Changeset 198 for palm/trunk/DOC/tec


Ignore:
Timestamp:
Sep 17, 2008 8:55:28 AM (14 years ago)
Author:
raasch
Message:

file headers updated for the next release 3.5

File:
1 edited

Legend:

Unmodified
Added
Removed
  • palm/trunk/DOC/tec/technical_documentation.html

    r197 r198  
    23512351prognostic_equations</td></tr><tr><td style="vertical-align: top;">&nbsp;</td><td style="vertical-align: top;">&nbsp;</td><td style="vertical-align: top;">&nbsp;</td><td style="vertical-align: top;">E</td><td style="vertical-align: top;">Bugfix: summation of<span style="font-family: Courier New,Courier,monospace;"> sums_l_l </span>(<span style="font-family: Courier New,Courier,monospace;">diffusivities</span>).<br><br>Several bugfixes in the ocean part: Initial density<span style="font-family: Courier New,Courier,monospace;"> rho </span>is calculated (<span style="font-family: Courier New,Courier,monospace;">init_ocean</span>). Error in initializing<span style="font-family: Courier New,Courier,monospace;"> u_init </span>and<span style="font-family: Courier New,Courier,monospace;"> v_init </span>removed (<span style="font-family: Courier New,Courier,monospace;">check_parameters</span>). Calculation of density flux now starts from<span style="font-family: Courier New,Courier,monospace;"> nzb+1 </span>(<span style="font-family: Courier New,Courier,monospace;">production_e</span>).<br><br>Bugfix: <span style="font-family: Courier New,Courier,monospace;">pleft</span>/<span style="font-family: Courier New,Courier,monospace;">pright</span> changed to <span style="font-family: Courier New,Courier,monospace;">pnorth</span>/<span style="font-family: Courier New,Courier,monospace;">psouth</span> in sendrecv of particle tail numbers along y, small bugfixes in the SGS part (<span style="font-family: Courier New,Courier,monospace;">advec_particles</span>).<br><br>Bugfix:<span style="font-family: Courier New,Courier,monospace;"> model_string </span>needed a default value (<span style="font-family: Courier New,Courier,monospace;">combine_plot_fields</span>).<br><br>Bugfix: wavenumber calculation for even<span style="font-family: Courier New,Courier,monospace;"> nx </span>in routines<span style="font-family: Courier New,Courier,monospace;"> maketri </span>(<span style="font-family: Courier New,Courier,monospace;">poisfft</span>).<br><br>Bugfix: assignment of fluxes at walls.<br><br>Bugfix: absolute value of<span style="font-family: Courier New,Courier,monospace;"> f </span>must be used when calculating the Blackadar mixing length (<span style="font-family: Courier New,Courier,monospace;">init_1d_model</span>).</td><td style="vertical-align: top;">advec_particles,
    23522352check_parameters, combine_plot_fields, diffusion_s, diffusivities,
    2353 init_ocean, init_1d_model, poisfft, production_e</td></tr><tr><td style="vertical-align: top;">16/09/08</td><td style="vertical-align: top;">SR</td><td style="vertical-align: top;">3.5</td><td style="vertical-align: top;">N</td><td style="vertical-align: top;"></td><td style="vertical-align: top;"></td></tr><tr><td align="undefined" valign="undefined">&nbsp;</td><td align="undefined" valign="undefined">&nbsp;</td><td align="undefined" valign="undefined">&nbsp;</td><td style="vertical-align: top;">C</td><td style="vertical-align: top;"></td><td align="undefined" valign="undefined"></td></tr><tr><td align="undefined" valign="undefined">&nbsp;</td><td align="undefined" valign="undefined">&nbsp;</td><td align="undefined" valign="undefined">&nbsp;</td><td style="vertical-align: top;">E</td><td style="vertical-align: top;"></td><td style="vertical-align: top;"></td></tr>
     2353init_ocean, init_1d_model, poisfft, production_e</td></tr><tr><td style="vertical-align: top;">16/09/08</td><td style="vertical-align: top;">SR</td><td style="vertical-align: top;">3.5</td><td style="vertical-align: top;">N</td><td style="vertical-align: top;">The
     2354turbulence recycling method of Kataoka and Mizuno (2002) has been
     2355implemented. A precursor run must be used to initialize the main run
     2356with a quasi-stationary turbulence state (see <span style="font-family: monospace;">initializing_actions</span> = <span style="font-style: italic;">'read_data_for_recycling'</span>).
