Nature's Archive Podcast

Michael Hawk owns copyright in and to all content in transcripts.

You are welcome to share the below transcript (up to 500 words but not more) in media articles (e.g., The New York Times, LA Times, The Guardian), on your personal website, in a non-commercial article or blog post (e.g., Medium), and/or on a personal social media account for non-commercial purposes, provided that you include attribution to “Nature’s Archive Podcast” and link back to the naturesarchive.com URL.

Transcript creation is automated and subject to errors. No warranty of accuracy is implied or provided.

[00:00:00] Michael Hawk: Weather is like your mood and climate is like your personality. That’s a great way to think about the differences between weather and climate. And those are the words of today’s guest, Dr. Marshall Shepherd. Despite the clear differences between these two concepts. There are many topics of confusion that persist for example, is El Nino, which we discussed a few weeks ago, a climate condition or a weather condition and how can forecasters be confident in their long term climate predictions when it’s so hard to predict weather just 10 days out. Dr.

[00:00:28] Shepherd is just the person to help us understand these concepts and why they’re important.

[00:00:33] If you’re unfamiliar with Dr. Shepherd, he’s the Director of the Atmospheric Sciences program at the University of Georgia and Georgia Athletic Association Distinguished Professor. He’s also host of the weather channel series, Weather Geeks, and previously was a research meteorologist for NASA Goddard space flight center. He also has multiple popular TED talks.

[00:00:53] Today. We talk about climate and weather and how they are predicted. We discussed the computer models used for both how they differ when looking at longer-term climate and how they’ve improved over the years. There are also areas of improvement that we discuss.

[00:01:07] Dr. Shepherd share some of his research on how urban areas affect and change weather and several other fascinating topics. This episode might sound slightly different than a typical nature’s archive interview. That’s because originally we were planning to use this conversation as part of an upcoming jumpstart nature podcast. But as we were talking, it pretty quickly became clear to me that you would all enjoy this too. So as a result, you might hear a few terms and concepts mentioned without explanation, but stick with it because we end up defining those at later points in the recording. If this is interesting to you, you might want to check episode number 80.

[00:01:41] As I mentioned, that has some background on the importance of oceans and El Nino. And going back about a year, episode, number 62 with Dr. Kenneth Librecht had an in-depth look at topics like snowflake, Genesis and condensation nuclei, which comes up a few times today. That might sound a bit complicated, but it’s basically the process that allows water droplets to form.

[00:02:02] And it’s surprisingly fascinating. You can find Dr. Shepherd at @drshepherd2013 on Twitter. So without further delay, Dr. Marshall Shepherd.

[00:02:14] If you are encountering somebody who is. Somehow unfamiliar with you and your work. how would you describe your job to people that you just met?

[00:02:21] Marshall Shepherd: . How I would describe my job is that I’m actually a scientist and a professor. I actually teach at the University of Georgia, but I also maintain a vigorous research program trying to understand aspects of weather, climate, and the ways they impact Nature, the way they impact infrastructure, the way they impact people.

[00:02:39] So, most people associate professors with just teaching. In fact, when I say I’m a professor at the University of Georgia, the next question typically is, what do you teach? But in fact, that a research one level university, many of the faculty like myself are also scholars and scientists and authors.

[00:02:56] And in my case, I have quite a bit of active research on weather and climate.

[00:03:01] Michael Hawk: And what led you down this path of weather and climate in the first place?

[00:03:06] Marshall Shepherd: Well, to talk about what got me into this field, we have to go all the way back to fifth and sixth grade. I mean, I was an adamant nature observer. I was fascinated by insects and creeks and streams and so forth. And so I would catch insects in the yard. In fifth grade, I got stung by a bee. I found out I was highly allergic to bee stings.

[00:03:28] And sixth grade science project, I said, I need a plan B. I don’t want to get stung by a bee anymore. And so I decided to do a science project on weather. And I did a science project called, can a sixth grader predict the weather? And even decided to make my own weather instruments out of things we had around the house.

[00:03:46] And so I was able to do that. I was able to develop little Weather models for my community and won the science fair and from that point on I knew I was bitten by the weather bug Instead and wanted to study weather and not just predict it or stand in front of a screen and do television meteorologies Which is what many people associate my field with even though it’s a fairly small percentage of the field I was more interested in how’s and why’s So that’s what sent me down the path of being a research scientist a research meteorologist

[00:04:19] Michael Hawk: Wow, that is, that’s really cool. , where did you grow up? Like what kind of weather were you experiencing?

