',&nbsp; </span><br style="font-style: italic;"> <span style="font-style: italic;">'mixing length in m',</span>&nbsp;
325<br>14 * <span style="font-style: italic;">' '</span></p>
326</ul> <p>This parameter can be used to assign x-axis
327labels to vertical
328profiles to be plotted with the plot software <span style="font-weight: bold;">profil </span>(for output
329of vertical
330profile data see <a href="#data_output_pr">data_output_pr</a>).<br>
331The labels are assigned to those coordinate systems (panels) defined by
332<a href="#cross_profiles">cross_profiles</a>
333according to their respective order (compare the default values of <b>cross_xtext</b>
334and <b>cross_profiles</b>). </p> <p>Umlauts
335are possible (write &ldquo; in front of, similar to TeX), as
336well as super- and subscripts (use "&gt;" or "&lt;" in front of
337each
338character), special characters etc. (see UNIRAS manuals) when using the
339plot software <a href="http://www.muk.uni-hannover.de/institut/software/profil_beschreibung.html#chapter3.2.6">profil</a>.</p>
340</td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="cycle_mg"></a><b>cycle_mg</b></p>
341</td> <td style="vertical-align: top;">C*1</td>
342<td style="vertical-align: top;"><i>'w'</i></td>
343<td style="vertical-align: top;"> <p>Type of cycle
344to be used with the multi-grid method.&nbsp; </p> <p>This
345parameter determines which type of cycle is applied in
346the multi-grid method used for solving the Poisson equation for
347perturbation pressure (see <a href="#psolver">psolver</a>).
348It defines in which way it is switched between the fine and coarse
349grids. So-called v- and w-cycles are realized (i.e. <b>cycle_mg</b>
350may be assigned the values <i>'v'</i> or <i>'w'</i>).
351The
352computational cost of w-cycles is much higher than that of v-cycles,
353however, w-cycles give a much better convergence. </p> </td>
354</tr> <tr> <td style="vertical-align: top;"><p><a name="data_output"></a><b>data_output</b></p>
355</td> <td style="vertical-align: top;">C * 10 (100)<br>
356</td> <td style="vertical-align: top;"><span style="font-style: italic;">100 * ' '</span><br>
357</td> <td style="vertical-align: top;">Quantities
358for which 2d cross section and/or 3d volume data are to be output.<br><br>PALM
359allows the output of instantaneous data as well as of temporally
360averaged data which is steered by the strings assigned to this
361parameter (see below).<br><br>By default, cross section
362data are output (depending on the selected cross sections(s), see
363below)&nbsp; to local files <a href="chapter_3.4.html#DATA_2D_XY_NETCDF">DATA_2D_XY_NETCDF</a>,
364<a href="chapter_3.4.html#DATA_2D_XZ_NETCDF">DATA_2D_XZ_NETCDF</a>
365and/or <a href="chapter_3.4.html#DATA_2D_YZ_NETCDF">DATA_2D_YZ_NETCDF</a>.
366Volume data are output to file <a href="chapter_3.4.html#DATA_3D_NETCDF">DATA_3D_NETCDF</a>.
367If the user has switched on the output of temporally averaged data,
368these are written seperately to local files <a href="chapter_3.4.html#DATA_2D_XY_AV_NETCDF">DATA_2D_XY_AV_NETCDF</a>,
369<a href="chapter_3.4.html#DATA_2D_XZ_AV_NETCDF">DATA_2D_XZ_AV_NETCDF</a>,
370<a href="chapter_4.3.html#DATA_2D_YZ_AV_NETCDF">DATA_2D_YZ_AV_NETCDF</a>,
371and <a href="chapter_3.4.html#DATA_3D_AV_NETCDF">DATA_3D_AV_NETCDF</a>,
372respectively.<br><br>The
373filenames already suggest that all files have NetCDF format.
374Informations about the file content (kind of quantities, array
375dimensions and grid coordinates) are part of the self describing NetCDF
376format and can be extracted from the NetCDF files using the command
377"ncdump -c &lt;filename&gt;". See chapter <a href="chapter_4.5.1.html">4.5.1</a> about processing
378the PALM NetCDF data.<br><br>The following quantities are
379available for output by default:<br><br><table style="text-align: left; width: 576px; height: 481px;" border="1" cellpadding="2" cellspacing="2"><tbody><tr><td style="width: 106px;"><span style="font-weight: bold;">quantity
380name</span></td><td style="width: 196px;"><span style="font-weight: bold;">meaning</span></td><td><span style="font-weight: bold;">unit</span></td><td><span style="font-weight: bold;">remarks</span></td></tr><tr><td style="width: 106px;"><span style="font-style: italic;">e</span></td><td style="width: 196px;">SGS TKE</td><td>m<sup>2</sup>/s<sup>2</sup></td><td></td></tr><tr><td style="width: 106px; vertical-align: top;"><span style="font-style: italic;">lwp*</span></td><td style="width: 196px; vertical-align: top;">liquid water path</td><td style="vertical-align: top;">m</td><td style="vertical-align: top;">only horizontal cross section
381is allowed,&nbsp;requires <a href="chapter_4.1.html#cloud_physics">cloud_physics</a>
382= <span style="font-style: italic;">.TRUE.</span></td></tr><tr><td style="width: 106px; vertical-align: top;"><span style="font-style: italic;">p</span></td><td style="width: 196px; vertical-align: top;">perturpation
383pressure</td><td style="vertical-align: top;">N/m<sup>2</sup>,
384Pa</td><td style="vertical-align: top;"></td></tr><tr><td style="width: 106px; vertical-align: top;"><span style="font-style: italic;">pc</span></td><td style="width: 196px; vertical-align: top;">particle/droplet
385concentration</td><td style="vertical-align: top;">#/gridbox</td><td style="vertical-align: top;">requires that particle
386advection is switched on by <span style="font-weight: bold;">mrun</span>-option
387"-p particles"</td></tr><tr><td style="width: 106px; vertical-align: top;"><span style="font-style: italic;">pr</span></td><td style="width: 196px; vertical-align: top;">mean
388particle/droplet radius </td><td style="vertical-align: top;">m</td><td style="vertical-align: top;">requires that particle
389advection is switched on by <span style="font-weight: bold;">mrun</span>-option
390"-p particles"</td></tr><tr><td style="width: 106px; vertical-align: top;"><span style="font-style: italic;">pt</span></td><td style="width: 196px; vertical-align: top;">potential
391temperature<br></td><td style="vertical-align: top;">K</td><td style="vertical-align: top;"></td></tr><tr><td style="width: 106px; vertical-align: top;"><span style="font-style: italic;">q</span></td><td style="width: 196px; vertical-align: top;">specific humidity
392(or total water content, if cloud physics is switched on)</td><td style="vertical-align: top;">kg/kg</td><td style="vertical-align: top;">requires <a href="chapter_4.1.html#moisture">moisture</a> = <span style="font-style: italic;">.TRUE.</span></td></tr><tr><td style="width: 106px; vertical-align: top;"><span style="font-style: italic;">ql</span></td><td style="width: 196px; vertical-align: top;">liquid water
393content</td><td style="vertical-align: top;">kg/kg</td><td style="vertical-align: top;">requires <a href="chapter_4.1.html#cloud_physics">cloud_physics</a>
394= <span style="font-style: italic;">.TRUE.</span>
395or <a href="chapter_4.1.html#cloud_droplets">cloud_droplets</a>
396= <span style="font-style: italic;">.TRUE.</span></td></tr><tr><td style="width: 106px; vertical-align: top;"><span style="font-style: italic;">ql_c</span></td><td style="width: 196px; vertical-align: top;">change in liquid
397water content due to condensation/evaporation during last timestep</td><td style="vertical-align: top;">kg/kg</td><td style="vertical-align: top;">requires <a href="chapter_4.1.html#cloud_droplets">cloud_droplets</a>
398= <span style="font-style: italic;">.TRUE.</span></td></tr><tr><td style="width: 106px; vertical-align: top;"><span style="font-style: italic;">ql_v</span></td><td style="width: 196px; vertical-align: top;">volume of liquid
399water</td><td style="vertical-align: top;">m<sup>3</sup>/gridbox</td><td style="vertical-align: top;">requires <a href="chapter_4.1.html#cloud_droplets">cloud_droplets</a>
400= <span style="font-style: italic;">.TRUE.</span></td></tr><tr><td style="width: 106px; vertical-align: top;"><span style="font-style: italic;">ql_vp</span></td><td style="width: 196px; vertical-align: top;">weighting factor</td><td style="vertical-align: top;"></td><td style="vertical-align: top;">requires <a href="chapter_4.1.html#cloud_droplets">cloud_droplets</a>
401= <span style="font-style: italic;">.TRUE.</span></td></tr><tr><td style="width: 106px; vertical-align: top;"><span style="font-style: italic;">qv</span></td><td style="width: 196px; vertical-align: top;">water vapor
402content (specific humidity)</td><td style="vertical-align: top;">kg/kg</td><td style="vertical-align: top;">requires <a href="chapter_4.1.html#cloud_physics">cloud_physics</a>
403= <span style="font-style: italic;">.TRUE.</span></td></tr><tr><td style="width: 106px; vertical-align: top;"><span style="font-style: italic;">s</span></td><td style="width: 196px; vertical-align: top;">concentration of
404the scalar</td><td style="vertical-align: top;">1/m<sup>3</sup></td><td style="vertical-align: top;">requires&nbsp;<a href="chapter_4.1.html#passive_scalar">passive_scalar</a>
405= <span style="font-style: italic;">.TRUE.</span></td></tr><tr><td style="width: 106px; vertical-align: top;"><span style="font-style: italic;">t*</span></td><td style="width: 196px; vertical-align: top;">(near surface)
406characteristic temperature</td><td style="vertical-align: top;">K</td><td style="vertical-align: top;">only horizontal cross section
407is allowed</td></tr><tr><td style="width: 106px; vertical-align: top;"><span style="font-style: italic;">u</span></td><td style="width: 196px; vertical-align: top;">u-component of
408the velocity</td><td style="vertical-align: top;">m/s</td><td style="vertical-align: top;"></td></tr><tr><td style="width: 106px; vertical-align: top;"><span style="font-style: italic;">u*</span></td><td style="width: 196px; vertical-align: top;">(near surface)
409friction velocity</td><td style="vertical-align: top;">m/s</td><td style="vertical-align: top;">only horizontal cross section
410is allowed</td></tr><tr><td style="width: 106px; vertical-align: top;"><span style="font-style: italic;">v</span></td><td style="width: 196px; vertical-align: top;">v-component of
411the velocity</td><td style="vertical-align: top;">m/s</td><td style="vertical-align: top;"></td></tr><tr><td style="width: 106px; vertical-align: top;"><span style="font-style: italic;">vpt</span></td><td style="width: 196px; vertical-align: top;">virtual potential
412temperature</td><td style="vertical-align: top;">K</td><td style="vertical-align: top;">requires <a href="chapter_4.1.html#moisture">moisture</a> = <span style="font-style: italic;">.TRUE.</span></td></tr><tr><td style="width: 106px; vertical-align: top;"><span style="font-style: italic;">w</span></td><td style="width: 196px; vertical-align: top;">w-component of
413the velocity</td><td style="vertical-align: top;">m/s</td><td style="vertical-align: top;"></td></tr></tbody></table><br>Multiple
414quantities can be assigned, e.g. <span style="font-weight: bold;">data_output</span>
415= <span style="font-style: italic;">'e'</span>, <span style="font-style: italic;">'u'</span>, <span style="font-style: italic;">'w'</span>.<br><br>By
416assigning the pure strings from the above table, 3d volume data is
417output. Cross section data can be output by appending the string <span style="font-style: italic;">'_xy'</span>, <span style="font-style: italic;">'_xz'</span>, or <span style="font-style: italic;">'_yz'</span> to the
418respective quantities. Time averaged&nbsp;output is created by
419appending the string <span style="font-style: italic;">'_av'
420</span>(for
421cross section data, this string must be appended after the cross
422section string). Cross section data can also be (additionally) averaged
423along the direction normal to the respective section (see below).
424Assignments of quantities can be given in arbitrary
425order:<br><br>Example:<br><br><div style="margin-left: 40px;"><span style="font-weight: bold;">data_output</span> = <span style="font-style: italic;">'u'</span>, <span style="font-style: italic;">'pt_xz_av'</span>, <span style="font-style: italic;">'w_xy'</span>, <span style="font-style: italic;">'u_av'</span>.<br></div><br>This
426example will create the following output: instantaneous 3d volume data
427of u-velocity component (by default on file DATA_3D_NETCDF), temporally
428averaged 3d volume data of u-velocity component (by default on file
429DATA_3D_AV_NETCDF), instantaneous horizontal cross section data of
430w-velocity component (by default on file DATA_2D_XY_NETCDF), and
431temporally averaged vertical cross section data of potential
432temperature (by default on file DATA_2D_XZ_AV_NETCDF).<br><br>The
433user is allowed to extend the above list of quantities by defining his
434own output quantities (see the user-parameter <a href="chapter_4.3.html#data_output_user">data_output_user</a>).<br><br>The
435time interval of the output times is determined via <a href="#dt_data_output">dt_data_output</a>.
436This is valid for all types of output quantities by default. Individual
437time intervals for instantaneous &nbsp;(!) 3d and section data can
438be
439declared using <a href="#dt_do3d">dt_do3d</a>, <a href="#dt_do2d_xy">dt_do2d_xy</a>, <a href="#dt_do2d_xz">dt_do2d_xz</a>, and <a href="#dt_do2d_yz">dt_do2d_yz</a>.<br><br>Also,
440an individual time interval for output of temporally averaged data can
441be assigned using parameter <a href="#dt_data_output_av">dt_data_output_av</a>.
442This applies to both 3d volume and cross section data. The length of
443the averaging interval is controlled via parameter <a href="#averaging_interval">averaging_interval</a>.<br><br>The
444parameter <a href="#skip_time_data_output">skip_time_data_output</a>
445can be used to shift data output activities for a given time interval.
446Individual intervals can be set using <a href="#skip_time_do3d">skip_time_do3d</a>,
447<a href="#skip_time_do2d_xy">skip_time_do2d_xy</a>, <a href="#skip_time_do2d_xz">skip_time_do2d_xz</a>, <a href="#skip_time_do2d_yz">skip_time_do2d_yz</a>, and <a href="#skip_time_data_output_av">skip_time_data_output_av</a>.<br><p>With
448the parameter <a href="chapter_4.2.html#nz_do3d">nz_do3d</a>&nbsp;
449the output can be limited in the vertical direction up to a certain
450grid point.<br> </p> Cross sections extend through the
451total model
452domain. In the two horizontal directions all grid points with 0
453&lt;= i
454&lt;= nx+1 and 0 &lt;= j
455&lt;= ny+1 are output so that in case of cyclic boundary conditions
456the
457complete total domain is represented. The location(s) of the cross
458sections can be defined with parameters <a href="#section_xy">section_xy</a>,
459<a href="#section_xz">section_xz</a>, and <a href="#section_yz">section_yz</a>. Assigning <span style="font-weight: bold;">section_..</span> = <span style="font-style: italic;">-1</span>
460causes&nbsp;the output data to be averaged along the direction
461normal to the respective section.<br><br><br><span style="font-weight: bold;">Output of user defined quantities:</span><br><br>Beside
462the standard quantities from the above list, the user can output any
463other quantities. These have to be defined and calculated within the
464user-defined code (see <a href="chapter_3.5.4.html">3.5.4</a>).
465They can be selected for output with the user-parameter <a href="chapter_4.3.html#data_output_user">data_output_user</a>
466for which the same rules apply as for <span style="font-weight: bold;">data_output</span>.
467Output of the user defined quantities (time interval, averaging,
468selection of cross sections, etc.) is controlled with the parameters
469listed above and data are written to the same file(s) as the standard
470quantities.<br><br><p style="font-weight: bold;">Output
471on parallel machines:</p><p>
472By default, with parallel runs, processors output only data
473of their respective subdomains into seperate local files (file names
474are
475constructed by appending the four digit processor ID, e.g.
476&lt;filename&gt;_0000, &lt;filename&gt;_0001, etc.).
477After PALM has
478finished, the contents of these individual
479files are sampled into one final file<span style="font-weight: bold;"></span>
480using the program <tt><font style="font-size: 11pt;" size="2">combine_plot_fields.x</font></tt>
481(to be started e.g. by a suitable OUTPUT command in the <span style="font-weight: bold;">mrun</span>
482configuration file).</p> <p>Alternatively, PALM is able to
483collect all grid points of a
484cross section on PE0 before output is done. In this case only
485one&nbsp;
486output file (DATA_2D_XY_NETCDF, etc.) is created and <tt><font style="font-size: 11pt;" size="2">combine_plot_fields.x</font></tt>
487does not have to be called. In case of very large numbers of horizontal
488gridpoints, sufficient
489memory is required on PE0.&nbsp; This method can be used by
490assigning <a href="chapter_4.2.html#data_output_2d_on_each_pe">data_output_2d_on_each_pe</a>
491= <i>.F.</i>.</p><p>3d volume data output is
492always handled seperately by each processor so that <span style="font-family: monospace;">combine_plot_fields.x</span>
493has to be called anyway after PALM has been finished.</p><p><br><span style="font-weight: bold;">Old formats:</span></p>
494<p>Beside
495the NetCDF format,&nbsp;2d cross section data and 3d volume data
496can
497also be output, for historical reasons, in a different (binary) format
498using parameter <a href="#data_output_format">data_output_format</a>.</p><p>By
499assigning <span style="font-weight: bold;">data_output_format
500</span>= <span style="font-style: italic;">'avs'</span>,
501the 3d volume data is output to the local file <a href="chapter_3.4.html#PLOT3D_DATA">PLOT3D_DATA</a>.
502Output is in FORTRAN binary format&nbsp;readable by
503the plot software <span style="font-weight: bold;">AVS</span>.&nbsp;
504The order of data on the file follows the order used in the assignment
505for <b>data_output</b> (e.g. <b>data_output</b>
506= <span style="font-style: italic;">'p'</span>, <span style="font-style: italic;">'v'</span>,...&nbsp;
507means that the file starts with the pressure data, followed by the
508v-component of the velocity, etc.). Both instantaneous and time
509averaged data are written on this file! Additional to this file, PALM
510creates
511a second binary file (local name <a href="chapter_3.4.html#PLOT3D_COOR">PLOT3D_COOR</a>)
512with coordinate information needed by <span style="font-weight: bold;">AVS</span>.
513As third and
514fourth file two ASCII files are created (AVS-FLD-format, local name <a href="chapter_3.4.html#PLOT3D_FLD">PLOT3D_FLD</a>
515and <a href="chapter_3.4.html#PLOT3D_FLD_COOR">PLOT3D_FLD_COOR</a>),
516which describe the contents of the data file and/or coordinate file
517and are used by AVS. However, AVS expects the content description in
518one file. This needs the local file PLOT3D_FLD_COOR to be appended to
519the file
520PLOT3D_FLD (by suitable OUTPUT command in the <span style="font-weight: bold;">mrun</span>
521configuration file: &ldquo;<span style="font-family: monospace;">cat
522PLOT3D_FLD_COOR &gt;&gt; PLOT3D_FLD</span>&rdquo;)
523after PALM has
524finished.&nbsp;To reduce the amount of data, output to this file
525can be done
526in
527compressed form (see <a href="chapter_4.2.html#do3d_compress">do3d_compress</a>).
528Further details about plotting 3d volume data with <span style="font-weight: bold;">AVS </span>can be found in
529<a href="chapter_4.5.5.html">chapter
5304.5.5</a>.</p>By assigning <span style="font-weight: bold;">data_output_format </span>=
531<span style="font-style: italic;">'iso2d'</span>,
532the cross section data is output to the local files <a href="chapter_3.4.html#PLOT2D_XY">PLOT2D_XY</a>, <a href="chapter_3.4.html#PLOT2D_XZ">PLOT2D_XZ</a>, and <a href="chapter_3.4.html#PLOT2D_YZ">PLOT2D_YZ</a>.
533Output is in FORTRAN binary format&nbsp;readable by
534the plot software&nbsp;<span style="font-weight: bold;">iso2d</span>.&nbsp;
535The order of data on the files follows the order used in the assignment
536for <b>data_output</b> (e.g. <b>data_output</b>
537= <span style="font-style: italic;">'p_xy'</span>, <span style="font-style: italic;">'v_xy_av'</span>,...&nbsp;
538means that the file containing the horizontal cross section data starts
539with the instantaneous pressure data, followed by the
540temporally averaged v-component of the velocity, etc.). Both
541instantaneous and time averaged data are written on this
542file!Additional to these binary files, PALM
543creates NAMELIST parameter files
544(local names <a href="chapter_3.4.html#PLOT2D_XY_GLOBAL">PLOT2D_XY_GLOBAL</a>,
545<a href="chapter_3.4.html#PLOT2D_XY_LOCAL">PLOT2D_XY_LOCAL</a>,
546<a href="chapter_3.4.html#PLOT2D_XZ_GLOBAL">PLOT2D_XZ_GLOBAL</a>,
547<a href="chapter_3.4.html#PLOT2D_XZ_LOCAL">PLOT2D_XZ_LOCAL</a>,
548<a href="chapter_3.4.html#PLOT2D_YZ_GLOBAL">PLOT2D_YZ_GLOBAL</a>,
549<a href="chapter_3.4.html#PLOT2D_YZ_LOCAL">PLOT2D_YZ_LOCAL</a>)
550which can be used as parameter input files for the plot software <a href="http://www.muk.uni-hannover.de/institut/software/iso2d_beschreibung.html">iso2d</a>.
551That needs local files with suffix _LOCAL to be appended to the
552respective files with suffix _GLOBAL (by
553suitable OUTPUT commands in the <span style="font-weight: bold;">mrun</span>
554configuration file, e.g.: &ldquo;<span style="font-family: monospace;">cat
555PLOT2D_XY_LOCAL &gt;&gt; PLOT2D_XY_GLOBAL</span>&rdquo;)
556after PALM has
557finished. Cross sections can be directly plotted with <span style="font-weight: bold;">iso2d</span> using the
558respective data and
559parameter file. The plot layout is steered via the parameter input
560file.
561The values of these <span style="font-weight: bold;">iso2d</span>
562parameters are determined by a set of mostly internal PALM parameters
563(exception: <a href="chapter_4.2.html#z_max_do2d">z_max_do2d</a>).
