Radar or Satellite - Which is more accurate?

[Xihuitl]

Hi

 

I'm developing a forecasting app that plots anticipated precipitation.  I'm trying to check the plots against images online from the MetOffice (radar) and AccuWeather (satellite).  They give quite different results.  Just take a look at Scotland on the satellite image (purple) and how close to the Pennines the rain comes.  Why so different?  I thought that it either rained in a particular spot, or it didn't.  The clouds match, so that's good...

 

So which do I go with?  My calculations are somewhere in between, but tending towards the satellite.

 

And if I were dealing in weather derivatives, whose data would determine payouts?

 

Many thanks

 

 

[BornFromTheVoid]

Hi

 

I'm developing a forecasting app that plots anticipated precipitation.  I'm trying to check the plots against images online from the MetOffice (radar) and AccuWeather (satellite).  They give quite different results.  Just take a look at Scotland on the satellite image (purple) and how close to the Pennines the rain comes.  Why so different?  I thought that it either rained in a particular spot, or it didn't.  The clouds match, so that's good...

 

So which do I go with?  My calculations are somewhere in between, but tending towards the satellite.

 

And if I were dealing in weather derivatives, whose data would determine payouts?

 

Many thanks

 

I can only guess that the satellite based estimates are derived from geostationary satellites, which aren't specifically designed for capturing rainfall data. In a sense, many of the satellites cannot really "see" the rain, but try to derive rainfall rates from things like cloud temperature, optical thickness (from IR and optical imagery) and then integrate that data into some kind of numerical model. This would be of use in remote regions where there aren't enough ground based data.

 

With Doppler radar, it's a more direct measurement of the actual rainfall and intensity, and thus I'd imagine is much more accurate. 

[knocker]

I can only guess that the satellite based estimates are derived from geostationary satellites, which aren't specifically designed for capturing rainfall data. In a sense, many of the satellites cannot really "see" the rain, but try to derive rainfall rates from things like cloud temperature, optical thickness (from IR and optical imagery) and then integrate that data into some kind of numerical model. This would be of use in remote regions where there aren't enough ground based data.

 

With Doppler radar, it's a more direct measurement of the actual rainfall and intensity, and thus I'd imagine is much more accurate. 

 

I agree. The doppler radar would definitely be more accurate.