[00:04:24] Marshall Shepherd: I grew up in, uh, suburban Atlanta. In fact, it was rural Georgia at that point. It’s a little town north of Atlanta called Canton, Georgia, about 40 miles north of Atlanta. It’s now just suburban Atlanta. And so, fairly typical weather that we get in the southeastern United States. One of the really interesting and rich things about being interested in weather and living in the state.

[00:04:46] South is we get all types of weather. We get tornadic storms, we get hurricanes that come in, we get heat waves, we get snow storms, wildfires, flooding, every single type of weather event we get here in the Southeast that not every part of the United States can say that. In fact, you don’t get. too many tornadic storms out in the western U.

[00:05:08] S. for example. You don’t get hurricanes in the northern tier, but here in the southeast we get a little bit of everything. And so, as someone that considers himself a card carrying weather geek, I grew up in the midst of all types of weather.

[00:05:21] Michael Hawk: Yeah, my, my experience, I grew up in Omaha, Nebraska, and it was the tornadic storms and the blizzards that really drew me in. But yeah, no hurricanes up there.

[00:05:30] Marshall Shepherd: Yes.

[00:05:31] Michael Hawk: From your area of study, what is one of the more amazing facts that you share with people, like with a friend or family member that, that you’re like, I’m just going to blow your mind with this one.

[00:05:41] Marshall Shepherd: Well, I’m quite known in the research world for my work on how urban environment cities affect weather. And so one of the things that we found over the years and confirmed is that sometimes cities can create their own thunderstorms and, that often, stuns people. But we have the data that show that cities like Atlanta or Houston.

[00:06:02] can either create, initiate, or modify existing thunderstorms, modify the rainfall and lightning distributions around them. So that’s certainly some, something that has been a focus of my research for many years. Another thing that we have, uh, in more recent times have talked and discussed in our research is something called the brown ocean effect.

[00:06:21] And this is the idea that hurricanes, when they move. Overland, they’re supposed to weaken because they, don’t have their fuel supply anymore, which is warm water. But over the years, some of my research group and studies have shown that in certain circumstances, if the soil is wet enough from previous rains, or perhaps a swampy or wetland.

[00:06:41] The storm can maintain its intensity. So you actually have a hurricane or a tropical system over land, and it’s still feeding off of that wet soil or that wetland or that marshy land. So that’s why we call it the brown ocean. Hurricanes aren’t intelligent beings. So it doesn’t necessarily know it’s not over the ocean anymore.

[00:07:00] If it’s over a wetland or wet soil.

[00:07:03] Michael Hawk: Yeah, let’s, I, I’m going to maybe go off on a little tangent here. I’ve seen people a lot of times back to the thunderstorms and cities for a moment, comment about how. They’ll be looking at the radar and they’ll see a strong storm heading right out of city and as it approaches the city, it tends to fall apart.

[00:07:20] So I’ve often heard the opposite of what you described where, where people believe that, that cities can actually, break down storms.

[00:07:27] Marshall Shepherd: Well, so first, some of that is a myth, by the way, but there are studies that show something called urban bifurcation of storms. This idea that as storms approach, uh, they tend to split or bifurcate. What we have found is not inconsistent with that at all.

[00:07:43] We actually have found that urban areas can create or induce their own thunderstorms and they move downwind. And often even with this bifurcated. Uh, bifurcation effect that we’ve seen and some have published on, they actually tend to come back together in many cases, downwind. So, um, you know, in many of my papers I’ve talked about the bifurcation of storms as well as the enhancement or the inducement of storms, by the urban environment.

[00:08:07] And by the way, the cities create and. Impact their own weather in many ways. Cities tend to be warmer because of the urban heat island. They have large buildings in them that which can create convergence, which gives you lift that you need for thunderstorm development. There are also particular aerosols and city environments, which service cloud condensation nuclei for clouds and then the bifurcation effect.

[00:08:30] Those are the 4 sort of. Physical processes and that we’ve talked about that lead to cities modifying precipitation in some way. So absolutely. But again, I, I went to school at Florida State University, and we used to always talk about the Tallahassee, Florida effect, because anytime there was a collection of people interested in whether the storm seemed to always break down before they came to that city again, some of that is a little bit myth, but there are some studies that talk about bifurcation due to the cities.

[00:08:59] Michael Hawk: yeah, I always. Thought it was a bit of a myth, just from the fact that it seems like in a lot of cases, the, primary radars are in the cities and when the storm is further away, you know, just the angle of the beam is higher up in the atmosphere. So you’re getting more into that core of the storm.

