The following transcript was generated using automated transcription software for the accessibility and convenience of our audience. While we strive for accuracy, the automated process may introduce errors, omissions, or misinterpretations. This transcript is intended as a helpful companion to the original audio and should not be considered a verbatim record. For the most accurate representation, please refer to the audio recording.
Michael Dunne: I'm Michael Dunne. Here's a riddle for you. What's something that's hard to predict, impossible to control, yet vital to everyone on the planet? It's weather. And for the vast majority of us, we know so little about the science, technology and experts who study weather and try to give us the best information about what it will do. Today on the show, we'll bring you just such an expert: a professor at OSU and the state's chief climatologist. He'll talk about how weather forecasts are made, why they vary greatly, and why Oregon is such a difficult state in which to make weather predictions. As a climatologist, he'll also explain the vital differences between weather and climate and how our changing climate is making weather forecasting even more difficult. Larry O'Neil is the state of Oregon's climatologist and an associate professor at Oregon State University. Larry, always good to talk to you. Thanks so much for coming on.
Larry O'Neil: Thank you for having me.
Dunne: Let's start with a real basic one. How is a weather forecast developed?
O'Neil: Great question. Modern weather forecasts basically start with observations of the weather. That seems pretty simple, but we use a lot of different types of observations. Those come from satellites, radars, ground-based weather stations, weather balloons and buoys out in the ocean. Basically, what we try to do with that is get a snapshot of what the current weather is right now. That is absolutely essential: We need to know what the weather is now before we can say what the forecast will do. Once we have that, modern weather forecasting really utilizes a lot of weather forecast models. Over time, those forecast models have become quite skillful, especially out to about five to seven, maybe even 10 days. What those models do is take all of our physics knowledge of how the atmosphere works. The atmosphere is a fluid. It's filled with air, and we know quite well how it changes its temperature, how it changes its wind and how it produces clouds and precipitation and all those sorts of things. Weather forecast models are absolutely essential to modern weather forecasting.
Dunne: All those diagnostic tools you just described, the buoys, satellite imagery, etc. What happens between when, say, that buoy says the wind speed is 12 miles an hour, and a couple days later? Because obviously weather isn't a linear progression. Things can change in a matter of hours, right?
O'Neil: That's right, and that's why we collect those observations initially: to figure out what the weather is doing at this minute. Based on our vast knowledge of how weather works, we're able to find out what the jet stream is doing, where the high- and low-pressure systems are. That tells us a great deal about what we can expect. If some part of our observing system sees a hurricane or a strong cold front coming through, based on all of our past experience with that sort of data, we can determine that areas to the east of a big cold front might experience that cold front, or areas in the path of a hurricane might experience hurricane conditions. It takes a lot of experience to interpret those observations. Those observations are also fed into our computer models for what we can expect in the next few days. That really helps our predictive capabilities. We initially diagnose what the initial state of the weather is, and then we try to predict it out a few days.
Dunne: How good are predictive models? I imagine that knowing, say, the average temperature for Eugene on March 9 has been 45 degrees for the last 30 years is helpful, but obviously just because it was that way historically doesn't mean it will be tomorrow.
O'Neil: What you're referring to is one of the really useful tools for weather forecasting: what we call climatology. That's exactly as you said, looking at what the weather has been on this day for the last 30 years or so. That gives us an expectation, even before we look at the weather maps and forecasts, of what we can expect historically and what the historical spread has been over time. That really tempers our expectations for a forecast. Climatology is actually very useful for interpreting both the observations and the model forecasts. To use an example: We had that heat dome event in June 2021. The forecast seven to 10 days out actually did quite well at foreseeing that event. However, as it approached, the temperature forecast of 115 or 116 degrees for Portland, the models kind of latched onto that really well. But it was so far outside of our historical norm that we were nervous the models were incorrect. It turns out they were actually quite good. However, there have been other instances where models have forecast something outside of our normal range that didn't pan out. Knowing the climatology of the region can help us interpret whether a forecast model may not be doing well.
Dunne: Does Oregon present its own set of challenges? Is it more interesting, more challenging to be a climatologist or meteorologist in Oregon versus somewhere like San Diego?
O'Neil: Absolutely. One of the things that makes Oregon so challenging is the vast difference in climate across the state. In the western part of the state, from the crest of the Cascades to the west, we have a very wet climate for the most part. The coastal mountains and the Cascades get over 100 to 150 inches of precipitation per year, and the Cascades typically get a lot of snowfall. Then once you go east of the Cascades, out to Bend and beyond, you're looking at roughly two-thirds of the state, and a lot of that is what we characterize as a semi-arid climate. It's not quite a desert, but it's getting there. It's really dry, and temperature variations can be quite vast. The coldest temperatures recorded in the state have been in the eastern part, with a record low of around minus 54 degrees Fahrenheit. The highest temperature, also occurring in eastern Oregon, is 121 degrees Fahrenheit. We also have a lot of mountain ranges beyond the Cascades, including the Blue Mountains, the Steens Mountains and the Ochocos. That creates a lot of variability. I would classify Oregon as the state with one of the most variable climates of all states in the country.
Dunne: Can we predict drought? And within that, can you explain the El Nino and La Nina effects and their typical impact on Oregon?