     2357The precursor run may have a smaller domain than the main run, which
     2358helps to spare CPU time. New parameters for steering the recycling
     2359method are <span style="font-family: monospace;">turbulent_inflow</span>, <span style="font-family: monospace;">recycling_width</span>, <span style="font-family: monospace;">inflow_damping_height</span>, and <span style="font-family: monospace;">inflow_damping_width</span>.<br><br>One
     2360new feature related with the recycling method is that within a job
     2361chain, the number of processors or the virtual processor grid is now
     2362allowed to&nbsp;change. Such a change modifies the domain decomposition
     2363and requires that the PEs within a restart job have to know from which
     2364file(s) they have to read their restart data (up to now there has been
     2365a one to one relation between the restart files and the PEs, i.e.
     2366processor 4 only reads data from file "_0004"). Informations about the
     2367subdomain index limits of all restart files are now written by PE0 (to
     2368file "_0000"). Also, all variables containing global values which do
     2369not depend on the subdomain are written to file "_0000" only. In a
     2370restart run, this file is opened by all PEs to read these global values
     2371and the informations about the subdomain index limits. Depending on
     2372these index limit information it is decided which restart file(s)
     2373has/have to be opened in order to read the complete data necessary for
     2374the current subdomain.<br><br>The plant canopy model has been extended
     2375to the prognostic equation for&nbsp;the potential temperature
     2376and&nbsp;passive scalar, as proposed by Watanabe (2004) and Shaw and
     2377Schumann (1992).<br><br>Particle advection adjusted for ocean runs.<br><br>Grid
     2378coarsening (multigrid method) is limited to subdomains (i.e. no
     2379collection of data on PE0 for further coarsening beyond subdomains) if <span style="font-family: monospace;">mg_switch_to_pe0_level</span> = <span style="font-style: italic;">-1</span>.<br><br>User-defined spectra are allowed.<br><br>Output of <span style="font-family: monospace;">q*2</span> profile and <span style="font-family: monospace;">w"u"0</span>, <span style="font-family: monospace;">w"v"0</span>, <span style="font-family: monospace;">w"q"0</span> timeseries are added.<br><br>Argument<span style="font-family: monospace;"> -c </span>introduced to most of the subjob calls, which allows the user to choose his own job catalog by setting the environment variable<span style="font-family: monospace;"> job_catalog </span>in the configuration file (default is <span style="font-family: monospace;">~/job_queue</span>).<br><br>This release contains first adjustments for the SGI-ICE system at HLRN.<br><br><span style="font-weight: bold;">Attention: On the SGI-ICE,&nbsp;runs are still aborted or are hanging for unknown reasons from time to time.</span></td><td style="vertical-align: top;">advec_particles, calc_spectra, check_open, check_parameters, data_output_spectra,
     2380header, init_3d_model, init_particles, init_pegrid, modules, netcdf, parin, particle_boundary_conds, plant_canopy_model,
     2381
     2382prognostic_equations, read_3d_binary, read_var_list, time_integration,
     2383user_interface, write_3d_binary, write_var_list<br><br><span style="font-weight: bold;">new:</span><br>inflow_turbulence<br></td></tr><tr><td align="undefined" valign="undefined">&nbsp;</td><td align="undefined" valign="undefined">&nbsp;</td><td align="undefined" valign="undefined">&nbsp;</td><td style="vertical-align: top;">C</td><td style="vertical-align: top;">Allow 100 spectra levels instead of 10 for consistency with define_netcdf_header. (<span style="font-family: monospace;">calc_spectra</span>, <span style="font-family: monospace;">data_output_spectra</span>, <span style="font-family: monospace;">header</span>, <span style="font-family: monospace;">modules</span>, <span style="font-family: monospace;">netcdf</span>)<br><br>Modification
     2384of the integrated version of the profile function for momentum for
     2385unstable stratification; more consistent flux definitions. (<span style="font-family: monospace;">wall_fluxes</span>, <span style="font-family: monospace;">production_e</span>)<br><br>Strict grid matching along z is not needed for mg-solver. (<span style="font-family: monospace;">check_parameters</span>)<br><br>Sharp
     2386lateral edges of the plant canopy introduced. Before, smaller values of
     2387the leaf area density than those applied in the prognostic equation for
     2388the vertical velocity were used in the prognostic equations for u and v
