Monday, July 15, 2013

Second Helping Of Radar Goodness

(Note: This post is a continuation of our discussion on radar. If you have not already, you can read the first part here.)

Doppler radar has proven its effectiveness multiple times over just this year by its ability to show forecasters where rotation is present in thunderstorms, indicating that a tornado may soon form.

This image from May 19, 2013 shows the Shawnee, Okla. Tornado, an EF-4, right after it crossed I-40. On the left side if the split screen, base reflectivity is displayed. You will notice a large hook or “C” shape. This indicates that the Rear Flank Downdraft (RFD) has descended and wrapped around the mesocyclone, which essentially guarantees that a tornado is on the ground.

The ball-shaped region of red returns (with even a dot of purple) is a Debris Ball. Remember, Doppler radar does not discriminate rain from anything else, it merely returns how much sound was echoed back by whatever the beam of energy hit. In this case, the tornado had just passed through a populated area, and the radar was picking up all of the debris that was being lofted by the tornado, producing that ball.

On the right side, you see wind velocities. The cool colors (green and blue) are showing wind moving away from the radar, whereas the warm colors (red and yellow) are showing winds moving towards it. Where these meet, a circulation is present.

In this case, you’re seeing winds of around 80 knots in each direction, which combined would indicate that the tornado had wind speeds of at least 160 knots (184 m.p.h.) at the level at which the radar is scanning. (Friction from the ground slows wind, sometimes by more than 20%, which is why it is important to point out that those winds were measured aloft.)

In the next image, which is actually an earlier scan, notice that there is now an area between the blue and yellow returns on the velocity side where there is no color at all. This indicates that winds there were too strong to be recorded by the radar.

On the left, you’ll notice that the Reflectivity image has changed. Certain radar software, like GRLevel3 (which I use), has an option to “smooth” reflectivity images. Most meteorologists prefer to keep “smoothing” turned off – which is what you see here, producing a more pixilated look – but sometimes hook echoes show up better when the image is “smoothed,” as it was in the first image I posted.

While smoothing can sometimes look nice, it actually displays less-specific data than the unfiltered unsmoothed image. (As a side note, when the local news tells you they can show you that it’s raining more heavily on your street than on your neighbor’s, they’re lying. They’re just smoothing the data too much.)

Hurricanes also show up nicely, like Hurricane Alex landfalling in Mexico in late June 2010…

…or Hurricane Earl north of Puerto Rico in August of that same year.

But Doppler radar can be used for so much more than tracking rain and violent storms. By virtue of it constantly scanning, it is always ready to pick something up.

In the image below, a line of thunderstorms on June 14, 2010, coincidentally also over Shawnee, Okla., just went outflow-dominant, producing an Outflow Boundary. On the left, you see a band of light green reflectivity out ahead of the thunderstorms. This is not rain. It is a combination of airborne dirt and bugs that are being blown towards one convergent line that is showing up like light rain.

Notice that on the right side (where velocity is displayed) there is a distinct wind shift that corresponds with this Outflow Boundary.

If you see this line of light reflectivity moving out away from a complex of thunderstorms, it is a sign that the storms are weakening rapidly.

But Outflow isn’t the only kind of boundary that can be picked up on radar. Any type of wind shift line where dirt and bugs are converging can be seen by radar. Below, KDIX is picking up the seabreeze moving west through Philadelphia on April 25, 2011. You can clearly see a line of light reflectivity that corresponds to that area where the wind shifts from being southwesterly (west of the line) to easterly (east of it).

To emphasize, none of this is rain that is falling. Yet, the boundary shows up clearly on radar, like a fingerprint that can only be seen under a blacklight.

And while a cool breeze in summer can be refreshing, we snow weenies hate this next mage: virga.

This image from Chicago’s radar shows an Alberta Clipper moving through the Great Lakes region on January 25, 2013. But notice that there is a circle around the radar where there is no reflectivity. This shows that precipitation is evaporating before it can hit the ground.

Remember that as the radar beam moves away from the source it rises above the ground, so farther from the radar site the beam will be higher. Eventually it will reach a layer where the precipitation has not yet evaporated, hence the circle. Despite it showing precipitation outside of that circle, almost none of it is reaching the ground. (The heavier returns may be.)

A meteorologist can then see what the evaporation level is, and how it's changing over time, by noticing the height AGL (above ground level) of the radar beam when it switches from seeing nothing to seeing something.

And lastly, sometimes radar picks up stuff that really is nothing at all. At night, when the air near the ground cools, it sinks, which brings the radar beam down toward the ground with it. This causes the beam to “bounce off” regular ground objects, like houses and trees, and show up as precipitation.

A similar phenomenon, only more pronounced, is happening in the image below, where a farm of windmills is distorting the radar beam.

You can always tell ground clutter from real precipitation by looping the radar. If the returns “dance,” they’re ground clutter. If they move, they’re real precipitation.