<p data-start="128" data-end="153"></p><span>Dear Community,</span><p data-start="155" data-end="562">I am currently running a simulation based on the <code data-start="204" data-end="217">cpl_aim_ocn</code> configuration. After completing a 10-year integration, I found that the simulation was quite time-consuming. To improve computational efficiency, I attempted to reduce the coupling frequency by changing the <code data-start="425" data-end="441">cpl_atmSendFrq</code> parameter in <code data-start="455" data-end="475">input_atm/data.cpl</code> from 6 hours to 24 hours, so that boundary conditions would be exchanged once per day.</p><p data-start="564" data-end="699">However, when I tried to restart the simulation with this new setting, I encountered the following error in <code data-start="672" data-end="698">rank_0/Coupler.0000.clog</code>:<br></p><div data-start="564" data-end="699" style="margin-top: 14px; margin-bottom: 14px;"><span>CPL_CHECK_CPLCONFIG: Nb of time-steps to run (ATM) = 356400</span><div><span>CPL_CHECK_CPLCONFIG: Nb of time-steps to run (OCN) = 1425600</span></div><div><span>*** ERROR *** CPL_CHECK_CPLCONFIG: wrong Nb of time-steps to run</span></div><div><span>*** ERROR *** CPL_CHECK_CPLCONFIG: inconsistent CPL/ATM/OCN config </span></div><div><span>FATAL ERROR for 1 Proc(s) ==> Stop here</span></div><span></span><br></div><div data-start="564" data-end="699" style="margin-top: 14px; margin-bottom: 14px;"><p data-start="1032" data-end="1233">From the error message, I assume this inconsistency is caused by the change in the frequency at which the atmosphere component sends data to the coupler, leading to a mismatch with the ocean component.</p><p data-start="1235" data-end="1465">I would like to ask:<br data-start="1255" data-end="1258">
If I want to change <code data-start="1278" data-end="1294">cpl_atmSendFrq</code>, is it necessary to restart the simulation from the beginning with a consistent configuration?<br data-start="1389" data-end="1392">
Or is there a way to adjust this parameter mid-run without starting over?</p><p data-start="1467" data-end="1531">Any guidance on how to handle this would be greatly appreciated.</p><p data-start="1533" data-end="1555">Best regards,<br data-start="1546" data-end="1549">
Fisher</p></div>