Re: weather blog and ListStreamPlot sampling
- To: mathgroup at smc.vnet.net
- Subject: [mg96296] Re: weather blog and ListStreamPlot sampling
- From: P_ter <petervansummeren at gmail.com>
- Date: Wed, 11 Feb 2009 05:19:07 -0500 (EST)
Hello Sjoerd,
I am also very much interested in the blog about weatherdata, especially for the Netherlands.
I used FullPolygon because the Polygon missed to much:
netherlands = ((CountryData[#, {"FullPolygon", "Mercator"}] & /@ {"Netherlands"}) /. Polygon -> Line);
The Dutch cities are not all tabled with their coordinates:
nedsteden = CityData[{All, "Netherlands"}];
nedstedencoords = {#, CityData[#, "Coordinates"]} & /@ nedsteden ;
rr = DeleteCases[nedstedencoords, {{_, _, _}, {_Missing, _Missing}}] ;
With the cities which do have coordinates, I want to know where the weather stations are within shortest reach of the Dutch cities. Some of them are in Belgium, Germany.
dutchstations = WeatherData[#] & /@ rr[[All, 2]];
Not all the weather stations do have a temperature record:
data = {#, WeatherData[#, "Temperature"]} & /@ dutchstations;
data1 = DeleteCases[data, {_, _Missing}];
data2 = {WeatherData[#[[1]], "Coordinates"], #[[2]]} & /@ data1;
( I splitted in smallest steps)
The weatherstations with available temperature and coordinates are now for use:
Graphics[{Gray, CountryData["Netherlands", "FullPolygon"], {PointSize[Large], Hue[WeatherData[#, "Temperature"]/30],Tooltip[Point[Reverse[WeatherData[#, "Coordinates"]]], WeatherData[#]]} & /@ ds}]
To make your first point: yes, the weatherstations are not uniformly spread. How to make a nice plot of the temperature, which we can see every day on the Royal Dutch Meteorological Institute? It is certainly not that simple for a small country.
with friendly greetings,
P_ter