From ovens Thu May 21 14:01:30 2015 To: Cliff Mass Subject: Google Maps interface MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline X-Mutt-Fcc: =sent-mail Status: RO Date: Thu, 21 May 2015 14:01:30 -0700 Content-Length: 4641 Lines: 132 Hi Cliff, I found what I did before: http://www.atmos.washington.edu/~ovens/javascript/overlay5.html And, I've heard back from the WSU folks. ----- Forwarded message 1 of 3 from farrenthorpe@wsu.edu ---- Date: Thu May 21 13:03:57 2015 From: farrenthorpe@wsu.edu Received: from mx11.s.uw.edu (mx11.s.uw.edu [140.142.234.149]) To: David Ovens , Subject: Re: Airpact Web and graphics code Hi David, I just wanted to mention a couple shortcomings of the method, which can cause complications. 1) The images are overlayed onto a Google Maps frame which inherently uses Mercator-like projection. 2) We generate the images with NCL which reads our netcdf files and plots them in mercator with no borders and use transparency for no data. It is not an exact perfect fit for image overlay onto the map, but for gridded output imagery it's pretty good (the cells look funky, though, in Mercator). 3) The google maps API loads all the imagery for the day that are available on the page, so as we added more overlay options (e.g. variables plotted) onto the page, it increased the load time when someone first visits the page (and the load time when selecting a different forecast date). Multiple URLs could be used for different pieces of WRF output... or perhaps there are other scripting options where people select categories before loading the Google Maps frame. But I wanted to mention this, since there appear to be more WRF outputs plotted than what we do in AIRPACT. 4) We use a lot of KML overlays... which are inherently more straightforward, but many functionalities do not work since the format is really for Google Earth. I have been curious about the possibility of using the Google Earth API, which is embedded into some webpages... but requires users to install Google Earth, which is problematic and not realistic what with all the restrictions on agency users in AIRQUEST. Food for thought. -Farren Jen Hinds is the person to talk to for the Google Maps web code. On 5/21/2015 12:12 PM, David Ovens wrote: > Hello Joe and others, > > Cliff has asked me to look into getting our standard WRF output into a > GoogleMaps interface for the Web. You folks have done some great work > with this for the Airpact site. I am wondering if someone would be > willing to share web code and answer a few questions I'll have about > implementing this type of thing? I'd also be interesting in finding > out how the graphics are generated, what software is used. > > Thanks in advance for any help you can give. > > Cheers, > > David > -- > David Ovens e-mail: ovens@atmos.washington.edu > Research Meteorologist phone: (206) 685-8108 > Dept of Atm. Sciences plan: Real-time MM5 forecasting for the > Box 351640 Pacific Northwest > Seattle, WA 98195 Weather Graphics and Loops > ---- End of forwarded message 1 of 3 ----- ----- Forwarded message 2 of 3 from jvaughan@wsu.edu ---- Date: Thu May 21 13:14:06 2015 From: jvaughan@wsu.edu Received: from mx13.s.uw.edu (mx13.s.uw.edu [140.142.32.138]) To: David Ovens Subject: Re: Airpact Web and graphics code David, Farren has given you a good summary. We could ask Jen Hinds to provide a second level of commentary, if that would help. Or do you want to send a list of specific questions? We're also happy to share example codes/scripts, I'm sure. Jen Hinds has worked for us on our web stuff for a long time. But Jen works at UI and we've had to make special arrangements, recently, to continue having her talents available to us! Joe Joseph K. Vaughan jvaughan@wsu.edu 509-335-2832 Research Associate Professor Laboratory for Atmospheric Research Civil & Environmental Engineering Washington State University Pullman, WA 99164-2910 ---- End of forwarded message 2 of 3 ----- ----- Forwarded message 3 of 3 from serena_chung@wsu.edu ---- Date: Thu May 21 13:39:29 2015 From: serena_chung@wsu.edu Received: from mx13.s.uw.edu (mx13.s.uw.edu [140.142.32.138]) To: "Vaughan, Joseph K" , Subject: Re: Airpact Web and graphics code Hi Dave, Just a little more about NCL, the plotting program we use to create jpeg images (of model results) that overlays on Google Map. There are many example NCL scripts and built-in functions specifically for WRF at see http://ncl.ucar.edu/Applications/wrf.shtml. ( We modified the examples for WRF to get them to work for CMAQ.) I've recently learned to use R to create *.htm of model output with Google Map directly, but haven't had enough time to get the graphics to look as I like them. I also have a very skeleton WRF to KMZ R script. I can provide R scripts if you like; but as I noted, they are very basic at this point and are not what we currently use for the web. Serena ---- End of forwarded message 3 of 3 ----- -- David Ovens e-mail: ovens@atmos.washington.edu Research Meteorologist phone: (206) 685-8108 Dept of Atm. Sciences plan: Real-time MM5 forecasting for the Box 351640 Pacific Northwest University of Washington http://www.atmos.washington.edu/mm5rt Seattle, WA 98195 Weather Graphics and Loops http://www.atmos.washington.edu/~ovens/loops