NASA, Michigan Tech working together on snow study


MARQUETTE COUNTY (Courtesy MTU) – In English there is only one word for snow, but it’s a well-known fact there are different types of snow. Especially in Michigan’s Upper Peninsula. And because snow is also big business in the UP, it warrants study and measurement to better measure accumulation as it’s snowing.

Mark Kulie, assistant professor in the Department of Geological and Mining Engineering and Sciences at Michigan Technological University, is deploying high tech precipitation gauges in Marquette County this week to improve satellite-based and ground-based radar quantitative snowfall estimates.

Kulie is collaborating with NASA Global Precipitation Measurement (GPM) mission researchers, the Marquette National Weather Service Weather Forecast Office and researchers at the University of Wisconsin-Madison. NASA is supplying the instrumentation, while Michigan Tech is responsible for selecting deployment sites as well as deploying and maintaining the instruments.

Kulie will be in Marquette County tomorrow, Nov. 14, through Friday, Nov. 17, testing and deploying a network of up to 12 precipitation gauges within a 15-mile “footprint” or spatial region. The precipitation gauges—called Pluvios—are weighing instruments, essentially high-tech buckets that continually collect and weigh precipitation and archive the data. A combination of Marquette County citizen scientists, businesses and local government facilities have volunteered to help by allowing the precipitation gauges to be installed on their premises.

Kulie, a GPM science team member, said a typical satellite “footprint” is about five to 15 kilometers for the GPM mission, which is helping scientists globally map rain and snow around the world and in turn improve forecasting of extreme weather events that may cause natural disasters.

Installing the instruments around Marquette County is also called “ground truthing”; scientists make measurements by satellite and from the ground and compare the results against each other to make sure the satellites are accurate.   

“We want to exhaustively study the spatial variability of snowfall within a typical satellite footprint to improve GPM’s snowfall estimates,” Kulie says.

Marquette County is the perfect place to do this work for a number of reasons, not least of which is that it on average receives a lot of snowfall. The local terrain enhances the snowfall, and the county receives a variety of different types of snowfall, including lake-effect, orographic, synoptic/frontal and lake- or orographically enhanced synoptic/frontal.

Lake-effect snow occurs when cold air passes over a large body of water that is warmer than the air, creating cloud structures capable of producing large volumes of heavy snow. Orographic snow occurs when moist air is lifted as it moves over hills and mountains; Marquette’s terrain causes the air to rise and most of the precipitation falls on the windward side of ridges.  Synoptic/frontal or “system” snow is associated with low-pressure weather systems with linked warm fronts, cold fronts or upper-level atmospheric disturbances. System snow can also be enhanced by interactions with large lakes and terrain.  

Lake-effect snow typically exhibits large spatial variability, which means large amounts of snow can fall in one isolated area, while adjacent areas receive very little snow.

The GPM science team additionally has an existing collaboration with the Marquette National Weather Service after installing the Micro Rain Radar and Precipitation Imaging Package ground instruments in 2014 to study snowfall.

“Our work really won’t say anything about what type of winter will occur in Marquette County—we’re passive observers of the weather and hope that it will be an active winter,” Kulie says. “But this ground-based snowfall measurement dataset will improve GPM snowfall estimates in not only our region, but around the world.”

Copyright 2019 Nexstar Broadcasting, Inc. All rights reserved. This material may not be published, broadcast, rewritten, or redistributed.