[00:09:15] That’s not, uh, not evaporating before it hits the ground. So my little amateurist view was, uh, some of that was the effect of the, of the radar, but I’m going to read your papers. I need to read those papers

[00:09:26] Marshall Shepherd: Yeah, no, you should, I, I, absolutely. Yeah, we have several papers in the scientific literature on urban effects. There’s a 2005 paper, which is kind of a review paper that I wrote on this whole idea of urban effects on rainfall, but I’ve certainly I’ve published things more, more recently than that as well, but yeah, there’s an entire body of literature on this.

[00:09:47] It’s not just me doing this type of work as well. Um, but yeah, the radar is an interesting tool to examine. Um, Rain radar meteorology is one of my areas of expertise. In fact, my master’s work before I went on to do my PhD was looking at some of the first ever next red radar and how it was able to detect storm centers and hurricanes.

[00:10:07] And yeah, there are some really interesting things about radar. There’s the radar cone of silence, typically near the radar itself. There’s a. Cone of silence because the radar scans at an angle and scans out a cone. And so, um, people don’t realize that you’re right. As the beam gets further away, it’s in a higher altitude and so forth.

[00:10:25] So, um, I teach an entire class on radar meteorology at the university of Georgia. And so we get into all of those really interesting things.

[00:10:31] Michael Hawk: And yeah, I need to stay on, track here and hit the other questions that I intend to hit. but just another side comment on radar is I I’ve recently come to learn how important of a tool it is to track, bird migration and, you know, other, even insects, and things like

[00:10:50] Marshall Shepherd: yeah, yeah. The buyer, many of the targets for radar, whether radar are non meteorological. In fact, meteorological targets were an accident for radar in itself when, the early war, world wars, world war two, and even a little bit before that, radar was, um, used for tracking things like airplanes and ships and so forth.

[00:11:10] And people started noticing this. Clutter as they were looking for ships and airplanes, and it turned out it was rain. So we, we knew that we could use a microwave energy to detect rainfall. And so over time, you know, radar has evolved that we can use the Doppler shift to see the motion of those raindrops or particles.

[00:11:29] We can use things like dual polar metric capabilities, which sends out pulses and horizontal and vertical orientations. And so that helps us fine tune sort of the. The stage of the rainfall, whether it’s snow or rain or ice, or it helps us with things like detecting debris from tornadoes, but also these radars are sensitive enough to detect things like birds and insects and so forth.

[00:11:52] And in fact, we purchased a radar here at my university. It’s an X-band radar. We use different bands and X-band radar systems about. Three centimeter wavelength. The radar typical radar that you may see on the news is either five or 10 centimeter X-band will be attenuated more because it’s a smaller wavelength.

[00:12:10] In other words, it means it might not. extend as far out because the rain will, uh, will attenuate the signal in the same way that maybe satellite TV if you had those back in the older days. Um, but yeah, and the radar is one of the tools that’s very important for what we, in my business call mesoscale meteorology, the scale of, of weather events that, you know, from about one hour to a few hours or so.

[00:12:34] Michael Hawk: All right, so let’s move into some of the core of what I was hoping to discuss with you and that’s, , weather and climate and the differences and how we can wrap our heads around those differences. So to maybe kick things off, how do you define climate?

[00:12:50] Marshall Shepherd: So, you know, oftentimes there’s been a misnomer and I learned it too as a younger scholar and people would say, Oh yeah, climate is just average weather, but that’s really short changing climate in so many ways. Climate really is more the statistical Properties of the atmosphere and of weather, if you will.

[00:13:08] So it’s not just average, it’s median, it’s maxima and minima, it’s the distributions over time, tendencies and so forth. So, weather and climate are very different, but they’re definitely related to each other. You know, it’s really interesting. I’m on, I’m active in social media and sometimes when it snows.

[00:13:27] Uh, someone will tweet me and say, well, what do you mean there’s climate change or global warming? It’s snowing here today. And that, that’s really an indication that the person may not fully understand the differences between weather and climate. And so I often use the analogy that weather is your mood and climate is your personality.

[00:13:43] And it tries to sort of convey that your mood today doesn’t necessarily tell me much about your personality. Uh, another one that I like to use is that weather is what you have on today in terms of your clothing. Climate is what you have in your closet. Because certainly in your closet you have a raincoat, or you have a heavy coat for a cold day, or maybe snow boots and so forth.

[00:14:04] So it’s, uh, for the range of possible weather outcomes. So, it’s an important distinction, but one of the things that we increasingly know, and I suspect we may talk a little bit about it in this podcast, is that climate changes are certainly in the DNA of today’s weather, if you will.

