Home


About us
PALM
PALM-4U
Installation
Release notes
Documentation
Getting Help
Gallery
Developers area

source: palm/trunk/DOC/app/chapter_4.4.1.html @ 1010

Last change on this file since 1010 was 244, checked in by letzel, 16 years ago
  • updated trunk/DOC/app to reflect the renaming of 'example' to 'example_cbl'
  • Property svn:keywords set to Id
File size: 16.1 KB
Line 
1<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
2<html><head>
3<meta content="text/html; charset=windows-1252" http-equiv="CONTENT-TYPE"><title>PALM chapter 4.4</title>
4<meta content="StarOffice 7 (Win32)" name="GENERATOR">
5<meta content="Siegfried Raasch" name="AUTHOR"> <meta content="20041013;13430732" name="CREATED"> <meta content="20041117;11162734" name="CHANGED"> <meta content="parallel LES model" name="KEYWORDS"> <style>
6<!--
7@page { size: 21cm 29.7cm }
8-->
9</style></head>
10<body style="direction: ltr;" lang="en-US"><h3 style="line-height: 100%;">4.4.1 A minimum
11parameter set for the CBL</h3>
12
13<p style="line-height: 100%;">In this chapter a brief,
14simple and
15complete parameter set is described, which can be used to simulate a&nbsp;quasi-stationary,
16convective, atmospheric boundary layer with&nbsp; <font color="#000000">zero
17mean horizontal
18wind.</font> For evaluation purposes, cross sections and
19horizontally averaged vertical
20profiles of typical boundary layer variables
21are output at the end of the run. The run shall be carried out in
22batch mode on the IBM Regatta "hanni" of the HLRN.</p>
23<p style="line-height: 100%;">The parameter file necessary
24to carry
25out a run must be provided to the model as an input file under the
26local name <a href="chapter_3.4.html#PARIN">PARIN</a>
27and has the following contents:</p>
28<pre style="line-height: 100%;">&amp;inipar <a href="chapter_4.1.html#nx">nx</a> = <span style="font-style: italic;">39</span>, <a href="chapter_4.1.html#ny">ny</a> = <span style="font-style: italic;">39</span>, <a href="chapter_4.1.html#nz">nz</a> = <span style="font-style: italic;">40</span>,<br> <a href="chapter_4.1.html#dx">dx</a> = <span style="font-style: italic;">50.0</span>, <a href="chapter_4.1.html#dy">dy</a> = <span style="font-style: italic;">50.0</span>, <a href="chapter_4.1.html#dz">dz</a> = <span style="font-style: italic;">50.0</span>,<br> <a href="chapter_4.1.html#dz_stretch_level">dz_stretch_level</a> = <span style="font-style: italic;">1200.0</span>,<br> <a href="chapter_4.1.html#fft_method">fft_method</a> = <span style="font-style: italic;">'temperton-algorithm'</span>,<br> <a href="chapter_4.1.html#initializing_actions">initializing_actions</a> = <span style="font-style: italic;">'set_constant_profiles'</span>,<br> <a href="chapter_4.1.html#ug_surface">ug_surface</a> = <span style="font-style: italic;">0.0</span>, <a href="chapter_4.1.html#vg_surface">vg_surface</a> = <span style="font-style: italic;">0.0</span>,<br> <a href="chapter_4.1.html#pt_vertical_gradient">pt_vertical_gradient</a> = <span style="font-style: italic;">0.0</span>, <span style="font-style: italic;">1.0</span>,<br> <a href="chapter_4.1.html#pt_vertical_gradient_level">pt_vertical_gradient_level</a> = <span style="font-style: italic;">0.0</span>, <span style="font-style: italic;">800.0</span>,<br> <a href="chapter_4.1.html#surface_heatflux">surface_heatflux</a> = <span style="font-style: italic;">0.1</span>, <a href="chapter_4.1.html#bc_pt_b">bc_pt_b</a> = <span style="font-style: italic;">'neumann'</span>,/<br><br>&amp;d3par <a href="chapter_4.2.html#end_time">end_time</a> = <span style="font-style: italic;">3600.0</span>,<br> <a href="chapter_4.2.html#create_disturbances">create_disturbances</a> = <span style="font-style: italic;">.T.