GFS temperature VS Weatheronline
-
The rendering effect of Weatheronline（https://www.weatheronline.co.uk/cgi-bin/expertcharts?LANG=en&MENU=0000000000&CONT=asie&MODELL=gfs25&MODELLTYP=1&BASE=202410101200&VAR=tmp2&HH=12&ZOOM=1&ARCHIV=0&RES=0&WMO=&PERIOD=） has many details, while Windy rendering is very smooth. Has Windy's data been smoothed
-
@zhaoy
You can change the color scale yourself for most of layers in Windy.
See this article
The color scale in Windy is smoothed but the way how the colors are staggered in your example gives more contrast between the temperature steps.
And as explained in the article you can design a non-smoothed color scale. -
@idefix37 still smooth,There are many small patches on Weatheronlie and connected areas on Windy
-
@zhaoy
Did you read the article I mentioned here above ? At the end I show how to make a non-smoothed color scale if you prefer it. -
@idefix37 There are 38 color classifications in the temperature legend of Weatheronline,I have modified the color according to legend on Windy, but there are still differences between the two images. So I guess Windy's data may have been processed.
-
@zhaoy
Could you show your code. -
@idefix37 I picked up steps with numerical values on color legend, and other steps were interpolated by numerical values
[[245,[136,136,136,255]],[246.5,[176,176,176,255]],[248,[204,204,204,255]],[249.5,[228,228,228,255]],[251,[119,51,119,255]],[253.4,[170,51,170,255]],[255.8,[204,51,204,255]],[258.2,[255,51,255,255]],[260.6,[255,153,255,255]],[263,[21,0,204,255]],[265,[51,102,255,255]],[267,[51,153,255,255]],[269,[51,204,255,255]],[271,[51,255,255,255]],[273,[0,119,0,255]],[275.4,[0,153,0,255]],[277.8,[0,187,0,255]],[280.2,[0,221,0,255]],[282.6,[0,255,0,255]],[285,[255,255,51,255]],[286.75,[255,238,51,255]],[288.5,[255,221,51,255]],[290.25,[255,204,51,255]],[292,[255,187,51,255]],[293.75,[255,170,0,255]],[295.5,[255,153,0,255]],[297.25,[255,119,0,255]],[299,[255,0,0,255]],[301,[238,0,0,255]],[303,[204,0,0,255]],[305,[187,0,0,255]],[307,[170,0,0,255]],[309,[153,0,0,255]],[311,[136,0,0,255]],[313.5,[114,0,0,255]],[316,[91,0,0,255]],[318.5,[66,0,0,255]],[321,[29,0,0,255]]] -
@zhaoy
Thank you.
Could you try[[245,[136,136,136,255]],[247.9,[176,176,176,255]],[248,[204,204,204,255]],[250.9,[228,228,228,255]], [251,[119,51,119,255]], …etc
It should improve the scale, but not sure you get exactly what you want.
-
@idefix37 thank you! the result is not yet which I wanted,I think the same data source (GFS 0.25°) should render the same results,but the Weatheronline image has many small patches,and the boundary is quite sharp.
WeatherOnline Result:
so,I guess Windy did some smoothing on the data, I don't know if it's appropriate.
such as this picture
-
@zhaoy
You are being picky. I find the granularity to be pretty good.
At boundary between the value ranges the limit is jagged. Can you try taking a screenshot of the Icing layer of the same device.
Then could you explain the 2 squares with numbers. How did you make them ? -
The temperature data(GFS 0.5°) render result on meteoblue,which also has many small patches,and the boundary is quite
https://www.meteoblue.com/en/weather/maps/#coords=5.65/37.766/99.822&map=temperature~hourly~GFS05~2 m above gnd~none
-
@idefix37 I'm just guessing if Windy smoothed the source data to make it look better
-
@zhaoy
I didn't realize you didn't apply the correct coding.
As already said, your coding should be like this theoretical diagram
-
@idefix37 I guess it's not a color rendering issue, Did Windy lose a lot of details in smoothing the source data(GFS 0.25°)
[[245,[136,136,136,255]],[246.4,[136,136,136,255]],[246.5,[176,176,176,255]],[247.9,[176,176,176,255]],[248,[204,204,204,255]],[249.4,[204,204,204,255]],[249.5,[228,228,228,255]],[250.9,[228,228,228,255]],[251,[119,51,119,255]],[253.3,[119,51,119,255]],[253.4,[170,51,170,255]],[255.7,[170,51,170,255]],[255.8,[204,51,204,255]],[258.1,[204,51,204,255]],[258.2,[255,51,255,255]],[260.5,[255,51,255,255]],[260.6,[255,153,255,255]],[262.9,[255,153,255,255]],[263,[21,0,204,255]],[264.9,[21,0,204,255]],[265,[51,102,255,255]],[266.9,[51,102,255,255]],[267,[51,153,255,255]],[268.9,[51,153,255,255]],[269,[51,204,255,255]],[270.9,[51,204,255,255]],[271,[51,255,255,255]],[272.9,[51,255,255,255]],[273,[0,119,0,255]],[275.3,[0,119,0,255]],[275.4,[0,153,0,255]],[277.7,[0,153,0,255]],[277.8,[0,187,0,255]],[280.1,[0,187,0,255]],[280.2,[0,221,0,255]],[282.5,[0,221,0,255]],[282.6,[0,255,0,255]],[284.9,[0,255,0,255]],[285,[255,255,51,255]],[286.65,[255,255,51,255]],[286.75,[255,238,51,255]],[288.4,[255,238,51,255]],[288.5,[255,221,51,255]],[290.15,[255,221,51,255]],[290.25,[255,204,51,255]],[291.9,[255,204,51,255]],[292,[255,187,51,255]],[293.65,[255,187,51,255]],[293.75,[255,170,0,255]],[295.4,[255,170,0,255]],[295.5,[255,153,0,255]],[297.15,[255,153,0,255]],[297.25,[255,119,0,255]],[298.9,[255,119,0,255]],[299,[255,0,0,255]],[300.9,[255,0,0,255]],[301,[238,0,0,255]],[302.9,[238,0,0,255]],[303,[204,0,0,255]],[304.9,[204,0,0,255]],[305,[187,0,0,255]],[306.9,[187,0,0,255]],[307,[170,0,0,255]],[308.9,[170,0,0,255]],[309,[153,0,0,255]],[310.9,[153,0,0,255]],[311,[136,0,0,255]],[313.4,[136,0,0,255]],[313.5,[114,0,0,255]],[315.9,[114,0,0,255]],[316,[91,0,0,255]],[318.4,[91,0,0,255]],[318.5,[66,0,0,255]],[320.9,[66,0,0,255]],[321,[29,0,0,255]]]
Windy (GFS 0.25°):
Meteoblue (GFS 0.5°):
WeatherOnline (GFS 0.25°):
-
@zhaoy
You already posted the same screenshots.
You can guess what you want.
End