564All parameter values can be changed by editing the parameter input
565file.&nbsp;Further details about plotting 2d cross sections with <span style="font-weight: bold;">iso2d </span>can be found
566in <a href="chapter_4.5.4.html">chapter
5674.5.4</a>.<br><br><span style="font-weight: bold;">Important:</span><br>There
568is no guarantee that iso2d- and avs-output will be available in future
569PALM versions (later than 3.0). </td> </tr> <tr> <td style="vertical-align: top;"><a name="data_output_format"></a><span style="font-weight: bold;">data_output_format</span><br>
570</td> <td style="vertical-align: top;">C * 10 (10) </td>
571<td style="vertical-align: top;"><span style="font-style: italic;">'netcdf'</span> </td>
572<td style="vertical-align: top;">Format of output data.<br><br>By
573default, all data (profiles, time
574series, spectra, particle data, cross sections, volume data) are output
575in NetCDF format (see chapter <a href="chapter_4.5.1.html">4.5.1</a>).
576Exception: restart data (local files <a href="chapter_3.4.html#BININ">BININ</a>, <a href="chapter_3.4.html#BINOUT">BINOUT</a>, <a href="chapter_3.4.html#PARTICLE_RESTART_DATA_IN">PARTICLE_RESTART_DATA_IN</a>,
577<a href="chapter_3.4.html#PARTICLE_RESTART_DATA_OUT">PARTICLE_RESTART_DATA_OUT</a>)
578are always output in FORTRAN binary format.<br><br>The
579numerical precision of the NetCDF output is determined with parameter <a href="#chapter_4.1.html#netcdf_precision">netcdf_precision</a>.<br><br>The
580maximum file size for NetCDF files is 2 GByte by default. Use the
581parameter <a href="#netcdf_64bit">netcdf_64bit</a>
582if larger files have to be created.<br><br>For historical
583reasons, other data formats are still available. Beside 'netcdf', <span style="font-weight: bold;">data_output_format</span>
584may be assigned the following values:<br><br><table style="text-align: left; width: 594px; height: 104px;" border="1" cellpadding="2" cellspacing="2"><tbody><tr><td style="vertical-align: top;"><span style="font-style: italic;">'profil'</span></td><td>output
585of profiles,&nbsp;time series and spectra in ASCII format to be
586read by the graphic software <span style="font-weight: bold;">profil
587</span>(see chapters <a href="chapter_4.5.2.html">4.5.2</a>,
588<a href="#chapter_4.5.3.html">4.5.3</a>)</td></tr><tr><td style="vertical-align: top;"><span style="font-style: italic;">'iso2d'</span></td><td>output
589of 2d cross-sections in FORTRAN binary format to be read by the graphic
590software <span style="font-weight: bold;">iso2d</span>
591(see chapter <a href="chapter_4.5.4.html">4.5.4</a>)</td></tr><tr><td style="vertical-align: top;"><span style="font-style: italic;">'avs'</span></td><td>output
592of 3d volume data in FORTRAN binary format to be read by the graphic
593software <span style="font-weight: bold;">AVS</span>
594(see chapter <a href="chapter_4.5.5.html">4.5.5</a>)</td></tr></tbody></table><br>Multiple
595values can be assigned to <span style="font-weight: bold;">data_output_format</span>,
596i.e. if the user wants to have both the "old" data format suitable for <span style="font-weight: bold;">iso2d</span> as well as
597cross section data in NetCDF format, then <span style="font-weight: bold;">data_output_format</span> =
598<span style="font-style: italic;">'iso2d'</span>, <span style="font-style: italic;">'netcdf'</span> has to be
599assigned.<br><br><span style="font-weight: bold;">Warning:</span>
600There is no guarantee that the "old" formats will be available in
601future PALM versions (beyond 3.0)!<br> </td> </tr> <tr>
602<td style="vertical-align: top;"> <p><a name="data_output_pr"></a><b>data_output_pr</b></p>
603</td> <td style="vertical-align: top;">C *
60410&nbsp; <br>
605(100)</td> <td style="vertical-align: top;"><i>100
606* ' '</i></td> <td style="vertical-align: top;">
607<p>Quantities for which vertical profiles (horizontally averaged)
608are to be output.&nbsp; </p> <p>By default vertical
609profile data is output to the local file <a href="chapter_3.4.html#DATA_1D_PR_NETCDF">DATA_1D_PR_NETCDF</a>.
610The file's format is NetCDF.&nbsp; Further details about processing
611NetCDF data are given in chapter <a href="chapter_4.5.1.html">4.5.1</a>.</p><p>For
612historical reasons, data can also be output in ASCII-format on local
613file <a href="chapter_3.4.html#PLOT1D_DATA">PLOT1D_DATA</a>
614which is readable by the graphic software <span style="font-weight: bold;">profil</span>. See
615parameter <a href="#data_output_format">data_output_format</a>
616for defining the format in which data shall be output.<br> </p>
617<p>For horizontally averaged vertical
618profiles always <span style="font-weight: bold;">all</span>
619vertical
620grid points (0 &lt;= k &lt;= nz+1) are output to file. Vertical
621profile data refers to the total domain but profiles for subdomains can
622also be output (see <a href="chapter_4.1.html#statistic_regions">statistic_regions</a>).&nbsp;
623</p> <p>The temporal interval of the output times of
624profiles is
625assigned via the parameter <a href="chapter_4.2.html#dt_dopr">dt_dopr</a>.
626Within the file <a href="chapter_3.4.html#PLOT1D_DATA">PLOT1D_DATA</a>,
627the profiles are ordered with respect to their
628output times.</p><p>Profiles can also be temporally
629averaged (see <a href="chapter_4.2.html#averaging_interval_pr">averaging_interval_pr</a>).&nbsp;<br>
630</p> <p>The following list shows the values which can be
631assigned to <span style="font-weight: bold;">data_output_pr</span>.
632The profile data is either defined on
633u-v-levels (variables marked in <font color="#ff6600">red</font>)
634or
635on w-levels (<font color="#33ff33">green</font>).
636According to this,
637the
638z-coordinates of the individual profiles vary. Beyond that, with a
639Prandtl layer switched on (<a href="chapter_4.1.html#prandtl_layer">prandtl_layer</a>)
640the lowest output
641level is z = zu(1) instead of z = zw(0) for profiles <i>w''
642u'',w''v"</i>, <i>wu</i> and <i>wv</i>
643.&nbsp; <br> </p> <table style="text-align: left; width: 100%;" cellpadding="2" cellspacing="2"> <tbody> <tr> <td style="vertical-align: top;"><font color="#ff6600"><i>u</i></font></td>
644<td style="vertical-align: top;">u-component of the
645velocity (in m/s).</td> </tr> <tr> <td style="vertical-align: top;"><font color="#ff6600"><i>v</i></font></td>
646<td style="vertical-align: top;">v-component of the
647velocity (in m/s).</td> </tr> <tr> <td style="vertical-align: top;"><font color="#33ff33"><i>w</i></font></td>
648<td style="vertical-align: top;">w-component of the
649velocity (in m/s).</td> </tr> <tr> <td style="vertical-align: top;"><font color="#ff6600"><i>pt</i></font></td>
650<td style="vertical-align: top;">Potential temperature (in
651K).</td> </tr> <tr> <td style="vertical-align: top;"><font color="#ff6600"><i>vpt</i></font></td>
652<td style="vertical-align: top;">Virtual potential
653temperature (in K).</td> </tr> <tr> <td style="vertical-align: top;"><font color="#ff6600"><i>lpt</i></font></td>
654<td style="vertical-align: top;">Potential liquid water
655temperature (in K).</td> </tr> <tr> <td style="vertical-align: top;"><font color="#ff6600"><i>q</i></font></td>
656<td style="vertical-align: top;">Total water content
657(in kg/kg).</td> </tr> <tr> <td style="vertical-align: top;"><font color="#ff6600"><i>qv</i></font></td>
658<td style="vertical-align: top;">Specific humidity (in
659kg/kg).</td> </tr> <tr> <td style="vertical-align: top;"><font color="#ff6600"><i>ql</i></font></td>
660<td style="vertical-align: top;">Liquid water content
661(in kg/kg).</td> </tr> <tr> <td style="vertical-align: middle;"><font color="#ff6600">s</font></td>
662<td style="vertical-align: top;">Scalar concentration (in
663kg/m<sup>3</sup>).</td> </tr> <tr> <td style="vertical-align: middle;"><font color="#ff6600"><i>e</i></font></td>
664<td style="vertical-align: top;">Turbulent kinetic energy
665(TKE, subgrid-scale) (in m<sup>2</sup>/s<sup>2</sup>).</td>
666</tr> <tr> <td style="vertical-align: middle;"><font color="#ff6600"><i>e*</i></font></td>
667<td style="vertical-align: top;">Perturbation energy
668(resolved) (in m<sup>2</sup>/s<sup>2</sup>).</td>
669</tr> <tr> <td style="vertical-align: middle;"><font color="#ff6600"><i>km</i></font></td>
670<td style="vertical-align: top;">Eddy diffusivity for
671momentum (in m<sup>2</sup>/s).</td> </tr> <tr>
672<td style="vertical-align: middle;"><font color="#ff6600"><i>kh</i></font></td>
673<td style="vertical-align: top;">Eddy diffusivity for heat
674(in m<sup>2</sup>/s).</td> </tr> <tr> <td style="vertical-align: top;"><font color="#ff6600"><i>l</i></font></td>
675<td style="vertical-align: top;">Mixing length (in m).</td>
676</tr> <tr> <td style="vertical-align: middle;"><font color="#33ff33"><i>w"u"</i></font></td>
677<td style="vertical-align: top;">u-component of the
678subgrid-scale vertical momentum flux (in m<sup>2</sup>/s<sup>2</sup>).</td>
679</tr> <tr> <td style="vertical-align: middle;"><font color="#33ff33"><i>w*u*</i></font></td>
680<td style="vertical-align: top;">u-component of the
681resolved vertical momentum flux (in m<sup>2</sup>/s<sup>2</sup>).</td>
682</tr> <tr> <td style="vertical-align: middle;"><font color="#33ff33"><i>wu</i></font></td>
683<td style="vertical-align: top;">u-component of the total
684vertical momentum flux (<i>w"u"</i> + <i>w*u*</i>)
685(in m<sup>2</sup>/s<sup>2</sup>).</td> </tr>
686<tr> <td style="vertical-align: middle;"><font color="#33ff33"><i>w"v"</i></font></td>
687<td style="vertical-align: top;">v-component of the
688subgrid-scale vertical momentum flux (in m<sup>2</sup>/s<sup>2</sup>).</td>
689</tr> <tr> <td style="vertical-align: middle;"><font color="#33ff33"><i>w*v*</i></font></td>
690<td style="vertical-align: top;">v-component of the
691resolved vertical momentum flux (in m<sup>2</sup>/s<sup>2</sup>).</td>
692</tr> <tr> <td style="vertical-align: middle;"><font color="#33ff33"><i>wv</i></font></td>
693<td style="vertical-align: top;">v-component of the total
694vertical momentum flux (<i>w"v"</i> + <i>w*v*</i>)
695(in m<sup>2</sup>/s<sup>2</sup>).</td> </tr>
696<tr> <td style="vertical-align: top;"><font color="#33ff33"><i>w"pt"</i></font></td>
697<td style="vertical-align: top;">Subgrid-scale vertical
698sensible heat flux (in K m/s).</td> </tr> <tr> <td style="vertical-align: top;"><font color="#33ff33"><i>w*pt*</i></font></td>
699<td style="vertical-align: top;">Resolved vertical
700sensible
701heat flux (in K m/s).</td> </tr> <tr> <td style="vertical-align: top;"><font color="#33ff33"><i>wpt</i></font></td>
702<td style="vertical-align: top;">Total vertical sensible
703heat flux (<i>w"pt"</i> + <i>w*pt*</i>)
704(in K
705m/s).</td> </tr> <tr> <td style="vertical-align: top;"><font color="#33ff33"><i>w*pt*BC</i></font></td>
706<td style="vertical-align: top;">Subgrid-scale vertical
707sensible heat flux using the
708Bott-Chlond scheme (in K m/s).</td> </tr> <tr> <td style="vertical-align: top;"><font color="#33ff33"><i>wptBC</i></font></td>
709<td style="vertical-align: top;">Total vertical sensible
710heat flux using the Bott-Chlond scheme
711(<i>w"pt"</i>
712+ <i>w*pt*BC</i>) (in K m/s).</td> </tr> <tr>
713<td style="vertical-align: top;"><font color="#33ff33"><i>w"vpt"</i></font></td>
714<td style="vertical-align: top;">Subgrid-scale vertical
715buoyancy flux (in K m/s).</td> </tr> <tr> <td style="vertical-align: top;"><font color="#33ff33"><i>w*pt*</i></font></td>
716<td style="vertical-align: top;">Resolved vertical
717buoyancy
718flux (in K m/s).</td> </tr> <tr> <td style="vertical-align: top;"><font color="#33ff33"><i>wvpt</i></font></td>
719<td style="vertical-align: top;">Total vertical buoyancy
720flux (w"vpt" + w*vpt*) (in K m/s).</td> </tr> <tr> <td style="vertical-align: top;"><font color="#33ff33"><i>w"q"</i></font></td>
721<td style="vertical-align: top;">Subgrid-scale vertical
722water flux (in kg/kg m/s).</td> </tr> <tr> <td style="vertical-align: top;"><font color="#33ff33"><i>w*q*</i></font></td>
723<td style="vertical-align: top;">Resolved vertical water
724flux (in kg/kg m/s).</td> </tr> <tr> <td style="vertical-align: top;"><font color="#33ff33"><i>wq</i></font></td>
725<td style="vertical-align: top;">Total vertical water flux
726(w"q" + w*q*) (in kg/kg m/s).</td> </tr> <tr> <td style="vertical-align: top;"><font color="#33ff33"><i>w"qv"</i></font></td>
727<td style="vertical-align: top;">Subgrid-scale vertical
728latent heat flux (in kg/kg m/s).</td> </tr> <tr> <td style="vertical-align: top;"><font color="#33ff33"><i>w*qv*</i></font></td>
729<td style="vertical-align: top;">Resolved vertical latent
730heat flux (in kg/kg m/s).</td> </tr> <tr> <td style="vertical-align: top;"><font color="#33ff33"><i>wqv</i></font></td>
731<td style="vertical-align: top;">Total vertical latent
732heat
733flux (w"qv" + w*qv*) (in kg/kg m/s).</td> </tr> <tr>
734<td style="vertical-align: middle;"><font color="#33ff33"><i>w"s"</i></font></td>
735<td style="vertical-align: top;">Subgrid-scale vertical
736scalar concentration flux (in kg/m<sup>3 </sup>m/s).</td>
737</tr> <tr> <td style="vertical-align: middle;"><font color="#33ff33"><i>w*s*</i></font></td>
738<td style="vertical-align: top;">Resolved vertical scalar
739concentration flux (in kg/m<sup>3</sup>)</td> </tr>
740<tr> <td style="vertical-align: middle;"><font color="#33ff33"><i>ws</i></font></td>
741<td style="vertical-align: top;">Total vertical scalar
742concentration flux (w"s" + w*s*) (in kg/m<sup>3 </sup>m/s).</td>
743</tr> <tr> <td style="vertical-align: top;"><font color="#33ff33"><i>w*e*</i></font></td>
744<td style="vertical-align: top;">Vertical flux of
745perturbation energy (resolved)</td> </tr> <tr> <td style="vertical-align: top;"><font color="#ff6600"><i>u*2</i></font></td>
746<td style="vertical-align: top;">Variance of the
747u-velocity
748component (resolved)</td> </tr> <tr> <td style="vertical-align: top;"><font color="#ff6600"><i>v*2</i></font></td>
749<td style="vertical-align: top;">Variance of the
750v-velocity
751component (resolved)</td> </tr> <tr> <td style="vertical-align: top;"><font color="#33ff33"><i>w*2</i></font></td>
752<td style="vertical-align: top;">Variance of the potential
753temperature (resolved)</td> </tr> <tr> <td style="vertical-align: top;"><font color="#ff6600"><i>pt*2</i></font></td>
754<td style="vertical-align: top;">Variance of the potential
755temperature (resolved)</td> </tr> <tr> <td style="vertical-align: top;"><font color="#33ff33"><i>w*3</i></font></td>
756<td style="vertical-align: top;">Third moment of the
757w-velocity component (resolved)</td> </tr> <tr> <td style="vertical-align: middle;"><font color="#33ff33"><i>Sw</i></font></td>
758<td style="vertical-align: top;">Skewness of the
759w-velocity
760component (resolved, S<sub>w</sub>
761= W<sup>3</sup>/(w<sup>2</sup>)<sup>1.5</sup>)</td>
762</tr> <tr> <td style="vertical-align: top;"><font color="#33ff33"><i>w*2pt*</i></font></td>
763<td style="vertical-align: top;">Third moment (resolved)</td>
764</tr> <tr> <td style="vertical-align: top;"><font color="#33ff33"><i>w*pt*2</i></font></td>
765<td style="vertical-align: top;">Third moment (resolved)</td>
766</tr> <tr> <td style="vertical-align: top;"><font color="#ff6666"><i>w*u*u*/dz</i></font></td>
767<td style="vertical-align: top;">Energy production by
768shear
769(resolved)</td> </tr> <tr> <td style="vertical-align: top;"><font color="#ff6666"><i>w*p*/dz</i></font></td>
770<td style="vertical-align: top;">Energy production by
771turbulent transport of pressure
772fluctuations (resolved)</td> </tr> <tr> <td style="vertical-align: top;"><font color="#ff6666"><i>w"e/dz</i></font></td>
773<td style="vertical-align: top;">Energy production by
774transport of resolved-scale TKE</td> </tr> </tbody>
775</table> <br>Beyond that, initial profiles (t=0) of some
776variables can be also be
777output (this output is only done once
778with the first plot output and not repeated with the profile output at
779later
780times). The names of these profiles result from the ones specified
781above leaded by a hash "#".&nbsp; Allowed values are:<br> <ul>
782<p><i>#u</i>, <i>#v</i>, <i>#pt</i>,
783<i>#km</i>, <i>#kh</i>, <i>#l</i></p>
784</ul> <p>These initial profiles have been either set by
785the user or
786have been calculated by a 1d-model prerun.<br> </p>In case
787of ASCII data output to local file PLOT1D_DATA,
788PALM additionally creates a NAMELIST parameter file (local name <a href="chapter_3.4.html#PLOT1D_PAR">PLOT1D_PAR</a>)
789which can be used as parameter input file for the plot software <a href="http://www.muk.uni-hannover.de/institut/software/profil_intro.html">profil</a>.
790Profiles can be directly plotted with <span style="font-weight: bold;">profil</span>
791using these two files. The
792plot layout is
793steered via the parameter input file. The values of these <span style="font-weight: bold;">profil</span>-parameters
794are determined by
795a set of PALM parameters (<a href="chapter_4.2.html#profile_columns">profile_columns</a>,
796<a href="chapter_4.2.html#profile_rows">profile_rows</a>,
797<a href="chapter_4.2.html#z_max_do1d">z_max_do1d</a>,
798<a href="chapter_4.2.html#cross_profiles">cross_profiles</a>,
799etc.) All parameter values can be changed by editing the parameter
800input
801file. <br><br>Further details about plotting vertical
802profiles with <span style="font-weight: bold;">profil </span>can
803be found in <a href="chapter_4.5.2.html">chapter
8044.5.2</a></td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="data_output_2d_on_each_pe"></a><b>data_output_2d_on</b>
805<br> <b>_each_pe</b></p> </td> <td style="vertical-align: top;">L<br> </td> <td style="vertical-align: top;"><span style="font-style: italic;">.T.</span><br> </td>
806<td style="vertical-align: top;">Output 2d cross section
807data by one or
808all processors.&nbsp; <p>In runs with several processors, by
809default, each processor
810outputs cross section data of its subdomain&nbsp;into an individual
811file. After PALM
812has finished, the contents of these files have to be sampled into one
813file<span style="font-weight: bold;"></span> using
814the program <tt>combine_plot_fields.x</tt>.&nbsp; </p>
815<p>Alternatively, by assigning <b>data_output_2d_on_each_pe</b>
816= <i>.F.,</i>
817the respective data is gathered on PE0 and output is done directly
818into one file, so <tt>combine_plot_fields.x</tt> does not
819have to be
820called. However, in case of very large numbers of horizontal
821gridpoints, sufficient
822memory is required on PE0. </p> </td> </tr>
823<tr> <td style="vertical-align: top;"> <p><a name="disturbance_amplitude"></a><b>disturbance<br>
824_amplitude</b></p> </td> <td style="vertical-align: top;">R</td> <td style="vertical-align: top;"><i>0.25</i></td>
825<td style="vertical-align: top;"> <p>Maximum
826perturbation amplitude of the random perturbations
827imposed to the horizontal velocity field (in m/s).&nbsp; </p>
828<p>The parameter <a href="#create_disturbances">create_disturbances</a>
829describes how to impose random perturbations to the horizontal velocity
830field. Since the perturbation procedure includes two filter operations,
831the amplitude assigned by <b>disturbance_amplitude</b> is
832only an
833approximate value of the real magnitude of the perturbation.</p> </td>
834</tr> <tr> <td style="vertical-align: top;"><p><a name="disturbance_energy_limit"></a><b>disturbance_energy</b>
835<br> <b>_limit</b></p> </td> <td style="vertical-align: top;">R</td> <td style="vertical-align: top;"><i>0.01</i></td>
836<td style="vertical-align: top;"> <p lang="en-GB">Upper
837limit value of the perturbation energy of
838the velocity field used as a criterion for imposing random
839perturbations (in m<sup>2</sup>/s<sup>2</sup>).&nbsp;
840</p> <p><span lang="en-GB"><font face="Thorndale, serif">The parameter </font></span><a href="#create_disturbances"><span lang="en-GB"><font face="Thorndale, serif">create_disturbances</font></span></a><font face="Thorndale, serif"><span lang="en-GB">
841describes how to impose
842random perturbations to the horizontal velocity field. The perturbation
843energy is defined as the volume average (over the total model domain)
844of the squares of the deviations of the velocity components from the
845mean flow (horizontal average). If the perturbation energy exceeds the
846assigned value, random perturbations to the fields of horizontal
847velocities are imposed no more. The value of this parameter usually
848must be determined by trial and error (it depends e.g. on the total
849number of grid points).</span> </font> </p> </td>
850</tr> <tr> <td style="vertical-align: top;"><p><a name="disturbance_level_b"></a><b>disturbance_level_b</b></p>
851</td> <td style="vertical-align: top;">R</td>
852<td style="vertical-align: top;"><i>zu(3)</i></td>
853<td style="vertical-align: top;"> <p lang="en-GB"><font face="Thorndale, serif"><font size="3">Lower
854limit of the vertical range for which random perturbations are to be
855imposed on the horizontal wind field (</font></font>in <font face="Thorndale, serif"><font size="3">m).&nbsp;
856</font></font> </p> <p><span lang="en-GB"><font face="Thorndale, serif">This
857parameter must hold the condition zu<i>(3)</i> &lt;= <b>disturbance_level_b</b>
858&lt;= <i>zu(</i></font></span><i><a href="chapter_4.1.html#nz"><span lang="en-GB"><font face="Thorndale, serif">nz-1</font></span></a><span lang="en-GB"><font face="Thorndale, serif">)</font></span></i><span lang="en-GB"><font face="Thorndale, serif">.
