<?xml version="1.0" encoding="UTF-8" standalone="yes"?><oembed><version><![CDATA[1.0]]></version><provider_name><![CDATA[Real Science]]></provider_name><provider_url><![CDATA[https://stevengoddard.wordpress.com]]></provider_url><author_name><![CDATA[stevengoddard]]></author_name><author_url><![CDATA[https://stevengoddard.wordpress.com/author/stevengoddard/]]></author_url><title><![CDATA[Models Are Useless For Six Hours, But Perfect For 50&nbsp;Years]]></title><type><![CDATA[link]]></type><html><![CDATA[<p>E-Mail from Andy DC</p>
<blockquote>
<div>Hi Steve,</div>
<div>I imagine you heard about the latest DC area snow fiasco.  They are already dubbing it &#8220;Snowquester&#8221;.</div>
<div>Just before the &#8220;storm&#8221; began, the weather models (about 10 of them) were predicting 8-12&#8243; of snow for downtown DC.  Even the most &#8220;bearish&#8221; model predicted 5&#8243;.  Same models likewise predicted 12-18&#8243; for western suburbs with the most bearish predicting 10&#8243;.  They shut down the government, all the schools, etc.</div>
<div>The net result was nothing in downtown DC, an inch of slush where I live in Greenbelt and only 2-3&#8243; in normally snowy western suburbs like Germantown and Dulles Airport.</div>
<div>So we are supposed to believe that similar models that can&#8217;t make an accurate 6 hr forecast can make a 50 year forecast?  Give me a break!</div>
<div>Andy</div>
</blockquote>
]]></html></oembed>