     2389at the boundaries of the plant_canopy. (<span style="font-family: monospace;">init_3d_model</span>)<br><br>-<span style="font-family: monospace;">cross_ts_*</span> (<span style="font-family: monospace;">read_var_list</span>, <span style="font-family: monospace;">write_var_list</span>), -<span style="font-family: monospace;">dopr_time_count</span>, old profil parameters (<span style="font-family: monospace;">cross_...</span>, <span style="font-family: monospace;">dopr_crossindex</span>,&nbsp;<span style="font-family: monospace;">profile_***</span>) removed. (<span style="font-family: monospace;">read_3d_binary</span>, <span style="font-family: monospace;">write_3d_binary</span>)<br><br>Reading/writing of <span style="font-family: monospace;">hom_sum</span>, <span style="font-family: monospace;">volume_flow_area</span>, <span style="font-family: monospace;">volume_flow_initial</span> is moved from <span style="font-family: monospace;">..._3d_binary</span> to <span style="font-family: monospace;">..._var_list</span>, <span style="font-family: monospace;">hom_sum</span> is now allocated in <span style="font-family: monospace;">parin</span> instead of <span style="font-family: monospace;">init_3d_model</span>.<br><br><span style="font-family: monospace;">npex</span>, <span style="font-family: monospace;">npey</span> moved from inipar- to d3par-namelist. (<span style="font-family: monospace;">parin</span>)<br><br><span style="font-family: monospace;">myid_char14</span>
     2390removed. This was used, since on IBM machines the process rank could be
     2391changed when the final communicator was defined, in order to save the
     2392preliminary processor-id for opening the binary output file for
     2393restarts (unit 14), because otherwise a mismatch occured when reading
     2394this file in a restart job. This mechanism is not needed any more since
     2395the subdomain informations are now contained in file _0000. (<span style="font-family: monospace;">parin</span>, <span style="font-family: monospace;">check_open</span>)<br><br>Transpositions for the 2D domain decomposition have been optimized by using <span style="font-family: monospace;">f_inv</span>
     2396as an automatic array instead of providing the memory by a dummy
     2397argument. This spares one copy loop per transposition. Order of indices
     2398in the 3D loops in some of the transpose routines have been rearranged
     2399for better cache utilization. Both have been suggested by Roland
     2400Richter (SGI) as part of the HLRN-II benchmark process. (<span style="font-family: monospace;">transpose</span>)<br><br>Workaround for getting information about the coupling mode, because the<span style="font-family: monospace;"> -env </span>Option (mpiexec) is not available on many systems (SGI, lcxt4 at&nbsp; Bergen Center for Computational Science). (<span style="font-family: monospace;">palm</span>, <span style="font-family: monospace;">mrun</span>)<br><br>Format adjustments in order to allow CPU# &gt; 999. (<span style="font-family: monospace;">cpu_statistics</span>)<br><br>Provisional solution for run_control_1d output: add 'CALL check_open( 15 )'. (<span style="font-family: monospace;">init_1d_model</span>)<br><br>Subdomain data are read into temporary arrays <span style="font-family: monospace;">pf_tmp</span> / <span style="font-family: monospace;">pf3d_tmp</span> in order to avoid INTEL compiler warnings about (automatic) creation of temporary arrays. (<span style="font-family: monospace;">combine_plot_fields</span>)<br><br>Time coordinate t=0 stored on netcdf-file only if an output is required for this time for at least one of the profiles.<br><br>Output of spline timeseries <span style="font-family: monospace;">splptx</span> etc. removed. (<span style="font-family: monospace;">flow_statistics</span>)</td><td style="vertical-align: top;">advec_s_ups,
     2401advec_u_ups, advec_v_ups, advec_w_ups, calc_spectra, check_open,
     2402check_parameters, cpu_statistics, data_output_profiles,
     2403data_output_spectra, flow_statistics, header, init_1d_model,
     2404init_3d_model, modules, netcdf, palm, parin, poisfft, production_e,
     2405read_3d_binary, read_var_list, transpose, wall_fluxes, write_3d_binary, write_var_list<br></td></tr><tr><td align="undefined" valign="undefined">&nbsp;</td><td align="undefined" valign="undefined">&nbsp;</td><td align="undefined" valign="undefined">&nbsp;</td><td style="vertical-align: top;">E</td><td style="vertical-align: top;">Bugfix:
     2406modification of the calculation of the vertical turbulent momentum
     2407fluxes u'w' and v'w'; the first usws that is computed corresponds to
     2408-u'w'/u* and not as priorily assumed to (-u'w')**0.5, the first vsws
     2409that is computed corresponds to -v'w'/u* and not as priorily assumed to
     2410(-v'w')**0.5. Therefore, the intermediate result for usws has to be
     2411multiplied by -u* instead by itself in order to get u'w'. Accordingly,
     2412the intermediate result for vsws has to be multiplied by -u* instead by