O'Neil: Those are great questions. On the drought question: We have difficulty predicting drought because we have difficulty predicting precipitation past about two to three weeks ahead of time. Drought is kind of like a slow-moving natural disaster. We can see drought develop in real time, but it's really difficult to say two or three weeks in advance whether we're going to be wetter or drier than normal. A drought deficit from the past couple of months could potentially be made up by one or two really good storms going forward. That said, there is one aspect of drought we can see coming: when we have really low snowpack. Low snowpack is often associated with spring and summer drought. Snowpack is a natural reservoir for water, so when it's not there come April, we can be reasonably certain that spring and summer will bring really low stream flows and low reservoir levels in snow-dominated basins. Right now, we have one of the lowest snowpacks in our state's recorded history, so if these conditions continue, we're fairly certain there will be water supply problems coming into spring and summer.
Dunne: And what about El Nino and La Nina?
O'Neil: We do struggle with these, but the reason we pay attention to El Nino and La Nina is that they represent about the only tool or relationship we have that can predict anything going out past a few weeks. For Oregon, when El Nino occurs, we tend to be drier and warmer than normal. When La Nina occurs, we tend to be wetter and cooler than normal. There's a lot of variation between different events and different years, but it gives us one of the only predictive capabilities we have for trying to see what's going to happen over the next couple of months. Our weather forecast models, when we try running them out past a few weeks, become extremely uncertain. Technically, we say they don't have very much skill, meaning we can't trust them past a few weeks. There have been a lot of advances trying to extend that predictability further out, because that would be a major benefit. But right now, it remains an elusive problem.
Dunne: Some people might turn on the TV for the weather report and then pull up an app on their phone and see different forecasts. Is that simply because the weather reports are constantly updating?
O'Neil: That's definitely part of it. The forecasts are updated constantly. Both the Weather Service and the models they rely on are run in real time and updated continuously, roughly every six to 12 hours, though some models are updated more frequently. There are also a number of large numerical weather prediction centers running different models, and different weather apps may rely on different models or different combinations of them. But the other thing about forecasts is that they are not absolute. They're what we call probabilistic, so we try to give a range. Weather apps often present things in what seem like absolute terms, but what we really have is a range. When we say it's going to be 60 degrees today, we really mean something like 57 to 63 degrees.
Dunne: Can you explain the difference between weather and climate?
O'Neil: Weather is basically what we experience and remember over short periods of time, on timescales of maybe a few weeks or a couple of months. Weather is all about the individual storm systems, individual heat waves and the individual events you would normally see. Climate is what you get when you zoom out to seasons, years and decades. Climate combines together a lot of different individual weather systems. When we talk about climate, we really mean the general weather patterns of a region: the general heating and cooling, the general precipitation, whether you get rain or snow, whether you occasionally get lightning or tornadoes. That's the climate timescale.
Dunne: Is climate change making weather forecasting harder?
O'Neil: Yes, it is. One of the key things in weather forecasting is the climatology of a region: what we expect in terms of temperature patterns, precipitation patterns, whether we get rain, snow, tornadoes and so on. With climate change, what we experienced in the past is not what we're experiencing now. That makes us a lot more uncertain when we get a forecast, such as that heat dome forecast, as to whether it's a realistic forecast or whether a model is picking up something about climate change that we're just not seeing yet. It's definitely making weather prediction harder, especially for extreme weather events. Extreme precipitation events, extreme temperature events and extreme phenomena like hurricanes or tornadoes are all harder to forecast now, because we're less able to rely on our past experiences to inform present and future decisions.
Dunne: Larry, my last question is really a two-parter. You and I have talked previously about concerns around cuts to federal programs that feed into the ability to forecast weather. How are things going with that? And can you remind folks how important weather modeling and forecasting is, not just for deciding whether to play golf, but for our way of life and economy?
O'Neil: I'll start by saying our weather observing system, which describes all the different assets that measure weather conditions, from ground-based observations to satellite observations to weather balloons and more, has been through a very challenging time in the last year. There's been a lot of uncertainty with the federal cuts. One of the things we've had to do is advocate for this system and explain how much it really impacts our society, our economy and our daily lives. There have been a number of threats to various parts of the system, coming from people who don't quite understand how vital weather observations are. That includes our weather balloons. Weather Service offices all across the country launch several weather balloons a day, and those are really important. We've had to discontinue some of those launches. There's some uncertainty about exactly how many balloons haven't been able to go up, but those reductions do affect our weather forecasts. There have also been satellite missions that have been delayed or canceled, and earth science satellite missions that have been under threat. That's been very worrying, and those threats continue.
Dunne: Larry O'Neil is the state of Oregon's climatologist and an associate professor at Oregon State University. Larry, it's always great, if not sobering, to talk to you. Thanks so much for jumping on and having a conversation with us.
O'Neil: You're welcome. Thank you.
Dunne: That's the show for today. All episodes of Oregon on the Record are available as a podcast at KLCC.org. Tomorrow on the show, a conversation with Aimee Green of The Oregonian about a bill that just passed the legislature that could be a vital lifeline to Oregon's struggling ski industry. I'm Michael Dunne, host of Oregon on the Record. Thanks for listening.
