As you may know by now, the WSR-88D radar at NWS Milwaukee/Sullivan was recently upgraded to use new dual-polarization (dual-pol) technology. Dual-pol's primary benefits are in distinguishing between different types of precipitation targets. This allows us to determine where in a thunderstorm hail is forming and also whether snow is melting before reaching the ground. However, there are additional benefits to dual-pol. Some are already known, but since the technology is still in the deployment phase, there will be additional benefits discovered over the next few years as more dual-pol radar sites come online and more types of events are observed.
On April 6, 2012, with sunny and dry weather across southern Wisconsin providing for good conditions, several controlled burns were performed in southern Waukesha and northwestern Walworth counties. These burns occurred in relatively close proximity to our radar. Prior to having dual-pol, we were able to observe smoke plumes on radar imagery at close ranges, but we did not see much distinction between the echoes caused by the smoke and those from other phenomena (precipitation, ground clutter, etc.). We once had to rely entirely on environmental clues to determine, for example, whether a suddenly appearing echo was a cloud that might grow into a rain shower, or simply a smoke plume. Dual-pol gives us more information that helps tell us what the radar "sees," instead of just telling us that it "sees" something. Take a look at the image below (and click on the image to load an animation, about 4 MB).
Clockwise from top left, the four panes show individual radar products base reflectivity (Z), base velocity (V), correlation coefficient (CC), and specific differential phase (KDP) from 1:11pm Friday, April 6. These four images show different ways of looking at the same portion of the atmosphere, within 10 to 20 miles of the radar site. The latter two products we did not have before dual-pol was installed at our site. The appearance of the light blue and green echo just to the right of center in the reflectivity image is similar in some ways to a traditional precipitation echo. Velocity is not of much help in this case since this particular echo is moving more or less with everything else "seen" by the radar. However, note the appearance of the two dual-pol products. The echo is distinctly different on the CC and KDP products. The large blue area in the CC image stands out from everything else around it.
CC indicates how similar the targets (particles) "seen" by the radar are within the area sampled by the radar beam. If a wide range of shapes and sizes of particles are present in one area, then the radar will indicate lower values of CC there than if all of the particles are the same shape/size. Pure rain or snow has a very high value of CC, close to a value of 1 and shown as a smooth area of purple. Even diverse mixes of raindrop sizes and even hailstones still have CC values above 0.9. On the other hand, smoke contains a wide array of particles, everything from microscopic droplets of water or oil up to fluttering flakes of ash large enough to be seen with the naked eye. These particles behave quite differently in the atmosphere. Because of this, smoke has relatively low values of CC (below about 0.6) which are shown in deep blue shades. There is no doubt that this is not a meteorological target, that is, not precipitation.
KDP has particular use in discerning heavy rainfall from light rain, hail or snow. Where CC values are low, KDP values are unreliable (prone to inaccuracy). Therefore, the radar automatically "blacks out" KDP in areas of low CC. This explains the lack of KDP for the smoke plumes.
The animation shows plumes from three different fires/burns. The biggest one is north of Mukwonago, while another is south of Mukwonago. A third plume of uncertain origin is visible as well. The smoke rising away from the fires is being carried by the steady easterly winds, moving westward.
National Weather Service Milwaukee/Sullivan