859Additionally, <b>disturbance_level_b</b>
860&lt;= </font></span><a href="#disturbance_level_t"><span lang="en-GB"><font face="Thorndale, serif">disturbance_level_t</font></span></a>
861<span lang="en-GB"><font face="Thorndale, serif">must
862also hold. <br> </font></span></p> <p><span lang="en-GB"><font face="Thorndale, serif">The
863parameter </font></span><a href="#create_disturbances"><span lang="en-GB"><font face="Thorndale, serif">create_disturbances</font></span></a><font face="Thorndale, serif"><span lang="en-GB">
864describes how to impose
865random perturbations to the horizontal velocity field</span></font><font face="Thorndale, serif"><span lang="en-GB">.</span>
866</font> </p> </td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="disturbance_level_t"></a><b>disturbance_level_t</b></p>
867</td> <td style="vertical-align: top;">R</td>
868<td style="vertical-align: top;"><i>zu(nz/3)</i></td>
869<td style="vertical-align: top;"> <p lang="en-GB"><font face="Thorndale, serif"><font size="3">Upper
870limit of the vertical range for which random perturbations are to be
871imposed on the horizontal wind field (</font></font>in <font face="Thorndale, serif"><font size="3">m).&nbsp;
872</font></font> </p> <p><span lang="en-GB"><font face="Thorndale, serif">This
873parameter must hold the condition <b>disturbance_level_t</b>
874&lt;= zu<i>(</i></font></span><i><a href="chapter_4.1.html#nz"><span lang="en-GB"><font face="Thorndale, serif">nz-1</font></span></a><span lang="en-GB"><font face="Thorndale, serif">)</font></span></i><span lang="en-GB"><font face="Thorndale, serif">.
875Additionally, </font></span><a href="#disturbance_level_b"><span lang="en-GB"><font face="Thorndale, serif">disturbance_level_b</font></span></a>
876<span lang="en-GB"><font face="Thorndale, serif">&lt;=
877<b>disturbance_level_t</b>
878must also hold.<br> </font></span></p> <p><span lang="en-GB"><font face="Thorndale, serif">The
879parameter </font></span><a href="#create_disturbances"><span lang="en-GB"><font face="Thorndale, serif">create_disturbances</font></span></a><font face="Thorndale, serif"><span lang="en-GB">
880describes how to impose
881random perturbations to the horizontal velocity field</span></font><font face="Thorndale, serif"><span lang="en-GB">.</span>
882</font> </p> </td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="do2d_at_begin"></a><b>do2d_at_begin</b></p>
883</td> <td style="vertical-align: top;">L<br> </td>
884<td style="vertical-align: top;">.F.<br> </td>
885<td style="vertical-align: top;"> <p>Output of 2d
886cross section data at the beginning of a run.&nbsp; </p> <p>The
887temporal intervals of output times of 2d cross section data (see <a href="chapter_4.2.html#data_output">data_output</a>)
888are usually determined with parameters <a href="chapter_4.2.html#dt_do2d_xy">dt_do2d_xy</a>, <a href="chapter_4.2.html#dt_do2d_xz">dt_do2d_xz</a>
889and <a href="chapter_4.2.html#dt_do2d_yz">dt_do2d_yz</a>.
890By assigning <b>do2d_at_begin</b> = <i>.T.</i>
891an additional output
892will be made at the
893beginning of a run (thus at the time t = 0 or at the respective
894starting times of restart runs).</p> </td> </tr> <tr>
895<td style="vertical-align: top;"> <p><a name="do3d_at_begin"></a><b>do3d_at_begin</b></p>
896</td> <td style="vertical-align: top;">L<br> </td>
897<td style="vertical-align: top;">.F.<br> </td>
898<td style="vertical-align: top;">Output of 3d volume data
899at the beginning
900of a run.<br><br>The temporal intervals of output times of
9013d volume data (see <a href="chapter_4.2.html#data_output">data_output</a>)
902is usually determined with parameter <a href="chapter_4.2.html#dt_do3d">dt_do3d</a>.
903By assigning <b>do3d_at_begin</b> = <i>.T.</i>
904an additional output
905will be made at the
906beginning of a run (thus at the time t = 0 or at the respective
907starting times of restart runs).</td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="do3d_compress"></a><b>do3d_compress</b></p>
908</td> <td style="vertical-align: top;">L<br> </td>
909<td style="vertical-align: top;">.F.<br> </td>
910<td style="vertical-align: top;"> <p>Output of data
911for 3d plots in compressed form.&nbsp; </p> <p>This
912parameter only applies for &nbsp;<a href="chapter_4.2.html#data_output_format">data_output_format</a>
913= <span style="font-style: italic;">'avs'</span>.</p><p>Output
914of 3d volume data may need huge amounts of disc storage
915(up to several Terabytes ore more). Data compression can serve to
916reduce this requirement. PALM is able to output 3d data in compressed
917form using 32-bit integers, if <span style="font-weight: bold;">do3d_compress</span>
918= <span style="font-style: italic;">.T.</span> is
919assigned. This
920yields a loss of accuracy, but the file size is clearly reduced. The
921parameter <a href="chapter_4.2.html#do3d_precision">do3d_precision</a>
922can be used to separately define the number of significant digits for
923each quantity.<br> </p> <p>So far compressed data
924output is only possible for Cray-T3E
925machines. Additional information for
926handling compressed data is given in <a href="chapter_4.5.6.html">chapter
9274.5.6</a>.</p> </td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="do3d_precision"></a><b>do3d_precision</b></p>
928</td> <td style="vertical-align: top;">C *
9297&nbsp; <br>
930&nbsp; (100)</td> <td style="vertical-align: top;">see
931right<br> </td> <td style="vertical-align: top;">
932<p>Significant digits in case of compressed data
933output.&nbsp; </p> <p>This parameter only applies for
934&nbsp;<a href="chapter_4.2.html#data_output_format">data_output_format</a>
935= <span style="font-style: italic;">'avs'</span>.</p><p>In
936case that data compression is used for output of 3d data
937(see <a href="chapter_4.2.html#do3d_compress">do3d_compress</a>),
938this parameter determines the number of significant digits
939which are to be output.<br> </p> <p>Fewer digits
940clearly reduce the amount
941of data. Assignments have to be given separately for each individual
942quantity via a character string of the form <span style="font-style: italic;">'&lt;quantity
943name&gt;&lt;number of
944significant digits&gt;'</span>, e.g. <span style="font-style: italic;">'pt2'</span>.
945Only those quantities listed in <a href="chapter_4.2.html#data_output">data_output</a>
946are admitted. Up to 9 significant digits are allowed (but large values
947are not very reasonable
948because they do not effect a significant compression).<br> </p>
949<p>The default assignment is <span style="font-weight: bold;">do3d_precision</span>
950= <span style="font-style: italic;">'u2'</span>, <span style="font-style: italic;">'v2'</span>, <span style="font-style: italic;">'w2'</span>, <span style="font-style: italic;">'p5'</span>, <span style="font-style: italic;">'pt2'</span>.</p> </td>
951</tr><tr> <td style="vertical-align: top;"> <p><a name="dt_laufparameter"></a><b>dt</b></p>
952</td> <td style="vertical-align: top;">R</td>
953<td style="vertical-align: top;"><i>variable</i></td>
954<td style="vertical-align: top;"> <p lang="en-GB"><font face="Thorndale, serif"><font size="3">Time
955step to be used by the 3d-model (</font></font>in <font face="Thorndale, serif"><font size="3">s).&nbsp;
956</font></font> </p> <p><span lang="en-GB"><font face="Thorndale, serif">This parameter</font></span>
957<font face="Thorndale, serif"><span lang="en-GB">is
958described in
959detail with the initialization parameters (see</span></font><span lang="en-GB"><font face="Thorndale, serif"> </font></span><a href="chapter_4.1.html#dt"><span lang="en-GB"><font face="Thorndale, serif">dt</font></span></a><font face="Thorndale, serif"><span lang="en-GB">).
960Additionally, it may be
961used as a run parameter and then applies to all restart runs (until it
962is changed again). A switch from a constant time step to a variable
963time step can be achieved with <b>dt</b> = <i>-1.0</i>.</span>
964</font> </p> </td> </tr> <tr> <td style="vertical-align: top;"><a name="dt_averaging_input"></a><span style="font-weight: bold;">dt_averaging_input</span><br>
965</td> <td style="vertical-align: top;">R<br> </td>
966<td style="vertical-align: top;"><span style="font-style: italic;">0.0</span><br> </td>
967<td style="vertical-align: top;">Temporal interval
968of&nbsp;data which are subject to temporal averaging (in s).<br><br>By
969default, data from each timestep within the interval defined by <a href="chapter_4.2.html#averaging_interval">averaging_interval</a>
970are used for calculating the temporal average. By choosing <span style="font-weight: bold;">dt_averaging_input</span>
971&gt; <span lang="en-GB"><font face="Thorndale, serif"> </font></span><a href="chapter_4.1.html#dt"><span lang="en-GB"><font face="Thorndale, serif">dt</font></span></a><font face="Thorndale, serif"><span lang="en-GB"></span></font><span lang="en-GB"></span><span style="font-style: italic;"></span>,
972the number of time levels entering the average can be minimized. This
973reduces the CPU-time of a run but may worsen the quality of the
974average's statistics.<br><br><font face="Thorndale, serif"><span lang="en-GB">With
975variable time step (see <span style="font-weight: bold;">dt</span>),
976the number of time levels entering the average can vary from one
977averaging interval to the next (for a more detailed explanation see </span></font><font><a href="#averaging_interval"><span lang="en-GB"><font face="Thorndale, serif">averaging_interval</font></span></a>)</font><font face="Thorndale, serif"><span lang="en-GB">. It
978is approximately given by the quotient of <span style="font-weight: bold;">averaging_interval</span> /
979MAX(<span style="font-weight: bold;"> dt_averaging_input</span>,
980<span style="font-weight: bold;">dt</span>) (which
981gives a more or less exact value if a fixed timestep is used and if
982this is an integral divisor of <span style="font-weight: bold;">dt_averaging_input</span>).</span></font>&nbsp;
983<br><br><span style="font-weight: bold;">Example:</span><br>With
984an averaging interval of 100.0 s and <span style="font-weight: bold;">dt_averaging_input</span> =
985<span style="font-style: italic;">10.0</span>,
986the time levels entering the average have a (minimum) distance of 10.0
987s (their distance may of course be larger if the current timestep is
988larger than 10.0 s), so the average is calculated from the data of
989(maximum) 10 time levels.<br><br><font face="Thorndale, serif"><span lang="en-GB">It
990is allowed
991to change <b>dt_averaging_input</b> during a job chain. If
992the last averaging
993interval of the run previous to the change could not be completed (i.e.
994has to be finished in the current run), the individual profiles and/or
995spectra entering the averaging are not uniformly distributed over the
996averaging interval.<br><br></span></font>Parameter&nbsp;<a href="#dt_averaging_input_pr">dt_averaging_input_pr</a>&nbsp;can
997be used to define&nbsp;a different temporal interval&nbsp;for
998vertical profile data and spectra.<br> </td> </tr>
999<tr> <td style="vertical-align: top;"> <p><a name="dt_averaging_input_pr"></a><b>dt_averaging_input_pr</b></p>
1000</td> <td style="vertical-align: top;">R</td>
1001<td style="vertical-align: top;"><span style="font-style: italic;">value of <a href="#dt_averaging_input">dt_<br>averaging_<br>input</a></span></td>
1002<td style="vertical-align: top;"> <p lang="en-GB">Temporal
1003interval of&nbsp;data which are subject to temporal averaging of <font face="Thorndale, serif"><font size="3">vertical
1004profiles and/or spectra&nbsp;(</font></font>in <font face="Thorndale, serif"><font size="3">s).&nbsp;
1005</font></font> </p> <p>By default, data from
1006each timestep within the interval defined by<font face="Thorndale, serif"><span lang="en-GB"> </span></font><a href="#averaging_interval_pr"><span lang="en-GB"><font face="Thorndale, serif">averaging_interval_pr</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><span lang="en-GB"><font face="Thorndale, serif">and </font></span><a href="#averaging_interval_sp"><span lang="en-GB"><font face="Thorndale, serif">averaging_interval_sp</font></span></a><span lang="en-GB"><font face="Thorndale, serif"> </font></span>are
1007used for calculating the temporal average.&nbsp;By choosing <span style="font-weight: bold;">dt_averaging_input_pr</span>
1008&gt; <span lang="en-GB"><font face="Thorndale, serif"> </font></span><a href="chapter_4.1.html#dt"><span lang="en-GB"><font face="Thorndale, serif">dt</font></span></a><font face="Thorndale, serif"><span lang="en-GB"></span></font><span lang="en-GB"></span><span style="font-style: italic;"></span>,
1009the number of time levels entering the average can be minimized. This
1010reduces the CPU-time of a run but may worsen the quality of the
1011average's statistics. <span lang="en-GB"><font face="Thorndale, serif"><span style="font-weight: bold;"></span><span style="font-weight: bold;"></span></font></span><a href="chapter_4.1.html#dt"><span lang="en-GB"></span></a><font face="Thorndale, serif"><span lang="en-GB"></span></font><span lang="en-GB"></span><br> </p><p>For
1012more explanations see parameter <a href="#dt_averaging_input">dt_averaging_input</a>.<a href="chapter_4.1.html#dt"><span lang="en-GB"></span></a><font face="Thorndale, serif"><span lang="en-GB"></span></font></p></td>
1013</tr> <tr> <td style="vertical-align: top;"><a name="dt_data_output"></a><span style="font-weight: bold;">dt_data_output</span><br>
1014</td> <td style="vertical-align: top;">R<br> </td>
1015<td style="vertical-align: top;"><span style="font-style: italic;">9999999.9</span><br>
1016</td> <td style="vertical-align: top;"><p lang="en-GB"><font face="Thorndale"><font face="Thorndale, serif">Temporal interval</font>
1017at which&nbsp;data (3d volume data (instantaneous or time
1018averaged),
1019cross sections (instantaneous or time averaged), vertical profiles,
1020spectra) shall be output (</font>in <font face="Thorndale">s).&nbsp;</font></p>
1021<span lang="en-GB"><font face="Thorndale">If
1022data output&nbsp;is switched on (see </font></span><a href="chapter_4.2.html#data_output"><span lang="en-GB"><font face="Thorndale">data_output</font></span></a><span lang="en-GB"><font face="Thorndale">, <a href="#data_output_pr">data_output_pr</a>, <a href="#data_output_sp">data_output_sp</a>, and </font></span><a href="chapter_4.2.html#section_xy"><span lang="en-GB"><font face="Thorndale">section_xy</font></span></a><span lang="en-GB"><font face="Thorndale">), this
1023parameter can be used to
1024assign the temporal interval at which these data shall be
1025output. </font></span><span lang="en-GB"><font face="Thorndale">Output can be skipped at the beginning of a
1026simulation using parameter <a href="#skip_time_data_output">skip_time_data_output</a>,
1027which has zero value by default. </font></span><span lang="en-GB"><font face="Thorndale">Reference
1028time is the beginning of the simulation, i.e. output
1029takes place at times t = <b>skip_time_data_output +
1030dt_data_output</b>, <span style="font-weight: bold;">skip_time_data_output</span>
1031+ 2*<b>dt_data_output</b>, <span style="font-weight: bold;">skip_time_data_output</span>
1032+ 3*<b>dt_data_output</b>,
1033etc. Since output is only done at the discrete time levels given by
1034the&nbsp;timestep used, the actual output times can slightly
1035deviate
1036from these theoretical values</font></span><a href="chapter_4.2.html#dt_dopr_zeitpunkte"><span lang="en-GB"></span></a><span lang="en-GB"><font face="Thorndale">.<br><br>Individual temporal
1037intervals for the different output quantities can be assigned using
1038parameters <a href="#dt_do3d">dt_do3d</a>, <a href="#dt_do2d_xy">dt_do2d_xy</a>, <a href="dt_do2d_xz">dt_do2d_xz</a>, <a href="#dt_do2d_yz">dt_do2d_yz</a>, <a href="#dt_dopr">dt_dopr</a>, <a href="#dt_dosp">dt_dosp</a>,
1039and <a href="#dt_data_output_av">dt_data_output_av</a>.</font></span>
1040</td> </tr> <tr> <td style="vertical-align: top;"><a name="dt_data_output_av"></a><span style="font-weight: bold;">dt_data_output_av</span><br>
1041</td> <td style="vertical-align: top;">R<br> </td>
1042<td style="vertical-align: top;"><i>value of
1043&nbsp;<a href="chapter_4.2.html#dt_data_output">dt_data_<br>output</a></i>
1044</td> <td style="vertical-align: top;"><p lang="en-GB"><font face="Thorndale"><font face="Thorndale, serif">Temporal interval</font>
1045at which time averaged 3d volume data and/or 2d cross section data
1046shall be output (</font>in <font face="Thorndale">s).&nbsp;</font></p><span lang="en-GB"><font face="Thorndale">If data
1047output of time averaged 2d and 3d data is switched on (see </font></span><a href="chapter_4.2.html#data_output"><span lang="en-GB"><font face="Thorndale">data_output</font></span></a>&nbsp;<span lang="en-GB"><font face="Thorndale">and </font></span><a href="chapter_4.2.html#section_xy"><span lang="en-GB"><font face="Thorndale">section_xy</font></span></a><span lang="en-GB"><font face="Thorndale">), this
1048parameter can be used to
1049assign the temporal interval at which they shall be
1050output. </font></span><span lang="en-GB"><font face="Thorndale">Output can be skipped at the beginning of a
1051simulation using parameter <a href="#skip_time_data_output_av">skip_time_data_output_av</a>,
1052which has zero value by default. </font></span><span lang="en-GB"><font face="Thorndale">Reference
1053time is the beginning of the simulation, i.e. output
1054takes place at times t = <b>skip_time_data_output_av +
1055dt_data_output_av</b>, <span style="font-weight: bold;">skip_time_data_output_av</span>
1056+ 2*<b>dt_data_output_av</b>, <span style="font-weight: bold;">skip_time_data_output_av</span>
1057+ 3*<b>dt_data_output_av</b>,
1058etc. Since output is only done at the discrete time levels given by
1059the&nbsp;timestep used, the actual output times can slightly
1060deviate from
1061these theoretical values</font></span><a href="chapter_4.2.html#dt_dopr_zeitpunkte"><span lang="en-GB"></span></a><span lang="en-GB"><font face="Thorndale">.<br><br></font></span>The
1062length of the averaging interval is controlled via parameter <a href="chapter_4.2.html#averaging_interval">averaging_interval</a>.</td>
1063</tr><tr> <td style="vertical-align: top;"> <p><a name="dt_disturb"></a><b>dt_disturb</b></p>
1064</td> <td style="vertical-align: top;">R</td>
1065<td style="vertical-align: top;"><i>9999999.9</i></td>
1066<td style="vertical-align: top;"> <p lang="en-GB"><font face="Thorndale"><font face="Thorndale, serif">Temporal
1067interval</font> at which random
1068perturbations are to be imposed on the horizontal velocity field
1069(</font>in <font face="Thorndale">s).&nbsp; </font>
1070</p> <p><span lang="en-GB"><font face="Thorndale, serif">The parameter </font></span><a href="#create_disturbances"><span lang="en-GB"><font face="Thorndale, serif">create_disturbances</font></span></a><font face="Thorndale, serif"><span lang="en-GB">
1071describes how to impose
1072random perturbations to the horizontal velocity field</span></font><font face="Thorndale, serif"><span lang="en-GB">.</span>
1073</font> </p> </td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="dt_dopr"></a><b>dt_dopr</b></p>
1074</td> <td style="vertical-align: top;">R</td>
1075<td style="vertical-align: top;"><i>value of
1076&nbsp;<a href="#dt_data_output">dt_data_<br>output</a></i></td>
1077<td style="vertical-align: top;"> <p><span lang="en-GB"><font face="Thorndale">Temporal
1078interval at
1079which data&nbsp;of vertical profiles shall be output (to local
1080file <a href="chapter_3.4.html#DATA_1D_PR_NETCDF">DATA_1D_PR_NETCDF</a>
1081or/and </font></span><a href="chapter_3.4.html#PLOT1D_DATA"><span lang="en-GB"><font face="Thorndale">PLOT1D_DATA</font></span></a><span lang="en-GB"><font face="Thorndale">) (</font></span>in
1082<span lang="en-GB"><font face="Thorndale">s).&nbsp;
1083</font></span> </p> <p><span lang="en-GB"><font face="Thorndale">If output of
1084horizontally averaged vertical profiles is switched on (see </font></span><a href="chapter_4.2.html#data_output_pr"><span lang="en-GB"><font face="Thorndale">data_output_pr</font></span></a><span lang="en-GB"><font face="Thorndale">), </font></span><span lang="en-GB"><font face="Thorndale">this
1085parameter can be used to
1086assign the temporal interval at which profile data shall be output.</font></span><span lang="en-GB"><font face="Thorndale"> </font></span><span lang="en-GB"><font face="Thorndale">Output can
1087be skipped at the beginning of a simulation using parameter <a href="#skip_time_dopr">skip_time_dopr</a>, which has
1088zero value by default. </font></span><span lang="en-GB"></span><span lang="en-GB"><font face="Thorndale">Reference
1089time is the beginning
1090of the simulation, thus t = 0,&nbsp;</font></span><span lang="en-GB"><font face="Thorndale">i.e. output
1091takes place at times t = <b>skip_time_dopr + dt_dopr</b>, <span style="font-weight: bold;">skip_time_dopr</span> + 2*<b>dt_dopr</b>,
1092<span style="font-weight: bold;">skip_time_dopr</span>
1093+ 3*<b>dt_dopr</b>,
1094etc.</font></span><span lang="en-GB"><font face="Thorndale"> Since
1095profiles can not be calculated for times lying within a time step
1096interval, the output times can deviate from these theoretical values.
