Results stored by Delf3D from SWANResults stored by Delf3D from SWANhttps://oss.deltares.nl/c/message_boards/find_thread?p_l_id=1806765&threadId=38792722024-03-29T07:30:43Z2024-03-29T07:30:43ZResults stored by Delf3D from SWANSamantha Matickahttps://oss.deltares.nl/c/message_boards/find_message?p_l_id=1806765&messageId=38792712021-03-09T18:59:32Z2021-03-09T18:58:08ZHello,<br />I'm trying to understand what results are actually written from SWAN, when delft3d runs a SWAN simulation. I've done some tests to compare, and it is still not clear. I'll say what I think I understand, and the tests I ran, and hope that someone can explain the rest!<br /><br />In the *.mdw file I specify <strong>TimePoints</strong>, <strong>TimeStep</strong>, and <strong>COMWriteInterval</strong>. The <u>TimePoints</u> determine when to run and record results from SWAN. For me they are 0, 3, and 6 hours. The <u>TimeStep</u> and <u>COMWriteInterval</u> appear to control the timestep and time interval that each SWAN simulation is run for. I vary these values in my tests, but for example, a TimeStep of 60 min and COMWriteInterval of 180 min produces a SWAN simulation performed from (for TimePoint=0hr) 0 hr to 3 hr in increments of 1 hr, and then (for TimePoint=3h) from 3 hr to 6 hr in increments of 1 hr. <br /><br />Specifically, my question is<strong> 'which SWAN time step calculation is actually stored, per TimePoint simulation that is conducted?'</strong> For example, at TimePoint=0h, is the data that is produced after 1 timestep recorded (1 hour for example above)? at the end of the simulation (3 h for example above)? something else?<br /><br />The problem for me is that in all of my simulations, each SWAN run does not converge until the 3rd timestep taken by SWAN (the 1st and 2nd end around 70% and 80% accuracy after 15 iterations), no matter the size of the timestep taken (30, 60, 180 min) so I would only want delft3D to retain data produced after the 3rd timestep. <br /><br />I ran some tests where I varied both TimeStep and COMWriteInterval to see how the results of significant wave height varied. I found the only way to get the same results between 2 simulations was to have the same TimeStep (COMWriteInterval) did not make a difference. <br /><br />For reference, the tests I ran wereTimeStep (minutes) : COMWriteInterval (minutes) as,<br /><br />30 : 60<br /><br />30 : 180<br /><br />60 : 60<br /><br />60 :180<br /><br />180 : 180<br /><br />Any insights into how data is stored from SWAN would be super helpful. I get results for hour 0, so I thought there must be some sort of assignment of data produced from a later time to the TimePoint of interest. <br /><br />Thank you in advance for any and all help with this <img alt="emoticon" src="https://oss.deltares.nl/o/deltares-fews-theme/images/emoticons/happy.gif" ><br /><br />Best Regards,<br /><br />SamanthaSamantha Maticka2021-03-09T18:58:08Z