« Back


Hello reader! My name is George Duffy! I am a graduate student from the University of Urbana Champaign, and I am excited to begin sharing my experience as a GPM graduate researcher of with you.

My research focuses on snow, or more specifically, snowfall retrieval. Technically, the GPM satellite doesn't measure precipitation, it measures radar echoes from its Dual Frequency Precipitation Radar (DPR) and radiation from it's General Microwave Imager (GMI). It's up to us to develop algorithms that can retrieve precipitation information from these radar images. It's also important to test these algorithms, which is why we regularly conduct GPM Ground Validation (GV) campaigns that directly compare ground and air measurements of precipitation to radar images. There have been a lot of studies to relate satellite radar style measurements to rainfall rates, but snowfall is brand new territory. No satellite has ever tried to measure snowfall with multiple wavelengths of radar the way GPM is doing right now, and not much research has been done on the subject. There has also only been one GV experiment conducted for snowfall, the GPM Cold Season Experiment (GCPEx) in Ontario, 2012. Because this information is so new, there are a lot of challenges and questions important to GPM measurements we are just now beginning to explore.

There is an old saying that "no two snowflakes look the same". While this is probably not the most scientifically accurate statement, it does good enough to explain the challenge we face trying to retrieve snowfall information from GPM measurements.  Wet snow, dry snow, single branching snowflakes, big clusters of flakes or needles, tiny snow spheres called graupel, and even some more exotic particles that can only be found inside of clouds like bullet rosettes, all of these different kinds of snowflakes will reflect radar waves in different ways. There is not a single equation we can apply that can relate snowfall rates to radar echoes of all of these different kinds of clouds and snow storms. Instead, we rely on different regimes of temperature, geographic location, and special algorithms that take advantage of the multiple radar frequencies aboard the GPM satellite to give us a clue as to the behavior of the snowflakes we are looking at. My ultimate goal is to help evaluate and improve these algorithms. Right now, I'm still just analyzing GCPEx data and developing methods to compare radar images to in-cloud snowfall measurements.

Examples of different snow "habits".
Snowflakes photographed by the University of Utah's Multiple Angle Snow Camera (MASC)

Personal info:

I've been interested in weather for as long as I can remember. In grade school I made my own precipitation stations, I got books on weather for Christmas, I used to watch the Weather Channel for hours straight just to watch the hurricane tracks move. Working on this project with NASA is a fulfillment of a nerdy lifelong dream for me, and I love the work that I do. Outside of research, most of my hobbies are related to music: listening, playing, dancing. I like guitar and I’m learning piano, but I'm pretty terrible at both. I’m also kind of obsessed with western movies, and I have a life-size John Wayne in my living room.


This is what I look like:

And this is what my cat looks like. 


That's all for now. Stay tuned!

Previous Next
Please log-in to post comments