<div dir="ltr">Thomas --<div><br></div><div>Correlation of each member to the others is part of the first step, and represents the distance of the members from each other. Another part is choosing the forecast variable most relevant for your clustering .... 500 hPa heights (this is a common one), sea level pressure, or some other variable. While a global domain can be used, you may even want to choose a smaller domain over which to cluster, say Europe or North America. </div><div><br></div><div>Once those decisions are made, things get a little fuzzy for me, but I can recommend you read Tracton and Kalnay 1993 (Tracton, M. S., and E. Kalnay, 1993: Operational ensemble prediction
at the National Meteorological Center: Practical aspects.<i>
Wea. Forecasting</i>, <b>8</b>, 379–398) or.Toth et al. (<a href="http://journals.ametsoc.org/doi/pdf/10.1175/MWR-D-11-00016.1" target="_blank">Toth, Z., E. Kalnay, S. M. Tracton, R. Wobus, and J. Irwin, 1997: A synoptic evaluation of the NCEP ensemble. Wea. Forecasting,12, 140–153</a>.) for details of using <b>anomaly</b> correlation as a distance metric in a clustering algorithm.</div><div><br></div><div>Good luck!</div></div><div class="gmail_extra"><br clear="all"><div><div class="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div>Dr. William R. (Bill) Bua<br>UCAR/COMET/NCEP/EMC<br>5830 University Research Court #2784<br>
College Park, MD 20740<br><br></div></div></div></div></div></div>
<br><div class="gmail_quote">On Wed, Feb 17, 2016 at 5:32 PM, Thomas Davies <span dir="ltr"><<a href="mailto:tsd.davies@gmail.com" target="_blank">tsd.davies@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">Thanks for your reply.<div><br></div><div>I looped over each GEFS ensemble member, calculating the correlation with every other member and stored these values like this:</div><div><br></div><div><div><b>p1p2 0.625844</b></div><div><b>p1p3 0.778631</b></div><div><b>p1p4 0.718804</b></div><div><b>p1p5 0.597726</b></div><div><b>p1p6 0.73876</b></div><div><b>p1p7 0.744276</b></div><div><b>p1p8 0.683437</b></div><div><b>p1p9 0.548653</b></div><div><b>p1p10 0.479272</b></div><div><b>p1p11 0.668351</b></div><div><b>p1p12 0.600638</b></div><div><b>p1p13 0.698219</b></div><div><b>p1p14 0.606607</b></div><div><b>p1p15 0.536851</b></div><div><b>p1p16 0.736465</b></div><div><b>p1p17 0.564005</b></div><div><b>p1p18 0.557531</b></div><div><b>p1p19 0.641763</b></div><div><b>p1p20 0.5872</b></div><div><b>p1p21 0.57592</b></div></div><div><b>...</b></div><div><div style="font-weight:bold">p21p1 0.57592</div><div style="font-weight:bold">p21p2 0.623274</div><div style="font-weight:bold">p21p3 0.630096</div><div style="font-weight:bold">p21p4 0.551386</div><div style="font-weight:bold">p21p5 0.561231</div><div style="font-weight:bold">p21p6 0.60267</div><div style="font-weight:bold">p21p7 0.638492</div><div style="font-weight:bold">p21p8 0.646961</div><div style="font-weight:bold">p21p9 0.5132</div><div style="font-weight:bold">p21p10 0.621424</div><div style="font-weight:bold">p21p11 0.684271</div><div style="font-weight:bold">p21p12 0.541928</div><div style="font-weight:bold">p21p13 0.502403</div><div style="font-weight:bold">p21p14 0.671402</div><div style="font-weight:bold">p21p15 0.637574</div><div style="font-weight:bold">p21p16 0.76546</div><div style="font-weight:bold">p21p17 0.564914</div><div style="font-weight:bold">p21p18 0.580842</div><div style="font-weight:bold">p21p19 0.644114</div><div style="font-weight:bold">p21p20 0.599859</div><div style="font-weight:bold"><br></div><div>However I am not sure how I should progress. I have read through the link you provided but couldn't find anything useful, all I know is that I need to firstly correlate between members </div></div></div><div class="HOEnZb"><div class="h5"><div class="gmail_extra"><br><div class="gmail_quote">On Wed, Feb 17, 2016 at 6:02 PM, Stephen McMillan <span dir="ltr"><<a href="mailto:smcmillan@planalytics.com" target="_blank">smcmillan@planalytics.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">Thomas,<div><br></div><div>In the GrADS Documentation Index (<a href="http://cola.gmu.edu/grads/gadoc/gadocindex.html" target="_blank">http://cola.gmu.edu/grads/gadoc/gadocindex.html</a>), under the letter "E" you can find several relevant topics, such as "ensemble dimension" and "eloop." One or more may answer your questions.</div><div><br></div><div>Stephen McMillan</div></div><div class="gmail_extra"><br><div class="gmail_quote"><div><div>On Wed, Feb 17, 2016 at 12:27 PM, Thomas Davies <span dir="ltr"><<a href="mailto:tsd.davies@gmail.com" target="_blank">tsd.davies@gmail.com</a>></span> wrote:<br></div></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div><div>Hi,<div><br></div><div>Does anybody know how to cluster ensemble members showing similar solutions and display? Is there a quick command to do this</div><div><br></div><div>Thanks in advance</div>
<br></div></div>_______________________________________________<br>
gradsusr mailing list<br>
<a href="mailto:gradsusr@gradsusr.org" target="_blank">gradsusr@gradsusr.org</a><br>
<a href="http://gradsusr.org/mailman/listinfo/gradsusr" rel="noreferrer" target="_blank">http://gradsusr.org/mailman/listinfo/gradsusr</a><br>
<br></blockquote></div><br></div>
<br>_______________________________________________<br>
gradsusr mailing list<br>
<a href="mailto:gradsusr@gradsusr.org" target="_blank">gradsusr@gradsusr.org</a><br>
<a href="http://gradsusr.org/mailman/listinfo/gradsusr" rel="noreferrer" target="_blank">http://gradsusr.org/mailman/listinfo/gradsusr</a><br>
<br></blockquote></div><br></div>
</div></div><br>_______________________________________________<br>
gradsusr mailing list<br>
<a href="mailto:gradsusr@gradsusr.org">gradsusr@gradsusr.org</a><br>
<a href="http://gradsusr.org/mailman/listinfo/gradsusr" rel="noreferrer" target="_blank">http://gradsusr.org/mailman/listinfo/gradsusr</a><br>
<br></blockquote></div><br></div>