<html><head>
<meta http-equiv="Content-Type" content="text/html; charset=Windows-1252"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; color: rgb(0, 0, 0); font-size: 14px; font-family: Calibri, sans-serif; "><div>We have been having some growing pains in migrating from 3.4.1 to 3.5.</div><div><br></div><div>We have preprocessed scripts that convert our UNIDATA LDM Metars and RAOBS into Little_R format for use in 3DVar and Obs Nudging (winds are set with the 127 QC flat) and we then process it with the RT_fdda_reformat_obsnud.pl for nudging… </div><div><br></div><div>We were good with this option for our realtime WRF until 3.5.</div><div><br></div><div>When using WRF 3.5 using obs nudging we were getting the following error…</div><div><br></div><div><div>Timing for main: time 2013-05-13_12:02:00 on domain 1: 1.89677 elapsed seconds</div><div> OBS NUDGING: Reading new obs for time window TBACK = 47.367 TFORWD = 48.700 for grid = 1</div><div> WARNING: NOBS = 999999 IS GREATER THAN NIOBF = 999999: INCREASE PARAMETER NIOBF</div><div> -------------- FATAL CALLED ---------------</div><div> FATAL CALLED FROM FILE: wrf_fddaobs_in.b LINE: 1144</div><div> wrf_fddaobs_in: in4dob STOP 122</div><div> -------------------------------------------</div><div>application called MPI_Abort(MPI_COMM_WORLD, 1) - process 0</div></div><div><br></div><div>(This error occurs in both serial and dmpar mode and the total number of obs in our OBS_DOMAIN files no where approaches 999999 nor does the error disappear when we change the appropriate field in our namelist.input file.</div><div><br></div><div>In comparing 3.4.1 and 3.5's versions of wrf_fddaobs_in.F, we found the following difference:</div><div><br></div><div>WRF 3.4.1 (circa line 72)</div><div><br></div><div><div> IF( (ktau.GT.krest.AND.MOD(ktau,infreq).EQ.0) &</div><div> .OR.(ktau.EQ.krest) ) then</div><div>!</div></div><div><br></div><div><br></div><div>WRF 3.5 (circa line 72)</div><div><br></div><div><div> IF( (ktau.GT.krest.AND.MOD(ktau,infreq).EQ.0) &</div><div> .OR.(ktau.EQ.krest) .AND. grid%xtime <= grid%fdda_end ) then</div></div><div><br></div><div><br></div><div>When I remove the additional criteria – a moot point since in our current application that demands nudging we want to nudge the model for as long as we have obs with which to nudge it (when we run it in real time this is about the first three hours of the simulation) WRF runs happily onward as it did with 3.4.1.</div><div><br></div><div>So the question is can we keep it as such for now or is there something "new" we need to do with our OBS_DOMAIN files to prevent this from happening with the native version of 3.5?</div><div><br></div><div>Thanks Much</div><div><div><div><font class="Apple-style-span" face="Courier New" size="3"><span class="Apple-style-span" style="font-size: 13px; "><span class="Apple-style-span" style="font-size: 12px; font-family: Consolas, monospace; "><div><div>================================================================</div><div>Bill Capehart <William.Capehart@sdsmt.edu> Associate Professor</div><div>Department of Atmospheric Sciences AES PhD Coordinator </div><div>213 Mineral Industries Building Skype/EVO: wcapehart</div><div>South Dakota School of Mines and Technology Ph: +1-605-394-1994</div><div>501 East Saint Joseph Street Fax: +1-605-394-6061</div><div>Rapid City, SD 57701-3995 Mobile: +1-605-484-5692</div><div>=================== http://capehart.sdsmt.edu ==================</div></div><div><br></div></span></span></font></div></div></div></body></html>