1097If a time step ranges from t = 1799.8 to t = 1800.2, then in the
1098example above the output would take place at t = 1800.2. In general,
1099the output always lie between t = 1800.0 and t = 1800.0 + </font></span><a href="chapter_4.1.html#dt"><span lang="en-GB"><font face="Thorndale">dt</font></span></a><span lang="en-GB"><font face="Thorndale">. If the
1100model uses a variable time step, these
1101deviations from the theoretical output times will of course be
1102different for each output time.<br> </font></span></p>
1103<p><span lang="en-GB"><font face="Thorndale">In
1104order to
1105guarantee an output of profile data at the end of a simulation (see </font></span><font><a href="chapter_4.1.html#end_time"><span lang="en-GB"><font face="Thorndale">end_time</font></span></a></font><span lang="en-GB"><font face="Thorndale">) in any way</font></span><span lang="en-GB"><font face="Thorndale">,&nbsp;
1106<span style="font-weight: bold;">end_time</span>
1107should be equal or a little bit
1108larger than the respective theoretical output time. For example, if <b>dt_dopr</b>
1109= <i>900.0</i><span style="font-style: italic;">
1110</span>and 3600.0
1111seconds are to be simulated, then <b>end_time</b>
1112&gt;= 3600.0 should be chosen.</font></span><a href="chapter_4.1.html#dt"><span lang="en-GB"></span></a><span lang="en-GB"><font face="Thorndale"><span style="font-weight: bold;"></span>&nbsp; </font></span>
1113</p> <p><span lang="en-GB"><font face="Thorndale">A selection of
1114profiles to be output can be done via parameter </font></span><a href="chapter_4.2.html#data_output_pr"><span lang="en-GB"><font face="Thorndale">data_output_pr</font></span></a><span lang="en-GB"><font face="Thorndale">.&nbsp;</font></span>
1115</p> </td> </tr> <tr> <td style="vertical-align: top;"><a name="dt_dopr_listing"></a><span style="font-weight: bold;">dt_dopr_listing</span><br>
1116</td> <td style="vertical-align: top;">R<br> </td>
1117<td style="vertical-align: top;"><i>9999999.9</i></td>
1118<td style="vertical-align: top;"> <p><span lang="en-GB"><font face="Thorndale, serif">Temporal
1119interval</font> at which data <font face="Thorndale">of
1120vertical
1121profiles shall be output (output for printouts, local file </font></span><a href="chapter_3.4.html#LIST_PROFIL"><span lang="en-GB"><font face="Thorndale">LIST_PROFIL</font></span></a><span lang="en-GB"><font face="Thorndale">) (</font></span>in
1122<span lang="en-GB"><font face="Thorndale">s).&nbsp;</font></span>
1123</p> <p>T<span lang="en-GB"></span><a href="chapter_4.2.html#pr1d"><span lang="en-GB"></span></a><span lang="en-GB"></span><span lang="en-GB"><font face="Thorndale">his
1124parameter can be used to
1125assign the temporal interval at which profile data shall be output.</font></span><span lang="en-GB"><font face="Thorndale"> Reference
1126time is the beginning
1127of the simulation, thus t = 0. For example if <b>dt_dopr_listing</b>
1128= 1800.0,
1129then output takes place at t = 1800.0, 3600.0, 5400.0, etc. Since
1130profiles can not be calculated for times lying within a time step
1131interval, the output times can deviate from these theoretical values.
1132If a time step ranges from t = 1799.8 to t = 1800.2, then in the
1133example above the output would take place at t = 1800.2. In general,
1134the output always lie between t = 1800.0 and t = 1800.0 + </font></span><a href="chapter_4.1.html#dt"><span lang="en-GB"><font face="Thorndale">dt</font></span></a> <span lang="en-GB"><font face="Thorndale">(numbers
1135are related to
1136the
1137example above). If the model uses a variable time step, these
1138deviations from the theoretical output times will of course be
1139different for each output time.<br> </font></span></p>
1140<p><span lang="en-GB"><font face="Thorndale">In
1141order to
1142guarantee an output of profile data at the end of a simulation (see </font></span><font><a href="chapter_4.1.html#end_time"><span lang="en-GB"><font face="Thorndale">end_time</font></span></a></font><span lang="en-GB"><font face="Thorndale">) in any way</font></span><span lang="en-GB"><font face="Thorndale">,&nbsp;
1143<span style="font-weight: bold;">end_time</span>
1144should be a little bit
1145larger than the respective theoretical output time. For example, if <b>dt_dopr_listing</b>
1146= <i>900.0</i><span style="font-style: italic;">
1147</span>and 3600.0
1148seconds are to be simulated, then it should be at least&nbsp; <b>end_time</b>
1149&gt; 3600.0 + </font></span><a href="chapter_4.1.html#dt"><span lang="en-GB"><font face="Thorndale">dt</font></span></a><span lang="en-GB"><font face="Thorndale">. If
1150variable time steps are used
1151(which is the default), <span style="font-weight: bold;">dt</span>
1152should be properly estimated.&nbsp; </font></span> </p>
1153<p><span lang="en-GB"><font face="Thorndale">Data
1154and output
1155format of the file </font></span><a href="chapter_3.4.html#LIST_PROFIL"><span lang="en-GB"><font face="Thorndale">LIST_PROFIL</font></span></a>
1156<span lang="en-GB"><font face="Thorndale">is
1157internally fixed. In this file
1158the profiles of the most important model variables are arranged in
1159adjacent columns.</font></span> </p> </td> </tr>
1160<tr> <td style="vertical-align: top;"> <p><a name="dt_dots"></a><b>dt_dots</b></p>
1161</td> <td style="vertical-align: top;">R</td>
1162<td style="vertical-align: top;"><span style="font-style: italic;">see right</span></td>
1163<td style="vertical-align: top;"> <p lang="en-GB"><font face="Thorndale"><font face="Thorndale, serif">Temporal
1164interval</font> at which&nbsp;time series data shall be
1165output (</font>in <font face="Thorndale">s).&nbsp;</font>
1166</p> <p>The default interval for the output of timeseries
1167is calculated as shown below (this tries to minimize the number of
1168calls of <span style="font-family: Courier New,Courier,monospace;">flow_statistics</span>)</p><p style="font-family: Courier New,Courier,monospace;">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
1169IF ( <a href="#averaging_interval_pr">averaging_interval_pr</a>
1170== 0.0 )&nbsp; THEN<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
1171<span style="font-weight: bold;">dt_dots</span> =
1172MIN( <a href="#dt_run_control">dt_run_control</a>, <a href="#dt_dopr">dt_dopr</a> )<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
1173ELSE<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
1174<span style="font-weight: bold;">dt_dots</span> =
1175MIN( dt_run_control, <a href="#dt_averaging_input_pr">dt_averaging_input_pr</a>
1176)<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
1177ENDIF</p><p>This parameter can be used to
1178assign the temporal interval at which data points shall be output. <span lang="en-GB"><font face="Thorndale">Reference
1179time is the beginning of
1180&nbsp;the simulation, i.e. output takes place at times t = <b>dt_dots</b>,
11812*<b>dt_dots</b>, 3*<b>dt_dots</b>, etc. The
1182actual output times can
1183deviate from these theoretical values (see </font></span><a href="#dt_dopr_zeitpunkte"><span lang="en-GB"><font face="Thorndale">dt_dopr</font></span></a><span lang="en-GB"><font face="Thorndale">).&nbsp;
1184Is <b>dt_dots</b> &lt; </font></span><a href="chapter_4.1.html#dt"><span lang="en-GB"><font face="Thorndale">dt</font></span></a><span lang="en-GB"><font face="Thorndale">, then data
1185of the time series are
1186written after each time step (if this is requested it should be <b>dt_dots</b>
1187= <i>0</i>).</font></span></p><p><span lang="en-GB"><font face="Thorndale">The default
1188value of <span style="font-weight: bold;">dt_dots</span>
1189is calculated as follows:</font></span></p>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
1190IF ( <a href="#averaging_interval_pr">averaging_interval_pr</a>
1191== 0.0 )&nbsp; THEN<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
1192<span style="font-weight: bold;">dt_dots</span> =
1193MIN( <a href="#dt_run_control">dt_run_control</a>, <a href="#dt_dopr">dt_dopr</a> )<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
1194ELSE<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
1195<span style="font-weight: bold;">dt_dots</span> =
1196MIN( <span style="font-weight: bold;">dt_run_control</span>,
1197<a href="#dt_averaging_input_pr">dt_averaging_input_pr</a>
1198)<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
1199ENDIF<br><br>(which minimizes the number of calls of
1200routine flow_statistics).<br><p>By default time series data
1201is output to the local file <a href="chapter_3.4.html#DATA_1D_TS_NETCDF">DATA_1D_TS_NETCDF</a>.
1202Because of the default settings of <span style="font-weight: bold;">dt_dots</span>,
1203it will&nbsp;generally be created for each model run. The file's
1204format is NetCDF.&nbsp; Further details about processing NetCDF
1205data are given in chapter <a href="chapter_4.5.1.html">4.5.1</a>.</p>The
1206file contains the following timeseries quantities (the first column
1207gives the name of the quantities as used in the NetCDF file):<br><table style="text-align: left; width: 100%;" cellpadding="2" cellspacing="2"> <tbody> <tr> <td style="font-style: italic; vertical-align: middle;">E<br>
1208</td> <td style="vertical-align: top;">Total
1209kinetic energy of
1210the flow (in m<sup>2</sup>/s<sup>2</sup>)
1211(normalized with respect to the total number of grid points).</td>
1212</tr> <tr> <td style="font-style: italic; vertical-align: middle;">E*<br>
1213</td> <td style="vertical-align: top;">Perturbation
1214kinetic
1215energy of the flow (in m<sup>2</sup>/s<sup>2</sup>)<sup>
1216</sup>(normalized
1217with respect to the total number of grid
1218points)</td> </tr> <tr> <td style="vertical-align: top; font-style: italic;">dt<br>
1219</td> <td style="vertical-align: top;">Time step
1220size (in s).</td> </tr> <tr> <td style="vertical-align: top; font-style: italic;">u<sub>*</sub></td>
1221<td style="vertical-align: top;">Friction velocity (in
1222m/s)
1223(horizontal average).</td> </tr> <tr> <td style="vertical-align: top; font-style: italic;">w<sub>*</sub></td>
1224<td style="vertical-align: top;">Vertical velocity scale
1225of
1226the CBL (in m/s) (horizontal average)</td> </tr> <tr>
1227<td style="vertical-align: top; font-style: italic;">th<sub>*</sub></td>
1228<td style="vertical-align: top;">Temperature
1229scale (Prandtl layer), defined as <i>w"pt"0
1230/&nbsp;</i><i>u<sub>*</sub></i>
1231(horizontal
1232average) (in K).</td> </tr> <tr> <td style="vertical-align: top; font-style: italic;">umax<br>
1233</td> <td style="vertical-align: top;">Maximum
1234u-component of the
1235velocity (in m/s).</td> </tr> <tr> <td style="vertical-align: top; font-style: italic;">vmax<br>
1236</td> <td style="vertical-align: top;">Maximum
1237v-component of the
1238velocity (in m/s).</td> </tr> <tr> <td style="vertical-align: top; font-style: italic;">wmax<br>
1239</td> <td style="vertical-align: top;">Maximum
1240w-component of the
1241velocity (in m/s).</td> </tr> <tr> <td style="vertical-align: top; font-style: italic;">div_old<br>
1242</td> <td style="vertical-align: top;">Divergence
1243of the velocity
1244field before the pressure
1245solver has been called (normalized with respect to the total number of
1246grid points) (in 1/s).</td> </tr> <tr> <td style="vertical-align: top; font-style: italic;">div_new</td>
1247<td style="vertical-align: top;">Divergence of the
1248velocity
1249field after the pressure
1250solver has been called (normalized with respect to the total number of
1251grid points) (in 1/s).</td> </tr> <tr> <td style="vertical-align: top; font-style: italic;">z_i_wpt</td>
1252<td style="vertical-align: top;">Height of the convective
1253boundary layer (horizontal average)
1254determined by the height of the minimum sensible heat flux (in m).</td>
1255</tr> <tr> <td style="vertical-align: top; font-style: italic;">z_i_pt</td>
1256<td style="vertical-align: top;">Height of the convective
1257boundary layer (horizontal average)
1258determined by the temperature profile (in m).</td> </tr> <tr>
1259<td style="vertical-align: top; font-style: italic;">w"pt"0</td>
1260<td style="vertical-align: top;">Subgrid-scale sensible
1261heat flux near the surface (horizontal
1262average)
1263between z = 0 and z = z<sub>p</sub> = zu(1) (there it
1264corresponds to
1265the total heat flux) (in K m/s).</td> </tr> <tr> <td style="vertical-align: top; font-style: italic;">w"pt"</td>
1266<td style="vertical-align: top;">Subgrid-scale heat flux
1267(horizontal average) for z = zw(1) (in K
1268m/s).</td> </tr> <tr> <td style="vertical-align: top; font-style: italic;">wpt</td>
1269<td style="vertical-align: top;">Total heat flux
1270(horizontal average) for z = zw(1) (in K m/s).</td> </tr> <tr>
1271<td style="vertical-align: top; font-style: italic;">pt(0)</td>
1272<td style="vertical-align: top;">Potential temperature at
1273the surface (horizontal average) (in K).</td> </tr> <tr>
1274<td style="vertical-align: top; font-style: italic;">pt(zp)</td>
1275<td style="vertical-align: top;">Potential temperature for
1276z = zu(1) (horizontal average) (in K).</td> </tr> <tr>
1277<td style="vertical-align: top; font-style: italic;">splptx</td>
1278<td style="vertical-align: top;">Percentage of grid points
1279using upstream scheme along x with
1280upstream-spline advection switched on.</td> </tr> <tr>
1281<td style="vertical-align: top; font-style: italic;">splpty</td>
1282<td style="vertical-align: top;">Percentage of grid points
1283using upstream scheme along y with
1284upstream-spline
1285advection switched on.</td> </tr> <tr> <td style="vertical-align: top; font-style: italic;">splptz</td>
1286<td style="vertical-align: top;">Percentage of grid points
1287using upstream scheme along z with
1288upstream-spline
1289advection switched on.<br> </td> </tr> <tr> <td style="vertical-align: top; font-style: italic;">L</td>
1290<td style="vertical-align: top;">Monin-Obukhov length.</td>
1291</tr> </tbody> </table><br>Additionally, the
1292user can add his own timeseries quantities to the file, by using the
1293user-interface subroutines<span style="font-family: Courier New,Courier,monospace;"> <a href="chapter_3.5.1.html#user_init">user_init</a> </span>and<span style="font-family: Courier New,Courier,monospace;"> <a href="chapter_3.5.1.html#user_statistics">user_statistics</a></span>.
1294These routines contain (as comment lines) a simple example how to do
1295this.<br><br>Time series data refers to the total
1296domain, but time series for subdomains can also be output (see <a href="chapter_4.1.html#statistic_regions">statistic_regions</a>).
1297However, the following time series always present the values of the
1298total model domain (even with output for subdomains): <i>umax</i>,
1299<i>vmax</i>, <i>wmax</i>, <i>div_old</i>,
1300<i>div_new</i>.</td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="dt_do2d_xy"></a><b>dt_do2d_xy</b></p>
1301</td> <td style="vertical-align: top;">R</td>
1302<td style="vertical-align: top;"><i>value of
1303&nbsp;<a href="chapter_4.2.html#dt_data_output">dt_data_<br>output</a></i></td>
1304<td style="vertical-align: top;"> <p lang="en-GB"><font face="Thorndale"><font face="Thorndale, serif">Temporal
1305interval</font> at which&nbsp;horizontal cross section data
1306shall be output (</font>in <font face="Thorndale">s).&nbsp;
1307</font> </p> <p><span lang="en-GB"><font face="Thorndale">If output of
1308horizontal cross sections is switched on (see </font></span><a href="#data_output"><span lang="en-GB"><font face="Thorndale">data_output</font></span></a>
1309<span lang="en-GB"><font face="Thorndale">and
1310</font></span><a href="#section_xy"><span lang="en-GB"><font face="Thorndale">section_xy</font></span></a><span lang="en-GB"><font face="Thorndale">), this
1311parameter can be used to
1312assign the temporal interval at which cross section data shall be
1313output. </font></span><span lang="en-GB"><font face="Thorndale">Output can be skipped at the beginning of a
1314simulation using parameter <a href="#skip_time_do2d_xy">skip_time_do2d_xy</a>,
1315which has zero value by default. </font></span><span lang="en-GB"><font face="Thorndale">Reference
1316time is the beginning of the simulation, i.e. output
1317takes place at times t = <b>skip_time_do2d_xy + dt_do2d_xy</b>,
1318<span style="font-weight: bold;">skip_time_do2d_xy</span>
1319+ 2*<b>dt_do2d_xy</b>, <span style="font-weight: bold;">skip_time_do2d_xy</span>
1320+ 3*<b>dt_do2d_xy</b>,
1321etc. The actual output times can deviate from these theoretical values
1322(see </font></span><a href="#dt_dopr_zeitpunkte"><span lang="en-GB"><font face="Thorndale">dt_dopr</font></span></a><span lang="en-GB"><font face="Thorndale">).<br>
1323</font></span></p> <p><span lang="en-GB"><font face="Thorndale">Parameter </font></span><a href="#do2d_at_begin"><span lang="en-GB"><font face="Thorndale">do2d_at_begin</font></span></a>
1324has to be used if an additional output is wanted at the start of a run <span lang="en-GB"><font face="Thorndale">(thus at
1325the time t = 0 or at the
1326respective starting times of restart runs).</font></span> </p>
1327</td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="dt_do2d_xz"></a><b>dt_do2d_xz</b></p>
1328</td> <td style="vertical-align: top;">R</td>
1329<td style="vertical-align: top;"><i>value of
1330&nbsp;<a href="chapter_4.2.html#dt_data_output">dt_data_<br>output</a></i></td>
1331<td style="vertical-align: top;"> <p lang="en-GB"><font face="Thorndale"><font face="Thorndale, serif">Temporal
1332interval</font> at which&nbsp;vertical cross sections data
1333(xz) shall be output (</font>in <font face="Thorndale">s).&nbsp;
1334</font> </p> <p><span lang="en-GB"><font face="Thorndale">If output of
1335horizontal cross sections is switched on (see </font></span><a href="#data_output"><span lang="en-GB"><font face="Thorndale">data_output</font></span></a>
1336<span lang="en-GB"><font face="Thorndale">and
1337</font></span><a href="#section_xz"><span lang="en-GB"><font face="Thorndale">section_xz</font></span></a><span lang="en-GB"><font face="Thorndale">),
1338this parameter can be used to assign the temporal interval at which
1339cross section data shall be output. </font></span><span lang="en-GB"><font face="Thorndale">Output can
1340be skipped at the beginning of a simulation using parameter <a href="#skip_time_do2d_xz">skip_time_do2d_xz</a>, which
1341has zero value by default. </font></span><span lang="en-GB"></span><span lang="en-GB"><font face="Thorndale">Reference time is the beginning of
1342the simulation, i.e. output takes place at times t = <b>skip_time_do2d_xz
1343+ dt_do2d_xz</b>,
1344<span style="font-weight: bold;">skip_time_do2d_xz</span>
1345+ 2*<b>dt_do2d_xz</b>, <span style="font-weight: bold;">skip_time_do2d_xz</span>
1346+ 3*<b>dt_do2d_xz</b>, etc. The actual output times
1347can deviate from these theoretical values (see </font></span><a href="#dt_dopr_zeitpunkte"><span lang="en-GB"><font face="Thorndale">dt_dopr</font></span></a><span lang="en-GB"><font face="Thorndale">).<br>
1348</font></span></p> <p><span lang="en-GB"><font face="Thorndale">Parameter </font></span><a href="#do2d_at_begin"><span lang="en-GB"><font face="Thorndale">do2d_at_begin</font></span></a>
1349has to be used if an additional output is wanted at the start of a run <span lang="en-GB"><font face="Thorndale">(thus at
1350the time t = 0 or at the
1351respective starting times of restart runs).</font></span> </p>
1352</td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="dt_do2d_yz"></a><b>dt_do2d_yz</b></p>
1353</td> <td style="vertical-align: top;">R</td>
1354<td style="vertical-align: top;"><i>value of
1355&nbsp;<a href="chapter_4.2.html#dt_data_output">dt_data_<br>output</a></i></td>
1356<td style="vertical-align: top;"> <p lang="en-GB"><font face="Thorndale"><font face="Thorndale, serif">Temporal
1357interval</font> at which&nbsp;vertical cross section data
1358(yz) shall be output (</font>in s<font face="Thorndale">).&nbsp;
1359</font> </p> <p><span lang="en-GB"><font face="Thorndale">If output of
1360horizontal cross sections is switched on (see </font></span><a href="#data_output"><span lang="en-GB"><font face="Thorndale">data_output</font></span></a>
1361<span lang="en-GB"><font face="Thorndale">and
1362</font></span><a href="#section_yz"><span lang="en-GB"><font face="Thorndale">section_yz</font></span></a><span lang="en-GB"><font face="Thorndale">),
1363this parameter can be used to assign the temporal interval at which
1364cross section data shall be output. </font></span><span lang="en-GB"><font face="Thorndale">Output can
1365be skipped at the beginning of a simulation using parameter <a href="#skip_time_do2d_yz">skip_time_do2d_yz</a>, which
1366has zero value by default. </font></span><span lang="en-GB"></span><span lang="en-GB"></span><span lang="en-GB"><font face="Thorndale">Reference
1367time is the beginning of
1368the simulation, i.e. output takes place at times t = <b>skip_time_do2d_yz
1369+ dt_do2d_yz</b>,
1370<span style="font-weight: bold;">skip_time_do2d_yz</span>
1371+ 2*<b>dt_do2d_yz</b>, <span style="font-weight: bold;">skip_time_do2d_yz
1372</span>+ 3*<b>dt_do2d_yz</b>, etc. The actual output
1373times
1374can deviate from these theoretical values (see </font></span><a href="#dt_dopr_zeitpunkte"><span lang="en-GB"><font face="Thorndale">dt_dopr</font></span></a><span lang="en-GB"><font face="Thorndale">).<br>
1375</font></span></p> <p><span lang="en-GB"><font face="Thorndale">Parameter </font></span><a href="#do2d_at_begin"><span lang="en-GB"><font face="Thorndale">do2d_at_begin</font></span></a>
1376has to be used if an additional output is wanted at the start of a run <span lang="en-GB"><font face="Thorndale">(thus at
1377the time t = 0 or at the
1378respective starting times of restart runs).</font></span> </p>
1379</td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="dt_do3d"></a><b>dt_do3d</b></p>
1380</td> <td style="vertical-align: top;">R</td>
1381<td style="vertical-align: top;"><i>value of
1382&nbsp;<a href="chapter_4.2.html#dt_data_output">dt_data_<br>output</a></i></td>
1383<td style="vertical-align: top;"> <p lang="en-GB"><font face="Thorndale"><font face="Thorndale, serif">Temporal
1384interval</font> at which 3d volume data shall be output (</font>in
1385<font face="Thorndale">s).&nbsp; </font> </p>
1386<p><span lang="en-GB"><font face="Thorndale">If
1387output of
13883d-volume data is switched on (see </font></span><font><a href="#data_output"><span lang="en-GB"><font face="Thorndale">data_output</font></span></a>)<span style="font-family: thorndale;">, this parameter can be used
1389to assign
1390th</span></font><span lang="en-GB"><font face="Thorndale">e temporal
1391interval at which 3d-data shall be output. </font></span><span lang="en-GB"><font face="Thorndale">Output can
1392be skipped at the beginning of a simulation using parameter <a href="#skip_time_do3d">skip_time_do3d</a>, which has
1393zero value by default. </font></span><span lang="en-GB"></span><span lang="en-GB"></span><span lang="en-GB"></span><span lang="en-GB"><font face="Thorndale">Reference
1394time is the
1395beginning of the simulation, i.e. output takes place at times t = <b>skip_time_do3d
1396+ dt_do3d</b>,
1397<span style="font-weight: bold;">skip_time_do3d</span>
1398+ 2*<b>dt_do3d</b>, <span style="font-weight: bold;">skip_time_do3d</span>
1399+ 3*<b>dt_do3d</b>, etc. The actual output times can
1400deviate from these theoretical values (see </font></span><a href="#dt_dopr_zeitpunkte"><span lang="en-GB"><font face="Thorndale">dt_dopr</font></span></a><span lang="en-GB"><font face="Thorndale">). <br>
1401</font></span></p> <p><span lang="en-GB"><font face="Thorndale">Parameter </font></span><a href="#do3d_at_begin"><span lang="en-GB"><font face="Thorndale">do3d_at_begin</font></span></a>
1402has to be used if an additional output is wanted at the start of a run <span lang="en-GB"><font face="Thorndale">(thus at
1403the time t = 0 or at the
1404respective starting times of restart runs).</font></span> </p>
1405</td> </tr> <tr><td style="vertical-align: top;"><a name="dt_max"></a><span style="font-weight: bold;">dt_max</span></td><td style="vertical-align: top;">R</td><td style="vertical-align: top;"><span style="font-style: italic;">20.0</span></td><td>Maximum
1406allowed value of the timestep (in s).<br><br>By default,
1407the maximum timestep is restricted to be 20 s. This might be o.k. for
1408simulations of any kind of atmospheric turbulence but may have to be
1409changed for other situations.</td></tr><tr> <td style="vertical-align: top;"> <p><a name="dt_restart"></a><b>dt_restart</b></p>
1410</td> <td style="vertical-align: top;">R</td>
1411<td style="vertical-align: top;"><i>9999999.9</i></td>
1412<td style="vertical-align: top;"> <p lang="en-GB"><font face="Thorndale"><font face="Thorndale, serif">Temporal
1413interval</font> at which a new
1414restart run is to be carried out (</font>in <font face="Thorndale">s). </font> </p> <p><span lang="en-GB"><font face="Thorndale">For a
1415description
1416how to assign restart times manually see run time parameter </font></span><a href="#restart_time"><span lang="en-GB"><font face="Thorndale">restart_time</font></span></a><span lang="en-GB"><font face="Thorndale">. <span style="font-weight: bold;">dt_restart</span>
1417does not show any effect, if <span style="font-weight: bold;">restart_time</span>
1418has not been set.</font></span> </p> </td> </tr>
1419<tr> <td style="vertical-align: top;"> <p><a name="dt_run_control"></a><b>dt_run_control</b></p>
1420</td> <td style="vertical-align: top;">R</td>
1421<td style="vertical-align: top;"><i>60.0</i></td>
1422<td style="vertical-align: top;"> <p lang="en-GB"><font face="Thorndale"><font face="Thorndale, serif">Temporal
1423interval</font> at which run control
1424output is to be made (</font>in <font face="Thorndale">s).&nbsp;
1425</font> </p> <p><span lang="en-GB"><font face="Thorndale">Run control
1426information is output to the local ASCII-file </font></span><a href="chapter_3.4.html#RUN_CONTROL"><span lang="en-GB"><font face="Thorndale">RUN_CONTROL</font></span></a><span lang="en-GB"><font face="Thorndale">. At each
1427output time, one line
1428with information about the size of the time step, maximum speeds, total
1429kinetic energy etc. is written to this file. Reference time is the
1430beginning of the simulation, i.e. output takes place at times t = <b>dt_run_control</b>,
14312*<b>dt_run_control</b>, 3*<b>dt_run_control</b>,
1432etc., and always at
1433the beginning of a model run (thus at the time t = 0 or at the
1434respective starting times of restart runs). The actual output times can
1435deviate from these theoretical values (see </font></span><a href="#dt_dopr_zeitpunkte"><span lang="en-GB"><font face="Thorndale">dt_dopr</font></span></a><span lang="en-GB"><font face="Thorndale">).<br>
1436</font></span></p> <p><span lang="en-GB"><font face="Thorndale">Run control
1437information is output after each time step can be achieved via <b>dt_run_control</b>
1438= <i>0.0</i>.</font></span> </p> </td>
1439</tr> <tr> <td style="vertical-align: top;"><p><a name="end_time"></a><b>end_time</b></p>
1440</td> <td style="vertical-align: top;">R</td>
1441<td style="vertical-align: top;"><i>0.0</i></td>
1442<td style="vertical-align: top;"> <p lang="en-GB"><font face="Thorndale">Simulation time of the 3D
1443model (</font>in <font face="Thorndale">s).&nbsp;
1444</font> </p> <p><span lang="en-GB"><font face="Thorndale">The simulation time
1445is starting from the beginning of the initialization run (t = 0), not
1446starting from the beginning of the respective restart run.</font></span>
1447</p> </td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="force_print_header"></a><b>force_print_header</b></p>
1448</td> <td style="vertical-align: top;">L</td>
1449<td style="vertical-align: top;"><i>.F.</i></td>
1450<td style="vertical-align: top;"> <p>Steering of
1451header output to the local file <a href="chapter_3.4.html#RUN_CONTROL">RUN_CONTROL</a>.&nbsp;
1452</p> <p>By default, informations about the model
1453parameters in use are
1454output to the beginning of file RUN_CONTROL for initial runs only
1455(these informations are identical to that which are output to the local
1456file <a href="chapter_3.4.html#HEADER">HEADER</a>).