[00:14:21] Michael Hawk: Yeah, I really, I like the maxima and minima terminology, you know, for me with an engineering background, I often think of kind of like, uh, bounding lines on a graph. Like you’re going to operate within this range of possibilities, you know, over time for a given system and your analogies, I think hit the nail on the head, uh, to explain that in easy to understand terms, non engineering terms.

[00:14:48] Marshall Shepherd: Yeah, no, it’s important. I mean, I think that. In, in the world of weather and climate, we, we have to, because of the nature, the stochastic nature of some processes that we deal with, precipitation, or modeling outcomes for hurricanes and so forth, we have to use things like probability and uncertainty, 30 percent chance of rain, or the hurricane cone of uncertainty, these are all trying to convey that we have some knowledge of the event, but also there’s uncertainty.

[00:15:15] Unfortunately, many of those concepts are often confused by the public. Many people don’t really know what They hear 30 percent chance of rain, but don’t really understand what it means. And so if it’s a 20 percent chance of rain and it rains, people will say, Oh, the forecast was wrong. They said it was 20 percent chance of rain and it rained.

[00:15:30] Well, it wasn’t a 0 percent chance of rain. And so, you know, that 20 percent is really representative of a forecaster’s confidence in a certain area, aerial coverage of rainfall. There was a 20 percent chance of rain today. You might be in the 20 percent aerial coverage that day. So there’s nothing wrong about the forecast.

[00:15:48] Uh, but the way people perceive the information, the same thing with the hurricane cone, people will see that hurricane cone and assume that it means the storm is going to go down the center of the cone. But in fact, that current hurricane cone of uncertainty saying that there’s a 66 percent chance that the center will be anywhere in that cone based on a previous year statistics of track forecasting.

[00:16:08] And so it’s really important to make those distinctions.

[00:16:12] Michael Hawk: Mm hmm. And this year. , There’s a lot in the news about the El Nino and, the Enso cycle in general. Uh, I did a kind of an introduction to Enso in a earlier episode. How do you see a, like a large scale? I don’t even know how to describe it. So I’ll let you describe it, but how do you see El Nino and Enso in general fitting into this definition of climate versus

[00:16:41] weather?

[00:16:41] Marshall Shepherd: Yeah, sure. So the El Nino Southern Oscillation or ENSO. is a naturally varying cyclical process in our atmosphere. It’s a short term climate variability, if you will, compared to longer term climate changes that we talk about often. Uh, there are many of these sort of short term climate variability signals in the atmosphere.

[00:17:03] El Nino is The one that many people are familiar with, but we have La Nina, we have things like the Pacific Decadal Oscillation, the North Atlantic Oscillation, the Madden Julian Oscillation. There’s so many, I could go on with this alphabet soup for, for some time. But what we know is that the El Nino Southern Oscillation, you get El Nino roughly every two to seven years, uh, it’s a warming of the Central Pacific, Eastern Central Pacific Oceans.

[00:17:30] The opposite phase of that is La Nina. which is a cooler phase, and this is all related to sort of changes in trade winds in the equatorial regions of the Pacific. When we have an El Nino pattern, you get a change in the winds in such a way that you don’t get as much upwelling of cold water, and in fact, you may have a situation where you get the anomalously warm water, and so that leads to changes in our, heating of the ocean and the atmosphere.

[00:18:00] And so you get changes in our jet stream patterns and things that affect our weather around the world. And so those are called teleconnection patterns. And so, you know, the El Nino is a really good example of how something seemingly so far away from us can actually affect weather literally around the world and it does.

[00:18:17] So it fits into sort of, I would say a continuum between day to day weather and longer term climate change, but they’re all connected.

[00:18:26] Michael Hawk: So speaking of longer term climate change, uh, you know, it’s well understood by most people anyway, that the climate is warming and the oceans are warming.

[00:18:37] Marshall Shepherd: Yeah,

[00:18:38] Michael Hawk: and

[00:18:38] Marshall Shepherd: and most of the warming is in the ocean, by the way.

[00:18:41] Michael Hawk: yeah, well, okay, let’s jump straight to that. because. As I understand it, the ocean and the atmosphere, it is a coupled system, and there’s a lot of exchange going on there. So I think this is a key point. Can you explain to me , why this is happening and what the implications are for short term weather and long term climate?

[00:19:01] Marshall Shepherd: Yes. Well, clearly climate changes naturally. And I, I say that because sometimes I’ll have contrarians or naysayers come up to me, a climate scientist, someone with a doctorate in atmospheric sciences and meteorology and say, well, climate changes naturally, or we’ve always had these. uh, storms. Absolutely.

