Technology: Method developed to predict onset of strong winds eight to 10 hours in advance
A team of researchers with Pyregence have developed a system for predicting when strong upper-air winds will descend to the surface 8 to 10 hours in advance. Strong wind is the environmental factor that is virtually always present during catastrophic wildfire events that destroy hundreds of structures and put thousands of residents at great risk. Fuel conditions, humidity and topography are also important factors but few fires become fire storms without strong winds. Predicting the onset of a wind event can affect the deployment of fire fighters, the tactics they employ on existing fires and allow better decisions about pre-emptive power shutoffs, community warnings and evacuations. A device called sodar blasts a very loud 91-decibel pulsing beep into the sky, which is then scattered by atmospheric turbulence back to the sodar, allowing profile calculations of wind speed, direction and height.
California experiences its largest and most destructive fires during periods of strong offshore winds, which are known as Diablo winds in Northern California and Santa Ana winds in Southern California. These winds occur as high pressure forces air over the Sierra Nevada Mountains and toward the Pacific Ocean. The winds accelerate through canyons, producing strong gusts, drying out vegetation and sometimes driving fast-spreading wildfires.
To protect Californians from wildfire, the Pyregence Extreme Weather Team is working to improve forecasts of these wind events. The team conducted a pilot test deployment of an upper-air wind profiler called a sodar to measure winds 80 to 600 metres above ground level.
Profilers offer distinct advantages over other data collection methods. Most upper-atmosphere weather data is collected using radiosondes, instruments carried aloft, generally by balloon, two times a day. Profilers, by contrast, gather data two or three times every hour and they also collect more detailed information throughout the lowest levels of the atmosphere, factors that allow for more accurate forecasts.
In 2003, for example, a profiler in New Mexico detected intensifying upper-air winds that had been missed by nearby radiosonde observations. The profiler helped forecasters accurately predict a midnight wind surge, giving fire crews the information they needed to rapidly contain the spread of the fire.
“Sodars have the ability to provide information that you can’t get from other instruments and that are not available in the surface meteorological network,” says Kenneth Craig, a senior atmospheric scientist and meteorologist with Sonoma Technology, an environmental consulting firm that conducted the study for Pyregence.
Detecting descending winds
For the Pyregence pilot test, the sodar system collected data from 25 July 2020 through 26 October 2020, north of Santa Rosa in Northern California.
Although a number of high-wind events occurred at the site during the pilot study period, a Diablo event that developed in late September proved particularly revealing.
Between 15h00 and 16h00 on 25 September, strong winds developed 300 to 600 metres above ground level. Then, just after 01h00am that night, surface wind gusts of about 35mph were recorded. The next day saw a similar pattern: strong winds developed aloft in the mid-afternoon and then gradually descended to the surface around midnight.
Both days, that is, saw high winds develop first in the upper atmosphere and then, about 8 to10 hours later, descend to the surface. That time gap offers a window of opportunity to improve wildfire preparedness, especially during active fire situations.
The ridgetop site for the deployment also proved critical. Two other sodars at lower elevations closer to San Francisco Bay detected the late September event but did not capture the distinctive pattern of strong winds descending from the upper air to the surface. In other words, the data differed dramatically at higher and lower elevations within the same region, highlighting the need for a robust monitoring network with observing stations at carefully chosen locations.
“Even with an instrument that only provides a few hundred metres of vertical range, we could still see very clear evidence of these winds descending toward the surface,” Craig says. “That has powerful implications in terms of planning future instrumentation.”
Better data means improved fire forecasts
The Extreme Weather Team concluded that a state-wide network of strategically placed upper-air profilers could improve short-term forecasts of surface winds and help scientists who model fire behaviour better understand the complex interactions of the atmosphere and wildfire.
How many sodars are needed? Although the scientists who led the study cautioned that they had not conducted a detailed analysis of this issue, they indicated that a relatively small number, perhaps in the range of 10 to 15 sodars carefully positioned across California, could dramatically improve the ability to predict strong winds.
“You don’t have to blanket every geographic area with instruments, there’s always a balance between the cost and the benefit,” Craig says. “But a handful of strategically placed sodars would fill gaps in our observing network and provide valuable information to support situational awareness and forecasting efforts.”