1457With <b>force_print_header</b> = <i>.T.</i>,
1458these informations are
1459also output to <a href="chapter_3.4.html#RUN_CONTROL">RUN_CONTROL</a>
1460at restart runs.</p> </td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="mg_cycles"></a><b>mg_cycles</b></p>
1461</td> <td style="vertical-align: top;">I</td>
1462<td style="vertical-align: top;"><i>-1</i></td>
1463<td style="vertical-align: top;"> <p>Number of
1464cycles to be used with the multi-grid scheme.<br> <br>
1465This parameter determines the number of cycles to be carried out in the
1466multi-grid method used for solving the Poisson equation for
1467perturbation pressure (see <a href="#psolver">psolver</a>).
1468The type of the cycles can be set with <a href="#cycle_mg">cycle_mg</a>.<br>
1469</p> <br>By default (<b>mg_cyles</b> = <i>-
14701</i>), the
1471number of cycles
1472depends on the requested accuracy of the scheme (see <a href="#residual_limit">residual_limit</a>)
1473and may vary from time step to time step. In this case, the CPU time
1474for a run will be difficult to estimate, since it heavily depends on
1475the total number of the cycles to be carried out.<br> <br>
1476By assigning <b>mg_cycles</b> a value (&gt;=<span style="font-style: italic;">1</span>), the number of
1477cycles can be
1478fixed so that the CPU time can be clearly estimated. <br> <br>
1479<b>Note:</b> When using a fixed number of cycles, the user
1480must
1481examine the local file <a href="chapter_3.4.html#RUN_CONTROL">RUN_CONTROL</a>
1482regularly to check whether the divergence of the velocity field is
1483sufficiently reduced by the pressure solver. It should be reduced at
1484least by two orders of magnitude. For cyclic boundary conditions along
1485both horizontal directions (see <a href="chapter_4.1.html#bc_lr">bc_lr</a>
1486and <a href="chapter_4.1.html#bc_ns">bc_ns</a>) <span style="font-weight: bold;">mg_cycles</span> = <span style="font-style: italic;">2</span> is typically a
1487good choice, for
1488non-cyclic lateral boundary conditions <span style="font-weight: bold;">mg_cycles</span>
1489= <span style="font-style: italic;">4</span> may be
1490sufficient.</td> </tr> <tr> <td style="vertical-align: top;"><a name="mg_switch_to_pe0_level"></a><b>mg_switch_to_pe0_<br>
1491level</b></td> <td style="vertical-align: top;">I</td>
1492<td style="vertical-align: top;"><br> </td> <td style="vertical-align: top;">Grid
1493level at which data shall be gathered on PE0.<br> <br>
1494In case of a run using several PEs and the multigrid method for solving
1495the Poisson equation for perturbation pressure (see <a href="#psolver">psolver</a>),
1496the value of this parameter defines on which grid level the data are
1497gathered on PE0 in order to allow for a further coarsening of the grid.
1498The finest grid defines the largest grid level. By default, the
1499gathering level is determined automatically and displayed in file <a href="chapter_3.4.html#RUN_CONTROL">RUN_CONTROL</a>.
1500It is only possible to gather data from a level larger than the one
1501determined automatically. A test run may be neccessary to determine
1502this level.</td> </tr> <tr> <td style="vertical-align: top;"><a name="netcdf_64bit"></a><span style="font-weight: bold;">netcdf_64bit</span><br>
1503</td> <td style="vertical-align: top;">L<br> </td>
1504<td style="vertical-align: top;"><span style="font-style: italic;">.F.</span><br> </td>
1505<td style="vertical-align: top;">NetCDF files will have 64
1506bit offset format.<br><br>By
1507default, the maximum file size of the NetCDF files opened by PALM is 2
1508GByte. Using netcdf_64bit = .TRUE. allows file sizes larger than 2
1509GByte.<br><br>The 64 bit offset format can be separately
1510switched off for those NetCDF files containing 3d volume date (<span style="font-family: Courier New,Courier,monospace;">DATA_3D_NETCDF</span>,
1511<span style="font-family: Courier New,Courier,monospace;">DATA_3D_AV_NETCDF</span>)
1512using <a href="#netcdf_64bit_3d">netcdf_64bit_3d</a>.<br><br><span style="font-weight: bold;">Warning:</span><br>Some
1513(PD or commercial) software may not support the 64 bit offset format.<br>
1514</td> </tr><tr><td style="vertical-align: top;"><a name="netcdf_64bit_3d"></a><span style="font-weight: bold;">netcdf_64bit_3d</span></td><td style="vertical-align: top;">L</td><td style="vertical-align: top;">.T.</td><td style="vertical-align: top;">NetCDF files containing 3d
1515volume data will have 64 bit offset format.<br><br>This
1516switch&nbsp;only comes into effect if <a href="#netcdf_64bit">netcdf_64bit</a>
1517= .TRUE.. It allows to switch off separately the 64 bit offset format
1518for those NetCDF files containing 3d volume data (<span style="font-family: Courier New,Courier,monospace;">DATA_3D_NETCDF</span>,
1519<span style="font-family: Courier New,Courier,monospace;">DATA_3D_AV_NETCDF</span>).</td></tr><tr>
1520<td style="vertical-align: top;"> <p><a name="ngsrb"></a><b>ngsrb</b></p> </td>
1521<td style="vertical-align: top;">I</td> <td style="vertical-align: top;"><i>2</i></td>
1522<td style="vertical-align: top;">Grid
1523level at which data shall be gathered on PE0.<br> <br>
1524In case of a run using several PEs and the multigrid method for solving
1525the Poisson equation for perturbation pressure (see <a href="#psolver">psolver</a>),
1526the value of this parameter defines on which grid level the data are
1527gathered on PE0 in order to allow for a further coarsening of the grid.
1528The finest grid defines the largest grid level. By default, the
1529gathering level is determined automatically and displayed in file <a href="chapter_3.4.html#RUN_CONTROL">RUN_CONTROL</a>.
1530It is only possible to gather data from a level larger than the one
1531determined automatically. A test run may be neccessary to determine
1532this level.</td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="normalizing_region"></a><b>normalizing_region</b></p>
1533</td> <td style="vertical-align: top;">I</td>
1534<td style="vertical-align: top;"><span style="font-style: italic;">0</span><br> </td>
1535<td style="vertical-align: top;"> <p>Determines the
1536subdomain from which the normalization
1537quantities are calculated.&nbsp; </p> <p>If output
1538data of the horizontally averaged vertical profiles
1539(see <a href="#data_output_pr">data_output_pr</a>)
1540is to be normalized (see <a href="#cross_normalized_x">cross_normalized_x</a>,
1541<a href="#cross_normalized_y">cross_normalized_y</a>),
1542the respective normalization quantities are by default calculated from
1543the averaged data of the total model domain (<b>normalizing_region</b>
1544= <i>0</i>) and are thus representative for the total
1545domain. Instead
1546of that, normalization quantities can also be calculated for a
1547subdomain. The wanted subdomain can be given with the parameter <span style="font-weight: bold;">normalizing_region</span>,
1548where <i>1</i>
1549&lt;= <b>normalizing_region</b> &lt;= <i>9 </i>must
1550hold. These
1551quantities are then used for normalizing of all profiles (even for that
1552of the total domain).</p> </td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="nsor"></a><b>nsor</b></p>
1553</td> <td style="vertical-align: top;">I</td>
1554<td style="vertical-align: top;"><i>20</i></td>
1555<td style="vertical-align: top;"> <p>Number of
1556iterations to be used with the SOR-scheme.&nbsp; </p> <p>This
1557parameter is only effective if the SOR-scheme is selected
1558as pressure solver (<a href="#psolver">psolver</a>
1559= <span style="font-style: italic;">'sor'</span>).
1560The number of
1561iterations necessary for a sufficient convergence of the scheme depends
1562on the grid point numbers and is to be determined by appropriate test
1563runs (the default value will not at all be sufficient for larger grid
1564point numbers). The number of iterations used for the first call of the
1565SOR-scheme (t = 0) is determined via the parameter <a href="chapter_4.1.html#nsor_ini">nsor_ini</a>.</p>
1566</td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="nz_do3d"></a><b>nz_do3d</b></p>
1567</td> <td style="vertical-align: top;">I</td>
1568<td style="vertical-align: top;"><i>nz+1</i></td>
1569<td style="vertical-align: top;"> Limits the output of 3d
1570volume data along the vertical direction (grid point index k).<br><br>By
1571default, data for all grid points along z are output. The parameter <span style="font-weight: bold;">nz_do3d</span>
1572can be used to limit the output up to a certain vertical grid point
1573(e.g. in order to reduce the amount of output data). It affects all
1574output of volume data ("normal" output to file, see <a href="#data_output">data_output</a>, as well as output
1575for <span style="font-weight: bold;">dvrp</span>-software,
1576see <a href="#mode_dvrp">mode_dvrp</a>).</td>
1577</tr> <tr> <td style="vertical-align: top;"><p><a name="omega_sor"></a><b>omega_sor</b></p>
1578</td> <td style="vertical-align: top;">R</td>
1579<td style="vertical-align: top;"><i>1.8</i></td>
1580<td style="vertical-align: top;"> <p>Convergence
1581factor to be used with the the SOR-scheme.&nbsp; </p> <p>If
1582the SOR-scheme is selected (<a href="#psolver">psolver</a>
1583= <span style="font-style: italic;">'sor'</span>),
1584this parameter
1585determines the value of the convergence factor, where <i>1.0</i>
1586&lt;= <b>omega_sor</b> &lt; <i>2.0 </i>.
1587The optimum value of <b>omega_sor</b>
1588depends on the number of grid points along the different directions in
1589space. For non-equidistant grids it can only be determined by
1590appropriate test runs.</p> </td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="prandtl_number"></a><b>prandtl_number</b></p>
1591</td> <td style="vertical-align: top;">R</td>
1592<td style="vertical-align: top;"><i>1.0</i></td>
1593<td style="vertical-align: top;"> <p>Ratio of the
1594eddy diffusivities for momentum and heat (K<sub>m</sub>/K<sub>h</sub>).&nbsp;
1595</p> <p>For runs with constant eddy diffusivity (see <a href="chapter_4.1.html#km_constant">km_constant</a>),
1596this parameter can be used to assign the Prandtl number (ratio K<sub>m</sub>
1597/ K<sub>h</sub>).</p> </td> </tr> <tr>
1598<td style="vertical-align: top;"> <p><a name="profile_columns"></a><b>profile_columns</b></p>
1599</td> <td style="vertical-align: top;">I</td>
1600<td style="vertical-align: top;"><i>3</i></td>
1601<td style="vertical-align: top;"> <p>Number of
1602coordinate systems to be plotted<span style="font-weight: bold;"></span>
1603in one row by <span style="font-weight: bold;">profil</span>.&nbsp;
1604</p> <p>This parameter only applies for &nbsp;<a href="chapter_4.2.html#data_output_format">data_output_format</a>
1605= <span style="font-style: italic;">'profil'</span>.</p><p>It
1606determines the layout of plots of
1607horizontally averaged profiles (<a href="#data_output_pr">data_output_pr</a>)
1608when plotted with the plot software <span style="font-weight: bold;">profil</span>.
1609Generally, the number and sequence of coordinate systems (panels) to be
1610plotted on one page are
1611determined by <a href="#cross_profiles">cross_profiles</a>.
1612<b>profile_columns</b>
1613determines how many panels are to be
1614arranged next to each other in one row (number of columns). The
1615respective number of rows on a page is assigned by <a href="#profile_rows">profile_rows</a>.
1616According to their order given by <a href="#data_output_pr">data_output_pr</a>,
1617the panels are arranged beginning in the top row from left to right and
1618then continued in the following row. If the number of panels cranz
1619&gt; <b>profile_columns</b> * <b>profile_rows</b>,
1620the remaining
1621panels are drawn on an additional page. If cranz &lt; <b>profile_columns</b>,
1622then <b>profile_columns</b> = cranz is automatically set.
1623If
1624row&nbsp; contains any panel, then the value of <b>profile_rows</b>
1625is reduced automatically.</p> </td> </tr> <tr>
1626<td style="vertical-align: top;"> <p><a name="profile_rows"></a><b>profile_rows</b></p>
1627</td> <td style="vertical-align: top;">I</td>
1628<td style="vertical-align: top;"><i>2</i></td>
1629<td style="vertical-align: top;"> <p>Number of rows
1630of coordinate systems to be plotted on one page
1631by <span style="font-weight: bold;">profil</span>.&nbsp;
1632</p> <p>This parameter only applies for &nbsp;<a href="chapter_4.2.html#data_output_format">data_output_format</a>
1633= <span style="font-style: italic;">'profil'</span>.</p><p>It
1634determines the layout of plots of horizontally averaged
1635profiles. See <a href="#profile_columns">profile_columns</a>.</p>
1636</td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="psolver"></a><b>psolver</b></p>
1637</td> <td style="vertical-align: top;">C * 10</td>
1638<td style="vertical-align: top;"><i>'poisfft'</i></td>
1639<td style="vertical-align: top;"> <p>Scheme to be
1640used to solve the Poisson equation for the
1641perturbation pressure.&nbsp; </p> <br>
1642The user can choose between the following schemes:<br> <table style="text-align: left; width: 100%;" cellpadding="2" cellspacing="2"> <tbody> <tr> <td style="vertical-align: top;"><i>poisfft</i></td>
1643<td style="vertical-align: top;">Direct method using FFT
1644along x and y, solution of a
1645tridiagonal matrix along z, and backward
1646FFT (see Siano, institute reports, volume 54). The FFT routines to be
1647used can be determined via the initialization parameter <a href="chapter_4.1.html#fft_method">fft_method</a>.<br>
1648This solver is specially optimized for 1d domain decompositions.
1649Vectorization is optimized for domain decompositions along x only.</td>
1650</tr> <tr> <td style="vertical-align: top;"><p><i>poisfft_</i>
1651<br> <i>hybrid</i></p>
1652</td> <td style="vertical-align: top;">Direct
1653method using FFT
1654along x and y, solution of a
1655tridiagonal matrix along z, and backward
1656FFT (see Siano, institute reports, volume 54). The FFT routines to be
1657used can be determined via the initialization parameter <a href="chapter_4.1.html#fft_method">fft_method</a>.<br>
1658This solver is specially optimized for 1d domain decompositions.
1659Vectorization is optimized for domain decompositions along x only.</td>
1660</tr> <tr> <td style="vertical-align: top;"><i>multigrid</i></td>
1661<td style="vertical-align: top;"> <p>Multi-grid
1662scheme (see Uhlenbrock, diploma thesis). v-
1663and
1664w-cycles (see <a href="#cycle_mg">cycle_mg</a>)
1665are implemented. The convergence of the iterative scheme can be
1666steered by the number of v-/w-cycles to be carried out for each call of
1667the scheme (<a href="#mg_cycles">mg_cycles</a>)
1668and by the number of Gauss-Seidel iterations (see <a href="#ngsrb">ngsrb</a>)
1669to be carried out on each grid level. Instead the requested accuracy
1670can be given via <a href="#residual_limit">residual_limit</a>.
