Published: March 4, 2024 By

Katja FriedrichKatja Friedrich is a professor and associate chair in CU Boulder’s Atmospheric and Oceanic Sciences department. She is known for her work in cloud seeding, a process used to generate precipitation from existing clouds. In 2017, she helped conduct the National Science Foundation-funded project SNOWIE (Seeded and Natural Orographic Wintertime Clouds: the Idaho Experiment), which was the first experiment to accurately measure the amount of snowfall caused by cloud seeding.

How do you best describe cloud seeding? 

Cloud seeding has been around for almost 100 years as a way to get more rain or precipitation out of a cloud. It was first discovered in a lab at MIT in 1946 that something similar to ice’s crystalline structure, like silver iodide, could be put in supercooled liquid to freeze the drops and create ice. People then applied this method to real clouds to generate precipitation. When we seed wintertime orographic clouds, we target clouds that contain supercooled liquid water, which are tiny water droplets that are too light to fall to the ground. After we seed these clouds with silver iodide, the droplets start to freeze into ice particles. These ice particles continue to grow and collect other droplets and ice particles and eventually form snow that is heavy enough to fall to the ground. 

Does it work? 

A problem with cloud seeding has always been showing how much more precipitation it can generate. We know it works because it works in the lab. However, we need to get the seeding material to the area that contains high amounts of supercooled liquid. It’s difficult to know where those areas are in a cloud, because we don’t have good measurements of supercooled liquid, and it’s difficult to fly in those areas because of aircraft icing. When we seed clouds, we often have to rely on numerical models which have a certain level of uncertainty. Also, once we seeded the clouds, we don’t really know how much precipitation a cloud would have produced without seeding. 

The other problem is that nature can be pretty efficient in producing precipitation, but not always. That’s why with our SNOWIE experiment in 2017 we wanted to gather enough information to run more accurate numerical models. Our idea was that because the models are now accurate enough to reproduce what’s going on in the cloud during cloud seeding, we could then run simulations with and without cloud seeding and see the precipitation produced for both. In SNOWIE, we were also able to show with our seeding line observations the entire chain of events from once we put the silver iodide into the cloud to how much snow we produced. No one had done that before.

The reason we are cloud seeding is because of water scarcity.

How much precipitation can one cloud-seeding event produce? 

We showed that you can produce additional snowfall. Based on our study that included seeding during three days, the total amount of water generated by cloud seeding was about the equivalent of the volume of water needed to fill 50 Olympic-sized swimming pools in 20 minutes over an area of about 7,500 square kilometers.

What are some concerns you hear about cloud seeding? 

The reason we are cloud seeding is because of water scarcity. It is becoming really important to show that we can produce some precipitation. Cloud seeding is not the holy grail if you think about how to generate water or mitigate droughts. But this is an important part because you can maybe produce additional water. I give this example of Lake Mead. Right now, the water levels are so low that hydropower can’t be run at full capacity. If we could cloud seed and raise the water levels just a little bit higher so we can still generate hydropower, this would have massive effects on large populations. 

The downside is putting materials in the atmosphere. Other people say we’re manipulating the weather, which is true. The other argument I say is if you get into your car or are flying on a commercial airplane, you are also manipulating the weather. Every airplane that flies through a cloud of super cold liquid is doing cloud seeding because they’re putting particles in the cloud that can generate snowfall. So people need to be aware that we are manipulating the weather and the climate with everything we are doing.

What is one of the most extreme situations in which you’ve conducted research? 

We are manipulating the weather and the climate with everything we are doing.

I have gone out for one hurricane, Hurricane Ike in 2008, when I had just started working at CU. This was one of the most amazing things that I’ve seen in respect to the weather. We were on these bridges and you saw the water coming in and everything was flooding around us, and we were in what felt like a carwash. We even deployed through the eyewall — for one hour it was totally quiet, and you could hear birds flying. Then came another five or six hours of this carwash feeling. The hurricane passed, and within half an hour you could see how the water trails out. And then I saw emergency boats coming in looking for people. … As for tornadoes, I have to say they look better on TV than in real life.

What else are you working on right now? 

I’m looking at Colorado’s Front Range and other high plateau regions where thunderstorms produce large amounts of hail — so much that we call these hail-accumulating thunderstorms “snowplowable hail.” We built a warning system for the weather service, so they know which thunderstorms are producing a lot of hail that will be dumped on the ground. But also we are trying to understand why that happens and whether there is a way we can forecast it perhaps an hour ahead of time so we can coordinate resources like snow plows, which aren’t always readily available in the summertime.

What do you do outside of work? 

When I’m not working, I like to ski. I like to mountain bike. I have two kids, so we are doing a lot of outdoorsy stuff. We like to camp. We like to travel. That’s what we do — things outside. 

Submit feedback to the editor


Photos by Matt Tyrie