There are actually 9 types of precipitation
Most of us generally think of precipitation in terms of three varieties: rain, snow, and sleet. But for meteorologists and climatologists, the trio is far from adequate. In fact, a team of researchers, including NASA engineers, spent nearly a decade analyzing weather data to refine these categories. After using machine learning to digest the mountains of data that followed, the group presented a total of nine different types of precipitation. As they explained in a study recently published in the journal Scientific advancesthey’re not trying to nitpick – they’re hoping to save lives.
It’s understandable to think that snow only enters a forecast when the temperature drops below freezing, but that’s actually not the case for meteorologists. According to the microphysics of a cloud, rain and snow are equally likely to occur whenever the temperature varies between 26.6 and 41 degrees Fahrenheit. This is one of many reasons why today’s most reliable weather models can struggle to predict precipitation. Meanwhile, satellite systems are good at tracking cloud movement from space, but they are not as strong at monitoring conditions on the ground.
In an effort to improve and strengthen the available data used in weather models, researchers from the University of Michigan have partnered with NASA on this multi-year project. To begin, they installed a specially designed NASA camera array called the Precipitation Imaging Package (PIP) at seven strategic sites across the United States, Canada and Europe. Once running, PIP recorded all surrounding precipitation with a brightly illuminated high-speed camera, while an instrument known as a disdrometer measured the speed and size distribution of liquid particles as they fell from the clouds.
After nine years, researchers have amassed about 1.5 million small-scale particle measurements, as well as measurements from the surface weather station, including temperature, dew point, relative humidity, pressure and wind speed. A basic calculator can’t squeeze all this information together, so the researchers relied on a statistical method called dimensionality reduction to simplify their data to identify all the patterns. They then built two machine learning models based on this technique – a conventional linear version measuring direct particle relationships and a nonlinear option that considered conditional relationships as the more subtle ways that particles interact and move.
After comparing the two models with independent weather data, the nonlinear method proved the clear winner. Not only did it track precipitation transitions in alignment with the radar data, but it also reduced ambiguity by 36% compared to the linear approach.
After some final computational touches, the team revealed its Uniform Manifold Approximation and Projection, or UMAP, system. In addition to reducing data complexity, UMAP highlights three main contributors to the final shape of a precipitation: particle characteristics, intensity and phase. The UMAP also allows a better understanding of how these precipitation types transform into each other.
So what are the nine technical categories looking for this fall and winter? According to the study authors, they are:
- Drizzle– Light, regular precipitation
- Heavy precipitation—Intense precipitation with many small drops
- Transition from rain to light mix—Light hindrances with dense ice pellets
- Provides rain to mix transition– Intense sleep with dense ice pellets
- Light mixed phase—A low volume of slushy and partially frozen particles
- In heavy mixed phase—A high volume of slushy and partially frozen particles
- Provides snow-mix transition– snowflakes and aggregation particles
- Light snow liner– Fluffy snow sheet
- Snowfall furs– an intense and heavy snowstorm
For University of Michigan climate scientist and study co-author Claire Pettersen, the benefits of UMAP are both immediate and far-reaching.
“In the short term, better forecasting can help people adjust their daily commute or prepare for large events like flooding or an ice storm,” she said in a statement. “On longer timescales, it can help predict how snowfall or runoff timing will change freshwater availability for a region.”
Pettersen and his colleagues don’t want their work to be limited to experienced scientists, however. To make the benefits of UMAP more accessible, they also released an interactive terrain to view the data, as well as a public-facing interface that is easier to use for the average weather enthusiast.
“Precipitation processes are very nonlinear. Many things influence precipitation as it falls that affect what we experience on the surface,” Pettersen added.
And for those who Really I want to dive into the results, everything is available on the deep blue data.
Shop Amazon’s First Day Sale