1671<span style="font-weight: bold;">This is the default!</span>
1672The
1673smaller this limit is, the more cycles have to be carried out in this
1674case and the number of cycles may vary from timestep to timestep.</p>
1675<br>If <a href="#mg_cycles">mg_cycles</a>
1676is set to its optimal value, the computing time of the
1677multi-grid scheme amounts approximately to that of the direct solver <span style="font-style: italic;">poisfft</span>, as long as
1678the number of
1679grid points in the three directions
1680of space corresponds to a power-of-two (2<sup>n</sup>)
1681where <i>n</i> &gt;= 5 must hold. With large <i>n,
1682</i>the
1683multi-grid scheme can even be faster than the direct solver (although
1684its accuracy is several orders of magnitude worse, but this does not
1685affect the accuracy of the simulation). Nevertheless, the user should
1686always carry out some test runs in order to find out the optimum value
1687for <a href="#mg_cycles">mg_cycles</a>,
1688because the CPU time of a run very critically depends on this
1689parameter. <p>This scheme requires that the number of grid
1690points of
1691the
1692subdomains (or of the total domain, if only one PE is uesd) along each
1693of the directions can at least be devided once by 2 without rest.</p>
1694With parallel runs, starting from a certain grid level the
1695data of the subdomains are possibly gathered on PE0 in order to allow
1696for a further coarsening of the grid. The grid level for gathering can
1697be manually set by <a href="#mg_switch_to_pe0_level">mg_switch_to_pe0_level</a>.<br>
1698<p>Using this procedure requires the subdomains to be of
1699identical size (see <a href="chapter_4.1.html#grid_matching">grid_matching</a>).</p>
1700</td> </tr> <tr> <td style="vertical-align: top;"><i>sor</i></td>
1701<td style="vertical-align: top;">Successive over
1702relaxation
1703method (SOR). The convergence of
1704this
1705iterative scheme is steered with the parameters <a href="#omega_sor">omega_sor</a>,
1706<a href="chapter_4.1.html#nsor_ini">nsor_ini</a>
1707and <a href="chapter_4.1.html#nsor">nsor</a>.&nbsp;
1708<br>Compared to the direct method and the multi-grid method, this
1709scheme
1710needs substantially
1711more computing time. It should only be used for test runs, e.g. if
1712errors in the other pressure solver methods are assumed.</td> </tr>
1713</tbody> </table> <br>In order to speed-up
1714performance, the Poisson equation is by default
1715only solved at the last substep of a multistep Runge-Kutta scheme (see <a href="#call_psolver_at_all_substeps">call_psolver
1716at_all_substeps</a> and <a href="chapter_4.1.html#timestep_scheme">timestep_scheme</a>).&nbsp;
1717</td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="rayleigh_damping_factor"></a><b>rayleigh_damping</b>
1718<br> <b>_factor</b></p> </td> <td style="vertical-align: top;">R</td> <td style="vertical-align: top;"><i>0.0 or</i><br>
1719<i>0.01</i></td> <td style="vertical-align: top;">
1720<p>Factor for Rayleigh damping.&nbsp; </p> <p>A
1721so-called Rayleigh damping is applied to all prognostic
1722variables if a non-zero value is assigned to <b>rayleigh_damping_factor</b>.&nbsp;
1723If switched on, variables are forced towards the value of their
1724respective basic states (e.g. the geostrophic wind). The intensity of
1725damping is controlled by the value the <b>rayleigh_damping_factor</b>
1726is assigned to.
1727The damping starts weakly at a height defined by <a href="#rayleigh_damping_height">rayleigh_damping_height</a>
1728and rises according to a sin<sup>2</sup>-function to its
1729maximum value
1730at
1731the top boundary. </p> <p>This method
1732effectively damps gravity waves, caused by boundary layer convection,
1733which may spread out vertically in the inversion layer and which are
1734reflected&nbsp; at the top
1735boundary. This particularly happens with the upstream-spline scheme
1736switched on (see <a href="chapter_4.1.html#momentum_advec">momentum_advec</a>
1737or <a href="chapter_4.1.html#scalar_advec">scalar_advec</a>).
1738Therefore, with this scheme the Rayleigh damping is switched on (<b>rayleigh_damping_factor</b>
1739= <i>0.01</i>) by default. Otherwise it remains switched
1740off.&nbsp; </p> <p>The Rayleigh damping factor must
1741hold the condition <i>0.0</i>
1742&lt;= <b>rayleigh_damping_factor</b>
1743&lt;= <i>1.0</i>. Large values (close to <span style="font-style: italic;">1.0</span>) can cause
1744numerical instabilities.</p> </td> </tr> <tr>
1745<td style="vertical-align: top;"> <p><a name="rayleigh_damping_height"></a><b>rayleigh_damping</b>
1746<br> <b>_height</b></p> </td> <td style="vertical-align: top;">R</td> <td style="vertical-align: top;"> <p><i>2/3 *</i>
1747<br><span style="font-style: italic;">zu</span><i style="font-style: italic;">(nz)</i></p>
1748</td> <td style="vertical-align: top;"> <p>Height
1749where the Rayleigh damping starts (in m).&nbsp; </p> <p>With
1750Rayleigh damping switched on (see <a href="#rayleigh_damping_factor">rayleigh_damping_factor</a>),
1751this parameter determines the range where damping is applied. By
1752default, Rayleigh damping will be applied in the upper third of the
1753model
1754domain.</p> </td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="residual_limit"></a><b>residual_limit</b></p>
1755</td> <td style="vertical-align: top;">R</td>
1756<td style="vertical-align: top;"><i>1.0E-6</i></td>
1757<td style="vertical-align: top;"> <p>Largest
1758residual permitted for the multi-grid scheme (in s<sup>-2</sup>m<sup>-3</sup>).&nbsp;
1759</p> <p>This is a parameter to steer the accuracy of the
1760multi-grid
1761scheme (see <a href="#psolver">psolver</a>).
1762The assigned cycle (v- or w-cycle, see <a href="#mg_cycles">mg_cycles</a>)
1763is passed through until the residual falls below the limit given by <span style="font-weight: bold;">residual_limit</span>. If
1764this
1765is not the case after 1000 cycles, the PALM aborts with a corresponding
1766error message.</p> <p>The reciprocal value of this
1767parameter can be interpreted as
1768a factor by the divergence of the provisional
1769velocity field is approximately reduced after the multi-grid scheme has
1770been applied (thus the default value causes a reduction of the
1771divergence by approx. 6 orders of magnitude).&nbsp; </p> </td>
1772</tr> <tr> <td style="vertical-align: top;"><p><a name="restart_time"></a><b>restart_time</b></p>
1773</td> <td style="vertical-align: top;">R</td>
1774<td style="vertical-align: top;"><i>9999999.9</i></td>
1775<td style="vertical-align: top;"> <p>Simulated time
1776after which a restart run is to be carried out
1777(in s). </p> <p>The simulated time refers to the
1778beginning of the
1779initial run (t = 0), not to the beginning of the respective
1780restart run. Restart runs can additionally be forced to be carried out
1781in regular intervals using the run time parameter <a href="#dt_restart">dt_restart</a>. </p> <p><span style="font-weight: bold;">Note:</span><br>
1782A successful operation of this parameter requires additional
1783modifications in the <span style="font-weight: bold;">mrun</span>-call
1784for the respective run (see <a href="chapter_3.3.html">chapter
17853.3</a>).<br> </p> <p>The choice of <b>restart_time</b>
1786or <b>dt_restart</b> does
1787not override the automatic start of restart runs in case that the job
1788runs out of CPU time. <br> </p> </td> </tr>
1789<tr> <td style="vertical-align: top;"> <p><a name="section_xy"></a><b>section_xy</b></p>
1790</td> <td style="vertical-align: top;">I(100)<br>
1791</td> <td style="vertical-align: top;"><span style="font-style: italic;">no section</span><br>
1792</td> <td style="vertical-align: top;"> <p lang="en-GB"><font face="Thorndale">Position
1793of&nbsp;cross section(s) for&nbsp;output of 2d horizontal cross
1794sections (grid point index k).&nbsp; </font> </p> <p><span lang="en-GB"><font face="Thorndale">If output
1795of
1796horizontal cross sections is selected (see </font></span><a href="chapter_4.2.html#data_output"><span lang="en-GB"><font face="Thorndale">data_output</font></span></a><span lang="en-GB"><font face="Thorndale">), this
1797parameter can be used to
1798define the position(s) of the cross section(s). Up to 100 positions of
1799cross sections can be selected by assigning <b>section_xy</b>
1800the
1801corresponding vertical grid point index/indices k of the requested
1802cross section(s). The exact location (height level) of the cross
1803section depends on the variable for which the output is made: zu(k) for
1804scalars and horizontal velocities, zw(k) for the vertical velocity.
1805Information about the exact location of the cross section is contained
1806in the NetCDF output file (if the default NetCDF output is switched on;
1807see <a href="#data_output_format">data_output_format</a>).</font></span></p><p><span lang="en-GB"><font face="Thorndale">Assigning <span style="font-weight: bold;">section_xy</span> = <span style="font-style: italic;">-1</span>
1808creates the output of horizontal cross sections averaged along z. In
1809the
1810NetCDF output file these (averaged) cross sections are given the
1811z-coordinate <span style="font-style: italic;">-1.0</span>.</font></span></p><p><span lang="en-GB"></span><span lang="en-GB"><font face="Thorndale">Assignments to <b>section_xy</b>
1812does not effect the output of horizontal cross sections of variable u<sub>*</sub>
1813and theta<sub>*</sub> and the liquid water path lwp*. For
1814these quantities always only one cross
1815section (for z=zu(1)) is output.</font></span></p><span lang="en-GB"><font face="Thorndale">In case of <span style="font-weight: bold;">data_output_format</span> =
1816<span style="font-style: italic;">'iso2d'</span> and
1817if several cross sections are selected (e.g. <b>section_xy</b>
1818= <i>1</i>, <i>10</i>, <i>15</i>),
1819then the respective data are
1820successively written to file. The output order follows the order given
1821by <b>section_xy</b>.&nbsp;</font></span></td>
1822</tr> <tr> <td style="vertical-align: top;"><p><a name="section_xz"></a><b>section_xz</b></p>
1823</td> <td style="vertical-align: top;">I(100)<br>
1824</td> <td style="vertical-align: top;"><span style="font-style: italic;">no section</span></td>
1825<td style="vertical-align: top;"> <p lang="en-GB"><font face="Thorndale">Position of&nbsp;cross section(s)
1826for&nbsp;output of 2d (xz) vertical cross sections (grid point
1827index j).&nbsp; </font> </p> <span lang="en-GB"><font face="Thorndale">If output of
1828vertical xz cross sections is selected (see </font></span><a href="chapter_4.2.html#data_output"><span lang="en-GB"><font face="Thorndale">data_output</font></span></a><span lang="en-GB"><font face="Thorndale">), this
1829parameter can be used to
1830define the position(s) of the cross section(s). Up to 100 positions of
1831cross sections can be selected by assigning <b>section_xz</b>
1832the
1833corresponding horizontal grid point index/indices j of the requested
1834cross section(s). The exact position (in y-direction) of the cross
1835section is given by j*</font></span><a href="chapter_4.1.html#dy"><span lang="en-GB"><font face="Thorndale">dy</font></span></a> <span lang="en-GB"><font face="Thorndale">or (j-0.5)*</font></span><a href="chapter_4.1.html#dy"><span lang="en-GB"><font face="Thorndale">dy</font></span></a><span lang="en-GB"><font face="Thorndale">, depending
1836on which grid the output quantity is defined. However, in
1837the&nbsp;NetCDF output file </font></span><span lang="en-GB"><font face="Thorndale">(if the
1838default NetCDF output is switched on; see <a href="chapter_4.2.html#data_output_format">data_output_format</a>)
1839no distinction is made between the quantities and j*<span style="font-weight: bold;">dy</span> is used for all
1840positions.<br><br>Assigning <span style="font-weight: bold;">section_xz</span> = <span style="font-style: italic;">-1</span>
1841creates the output of vertical cross sections averaged along y. In the
1842NetCDF output file these (averaged) cross sections are given the
1843y-coordinate <span style="font-style: italic;">-1.0</span>.<br></font></span><span lang="en-GB"><font face="Thorndale"><br></font></span><span lang="en-GB"><font face="Thorndale">In case of <span style="font-weight: bold;">data_output_format</span> =
1844<span style="font-style: italic;">'iso2d'</span> and
1845</font></span><span lang="en-GB"><font face="Thorndale">if several cross sections are
1846selected (e.g. <b>section_xz</b> = <i>0</i>, <i>12</i>,
1847<i>27</i>),
1848then the respective data are successively written to file. The output
1849order follows the order given by <b>section_xz</b>.</font></span></td>
1850</tr> <tr> <td style="vertical-align: top;"><p><a name="section_yz"></a><b>section_yz</b></p>
1851</td> <td style="vertical-align: top;">I(100)<br>
1852</td> <td style="vertical-align: top;"><span style="font-style: italic;">no section</span></td>
1853<td style="vertical-align: top;"> <p lang="en-GB"><font face="Thorndale">Position of&nbsp;cross section(s)
1854for&nbsp;output of 2d (yz) vertical cross sections (grid point
1855index i).&nbsp; </font> </p> <span lang="en-GB"><font face="Thorndale">If output of
1856vertical yz cross sections is selected (see </font></span><a href="chapter_4.2.html#data_output"><span lang="en-GB"><font face="Thorndale">data_output</font></span></a><span lang="en-GB"><font face="Thorndale">),
1857this parameter can be used to define the position(s) of the cross
1858section(s). Up to 100 positions of cross sections can be selected by
1859assigning <b>section_yz</b> the corresponding horizontal
1860grid point
1861index/indices i of the requested cross section(s). The exact position
1862(in x-direction) of the cross section is given by i*</font></span><a href="chapter_4.1.html#dx"><span lang="en-GB"><font face="Thorndale">dx</font></span></a> <span lang="en-GB"><font face="Thorndale">or (i-0.5)*</font></span><a href="chapter_4.1.html#dx"><span lang="en-GB"><font face="Thorndale">dx</font></span></a><span lang="en-GB"><font face="Thorndale">, depending
1863on which grid the output quantity is defined.&nbsp;</font></span><span lang="en-GB"><font face="Thorndale">However, in
1864the&nbsp;NetCDF output file </font></span><span lang="en-GB"><font face="Thorndale">(if the
1865default NetCDF output is switched on; see <a href="chapter_4.2.html#data_output_format">data_output_format</a>)
1866no distinction is made between the quantities and i*<span style="font-weight: bold;">dx</span> is used for all
1867positions.<br><br></font></span><span lang="en-GB"><font face="Thorndale">Assigning <span style="font-weight: bold;">section_yz</span> = <span style="font-style: italic;">-1</span>
1868creates the output of vertical cross sections averaged along x. In the
1869NetCDF output file these (averaged) cross sections are given the
1870x-coordinate <span style="font-style: italic;">-1.0</span>.</font></span><br><span lang="en-GB"></span><span lang="en-GB"><font face="Thorndale"> <br></font></span><span lang="en-GB"><font face="Thorndale">In case of <span style="font-weight: bold;">data_output_format</span> =
1871<span style="font-style: italic;">'iso2d'</span> and
1872</font></span><span lang="en-GB"><font face="Thorndale">if several cross sections are
1873selected (e.g. <b>section_yz</b> = <span style="font-style: italic;">3</span>, <span style="font-style: italic;">27</span>, 19), then the
1874respective data are successively written to file. The output order
1875follows the order given by <b>section_yz</b>.</font></span></td>
1876</tr> <tr> <td style="vertical-align: top;"><a name="skip_time_data_output"></a><span style="font-weight: bold;">skip_time_data_output</span><br>
1877</td> <td style="vertical-align: top;">R<br> </td>
1878<td style="vertical-align: top;"><span style="font-style: italic;">0.0</span><br> </td>
1879<td style="vertical-align: top;">No data output before
1880this interval has passed (in s).<br><br>This
1881parameter causes that data output activities are starting not before
1882this interval
1883(counting from the beginning of the simulation, t=0) has passed. By
1884default, this
1885applies for output of instantaneous 3d volume data, cross section data,
1886spectra and vertical profile data as well as for temporally averaged 2d
1887and 3d data. Individual intervals can be assigned using parameters <a href="#skip_time_do3d">skip_time_do3d</a>, <a href="#skip_time_do2d_xy">skip_time_do2d_xy</a>, <a href="#skip_time_do2d_xz">skip_time_do2d_xz</a>, <a href="#skip_time_do2d_yz">skip_time_do2d_yz</a>, <a href="#skip_time_dosp">skip_time_dosp</a>,&nbsp;<a href="#skip_time_dopr">skip_time_dopr</a>, and <a href="#skip_time_data_output_av">skip_time_data_output_av</a>.<br><br><span style="font-weight: bold;">Example:</span><br>If
1888the user has set <a href="#dt_data_output">dt_data_output</a>
1889= <span style="font-style: italic;">3600.0</span>
1890and <span style="font-weight: bold;">skip_time_data_output</span>
1891= <span style="font-style: italic;">1800.0</span>,
1892then the first output will be done at t = 5400 s.<br> </td>
1893</tr> <tr><td style="vertical-align: top;"><a name="skip_time_data_output_av"></a><span style="font-weight: bold;">skip_time_data_output_av</span></td><td style="vertical-align: top;">R</td><td style="vertical-align: top;"><span style="font-style: italic;">value of <a href="#skip_time_data_output">skip_time_<br>data_output</a></span></td><td style="vertical-align: top;">No output of temporally
1894averaged 2d/3d data before this interval has passed (in s).<br><br>This
1895parameter causes that data output activities are starting not before
1896this interval
1897(counting from the beginning of the simulation, t=0) has passed. <br><br><span style="font-weight: bold;">Example:</span><br>If
1898the user has set <a href="#dt_data_output_av">dt_data_output_av</a>
1899= <span style="font-style: italic;">3600.0</span>
1900and <span style="font-weight: bold;">skip_time_data_output_av</span>
1901= <span style="font-style: italic;">1800.0</span>,
1902then the first output will be done at t = 5400 s.</td></tr><tr>
1903<td style="vertical-align: top;"><a name="skip_time_dopr"></a><span style="font-weight: bold;">skip_time_dopr</span><br>
1904</td> <td style="vertical-align: top;">R<br> </td>
1905<td style="vertical-align: top;"><span style="font-style: italic;">value of <a href="chapter_4.2.html#skip_time_data_output">skip_time_<br>data_output</a></span>
1906</td> <td style="vertical-align: top;">No output of
1907vertical profile data before this interval has passed (in s).<br><br>This
1908parameter causes that data output activities are starting not before
1909this interval
1910(counting from the beginning of the simulation, t=0) has passed. <br><br><span style="font-weight: bold;">Example:</span><br>If
1911the user has set <a href="#dt_dopr">dt_dopr</a> = <span style="font-style: italic;">3600.0</span> and <span style="font-weight: bold;">skip_time_dopr</span> = <span style="font-style: italic;">1800.0</span>, then the
1912first output will be done at t = 5400 s. </td> </tr> <tr>
1913<td style="vertical-align: top;"><a name="skip_time_do2d_xy"></a><span style="font-weight: bold;">skip_time_do2d_xy</span><br>
1914</td> <td style="vertical-align: top;">R<br> </td>
1915<td style="vertical-align: top;"><span style="font-style: italic;">value of <a href="chapter_4.2.html#skip_time_data_output">skip_time_<br>data_output</a></span>
1916</td> <td style="vertical-align: top;">No output of
1917instantaneous horizontal cross section data before this interval has
1918passed (in s).<br><br>This
1919parameter causes that data output activities are starting not before
1920this interval
1921(counting from the beginning of the simulation, t=0) has passed. <br><br><span style="font-weight: bold;">Example:</span><br>If
1922the user has set <a href="#dt_do2d_xy">dt_do2d_xy</a>
1923= <span style="font-style: italic;">3600.0</span>
1924and <span style="font-weight: bold;">skip_time_do2d_xy</span>
1925= <span style="font-style: italic;">1800.0</span>,
1926then the first output will be done at t = 5400 s. </td> </tr>
1927<tr> <td style="vertical-align: top;"><a name="skip_time_do2d_xz"></a><span style="font-weight: bold;">skip_time_do2d_xz</span><br>
1928</td> <td style="vertical-align: top;">R<br> </td>
1929<td style="vertical-align: top;"><span style="font-style: italic;">value of <a href="chapter_4.2.html#skip_time_data_output">skip_time_<br>data_output</a></span>
1930</td> <td style="vertical-align: top;">No output of
1931instantaneous vertical (xz) cross section data before this interval has
1932passed (in s).<br><br>This
1933parameter causes that data output activities are starting not before
1934this interval
1935(counting from the beginning of the simulation, t=0) has passed. <br><br><span style="font-weight: bold;">Example:</span><br>If
1936the user has set <a href="#dt_do2d_xz">dt_do2d_xz</a>
1937= <span style="font-style: italic;">3600.0</span>
1938and <span style="font-weight: bold;">skip_time_do2d_xz</span>
1939= <span style="font-style: italic;">1800.0</span>,
1940then the first output will be done at t = 5400 s. </td> </tr>
1941<tr> <td style="vertical-align: top;"><a name="skip_time_do2d_yz"></a><span style="font-weight: bold;">skip_time_do2d_yz</span><br>
1942</td> <td style="vertical-align: top;">R<br> </td>
1943<td style="vertical-align: top;"><span style="font-style: italic;">value of <a href="chapter_4.2.html#skip_time_data_output">skip_time_<br>data_output</a></span>
1944</td> <td style="vertical-align: top;">No output of
1945instantaneous vertical (yz) cross section data before this interval has
1946passed (in s).<br><br>This
1947parameter causes that data output activities are starting not before
1948this interval
1949(counting from the beginning of the simulation, t=0) has passed. <br><br><span style="font-weight: bold;">Example:</span><br>If
1950the user has set <a href="#dt_do2d_yz">dt_do2d_yz</a>
1951= <span style="font-style: italic;">3600.0</span>
1952and <span style="font-weight: bold;">skip_time_do2d_yz</span>
1953= <span style="font-style: italic;">1800.0</span>,
1954then the first output will be done at t = 5400 s. </td> </tr>
1955<tr> <td style="vertical-align: top;"><a name="skip_time_do3d"></a><span style="font-weight: bold;">skip_time_do3d</span><br>
1956</td> <td style="vertical-align: top;">R<br> </td>
1957<td style="vertical-align: top;"><span style="font-style: italic;">value of <a href="chapter_4.2.html#skip_time_data_output">skip_time_<br>data_output</a></span>
1958</td> <td style="vertical-align: top;">No output of
1959instantaneous 3d volume data before this interval has passed (in s).<br><br>This
1960parameter causes that data output activities are starting not before
1961this interval
1962(counting from the beginning of the simulation, t=0) has passed. <br><br><span style="font-weight: bold;">Example:</span><br>If
1963the user has set <a href="#dt_do3d">dt_do3d</a> = <span style="font-style: italic;">3600.0</span> and <span style="font-weight: bold;">skip_time_do3d</span> = <span style="font-style: italic;">1800.0</span>, then the
1964first output will be done at t = 5400 s. </td> </tr>
1965<tr> <td style="vertical-align: top;"> <p><a name="termination_time_needed"></a><b>termination_time</b>
1966<br> <b>_needed</b></p> </td> <td style="vertical-align: top;">R<br> </td> <td style="vertical-align: top;"><span style="font-style: italic;">35.0</span><br> </td>
1967<td style="vertical-align: top;"> <p>CPU time
1968needed for terminal actions at the end of a run in
1969batch mode (in s).<br> </p> <p>If the environment
1970variable <b>write_binary </b>is
1971set <i>true</i> (see <a href="chapter_3.3.html">chapter
19723.3</a>), PALM checks the remaining CPU time of the job after
1973each
1974timestep. Time integration must not consume the CPU time completely,
1975since several actions still have to be carried out after time
1976integration has finished (e.g. writing of binary data for the restart
1977run, carrying out output commands, copying of local files to their
1978permanent destinations, etc.) which also takes some time. The maximum
1979possible time needed for these activities plus a reserve is to be given
1980with the parameter <b>termination_time_needed</b>. Among
1981other things,
1982it depends on
1983the number of grid points used. If its value is selected too small,
1984then the
1985respective job will be prematurely aborted by the queuing system, which
1986may result in a data loss and will possibly interrupt the job chain.<br>
1987</p> <p>An abort happens in any way, if the environment
1988variable <span style="font-weight: bold;">write_binary</span>
1989is not set to <span style="font-style: italic;">true</span>
1990and if moreover the job has
1991been assigned an insufficient CPU time by <b>mrun</b>
1992option <tt><tt>-t</tt></tt>. <i><br>
1993</i> </p> <p><span style="font-weight: bold;">Note:</span><br>
1994On the IBM computers of the HLRN the time used by the job <span style="font-weight: bold;">before</span> the start of
1995PALM
1996have also to be accounted for (e.g. for
1997compilation and copying of input files).</p> </td> </tr>
1998<tr> <td style="vertical-align: top;"> <p><a name="use_prior_plot1d_parameters"></a><b>use_prior_plot1d</b>
1999<br> <b>_parameters</b></p> </td> <td style="vertical-align: top;">L</td> <td style="vertical-align: top;"><i>.F.</i></td>
2000<td style="vertical-align: top;"> <p>Additional
2001plot of vertical profile data with <span style="font-weight: bold;">profil</span>
2002from preceding runs of the
2003job chain.&nbsp; </p> <p>This parameter only applies
2004for &nbsp;<a href="chapter_4.2.html#data_output_format">data_output_format</a>
2005= <span style="font-style: italic;">'profil'</span>.</p><p>By
2006default, plots of horizontally averaged vertical profiles
2007(see <a href="#data_output_pr">data_output_pr</a>)
2008only contain profiles of data produced by the model
2009run. If profiles of prior times (i.e. data of preceding jobs of a
2010job chain) shall be plotted additionally (e.g. for comparison
2011purposes), <b>use_prior_plot1d_parameters</b> = <i>.T</i>.