</span>,<br> <a href="chapter_4.2.html#dt_disturb">dt_disturb</a> = <span style="font-style: italic;">150.0</span>, <a href="chapter_4.2.html#disturbance_energy_limit">disturbance_energy_limit</a> = <span style="font-style: italic;">0.01</span>,<br> <a href="chapter_4.2.html#dt_run_control">dt_run_control</a> = <span style="font-style: italic;">0.0</span>,<br> <a href="chapter_4.2.html#data_output">data_output</a> = <span style="font-style: italic;">'w_xy'</span>, <span style="font-style: italic;">'w_xz'</span>, <span style="font-style: italic;">'w_xz_av'</span>, <span style="font-style: italic;">'pt_xy'</span>, <span style="font-style: italic;">'pt_xz'</span>,<br> <a href="chapter_4.2.html#dt_data_output">dt_data_output</a> = <span style="font-style: italic;">900.0</span>,<br> <a href="chapter_4.2.html#dt_data_output_av">dt_data_output_av</a> = <span style="font-style: italic;">1800.0</span>,<br> <a href="chapter_4.2.html#averaging_interval">averaging_interval</a> = <span style="font-style: italic;">900.0</span>,<br> <a href="chapter_4.2.html#dt_averaging_input">dt_averaging_input</a> = <span style="font-style: italic;">10.0</span>,<br> <a href="chapter_4.2.html#section_xy">section_xy</a> = <span style="font-style: italic;">2</span>, <span style="font-style: italic;">10</span>, <a href="chapter_4.2.html#section_xz">section_xz</a> = <span style="font-style: italic;">20</span>,<br> <a href="chapter_4.2.html#data_output_2d_on_each_pe">data_output_2d_on_each_pe</a> = <span style="font-style: italic;">.F.</span>,<br> <a href="chapter_4.2.html#dt_dopr">dt_dopr</a> = <span style="font-style: italic;">900.0</span>, <a href="chapter_4.2.html#averaging_interval_pr">averaging_interval_pr</a> = <span style="font-style: italic;">600.0</span>,<br> <a href="chapter_4.2.html#dt_averaging_input_pr">dt_averaging_input_pr</a> = <span style="font-style: italic;">10.0</span>,<br> <a href="chapter_4.2.html#data_output_pr">data_output_pr</a> = <span style="font-style: italic;">'#pt'</span>, <span style="font-style: italic;">'w&rdquo;pt&rdquo;'</span>, <span style="font-style: italic;">'w*pt*'</span>, <span style="font-style: italic;">'wpt'</span>, <span style="font-style: italic;">'w*2'</span>, <span style="font-style: italic;">'pt*2'</span>,<br> <a href="chapter_4.2.html#cross_profiles">cross_profiles</a> = <span style="font-style: italic;">' pt '</span>, <span style="font-style: italic;">' w"pt" w*pt* wpt '</span>, <span style="font-style: italic;">' w*2 '</span>, <span style="font-style: italic;">' pt*2 '</span>,<br> <a href="chapter_4.2.html#cross_xtext">cross_xtext</a> = <span style="font-style: italic;">'pot. temperature in K'</span>,<br> <span style="font-style: italic;">'heat flux in K ms&gt;-&gt;1'</span>,<br> <span style="font-style: italic;">'velocity variance in m&gt;2s&gt;-&gt;2'</span>,<br> <span style="font-style: italic;">'temperature variance in K&gt;2'</span>,<br> <a href="chapter_4.2.html#z_max_do1d">z_max_do1d</a> = <span style="font-style: italic;">1500.0</span>, /</pre>
29<p style="line-height: 100%;">The initialization
30parameters (<tt><font style="font-size: 10pt;" size="2">&amp;inipar</font></tt>)
31are located at the beginning of the file. For analysis of a
32convective boundary layer of approx. 1000 m thickness the horizontal
33size of the model domain should amount to at least 2 km x 2 km. In
34order to resolve the convective structures a grid spacing of <b>dx</b>
35=
36<b>dy</b> = <b>dz</b> = <i>50 m</i>
37is enough, since the typical
38diameter of convective plumes is more than 100 m. Thereby the
39upper array index in the two horizontal directions needs to be <b>nx</b>
40= <b>ny</b> = <i>39</i>. <font color="#000000">Since in
41each case the lower array index has the value 0, 40 grid points are
42used along both horizontal directions.</font> In the vertical
43direction
44the domain must be high enough to include the entrainment processes at
45the top of the boundary layer as well as the propagation of gravity
46waves, which were stimulated by
47the convection. However, in the stably stratified region the grid
48resolution has not necessarily to be as high as within the boundary
49layer. This can be obtained by a vertical stretching of the grid
50starting
51from 1200 m via <b>dz_stretch_level</b> = <i>1200.0
52m.</i> This saves
53grid points and computing time. <font color="#800000">T</font><font color="#000000">he
54upper boundary of the model is located at (see </font><a href="chapter_4.1.html#dz_stretch_factor"><font color="#000000">dz_stretch_factor</font></a><font color="#000000">)
55&hellip; m (computed by the model)</font>.</p><p style="line-height: 100%;">Fast Fourier transformations are
56calculated using the Temperton-algorithm, which -on the IBM Regatta- is
57faster than the default system-specific algorithm (from IBM essl
58library).</p><p style="line-height: 100%;">The
59initial profiles for
60wind and temperature can be assigned via <b>initializing_actions</b>
61= <span style="font-style: italic;">'set_constant_profiles'</span>.
