tag:blogger.com,1999:blog-7729093380675162051.post5797753482191772927..comments2024-03-28T13:56:47.604+11:00Comments on moyhu: GHCN V4 Monthly temperature data displayed on an active sphereNick Stokeshttp://www.blogger.com/profile/06377413236983002873noreply@blogger.comBlogger12125tag:blogger.com,1999:blog-7729093380675162051.post-12988871572539069712019-06-10T23:29:35.360+10:002019-06-10T23:29:35.360+10:00That's funny since the global temperature vari...That's funny since the global temperature variability observed can be reasonably well modeled by observations from a single location at a met station in Australia.pphttps://www.blogger.com/profile/15737287219806254245noreply@blogger.comtag:blogger.com,1999:blog-7729093380675162051.post-82164477477765566482019-06-10T13:30:48.793+10:002019-06-10T13:30:48.793+10:00Well Willis that's why you check the answer ag...Well Willis that's why you check the answer against data that has more stations.stevenhttps://www.blogger.com/profile/06920897530071011399noreply@blogger.comtag:blogger.com,1999:blog-7729093380675162051.post-58106061490046288802019-06-01T11:35:08.277+10:002019-06-01T11:35:08.277+10:00Thanks, Bryan.
On the Arctic, it's true that ...Thanks, Bryan. <br />On the Arctic, it's true that one station, OSTROV_KOTELNYJ, carries a lot of weight. Of the pentagon shape covering most of that ocean, it counts for about a third in the mesh method of TempLS. It can't be more, because each triangle weights its nodes equally for that area. In the <a href="https://moyhu.blogspot.com/2019/04/new-methods-of-integration-in-templs-v4.html" rel="nofollow">LOESS method</a>, it gets much less, because the stations further south also get quite a lot of weight. Then it becomes important that those stations are in agreement, as they are for that region. That doesn't certify that it's right for the N Pole, but it does safeguard against an outlier reading at the weighted location.<br />Nick Stokeshttps://www.blogger.com/profile/06377413236983002873noreply@blogger.comtag:blogger.com,1999:blog-7729093380675162051.post-3049326992039830092019-06-01T11:01:11.212+10:002019-06-01T11:01:11.212+10:00Same with much of the Arctic influenced by one sta...Same with much of the Arctic influenced by one station. I agree on the excellent work, great for visualizing the temperature data. Too bad most of the land masses don't look like the US.Bryan - oz4casterhttps://www.blogger.com/profile/18027990322659101002noreply@blogger.comtag:blogger.com,1999:blog-7729093380675162051.post-43847979860628210142019-06-01T11:00:02.473+10:002019-06-01T11:00:02.473+10:00Thanks, JohnThanks, JohnNick Stokeshttps://www.blogger.com/profile/06377413236983002873noreply@blogger.comtag:blogger.com,1999:blog-7729093380675162051.post-35623781015404761472019-06-01T10:59:39.888+10:002019-06-01T10:59:39.888+10:00Thanks, Willis
Yes, Central Africa is a problem.Thanks, Willis<br />Yes, Central Africa is a problem.Nick Stokeshttps://www.blogger.com/profile/06377413236983002873noreply@blogger.comtag:blogger.com,1999:blog-7729093380675162051.post-32034624861755075482019-06-01T10:28:06.062+10:002019-06-01T10:28:06.062+10:00Nicely done.Nicely done.John Palkovichttps://www.blogger.com/profile/08839241538045673154noreply@blogger.comtag:blogger.com,1999:blog-7729093380675162051.post-78009423201155815522019-06-01T10:13:30.792+10:002019-06-01T10:13:30.792+10:00Excellent work, thoughtfully done and well display...Excellent work, thoughtfully done and well displayed. <br /><br />I did have to laugh at estimating about a third of Africa based on one temperature station ...<br /><br />w.Willis Eschenbachhttps://www.blogger.com/profile/14276840691598976175noreply@blogger.comtag:blogger.com,1999:blog-7729093380675162051.post-24881762217988649172019-06-01T05:50:49.546+10:002019-06-01T05:50:49.546+10:00Thanks, StevenThanks, StevenNick Stokeshttps://www.blogger.com/profile/06377413236983002873noreply@blogger.comtag:blogger.com,1999:blog-7729093380675162051.post-51031736640937987022019-05-31T21:49:35.963+10:002019-05-31T21:49:35.963+10:00Nice work as alwaysNice work as alwaysstevenhttps://www.blogger.com/profile/06920897530071011399noreply@blogger.comtag:blogger.com,1999:blog-7729093380675162051.post-56214038509158462002019-05-31T21:07:39.241+10:002019-05-31T21:07:39.241+10:00William,
It's a convex hull mesh. At the equat...William,<br />It's a convex hull mesh. At the equator, four SST points can be coplanar, so the mesh can validly connect them in two ways, and it seems to alternate. The roaring forties are due to my SST culling policy, described <a href="https://moyhu.blogspot.com/2017/03/making-even-sst-mesh-on-globe.html" rel="nofollow">here</a>. I try to thin out nodes as the longitudes converge, to retain reasonably well-shaped triangles. But it does mess up the regularity of the mesh in transition regions.<br /><br />The meshes come from convhulln() in the R gometry package.<br />Nick Stokeshttps://www.blogger.com/profile/06377413236983002873noreply@blogger.comtag:blogger.com,1999:blog-7729093380675162051.post-6284708964183930062019-05-31T20:19:10.017+10:002019-05-31T20:19:10.017+10:00Nice. What are the meshing artifacts on the equato...Nice. What are the meshing artifacts on the equator and ?75? N/S?William M. Connolleyhttps://www.blogger.com/profile/05836299130680534926noreply@blogger.com