2012must be
2013set.<br> </p> <p>For further explanation see <a href="chapter_4.5.2.html">chapter
20144.5.2</a>.</p> </td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="z_max_do1d"></a><b>z_max_do1d</b></p>
2015</td> <td style="vertical-align: top;">R</td>
2016<td style="vertical-align: top;"><i>zu(nzt+1) (model
2017top)</i></td> <td style="vertical-align: top;">
2018<p>Height level up to which horizontally averaged profiles are to
2019be
2020plotted with <span style="font-weight: bold;">profil</span>
2021(in
2022m).&nbsp; </p> <p>This parameter only applies for
2023&nbsp;<a href="chapter_4.2.html#data_output_format">data_output_format</a>
2024= <span style="font-style: italic;">'profil'</span>.</p><p>It
2025affects plots of horizontally averaged profiles
2026(<a href="#data_output_pr">data_output_pr</a>)
2027when plotted with the plot software <span style="font-weight: bold;">profil</span>.
2028By default, profiles are plotted up to the top boundary. The height
2029level up to which profiles are plotted can be decreased by assigning <span style="font-weight: bold;">z_max_do1d</span> a smaller
2030value.
2031Nevertheless, <span style="font-weight: bold;">all</span>
2032vertical
2033grid points (0 &lt;= k &lt;= nz+1) are still output to file <a href="chapter_3.4.html#PLOT1D_DATA">PLOT1D_DATA</a>.</p>
2034<p>If a normalization for the vertical axis was selected (see <a href="#cross_normalized_y">cross_normalized_y</a>), <b>z_max_do1d</b>
2035has no effect. Instead, <a href="#z_max_do1d_normalized">z_max_do1d_normalized</a>
2036must be used.</p> </td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="z_max_do1d_normalized"></a><b>z_max_do1d</b>
2037<br> <b>_normalized</b></p> </td> <td style="vertical-align: top;">R</td> <td style="vertical-align: top;"><i>determined by plot</i>
2038<br> <i>data</i> <br> </td> <td style="vertical-align: top;"> <p>Normalized height
2039level up to which horizontally averaged
2040profiles are to be plotted with <span style="font-weight: bold;">profil</span>.&nbsp;
2041</p> <p>This parameter only applies for &nbsp;<a href="chapter_4.2.html#data_output_format">data_output_format</a>
2042= <span style="font-style: italic;">'profil'</span>.</p><p>It
2043affects plots of horizontally averaged profiles
2044(<a href="#data_output_pr">data_output_pr</a>)
2045when plotted with the plot software <span style="font-weight: bold;">profil</span>,
2046if a normalization for the vertical axis is selected
2047(see <a href="#cross_normalized_y">cross_normalized_y</a>).
2048If e.g. the boundary layer height is used for normalization, then <b>z_max_do1d_normalized</b>
2049= <i>1.5</i> means that all profiles up to the height
2050level of z =
20511.5* z<sub>i </sub>are plotted.</p> </td> </tr>
2052<tr> <td style="vertical-align: top;"> <p><a name="z_max_do2d"></a><b>z_max_do2d</b></p>
2053</td> <td style="vertical-align: top;">R<br> </td>
2054<td style="vertical-align: top;"><span style="font-style: italic;">zu(nz)</span><br> </td>
2055<td style="vertical-align: top;"> <p>Height level
2056up to which 2d cross sections are to be plotted
2057with <span style="font-weight: bold;">iso2d</span>
2058(in m).&nbsp; </p> <p>This parameter only applies for
2059&nbsp;<a href="#data_output_format">data_output_format</a>
2060= <span style="font-style: italic;">'iso2d'</span>.</p><p>It
2061affects plots of&nbsp; 2d vertical cross
2062sections (<a href="#data_output">data_output</a>)
2063when plotted with <span style="font-weight: bold;">iso2d</span>.
2064By
2065default, vertical sections are plotted up to the top boundary. <span style="font-weight: bold;"></span>In contrast, with <b>z_max_do2d
2066</b>the
2067visualization within
2068the plot can be limited to a certain height level (0 &lt;= z
2069&lt;= <b>z_max_do2d</b>).
2070Nevertheless, <span style="font-weight: bold;">all</span>
2071grid points
2072of the complete cross section are still output to the local files <a href="chapter_3.4.html#PLOT2D_XZ">PLOT2D_XZ</a>
2073or <a href="chapter_3.4.html#PLOT2D_YZ">PLOT2D_YZ</a>.
2074The level up to which the section is visualized can later be changed by
2075manually editing the
2076file <a href="chapter_3.4.html#PLOT2D_XZ_GLOBAL">PLOT2D_XZ_GLOBAL</a>
2077or <a href="chapter_3.4.html#PLOT2D_YZ_GLOBAL">PLOT2D_YZ_GLOBAL</a>
2078(the respective <span style="font-weight: bold;">iso2d</span>-parameter
2079is <a href="http://www.muk.uni-hannover.de/institut/software/iso2d_beschreibung.html#YRIGHT">yright</a>).</p>
2080</td> </tr> </tbody></table><br>
2081<br><h3 style="line-height: 100%;"><a name="Paketparameter"></a>Package
2082parameters: </h3>
2083Package (<span style="font-weight: bold;">mrun</span>
2084option -p): <span style="font-weight: bold;"><a name="particles_package"></a>particles</span>&nbsp;&nbsp;&nbsp;&nbsp;
2085NAMELIST group name: <span style="font-weight: bold;">particles_par<br>
2086</span><table style="text-align: left; width: 100%;" border="1" cellpadding="2" cellspacing="2"> <tbody><tr>
2087<td style="vertical-align: top;"><font size="4"><b>Parameter
2088name</b></font></td>
2089<td style="vertical-align: top;"><font size="4"><b>Type</b></font></td>
2090<td style="vertical-align: top;"> <p><b><font size="4">Default</font></b> <br> <b><font size="4">value</font></b></p> </td>
2091<td style="vertical-align: top;"> <p><font size="4"><b>Explanation</b></font></p>
2092</td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="dt_prel"></a><b>dt_prel</b></p>
2093</td> <td style="vertical-align: top;">R</td>
2094<td style="vertical-align: top;"><i>9999999.9</i></td>
2095<td style="vertical-align: top;"> <p><font face="Thorndale, serif"><span lang="en-GB">Temporal
2096interval at
2097which particles are to be released <span lang="en-GB">from
2098a particle
2099source </span>(</span></font>in <font face="Thorndale, serif"><span lang="en-GB">s).</span>&nbsp;
2100</font> </p> <p><span lang="en-GB"><font face="Thorndale, serif">By default
2101particles are released only at the beginning of a simulation
2102(t_init=0). The time of the first release (t_init) can be changed with
2103package parameter </font></span><span lang="en-GB"></span><font><a href="#particle_advection_start"><font face="Thorndale, serif">particle_advection_start</font></a>.
2104</font><span lang="en-GB"><font face="Thorndale, serif">The time of the last release can be
2105set with the package parameter <a href="#end_time_prel">end_time_prel</a>.
2106If <span style="font-weight: bold;">dt_prel</span>
2107has been set, additional
2108releases will be at t = t_init+<span style="font-weight: bold;">dt_prel</span>,
2109t_init+2*<span style="font-weight: bold;">dt_prel</span>,
2110t_init+3*<span style="font-weight: bold;">dt_prel</span>,
2111etc.. Actual release times
2112may slightly deviate from thesel values (</font></span><span lang="en-GB"><font face="Thorndale, serif">see
2113e.g. </font></span><a href="#dt_dopr"><span lang="en-GB"><font face="Thorndale, serif">dt_dopr</font></span></a><span lang="en-GB"><font face="Thorndale, serif">).</font></span></p>
2114<p><span lang="en-GB"><font face="Thorndale, serif"> The domain
2115of the particle <span lang="en-GB"><font face="Thorndale, serif">source </font></span>as
2116well as the distance of&nbsp; released particles
2117within this source </font></span><span lang="en-GB"><font face="Thorndale, serif">are determined via package
2118parameters </font></span><a href="#pst"><span lang="en-GB"><font face="Thorndale, serif">pst</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><a href="#psl"><span lang="en-GB"><font face="Thorndale, serif">psl</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><a href="#psr"><span lang="en-GB"><font face="Thorndale, serif">psr</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><a href="#pss"><span lang="en-GB"><font face="Thorndale, serif">pss</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><a href="#psn"><span lang="en-GB"><font face="Thorndale, serif">psn</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><a href="#psb"><span lang="en-GB"><font face="Thorndale, serif">psb</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><a href="#pdx"><span lang="en-GB"><font face="Thorndale, serif">pdx</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><a href="#pdy"><span lang="en-GB"><font face="Thorndale, serif">pdy</font></span></a>
2119<span lang="en-GB"><font face="Thorndale, serif">and
2120</font></span><a href="#pdz"><span lang="en-GB"><font face="Thorndale, serif">pdz</font></span></a><span lang="en-GB"><font face="Thorndale, serif">.</font></span><span lang="en-GB"><font face="Thorndale, serif"> By
2121default, one particle is released at all points defined by these
2122parameters. The package parameter <a href="#particles_per_point">particles_per_point</a>
2123can be used to start more than one particle per point.<br>
2124</font></span></p> <p><span lang="en-GB"><font face="Thorndale, serif">Up to 10
2125different groups of particles can be released at the same time (see </font></span><a href="chapter_4.2.html#number_of_particle_groups"><span lang="en-GB"><font face="Thorndale, serif">number_of_particle_groups</font></span></a><span lang="en-GB"><font face="Thorndale, serif">)
2126where each group may have a different source. All particles belonging
2127to one group have the same density ratio and the same radius. All other
2128particle features (e.g. location of the source) are
2129identical for all groups of particles.</font></span></p>Subgrid
2130scale velocities can (optionally) be included for calculating the
2131particle advection, using the method of Weil et al. (2004, JAS, 61,
21322877-2887). This method is switched on by the package
2133parameter <a href="#use_sgs_for_particles">use_sgs_for_particles</a>.
2134This also forces the Euler/upstream method to be used for time
2135advancement of the TKE (see initialization parameter <a href="chapter_4.1.html#use_upstream_for_tke">use_upstream_for_tke</a>).
2136The minimum timestep during the sub-timesteps is controlled by package
2137parameter <a href="#dt_min_part">dt_min_part</a>. <p><span lang="en-GB"><font face="Thorndale, serif">By
2138default, particles are weightless and transported passively with the
2139resolved scale flow. Particles can be given a mass and thus an inertia
2140by assigning the
2141package parameter </font></span><a href="#density_ratio"><span lang="en-GB"><font face="Thorndale, serif">density_ratio</font></span></a><span lang="en-GB"><font face="Thorndale, serif"> a
2142non-zero value (it
2143defines the ratio of the density of the fluid and the density of the
2144particles). In these cases their </font></span><a href="#radius"><span lang="en-GB"><font face="Thorndale, serif">radius</font></span></a><span lang="en-GB"></span><span lang="en-GB"><font face="Thorndale, serif">
2145must also be defined, which affects their flow resistance. </font></span><a href="#diameter"><span lang="en-GB"></span></a><span lang="en-GB"></span><span lang="en-GB"><font face="Thorndale, serif"> </font></span> </p>
2146<p><span lang="en-GB"><font face="Thorndale, serif">Boundary
2147conditions for the particle transport can be defined with package
2148parameters </font></span><a href="#bc_par_t"><span lang="en-GB"><font face="Thorndale, serif">bc_par_t</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><a href="#bc_par_lr"><span lang="en-GB"><font face="Thorndale, serif">bc_par_lr</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><a href="bc_par_ns"><span lang="en-GB"><font face="Thorndale, serif">bc_par_ns</font></span></a>
2149<span lang="en-GB"><font face="Thorndale, serif">and
2150</font></span><a href="#bc_par_b"><span lang="en-GB"><font face="Thorndale, serif">bc_par_b</font></span></a><span lang="en-GB"><font face="Thorndale, serif">.</font></span></p><span lang="en-GB"><font face="Thorndale, serif">Timeseries
2151of particle quantities in NetCDF format can be output to local file <a href="chapter_3.4.html#DATA_1D_PTS_NETCDF">DATA_1D_PTS_NETCDF</a>
2152by using package parameter <a href="#dt_dopts">dt_dopts</a>.<br></font></span><p>For
2153analysis, additional output of
2154particle
2155information in equidistant temporal intervals can be carried out using <a href="#dt_write_particle_data">dt_write_particle_data</a>
2156(file <a href="chapter_3.4.html#PARTICLE_DATA">PARTICLE_DATA</a>).<br>
2157</p> <p><span style="font-family: thorndale,serif;">Statistical
2158informations</span> (e.g. the total number of particles used, the
2159number of particles exchanged between the PEs, etc.) are output to the
2160local file <a href="chapter_3.4.html#PARTICLE_DATA">PARTICLE_INFOS</a>,
2161if switched on by the parameter <a href="#write_particle_statistics">write_particle_statistics</a>.
2162<br> </p> <p><span lang="en-GB"><font face="Thorndale, serif">If a job
2163chain is to be carried out, particle
2164informations </font></span><span lang="en-GB"><font face="Thorndale, serif">for the restart run (e.g. current
2165location of
2166all
2167particles at the end of the
2168run) is output to
2169the local file</font></span> <font><a href="chapter_4.2.html#dt_dvrp"><span lang="en-GB"></span></a></font><a href="chapter_3.4.html#PARTICLE_RESTART_DATA_OUT">PARTICLE_RESTART_DATA_OUT</a><font><a href="chapter_4.2.html#dt_dvrp"><span lang="en-GB"></span></a></font>,
2170<span lang="en-GB"><font face="Thorndale, serif">which
2171must be saved at the
2172end of the run <tt><span lang="en-GB"></span></tt>and
2173given as an
2174input file to the restart run
2175under local file name</font></span> <a href="chapter_3.4.html#PARTICLE_RESTART_DATA_IN">PARTICLE_RESTART_DATA_IN</a>
2176u<span lang="en-GB"><font face="Thorndale, serif">sing
2177respective file
2178connection statements in the <span style="font-weight: bold;">mrun</span>
2179configuration file. </font></span> <span lang="en-GB"></span></p><p><span lang="en-GB"></span><span lang="en-GB"><font face="Thorndale, serif">The output of
2180particles for visualization with the graphic software <span style="font-weight: bold;">dvrp</span> is steered by
2181the package
2182parameter </font></span><a href="chapter_4.2.html#dt_dvrp"><span lang="en-GB"><font face="Thorndale, serif">dt_dvrp</font></span></a><font face="Thorndale, serif"><span lang="en-GB">.
2183For visualization
2184purposes particles can be given a
2185diameter by the parameter <a href="chapter_4.2.html#dvrp_psize">dvrp_psize</a>
2186(this diameter only affects the visualization). All particles have the
2187same size. Alternatively, particles can be given an individual size and
2188a </span>color <span lang="en-GB">by modifying the
2189user-interface (subroutine</span></font> <span style="font-family: monospace;">user_init_particles</span>)<span lang="en-GB"><font face="Thorndale, serif">.
2190Particles can pull a
2191&ldquo;tail&rdquo; behind themselves to improve their
2192visualization.
2193This is steered via the parameter&nbsp;<a href="chapter_4.2.html#use_particle_tails">use_particle_tails</a>.</font></span><a href="chapter_4.2.html#maximum_number_of_tailpoints"><span lang="en-GB"></span></a></p> <span lang="en-GB"></span><p><b>So far, the
2194particle transport realized in PALM does only
2195work
2196duly in case of a constant vertical grid spacing!</b></p> </td>
2197</tr> <tr> <td style="vertical-align: top;"><p><a name="bc_par_b"></a><b>bc_par_b</b></p>
2198</td> <td style="vertical-align: top;">C*15</td>
2199<td style="vertical-align: top;"><i>&acute;reflect&acute;</i></td>
2200<td style="vertical-align: top;"> <p>Bottom
2201boundary condition for particle transport. </p> <p>By
2202default, particles are reflected at the bottom boundary.
2203Alternatively, a particle absorption can set by <b>bc_par_b</b>
2204= <i>&acute;absorb&acute;</i>.</p> </td>
2205</tr> <tr> <td style="vertical-align: top;"><p><a name="bc_par_lr"></a><b>bc_par_lr</b></p>
2206</td> <td style="vertical-align: top;">C*15</td>
2207<td style="vertical-align: top;"><i>&acute;cyclic&acute;</i></td>
2208<td style="vertical-align: top;"> <p>Lateral
2209boundary condition (x-direction) for particle
2210transport. </p> <p>By default, cyclic boundary conditions
2211are used along x.
2212Alternatively, reflection (<b>bc_par_lr</b>
2213= <i>&acute;reflect&acute;</i>) or absorption (<b>bc_par_lr</b>
2214= <i>&acute;absorb&acute;</i>)
2215can be set. <br> </p> <p>This lateral boundary
2216conditions should correspond to the
2217lateral boundary condition used for the flow (see <a href="chapter_4.1.html#bc_lr">bc_lr</a>).</p> </td>
2218</tr> <tr> <td style="vertical-align: top;"><p><a name="bc_par_ns"></a><b>bc_par_ns</b></p>
2219</td> <td style="vertical-align: top;">C*15</td>
2220<td style="vertical-align: top;"><i>&acute;cyclic&acute;</i></td>
2221<td style="vertical-align: top;"> <p>Lateral
2222boundary condition (y-direction) for particle
2223transport. </p> <p>By default, cyclic boundary conditions
2224are used along y.
2225Alternatively, reflection (<b>bc_par_ns</b>
2226= <i>&acute;reflect&acute;</i>) or absorption (<b>bc_par_ns</b>
2227= <i>&acute;absorb&acute;</i>)
2228can be set.<br> </p>
2229This lateral boundary conditions should correspond to the lateral
2230boundary condition used for the flow (see <a href="chapter_4.1.html#bc_ns">bc_ns</a>).</td> </tr>
2231<tr> <td style="vertical-align: top;"> <p><a name="bc_par_t"></a><b>bc_par_t</b></p>
2232</td> <td style="vertical-align: top;">C*15</td>
2233<td style="vertical-align: top;"><i>&acute;absorb&acute;</i></td>
2234<td style="vertical-align: top;"> <p>Top boundary
2235condition for particle transport. </p> <p>By default,
2236particles are absorbed at the top boundary.
2237Alternatively, a reflection condition can be set by <b>bc_par_t</b>
2238= <i>&acute;reflect&acute;</i>.</p> </td>
2239</tr> <tr> <td style="vertical-align: top;"><p><a name="density_ratio"></a><b>density_ratio</b></p>
2240</td> <td style="vertical-align: top;">R (10)</td>
2241<td style="vertical-align: top;"> <p><i>0.0, 9</i>
2242*<br> <i>9999999.9</i></p> </td> <td style="vertical-align: top;"> <p>Ratio of the density
2243of the fluid and the density of the
2244particles. </p> <p>With the default value<i> </i>the
2245particles are weightless and transported passively with the resolved
2246scale flow.
2247In case of <span style="font-weight: bold;">density_ratio</span>
2248/=
22490.0 particles have a mass and hence inertia so that their velocity
2250deviates more or less from the velocity of the surrounding flow.
2251Particle velocity is calculated analytically and depends on (besides
2252the density ratio and the current velocity difference between particles
2253and surrounding fluid) the
2254particle radius which is determined via <a href="#radius">radius</a>
2255as well as on the molecular viscosity (assumed as 1.461E-5 m<sup>2</sup>/s).
2256</p> <p>If <b>density_ratio</b> = <i>1.0</i>,
2257the particle density
2258corresponds to the density of the surrounding fluid and the particles
2259do not feel any buoyancy. Otherwise, particles will be accelerated
2260upwards (<b>density_ratio</b> &gt; <i>1.0</i>)
2261or downwards (<b>density_ratio</b> &lt; <i>1.0</i>).<br>
2262</p> <p>With several groups of particles (see <a href="chapter_4.2.html#number_of_particle_groups">number_of_particle_groups</a>),
2263each group can be assigned a different value. If the number of values
2264given for <span style="font-weight: bold;">density_ratio</span>
2265is less than the number of
2266groups defined by <span style="font-weight: bold;">number_of_particle_groups</span>,
2267then the last assigned value is used for all remaining groups. This
2268means that by default the particle density ratio for all groups will be
2269<span style="font-style: italic;">0.0</span>.</p>
2270</td> </tr> <tr><td align="left" valign="top"><a name="dt_dopts"></a><span style="font-weight: bold;">dt_dopts</span></td><td align="left" valign="top">R</td><td align="left" valign="top"><i>value of &nbsp;<a href="chapter_4.2.html#dt_data_output">dt_data_<br>output</a></i></td><td align="left" valign="top"><p lang="en-GB"><font face="Thorndale"><font face="Thorndale, serif">Temporal
2271interval</font> at which time series data of particle quantities
2272shall be output (</font>in <font face="Thorndale">s).&nbsp;</font></p>
2273<span lang="en-GB"><font face="Thorndale">If
2274particle advection is switched on (see</font></span><font><span style="font-family: thorndale;"> <a href="#dt_prel">dt_prel</a>)
2275this parameter can be used to assign
2276th</span></font><span lang="en-GB"><font face="Thorndale">e temporal
2277interval at which time series of particle quantities shall be output.
