<div dir="ltr">Very likely that's where the difference comes from. Have you checked the Ferret documentation to see how the EOFs are normalized there? The variance is different as well. How does Ferret compute the singular value decomposition?<div><br></div><div>Again, have you looked at the time series? Do those have a different amplitude?</div><div><br></div><div>In any case for both Ferret and NCL you should be able to reconstruct the whole data set using the EOFs, PCs, and singular values. Have you tried doing that?</div><div><br></div><div>Maria</div></div><div class="gmail_extra"><br><div class="gmail_quote">On Fri, Sep 4, 2015 at 7:49 AM, Vanúcia Schumacher <span dir="ltr"><<a href="mailto:vanucia-schumacher@hotmail.com" target="_blank">vanucia-schumacher@hotmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div><div dir="ltr"><font face="Calibri,sans-serif" color="#000000"><div><span style="font-size:21.3333339691162px">The difference between results generated by NCL and Fortran or ferret is in the amplitude! Using the same data set!</span></div><div><span style="font-size:21.3333339691162px">I wonder if this difference in amplitude (very low values for NCL) is due to normalization inserted in eofunc_wrap function, would it?</span></div><div><span style="font-size:21.3333339691162px"><br></span></div><div><span style="font-size:21.3333339691162px"><br></span></div></font><br><br><div><hr>From: <a href="mailto:maria.gehne@noaa.gov" target="_blank">maria.gehne@noaa.gov</a><br>Date: Thu, 3 Sep 2015 16:55:09 -0600<span class=""><br>Subject: Re: [ncl-talk] eof function<br></span>To: <a href="mailto:vanucia-schumacher@hotmail.com" target="_blank">vanucia-schumacher@hotmail.com</a><br>CC: <a href="mailto:ncl-talk@ucar.edu" target="_blank">ncl-talk@ucar.edu</a><div><div class="h5"><br><br><div dir="ltr">I don't think you attached the figure you meant to attach. <div><br></div><div>EOFs have no fixed normalization. The value of the patterns (EOFs) and principal components (the time series associated with the patterns) depends on how you normalize. The conventions on how to normalize differ between applications.The only value that is fixed coming out of an EOF analysis are the singular values (the percentage of variance of the EOFs).</div><div><br></div><div>Are the spatial patterns the same between Fortran, Ferret and NCL, even if they don't have the same amplitude? What about the time series? If both patterns and time series have a different amplitude maybe that's where the normalization is different.</div><div><br></div><div>For NCL the website for the EOF function states:</div><div><span style="color:rgb(51,51,51);font-family:verdana,sans-serif;font-size:13px;line-height:16px">The returned values are </span><b style="padding:0px;color:rgb(51,51,51);font-family:verdana,sans-serif;font-size:13px;line-height:16px">normalized</b><span style="color:rgb(51,51,51);font-family:verdana,sans-serif;font-size:13px;line-height:16px"> such that the sum of squares for each EOF pattern equals one. To </span><b style="padding:0px;color:rgb(51,51,51);font-family:verdana,sans-serif;font-size:13px;line-height:16px">denormalize</b><span style="color:rgb(51,51,51);font-family:verdana,sans-serif;font-size:13px;line-height:16px"> the returned EOFs multiply by the square root of the associated eigenvalue (</span><i style="padding:0px;color:rgb(51,51,51);font-family:verdana,sans-serif;font-size:13px;line-height:16px">aka</i><span style="color:rgb(51,51,51);font-family:verdana,sans-serif;font-size:13px;line-height:16px">, the singular value).</span><br></div><div> </div><div><br></div><div>Check if the Fortran and Ferret EOF patterns are normalized in the same way, or not.<br></div><div><br></div><div>Hope this helps,</div><div>Maria</div></div><div><br><div>On Thu, Sep 3, 2015 at 2:59 PM, Vanúcia Schumacher <span dir="ltr"><<a href="mailto:vanucia-schumacher@hotmail.com" target="_blank">vanucia-schumacher@hotmail.com</a>></span> wrote:<br><blockquote style="border-left:1px #ccc solid;padding-left:1ex">
<div><div dir="ltr"><font face="Calibri,sans-serif" size="4" style="font-size:16pt" color="#000000"><br></font><span style="font-size:21.3333339691162px">The values of EOF (output) are different, as attached (figure)</span><br><br><div><hr>From: <a href="mailto:vanucia-schumacher@hotmail.com" target="_blank">vanucia-schumacher@hotmail.com</a><br>To: <a href="mailto:shea@ucar.edu" target="_blank">shea@ucar.edu</a><br>Subject: RE: [ncl-talk] eof function<br>Date: Thu, 3 Sep 2015 17:57:45 -0300<br><br>
<div dir="ltr"><font face="Calibri,sans-serif" size="4" style="font-size:16pt" color="#000000">The values of EOF output are different, as attached (figure)<br></font><br><br><div>> Date: Thu, 3 Sep 2015 14:51:19 -0600<br>> Subject: Re: [ncl-talk] eof function<br>> From: <a href="mailto:shea@ucar.edu" target="_blank">shea@ucar.edu</a><br>> To: <a href="mailto:vanucia-schumacher@hotmail.com" target="_blank">vanucia-schumacher@hotmail.com</a><br>> CC: <a href="mailto:ncl-talk@ucar.edu" target="_blank">ncl-talk@ucar.edu</a><span><br>> <br>> Different normalizations by the EOFs?<br>> <br>> On Thu, Sep 3, 2015 at 1:44 PM, Vanúcia Schumacher<br>> <<a href="mailto:vanucia-schumacher@hotmail.com" target="_blank">vanucia-schumacher@hotmail.com</a>> wrote:<br>> > Dear NCL users,<br>> ><br>> > I calculated the EOF for the Pacific region ({-20:20},{140:290}) using the<br>> > function of NCL (eofunc_Wrap) and tested with other programs (ferret and<br>> > fortran ) and the values of EOF using the NCL are very small compared with<br>> > the other tested outputs.<br>> ><br>> > Using Fortran and Ferret - values between 0.6 to -0.2<br>> > Using NCL function - values between 0.14 to -0.10<br>> ><br>> ><br>> > Could someone explain to me why this happens?<br>> ><br>> > _______________________________________________<br>> > ncl-talk mailing list<br>> > <a href="mailto:ncl-talk@ucar.edu" target="_blank">ncl-talk@ucar.edu</a><br>> > List instructions, subscriber options, unsubscribe:<br>> > <a href="http://mailman.ucar.edu/mailman/listinfo/ncl-talk" target="_blank">http://mailman.ucar.edu/mailman/listinfo/ncl-talk</a><br>> ><br></span></div> </div></div> </div></div>
<br>_______________________________________________<br>
ncl-talk mailing list<br>
<a href="mailto:ncl-talk@ucar.edu" target="_blank">ncl-talk@ucar.edu</a><br>
List instructions, subscriber options, unsubscribe:<br>
<a href="http://mailman.ucar.edu/mailman/listinfo/ncl-talk" rel="noreferrer" target="_blank">http://mailman.ucar.edu/mailman/listinfo/ncl-talk</a><br>
<br></blockquote></div><br></div></div></div></div> </div></div>
<br>_______________________________________________<br>
ncl-talk mailing list<br>
<a href="mailto:ncl-talk@ucar.edu">ncl-talk@ucar.edu</a><br>
List instructions, subscriber options, unsubscribe:<br>
<a href="http://mailman.ucar.edu/mailman/listinfo/ncl-talk" rel="noreferrer" target="_blank">http://mailman.ucar.edu/mailman/listinfo/ncl-talk</a><br>
<br></blockquote></div><br></div>