Doppler Weather Radar: When Storms Aren't Storms
(Published August 2014)
Doppler Weather Radar
Doppler weather radar has a lot to offer a weather-savvy boater. Before leaving the security of your dock, a quick look at the wide array of radar products offered by the National Weather Service (NWS) can help you determine if thunderstorms are occurring nearby. With a little knowledge and practice, you can use radar to identify where the strongest storms are located, the speed and the direction in which they are moving, and if they possess rotation which raises the possibility of a tornado or waterspout. But a quick look doesn’t always tell the full story.
An Evening Sail
I was contemplating an evening sail in western Lake Erie and decided to check the radar before heading to the marina (figure 1). The radar showed a continuous line of storms stretching along the northern shore of Lake Erie, from the mouth of the Detroit River all the way to Long Point. The green areas indicated light rain, while the yellow and red areas suggested that thunderstorms were occurring over the area.
However, something about the radar image didn’t seem quite right. It was odd that the storms neatly conformed to the shape of the coast, and only occasionally extended over the lake. A review of the animated radar loop indicated that the storms had been stationary for more than 30 minutes. In such situations, it’s helpful to view the area from another perspective, so I switched to the NWS radar station in Detroit, MI.
The radar image from Detroit didn’t show the slightest hint of rain or storms along the southern shore of Ontario (figure 2). Not knowing which station to believe, I checked the most recent visible satellite image of Lake Erie (figure 3) and it supported the radar image from Detroit – no storms or precipitation were occurring along Lake Erie’s northern shore.
Radar Isn’t Perfect
Why did the Cleveland radar station show rain over southern Ontario? A little background on how radar works, and a review of the atmosphere above Lake Erie, will help solve the mystery.
As a radar station’s antennae spins, it emits a stream of electromagnetic pulses that travel outward from the station at the speed of light. This stream of pulses, or beam, is aimed slightly above the surface of the Earth in order to sample the lowest part of the atmosphere while avoiding the buildings, trees, and other obstructions surrounding the station. Some of these pulses hit objects, such as dust, birds, raindrops, hail, snow, etc., and bounce back to the radar station—a process called backscattering. Backscattered pulses contain a wealth of information, and radar stations are designed to collect and analyze this valuable data.
When returned pulses are captured, the station extracts the data related to the object responsible for the backscattering. This data includes the object’s size, distance and direction from the station, altitude, direction of motion (if it’s moving), and general characteristics such as shape and physical composition. For example, radar stations can distinguish between dust and a raindrop, a raindrop and a hailstone, or a hailstone and a snowflake.
The solution to our mystery lies in understanding how radar beams behave as they travel through the atmosphere. As the distance from the radar station increases, the radar beam tends to widen and gently curve down toward the Earth (figure 4). The curvature of the beam is influenced by the temperature and moisture characteristics of the atmospheric column it encounters on its journey.
These atmospheric properties aren’t known when the beam is transmitted, therefore the exact path it will take is also unknown. To compensate for this shortcoming, meteorologists assume that the air temperature and moisture content of the atmosphere the beam encounters will slowly and steadily decrease as altitude increases.
The temperature and moisture characteristics of the atmosphere don’t always correspond to this assumption, therefore radar beams often behave unpredictably. In some circumstances the beam curves less than expected (subrefraction) while at other times it curves more than expected (superrefraction). The radar station cannot determine the atmospheric properties the beam is encountering, and therefore does not compensate for the occurrence of subrefraction, refraction, or superrefraction.
Radar Beam Ducting
In unusual circumstances, the beam curves abruptly toward the ground in a process called ducting (click here for image). Because the speed of a radar pulse is known and constant (the speed of light), the elapsed time between transmission and the backscattered pulse’s return to the station can be used to determine the distance the pulse has travelled. When ducting occurs, the beam actually travels a much shorter distance than that calculated by the radar station based on the duration of the pulse’s round trip.
An analysis of the atmosphere above Lake Erie on this particular evening suggests that ducting of the radar beam was responsible for the appearance of the erroneous storms and precipitation on the Cleveland radar image. Figure 5 shows the air temperature with increasing height above Lake Erie, and indicates a weak temperature inversion between 330 and 555 meters above the surface (annotated image). A temperature inversion is an atmospheric condition in which the temperature in an air column warms instead of cools with increased height.
The relative humidity in the air column above Lake Erie is shown in figure 6, and indicates a thick layer of very low relative humidity (dry air) extending from 330 to 1,503 meters. Neither a temperature inversion nor a layer of dry air existed in the air column above the radar station.
It’s clear that the radar beam from the Cleveland radar site was abruptly deflected toward the surface of the lake as it encountered the combination of the weak temperature inversion and a thick layer of very dry air over the water. As the pulses hit the surface of the lake, they were very effectively backscattered to the radar station. Based upon the duration of the roundtrip, the radar station presumed that strong returns were associated with activity further from the site, and plotted thunderstorms across southern Ontario.
Radar imagery is susceptible to atmospheric anomalies, and it pays to be a skeptical user. If something doesn’t look quite right, or storms aren’t behaving normally, a comparison of data from another radar station or a review of satellite imagery can help identify local aberrations. Consulting more than one resource can increase your understanding of local conditions, and your time on the water.
As I was working on this article, an interesting image was published from data collected from the Cleveland, Ohio radar station (figure 7). A quick check of the Detroit radar station (figure 8) confirmed my suspicion that the activity directly north of Cleveland in southern Ontario was a radar anomaly while the storm southwest of Detroit, MI was quite real.