2278Output is written in NetCDF format on local file <a href="chapter_3.4.html#DATA_1D_PTS_NETCDF">DATA_1D_PTS_NETCDF</a>.<br><br>The
2279following list gives a short description of the&nbsp;quantities
2280available. Most quantities are averages over all particles. The
2281quantity name given in the first column is identical to the respective
2282name of the variable on the NetCDF file (see section <a href="chapter_4.5.1.html">4.5.1</a> for a general
2283description of the NetCDF files).<br><br>In case of using
2284more than one particle group (see <a href="#number_of_particle_groups">number_of_particle_groups</a>),
2285seperate time series are output for each of the groups. The long names
2286of the variables in the NetCDF file containing the respective
2287timeseries all end with the string</font><span style="font-style: italic; font-family: monospace;">' PG ##'</span><font face="Thorndale">, where ## is the number of the particle
2288group (<span style="font-style: italic;">01</span>, <span style="font-style: italic;">02</span>, etc.). <br>&nbsp;</font></span><table style="text-align: left; width: 631px; height: 652px;" border="1" cellpadding="2" cellspacing="2"><tbody><tr><td align="undefined" valign="undefined"><span style="color: rgb(255, 0, 0);">tnpt</span></td><td align="undefined" valign="undefined">total number of
2289particles</td></tr><tr><td align="undefined" valign="undefined"><span style="color: rgb(255, 0, 0);">x_</span></td><td align="undefined" valign="undefined">particle
2290x-coordinate&nbsp;with respect to the particle origin (in m)</td></tr><tr><td align="undefined" valign="undefined"><span style="color: rgb(255, 0, 0);">y_</span></td><td align="undefined" valign="undefined">particle
2291y-coordinate&nbsp;with respect to the particle origin (in m)</td></tr><tr><td align="undefined" valign="undefined"><span style="color: rgb(255, 0, 0);">z_</span></td><td align="undefined" valign="undefined">particle
2292z-coordinate&nbsp;with respect to the particle origin (in m)</td></tr><tr><td align="undefined" valign="undefined"><span style="color: rgb(255, 0, 0);">z_abs</span></td><td align="undefined" valign="undefined">absolute
2293particle z-coordinate (in m)</td></tr><tr><td align="undefined" valign="undefined"><span style="color: rgb(255, 0, 0);">u</span></td><td align="undefined" valign="undefined">u particle
2294velocity component (in m/s)</td></tr><tr><td align="undefined" valign="undefined"><span style="color: rgb(255, 0, 0);">v</span></td><td align="undefined" valign="undefined">v particle
2295velocity component (in m/s)</td></tr><tr><td align="undefined" valign="undefined"><span style="color: rgb(255, 0, 0);">w</span></td><td align="undefined" valign="undefined">w particle
2296velocity component (in m/s)</td></tr><tr><td align="undefined" valign="undefined"><span style="color: rgb(255, 0, 0);">u"</span></td><td align="undefined" valign="undefined">subgrid-scale u
2297particle velocity component (in m/s)</td></tr><tr><td align="undefined" valign="undefined"><span style="color: rgb(255, 0, 0);">v"</span></td><td align="undefined" valign="undefined">subgrid-scale v
2298particle velocity component (in m/s)</td></tr><tr><td align="undefined" valign="undefined"><span style="color: rgb(255, 0, 0);">w"</span></td><td align="undefined" valign="undefined">subgrid-scale w
2299particle velocity component (in m/s)</td></tr><tr><td align="undefined" valign="undefined"><span style="color: rgb(255, 0, 0);">npt_up</span></td><td align="undefined" valign="undefined">total number of
2300upward moving particles</td></tr><tr><td align="undefined" valign="undefined"><span style="color: rgb(255, 0, 0);">w_up</span></td><td align="undefined" valign="undefined">vertical
2301velocity of the upward moving particles (in m/s)</td></tr><tr><td align="undefined" valign="undefined"><span style="color: rgb(255, 0, 0);">w_down</span></td><td align="undefined" valign="undefined">vertical
2302velocity of the downward moving particles (in m/s)</td></tr><tr><td align="undefined" valign="undefined"><span style="color: rgb(255, 0, 0);">npt_max</span></td><td align="undefined" valign="undefined">maximum number
2303of particles in a subdomain (=tnpt for non-parallel runs)</td></tr><tr><td align="undefined" valign="undefined"><span style="color: rgb(255, 0, 0);">npt_min</span></td><td align="undefined" valign="undefined">minimum number
2304of particles in a subdomain (=tnpt for non-parallel runs)</td></tr><tr><td align="undefined" valign="undefined"><span style="color: rgb(255, 0, 0);">x*2</span></td><td align="undefined" valign="undefined">variance of the
2305particle x-coordinate&nbsp;with respect to <span style="color: rgb(255, 0, 0);">x_ </span>(in m<sup>2</sup>)</td></tr><tr><td align="undefined" valign="undefined"><span style="color: rgb(255, 0, 0);">y*2</span></td><td align="undefined" valign="undefined">variance of the
2306particle y-coordinate&nbsp;with respect to <span style="color: rgb(255, 0, 0);">y_</span> (in m<sup>2</sup>)</td></tr><tr><td align="undefined" valign="undefined"><span style="color: rgb(255, 0, 0);">z*2</span></td><td align="undefined" valign="undefined">variance of the
2307particle z-coordinate&nbsp;with respect to <span style="color: rgb(255, 0, 0);">z_</span> (in m<sup>2</sup>)</td></tr><tr><td align="undefined" valign="undefined"><span style="color: rgb(255, 0, 0);">u*2</span></td><td align="undefined" valign="undefined">variance of the
2308u particle velocity component with respect to <span style="color: rgb(255, 0, 0);">u </span>(in m<sup>2</sup>/s<sup>2</sup>)</td></tr><tr><td align="undefined" valign="undefined">v*2</td><td align="undefined" valign="undefined">variance of the
2309v particle velocity component with respect to&nbsp;<span style="color: rgb(255, 0, 0);">v </span>(in m<sup>2</sup>/s<sup>2</sup>)</td></tr><tr><td align="undefined" valign="undefined">w*2</td><td align="undefined" valign="undefined">variance of the
2310w particle velocity component with respect to&nbsp;<span style="color: rgb(255, 0, 0);">w </span>(in m<sup>2</sup>/s<sup>2</sup>)</td></tr><tr><td align="undefined" valign="undefined">u"2</td><td align="undefined" valign="undefined">variance of the
2311subgrid-scale u particle velocity component with respect to <span style="color: rgb(255, 0, 0);">u" </span>(in m<sup>2</sup>/s<sup>2</sup>)</td></tr><tr><td align="undefined" valign="undefined">v"2</td><td align="undefined" valign="undefined">variance of the
2312subgrid-scale v particle velocity component with respect to <span style="color: rgb(255, 0, 0);">v" </span>(in m<sup>2</sup>/s<sup>2</sup>)</td></tr><tr><td align="undefined" valign="undefined">w"2</td><td align="undefined" valign="undefined">variance of the
2313subgrid-scale w particle velocity component with respect to <span style="color: rgb(255, 0, 0);">w" </span>(in m<sup>2</sup>/s<sup>2</sup>)</td></tr><tr><td align="undefined" valign="undefined">npt*2</td><td align="undefined" valign="undefined">variance of the
2314number of particles with respect to the average number of particles per
2315subdomain</td></tr></tbody></table><span lang="en-GB"></span><span lang="en-GB"></span></td></tr><tr><td align="left" valign="top"><a name="dt_min_part"></a><span style="font-weight: bold;">dt_min_part</span></td><td align="left" valign="top">R</td><td align="left" valign="top"><span style="font-style: italic;">0.0002</span></td><td align="left" valign="top">Minimum value for the
2316particle timestep when SGS velocities are used (in s).<br><br>For
2317a further explanation see package parameter <a href="#use_sgs_for_particles">use_sgs_for_particles</a>.</td></tr><tr>
2318<td style="vertical-align: top;"> <p><a name="dt_write_particle_data"></a><b>dt_write_particle_</b>
2319<b>data</b></p> </td> <td style="vertical-align: top;">R<br> </td> <td style="vertical-align: top;"><i>9999999.9</i></td>
2320<td style="vertical-align: top;"> <p>Temporal
2321interval for output of particle data (in s). </p> <p>T<span lang="en-GB"></span><a href="#pr1d"><span lang="en-GB"></span></a><span lang="en-GB"></span><span lang="en-GB"><font face="Thorndale">his
2322parameter can be used to
2323assign the temporal interval at which particle data shall be output.</font></span>
2324Data are output to
2325the local file <a href="chapter_3.4.html#PARTICLE_DATA">PARTICLE_DATA</a>.
2326<span style="font-family: mon;">See the file description
2327for more
2328details about its format</span>. </p> <p>By
2329default, no particle data are output.</p> </td> </tr>
2330<tr> <td style="vertical-align: top;"> <p><a name="dvrp_psize"></a><b>dvrp_psize</b></p>
2331</td> <td style="vertical-align: top;">R<br> </td>
2332<td style="vertical-align: top;">0.2 * dx<br> </td>
2333<td style="vertical-align: top;"> <p>Diameter that
2334the particles is given in visualizations with
2335the <span style="font-weight: bold;">dvrp</span>
2336software (in
2337m).&nbsp; </p> <p>In case that particles are
2338visualized with the <span style="font-weight: bold;">dvrp</span>
2339software (see <a href="chapter_4.5.7.html">chapter
23404.5.7</a>), their size can be set by parameter <b>dvrp_psize</b>.
2341All
2342particles are displayed with this same size.<br> </p> <p>Alternatively,
2343the particle diameters can be set with the
2344user-interface in routine <span style="font-family: monospace;">user_init_particles</span>
2345(at the beginning of the simulation) and/or can be redefined after each
2346timestep in routine <tt>user<font style="font-size: 11pt;" size="2">_particle_attributes</font></tt>
2347(both routines can be found in file <tt><font style="font-size: 11pt;" size="2">user_interface.f90</font></tt><font style="font-size: 11pt;" size="2">)</font>.&nbsp;</p>
2348<p><b>Note:</b> This parameter determines exclusively
2349the size
2350under which particles appear in the <span style="font-weight: bold;">dvrp</span>
2351visualization. The flow relevant particle radius is determined via the
2352particle package parameter <a href="#radius">radius</a>!</p>
2353</td> </tr> <tr><td align="left" valign="top"><a name="end_time_prel"></a><span style="font-weight: bold;">end_time_prel</span></td><td align="left" valign="top">R</td><td align="left" valign="top"><span style="font-style: italic;">9999999.9</span></td><td align="left" valign="top">Time of the last release of
2354particles (in s).<br><br>See also <a href="#particle_advection_start">particle_advection_start</a>.</td></tr><tr>
2355<td style="vertical-align: top;"><span style="font-weight: bold;"><a name="initial_weighting_factor"></a>initial_weighting_factor</span></td>
2356<td style="vertical-align: top;">R<br> </td> <td style="vertical-align: top;"><span style="font-style: italic;">1.0</span><br> </td>
2357<td style="vertical-align: top;">Factor to define the real
2358number of initial droplets in a grid box.<br> <br>
2359In case of explicitly simulating cloud droplets (see <a href="chapter_4.1.html#cloud_droplets">cloud_droplets</a>),
2360the real number of initial droplets in a grid box is equal to the
2361initial number of droplets in this box (defined by the particle source
2362parameters <span lang="en-GB"><font face="Thorndale, serif"> </font></span><a href="chapter_4.2.html#pst"><span lang="en-GB"><font face="Thorndale, serif">pst</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><a href="chapter_4.2.html#psl"><span lang="en-GB"><font face="Thorndale, serif">psl</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><a href="chapter_4.2.html#psr"><span lang="en-GB"><font face="Thorndale, serif">psr</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><a href="chapter_4.2.html#pss"><span lang="en-GB"><font face="Thorndale, serif">pss</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><a href="chapter_4.2.html#psn"><span lang="en-GB"><font face="Thorndale, serif">psn</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><a href="chapter_4.2.html#psb"><span lang="en-GB"><font face="Thorndale, serif">psb</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><a href="chapter_4.2.html#pdx"><span lang="en-GB"><font face="Thorndale, serif">pdx</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><a href="chapter_4.2.html#pdy"><span lang="en-GB"><font face="Thorndale, serif">pdy</font></span></a>
2363<span lang="en-GB"><font face="Thorndale, serif">and
2364</font></span><a href="chapter_4.2.html#pdz"><span lang="en-GB"><font face="Thorndale, serif">pdz</font></span></a><span lang="en-GB"></span><span lang="en-GB"></span>)
2365times the <span style="font-weight: bold;">initial_weighting_factor</span>.</td>
2366</tr><tr> <td style="vertical-align: top;"> <p><a name="maximum_number_of_particles"></a><b>maximum_number_of_</b>
2367<br> <b>particles</b></p> </td> <td style="vertical-align: top;">I</td> <td style="vertical-align: top;"><i>1000</i></td>
2368<td style="vertical-align: top;"> <p>Maximum number
2369of particles (on a PE).&nbsp; </p> <p>This parameter
2370allows to set the number of particles for which
2371memory must be allocated at the beginning of the run.
2372If this memory becomes insufficient during the run, due to the
2373release of further particles (see <a href="#dt_prel">dt_prel</a>),
2374then more memory is automatically allocated.<br> </p>
2375For runs on several processors, <span style="font-weight: bold;">maximum_number_of_particles</span>
2376defines
2377the maximum number on each PE. This number must be larger than the
2378maximum number of particles initially released in a subdomain.</td>
2379</tr> <tr> <td style="vertical-align: top;"><p><a name="maximum_number_of_tailpoints"></a><b>maximum_number_of_</b>
2380<br> <b>tailpoints</b></p> </td> <td style="vertical-align: top;">I</td> <td style="vertical-align: top;"><i>100</i></td>
2381<td style="vertical-align: top;"> <p>Maximum number
2382of tailpoints that a particle tail can
2383have.&nbsp; </p> <p>&nbsp;<b>maximum_number_of_tailpoints</b>
2384sets the number of descrete points the tail consists of. A new point is
2385added to the particle tails after each time step. If the maximum number
2386of tail
2387points is reached after the corresponding number of timesteps, the
2388oldest respective tail points is deleted within the following
2389timestep.&nbsp; </p> <p>All particle tails have the
2390same number of points. The maximum
2391length of
2392these
2393tails is determined by the value of <b>maximum_number_of_tailpoints</b>
2394and by the minimum distance between each of the adjoining
2395tailpoints,&nbsp; which can be set by <a href="#minimum_tailpoint_distance">minimum_tailpoint_distance</a>.
2396Additionally, it can be determined that the temporal displacement
2397between the current position of the particle and the oldest point of
2398the tail may become not larger than a value to be assigned by <a href="#maximum_tailpoint_age">maximum_tailpoint_age</a>.</p>
2399</td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="maximum_tailpoint_age"></a><b>maximum_tailpoint_</b>
2400<br> <b>age</b></p> </td> <td style="vertical-align: top;">R</td> <td style="vertical-align: top;">100000.0</td> <td style="vertical-align: top;"> <p>Maximum age that the
2401end point of a particle tail is allowed to have (in s).&nbsp; </p>
2402<p>If the temporal displacement between the oldest point of a
2403particle tail and the current position of the particle becomes larger
2404than the value given by <b>maximum_tailpoint_age</b>, this
2405oldest
2406point (which defines the end of the tail) is
2407removed. If this time is so small that the number of points defining
2408the particle tail do not exceed the value given by <b>maximum_number_of_tailpoints</b>,
2409then the length the particle tails is a measure for the distance the
2410particle travelled along during the time interval defined via <b>maximum_tailpoint_age</b>,
2411i.e. for the
2412particle velocity. Fast particles will have long tails, slow particles
2413shorter ones (note: this will not neccessarily hold if <a href="#minimum_tailpoint_distance">minimum_tailpoint_distance</a>
2414= <i>0.0</i>).</p> </td> </tr> <tr>
2415<td style="vertical-align: top;"> <p><a name="minimum_tailpoint_distance"></a><b>minimum_tailpoint_distance</b></p>
2416</td> <td style="vertical-align: top;">R</td>
2417<td style="vertical-align: top;"><i>0.0</i></td>
2418<td style="vertical-align: top;"> <p>Minimum
2419distance allowed between two adjacent points of a
2420particle tail (in m).&nbsp; </p> <p>In case of <b>minimum_tailpoint_distance</b>
2421&gt; <i>0.0 </i>the
2422particle tail is extended by a new point only if the distance between
2423its current position and the most recent tail point exceed the
2424distance given via <b>minimum_tailpoint_distance</b>.<br>
2425</p> <p>If the length of the particle tails shall be
2426proportional to
2427the respective particle velocity, the parameter <a href="#maximum_tailpoint_age">maximum_tailpoint_age</a>
2428must also be set appropriately. </p> <b>Note:</b><br>
2429A suitable choice of <b>minimum_tailpoint_distance</b>
2430&gt; <i>0.0</i> is recommended, because then the tail
2431coordinates of
2432slowly moving particles require less memory and can also be drawn
2433faster. The upper limit of <b>minimum_tailpoint_distance</b>
2434should be chosen in a way that the visualized particle
2435tails still appear as smooth lines. Example: with a model domain of
24361000 m and a monitor resolution of 1280 * 1024 pixels it
2437should be sufficient to set <b>minimum_tailpoint_distance</b>
2438= <i>5.0</i>
2439(m). </td> </tr> <tr> <td style="vertical-align: top;"><a name="number_of_particle_groups"></a><span style="font-weight: bold;">number_of_particle_groups</span><br>
2440</td> <td style="vertical-align: top;">I<br> </td>
2441<td style="vertical-align: top;">1<br> </td> <td style="vertical-align: top;">Number of particle groups to be
2442used.<br> <br>
2443Each particle group can be assigned its own source region (see <a href="#pdx">pdx</a>, <a href="#psl">psl</a>,
2444<a href="#psr">psr</a>, etc.), particle diameter (<a href="#radius">radius</a>) and particle density ratio (<a href="density_ratio">density_ratio</a>).<br><br>If
2445less values are given for <a href="#pdx">pdx</a>, <a href="#psl">psl</a>,
2446etc. than the number of particle groups, then the last value is used
2447for the remaining values (or the default value, if the user did not set
2448the parameter).<br> <br>
2449The maximum allowed number of particle groups is limited to <span style="font-style: italic;">10</span>.<br> </td>
2450</tr><tr><td align="left" valign="top"><a name="particles_per_point"></a><span style="font-weight: bold;">particles_per_point</span></td><td align="left" valign="top">I</td><td align="left" valign="top">1</td><td align="left" valign="top">Number of particles to be
2451started per point.<br><br>By default, one particle is
2452started at all points of the particle source,&nbsp;defined by <span style="font-family: Thorndale,serif;">the </span><span lang="en-GB"><font face="Thorndale, serif">package
2453parameters </font></span><a href="chapter_4.2.html#pst"><span lang="en-GB"><font face="Thorndale, serif">pst</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><a href="chapter_4.2.html#psl"><span lang="en-GB"><font face="Thorndale, serif">psl</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><a href="chapter_4.2.html#psr"><span lang="en-GB"><font face="Thorndale, serif">psr</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><a href="chapter_4.2.html#pss"><span lang="en-GB"><font face="Thorndale, serif">pss</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><a href="chapter_4.2.html#psn"><span lang="en-GB"><font face="Thorndale, serif">psn</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><a href="chapter_4.2.html#psb"><span lang="en-GB"><font face="Thorndale, serif">psb</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><a href="chapter_4.2.html#pdx"><span lang="en-GB"><font face="Thorndale, serif">pdx</font></span></a><span lang="en-GB"><font face="Thorndale, serif">, </font></span><a href="chapter_4.2.html#pdy"><span lang="en-GB"><font face="Thorndale, serif">pdy</font></span></a>
2454<span lang="en-GB"><font face="Thorndale, serif">and
2455</font></span><a href="chapter_4.2.html#pdz"><span lang="en-GB"><font face="Thorndale, serif">pdz</font></span></a><span lang="en-GB"><font face="Thorndale, serif">.</font></span><span lang="en-GB"></span></td></tr><tr> <td style="vertical-align: top;"> <p><a name="particle_advection_start"></a><b>particle_advection_</b>
2456<br> <b>start</b></p> </td> <td style="vertical-align: top;">R </td> <td style="vertical-align: top;">0.0 </td> <td style="vertical-align: top;"> <p>Time of the first
2457release of particles (in s). </p> <p>If particles are not
2458to be released at the beginning of the
2459run, the release time can be set via <b>particle_advection_start</b>.<br>
2460If particle transport is switched on in a restart run, then <a href="#read_particles_from_restartfile">read_particles_from_restartfile</a>
2461= <span style="font-style: italic;">.F.</span> is
2462also required.</p><p>See also <a href="#end_time_prel">end_time_prel</a>.
2463</p> </td> </tr> <tr> <td style="vertical-align: top;"> <p><a name="particle_maximum_age"></a><b>particle_maximum_age</b></p>
2464</td> <td style="vertical-align: top;">R </td>
2465<td style="vertical-align: top;"><i>9999999.9</i>
2466</td> <td style="vertical-align: top;"> <p>Maximum
2467allowed age of particles (in s).&nbsp; </p> <p>If the
2468age of a particle exceeds the time set by <b>particle_maximum_age</b>,
2469the particle as well as its tail is deleted.</p> </td> </tr>
2470<tr> <td style="vertical-align: top;"> <p><a name="pdx"></a><b>pdx</b></p> </td>
2471<td style="vertical-align: top;">R (10)<br> </td>
2472<td style="vertical-align: top;"><i>10 * dx</i>
2473</td> <td style="vertical-align: top;"> <p>Distance
2474along x between particles within a particle source
2475(in m).&nbsp; </p> <p>If the particle source shall be
2476confined to one grid point,
2477the distances given by <span style="font-weight: bold;">pdx</span>,
2478<a href="#pdy">pdy</a>
2479and <a href="#pdz">pdz</a>
2480must be set larger than the respective domain size or <a href="#psl">psl</a>
2481= <a href="#psr">psr</a> has to be set
2482alternatively.<br>
2483</p> <p><span style="font-weight: bold;">pdx</span>
2484can be assigned a different value for each particle group (see <a href="#number_of_particle_groups">number_of_particle_groups</a>