     2413itself in order to get v'w'. As u* is calculated for the position of a
     2414scalar an additional interpolation of u* to the position of u and v,
     2415respectively, is necessary. As u* is not determined for the ghost
     2416points on each PE, an additional exchange of information from
     2417neighbouring PEs is necessary. Tests showed that this
     2418decreases u* by some 10% and increases TKE and momentum fluxes by some
     241910% because friction was underestimated and momentum fluxes were
     2420wrongly calculated due to the bug. (<span style="font-family: monospace;">prandtl_fluxes</span>, <span style="font-family: monospace;">wall_fluxes</span>)<br><br><span style="font-weight: bold;">Attention:</span> <br><span style="font-weight: bold;">
     2421The content of the testresult-file (run control output) has changed
     2422due to the above bugfix.</span><br><br>Bugfix:
     2423change definition of us_wall from 1D to 2D.&nbsp;<br><br>Bugfix: exponent = 1.0 / ( ny + 1.0 ) in calc_spectra_x. (<span style="font-family: monospace;">calc_spectra</span>)<br><br>Bugfix: calculation of horizontal fluxes at vertical walls. (<span style="font-family: monospace;">diffusion_s</span>)<br><br>Bugfix: zero assignments to tendency arrays in case of restarts. (<span style="font-family: monospace;">init_3d_model</span>)<br><br>Bugfix: wrong grid level used in multigrid solver, routine <span style="font-family: monospace;">restrict</span>. (<span style="font-family: monospace;">poismg</span>)<br><br>Bugfix:
     2424Construction of vertical profiles when 10 gradients have been specified
     2425in the parameter list (ug, vg, pt, q, sa, lad). (<span style="font-family: monospace;">check_parameters</span>)<br><br>Bugfix: Determination of the number of vertical levels for which spectra shall be output - 100 replaced by 10. (<span style="font-family: monospace;">netcdf</span>)<br><br>Bugfix:
     2426Lengths of those characters that are needed for the output of the
     2427characteristic levels of potential temperature, the geostrophic wind,
     2428scalar concentration, humidity and leaf area density to the header file
     2429has been increased. (<span style="font-family: monospace;">header</span>)<br><br>Bugfix: add ')' in <span style="font-family: monospace;">user_read_restart_data</span>.<br><br>The leaf area density is explicitly set to its surface value at k=0. (<span style="font-family: monospace;">check_parameters</span>)<br><br>Introduce<span style="font-family: monospace;"> prefix_chr</span> to ensure unique dvrp_file path.<br><br>small bugfixes for <span style="font-family: monospace;">user_interface</span> sample code (comments):<br>- initialize ustvst with 0.0 as it is now computed only until <span style="font-family: monospace;">nxr</span> and <span style="font-family: monospace;">nyn</span><br>- two ALLOCATE statements moved from <span style="font-family: monospace;">user_read_restart_data</span> back to <span style="font-family: monospace;">user_init</span><br>- remove 'READ (13) u2_av' statement in <span style="font-family: monospace;">user_read_restart_data</span><br><br>Bugfix: remove IF statement in <span style="font-family: monospace;">plant_canopy_model_ij</span>.<br><br>Bugfix: divide sums(k,8) (e) and sums(k,34) (e*) by ngp_2dh_s_inner(k,sr) (like other scalars). (<span style="font-family: monospace;">flow_statistics</span>)<br><br>Bugfix: <span style="font-family: monospace;">dopr_time_count</span>
     2430was written on the binary file, which caused that NetCDF files newly
     2431created by restart files (no append of existing files!) contained
     2432uneccessary time levels. (<span style="font-family: monospace;">read_3d_binary</span>, <span style="font-family: monospace;">write_3d_binary</span>)<br><br>Bugfix: extra '*' removed in user_statistics sample code. (<span style="font-family: monospace;">user_interface</span>)<br><br>Bugfix: a stop command was missing in some cases of the parallel branch. (<span style="font-family: monospace;">local_stop</span>)<br><br>Bugfix in volume flow control for non-cyclic boundary conditions. (<span style="font-family: monospace;">pres</span>)<br><br>Bugfix: misplaced #endif directives. (<span style="font-family: monospace;">combine_plot_fields</span>)<br><br>Data are collected from PE0 in an ordered sequence which seems to avoid hanging of processes on SGI-ICE. (<span style="font-family: monospace;">cpu_statistics</span>)</td><td style="vertical-align: top;">calc_spectra,
     2433check_parameters, cpu_statistics, diffusion_s, flow_statistics,
     2434init_3d_model, init_dvrp, local_stop, plant_canopy_model, poismg,
     2435prandtl_fluxes, pres, read_3d_binary, user_interface, wall_fluxes,
     2436write_3d_binary<br></td></tr>
    23542437</tbody>
    23552438</table>&nbsp;<b><blink>Attention:</blink></b>
Note: See TracChangeset for help on using the changeset viewer.