Fun with RADAR
28 September 2004





Over the course of last weekend’s hurricane Jeanne, I set up a couple of scripts to grab all of the radar scans from the area and store them for fun later. My Mom was in Tampa through the storm, and I wanted to have some record to show her what the beast looked like from this side. She said they made it through everything just fine, with not much more than a lot of rain and some heavy wind. Using ImageMagick’s convert function, I combined all of the images into an animated gif. Clicking on the image above will take you there, but beware: the file size is about 12MB.

  1. A few questions... I know next to nothing about how weather RADAR works. On the western side of Florida, it seems like the storm just disappears whenever it gets to that point. Like... as the fingers of the storm swirl around and the whole thing shifts slowly to the west, just off the shore of florida it seems to dissolve. Is this because the storm is actually disppearing, or because the RADAR isn't capable of reading in that particular spot? Secondly, in the middle of the animation, the entire storm disappears for a few frames, and then comes back. Is this a problem with my browser, or is the animation actually that way. If it is that way... did you mess up when you built it, or is this the actual data? If this is the actual data, is this because the RADAR went down or got shut off for some reason? It's hard to believe that the entire storm just disappeared and then came back.
    — Jim    2687 days ago    #
  2. RADAR (RAdio Distance And Ranging) works by sending out very fast pulses of RF energy and listening to the return signals. RADAR [usually] operates at microwave and millimeter-wave frequencies from 5GHz through about 100GHz. The RADAR transmitter uses a scanning antenna with a very narrow beamwidth(<0.5 degrees) in both the horizontal and vertical planes. The pulses are reflected off of objects, but since the RADAR is slightly pointing up (0.5-4.5 degrees), most of the energy goes out into space. On a rainy day, some of this energy reflects off of the rain falling from a storm and is recieved by the RADAR (and is called an 'echo'). By timing how long it takes the signal to travel round-trip, the RADAR can calculate the distance to the reflected object, and its direction (the direction the radar antenna was pointing when it received the echo). Since the time between these pulses is fixed, there is a maximum time that the radar waits to hear an echo before it transmits the next pulse. This waiting time determines the maximum range, along with the transmitter power. The blanking you see on the western edge of the storm is this principle in effect; the storm at this point extended beyond the range of the Tampa RADAR. Also note that this image is a compilation of all of the NWS radars in the region, known as a NEXRAD Composite. Doppler Radar gives more information than that. Since radio energy reflected off of moving objects causes a Doppler Shift to be introduced on the return signal, the RADAR can calculate the echo's frequency shift with respect to the transmitted frequency and thereby calculate the vector radial velocity (the velocity toward or away from the radar site) of the object that reflected the signal. This is called the "Relative Velocity". In practice, few people actually look at Velocity images, but they are available. I'll try to post a velcity signal alongside a reflectivity signal later today. To answer your questions: All of the images you see in the animation are the actual data as grabbed every 5 minutes from the NWS. The blank frames are because, for some reason or another, the radar site didn't upload the data to the central NWS server during those frames. This could have been something as severe as the radar site loosing power, or something as simple as network congestion.
    — justinm    2687 days ago    #