62The wind speed, constant with
63height, amounts to <b>ug_surface</b> = <b>vg_surface</b>
64= <i>0.0 m/s</i>. In order
65to allow for a fast onset of convection, a neutral stratified layer up
66to z
67= 800 m capped by an inversion with dtheta/dz = 1K/100 m is given:
68<b>pt_vertical_gradient</b> = <i>0.0, 1.0</i>,
69<b>pt_vertical_gradient_level</b> = <i>0.0, 800.0.</i>
70The surface
71temperature, which by default amounts to 300 K, provides the fixed
72point for the temperature profile (see <a href="chapter_4.1.html#pt_surface">pt_surface</a>).
73Convection is driven by a given, near-surface sensible heat flux via <b>surface_heatflux</b>
74= <i>0.1 K m/s.</i> A given surface sensible heta flux
75requires the
76bottom boundary condition for potential temperature to be <b>bc_pt_b</b>
77=
78<span style="font-style: italic;">'neumann'</span> .
79Thus
80all initialization parameters are determined. These can not be
81changed during the run (also not for restart runs). </p>
82<p style="line-height: 100%;">Now the run parameters (<tt><font style="font-size: 10pt;" size="2">&amp;d3par</font></tt>)
83must be specified. To produce a quasi stationary boundary layer the
84simulated time should be at least one hour, i.e. <b>end_time</b>
85= <i>3600
86s.</i> To stimulate convection, the initially homogeneous (zero)
87wind
88field must be disturbed (<b>create_disturbances</b> = <i>.T.</i>).
89These perturbations should be repeated in a temporal interval of
90<b>dt_disturb</b> = <i>150.0 s</i> until the
91energy of the
92perturbations exceeds the value <b>disturbance_energy_limit</b>
93= 0.<i>01
94m<sup>2</sup>/s<sup>2</sup></i>. After
95each time step run time
96informations (e.g. size of the timestep, maximum velocities, etc.) are
97to be written to the local file <a href="chapter_3.4.html#RUN_CONTROL">RUN_CONTROL</a>
98(<b>dt_run_control</b> = <i>0.0 s</i>).</p><p style="line-height: 100%;">Instantaneous cross section data
99of vertical velocity (<span style="font-style: italic;">w</span>)
100and potential temperature (<span style="font-style: italic;">pt</span>)
101are to be output for horizontal (<span style="font-style: italic;">xy</span>)
102and vertical (<span style="font-style: italic;">xz</span>)
103cross sections, and additionally, time averaged (<span style="font-style: italic;">av</span>) vertical cross
104section data are to be output for the vertical velocity: <span style="font-weight: bold;">data_output</span> = <span style="font-style: italic;">'w_xy'</span>, <span style="font-style: italic;">'w_xz'</span>, <span style="font-style: italic;">'w_xz_av'</span>, <span style="font-style: italic;">'pt_xy'</span>, <span style="font-style: italic;">'pt_xz'</span>. Output of
105instantaneous (time averaged) data is done after each 900 (1800)s: <span style="font-weight: bold;">dt_data_output</span> = <span style="font-style: italic;">900.0</span>, <span style="font-weight: bold;">dt_data_output_av</span> = <span style="font-style: italic;">1800.0</span>. The
106averaged data are time averaged over the last 900.0 s, where the
107temporal interval of data entering the average is 10 s: <span style="font-weight: bold;">averaging_interval</span> =
108<span style="font-style: italic;">900.0</span>, <span style="font-weight: bold;">dt_averaging_input</span> =
109<span style="font-style: italic;">10.0</span>.
110Horizontal cross sections are output for vertical levels with grid
111index k=2 and k=10, vertical cross sections are output for index j=20: <span style="font-weight: bold;">section_xy</span> = <span style="font-style: italic;">2</span>, <span style="font-style: italic;">10</span>, <span style="font-weight: bold;">section_xz</span> = <span style="font-style: italic;">20</span>. For runs on
112more than one processor, cross section data are collected and output on
113PE0: <span style="font-weight: bold;">data_output_2d_on_each_pe</span>
114= <span style="font-style: italic;">.F.</span>.</p><p style="line-height: 100%;">Output
115of vertical profiles is to be done after each 900 s. The profiles shall
116be temporally averaged<font color="#000000"> over the last
117<font color="#000000">600 </font>seconds, </font>whereby
118the temporal interval of the profiles entering the average has to be
11910 s: <b>dt_dopr</b> = <i>900.0 s</i>, <b>averaging_interval_pr</b>
120=
121<i>600.0 s</i>, <b>dt_averaging_input_pr</b> =
122<i>10.0 s.</i> The temperature
123profile including the initial temperature profile (therefore <span style="font-style: italic;">'#pt'</span>),
124the subgrid scale, resolved and total vertical sensible heat flux as
125well as the variances of the vertical velocity and the potential
126temperature are to be output:&nbsp; <b>data_output_pr</b>
127= <span style="font-style: italic;">'#pt'</span><i>,
128'w"pt&rdquo;',
129'w*pt*', 'wpt', 'w*2', 'pt*2'</i>.</p><p style="line-height: 100%;">If the data output format for
130graphic software <span style="font-weight: bold;">profil</span>
131is selected (see <a href="chapter_4.2.html#data_output_format">data_output_format</a>),
132the temperature
133profile and the individual variances are to be drawn into independent
134coordinate systems, and in contrast to this all heat flux profiles are
135to
136be
137drawn into the same system: <b>cross_profiles</b> = <span style="font-style: italic;">'pt'</span><i>,
138'w"pt"w*pt*wpt', 'w*2', 'pt*2'</i>. The legend of the x
139axes of these systems is set to <b>cross_xtext</b>= <i>'pot.
140temperature in K', 'heat flux in K ms&gt;-&gt;1', 'velocity
141variance
142in m&gt;2s&gt;-&gt;2', 'temperature variance in K&gt;2'</i>.
143The profiles are to be drawn up to a height level of <b>z_max_do1d</b>
144=
145<i>1500.0 m</i>. </p>
146<p style="line-height: 100%;">Before starting the mo<font color="#000000">del
147on the parallel computer, the number of processing elements must be
148specified.</font> Since relatively few grid points are used for
149this run, choosing of e.g. 8 PEs is sufficient. By default, a 1d domain
150decomposition along x is used on the IBM-Regatta, which means that a
151virtual processor topology (grid) of 8*1 (x*y) is used. (<span style="font-weight: bold;">Note:</span> the user may
152adjust this
153default domain decomposition with the help of the parameters <a href="chapter_4.2.html#npex">npex</a>
154and <a href="chapter_4.2.html#npey">npey</a>).
155</p><p style="line-height: 100%;">Provided that the
156parameters
157file described above are set within the file </p>
158<ul> <pre style="margin-bottom: 0.5cm; line-height: 100%;"><font style="font-size: 10pt;" size="2">~/palm/current_version/JOBS/example/INPUT/example_cbl_p3d</font></pre></ul><p style="line-height: 100%;">and that the conditions
159mentioned in the
160first sections of <a href="chapter_3.2.html">chapter
1613.2</a> are met, the model run can be started with the command </p>
162<p style="line-height: 100%;"><font face="Cumberland, monospace"><font style="font-size: 10pt;" size="2">mrun
163-d example_cbl -h ibmh -K parallel -X 8 -T 8 -t 1800 -q cdev -r
164&ldquo;d3# xy# xz# pr#&rdquo;</font></font></p>
165<p style="line-height: 100%;">The output files will appear
166in the
167directories </p>
168<blockquote style="line-height: 100%;"><tt><font style="font-size: 10pt;" size="2">~/palm/current_version/JOBS/example_cbl/MONITORING</font></tt><font style="font-size: 10pt;" size="2"><br> </font><tt><font style="font-size: 10pt;" size="2">~/palm/current_version/JOBS/example_cbl/OUTPUT
169,</font></tt></blockquote>
170<p style="line-height: 100%;">while the job protocol will
171appear in
172directory <font style="font-size: 10pt;" size="2"><font face="Cumberland, monospace">~/</font></font><tt><font style="font-size: 10pt;" size="2"><font face="Cumberland, monospace">job_queue</font></font></tt>.
173<br>
174&nbsp; </p>
175<hr><p style="line-height: 100%;"><br><font color="#000080"><font color="#000080"><a href="chapter_4.4.html"><font color="#000080"><img style="border: 2px solid ; width: 32px; height: 32px;" alt="" name="Grafik1" src="left.gif"></font></a><a href="index.html"><font color="#000080"><img name="Grafik2" src="up.gif" align="bottom" border="2" height="32" width="32"></font></a><a href="chapter_4.4.2.html"><font color="#000080"><img style="border: 2px solid ; width: 32px; height: 32px;" alt="" name="Grafik3" src="right.gif"></font></a></font></font></p><p style="line-height: 100%;"><i>Last change:&nbsp;
176</i>$Id: chapter_4.4.1.html 244 2009-02-24 22:42:38Z raasch $
177<br>&nbsp; <br>
178&nbsp; </p>
179</body></html>
Note: See TracBrowser for help on using the repository browser.
                                                                                                                                                                                                                                                                                                                                                                               
  | Impressum | ©Leibniz Universität Hannover |