[00:19:18] but we know that as we have seen an increase in greenhouse gases, something’s changing. So let me just step back. I’m sure most of your listeners have heard of the greenhouse effect. The greenhouse effect is why we survive every night. If we did not have greenhouse gases in our atmosphere, we would get too cold and we wouldn’t be able to sustain life on this planet, at least for us. But we have greenhouse gases and because of the radiative properties of those gases, they absorb long wave radiation emitted. By the planet and then absorbed and coming in from solar, uh, sort of the sun and so forth and re radiates that back to a way that we have a comfortable temperature to live.

[00:19:59] However, you had all of these sort of fossil fuels and sort of things that are been sort of compacted in the earth for many, many, many millions of years that now we, since 1850, the industrial revolution, we started figuring out how to utilize and burn. And so we’re rapidly increasing the carbon.

[00:20:17] amounts in the atmosphere. And so we’re out of balance. And so we’re warming it near the surface. And that is leading to changes throughout our earth system, whether it be in the oceans where much of the warming is occurring, but also in our atmosphere and so other aspects of the earth system respond, I spent the first 12 years of my career at NASA.

[00:20:37] Uh, NASA Goddard Space Flight Center, where we develop satellite missions to look at Earth as a system because the atmosphere, the ocean, cryosphere, the biosphere, they’re all connected. And so, um, For example, when you get warming in the ocean, that manifests itself in stronger hurricanes or it manifests itself in, uh, expand thermal expansion of water and perhaps in the cryosphere, you get glaciers melt and you get sea level rise and so forth.

[00:21:04] So all of these things are interconnected and they, they, they end up having effects on all aspects of society in so many ways.

[00:21:10] Michael Hawk: I think something that a lot of people struggle with is they hear some of the projections about One degree Celsius or two degrees Celsius warming. And what a dramatic impact even something seemingly small like that could have. Uh, can you help us understand why just a couple degrees make such a big difference?

[00:21:31] Marshall Shepherd: sure. And, and, and, and first of all, I don’t, even though everybody says two degrees is that that’s what the Paris agreement is trying to cap at two degrees, I don’t like to talk about it in terms of Celsius. I imagine many of your listeners are in the United States. We use Celsius around the globe, but.

[00:21:47] It’s 3. 6 degrees of warming in Fahrenheit. And I think that’s what most people in the United States understand. Two degrees and one degree don’t resonate as much as when I speak about the terms in Fahrenheit. But even when we talk about it in those terms, I put it this way. Imagine if the human body, the human body has a temperature of 98.

[00:22:05] 6 on average. Imagine if your human body temperature was 3. 6 degrees warmer. Uh, for a sustained period of time, you’d be running a significant fever and your body would start to shut down or your body was starting to feel that in many ways. that’s kind of my analogy, 3. 6 degrees is a fever for the planet and its systems start to respond in so many ways.

[00:22:28] so it’s not as so much, it’s, it’s really, as they say, quality, not quantity. I mean, I always say, uh, it doesn’t take very much of the venom from a black mamba snake to kill you. and, and so the carbon dioxide and other greenhouse gases, carbon dioxide being the most prevalent and dominant, uh, have such a significant radiative.

[00:22:47] Impact or forcing on our atmospheric processes and our, climate system that, uh, one, two degrees and that’s really, you know, what we’re hoping to keep it to some of the climate models project warming well past that, uh, but even at one to two degrees, we’re already starting to see dramatic impacts.

[00:23:05] So imagine that 3 and 4 degrees what we’re expecting to see

[00:23:09] Michael Hawk: I like that analogy too, is I’m thinking about the implications of the human body and our temperature and you know, you can go exercise, you can go on a run and Your body temperature might spike during that exercise, but you’re going to bring it back down when you’re, when you’re done. But if you sustain it over your whole system for a period of time, you’re going to have heat stroke, you know, like it’s going to be bad.

[00:23:31] Marshall Shepherd: Yeah, absolutely. I mean, again, if your, body is running a 3. 6 degree Fahrenheit fever for a sustained period of time, I mean, I, none of us can live at 103 degree temperature for too long if we’re sustaining that for hours to days to years, our planet is essentially the same analogy.

[00:23:51] Michael Hawk: I don’t know if this is a good example or not, but in the news just this week was hurricane Otis. which made landfall near Acapulco, and it surprised a lot of meteorologists with its rapid intensification. And I’ve read a little bit about it, but does, does this scenario fit in with, with this landscape of warming oceanic temperatures and atmospheric

[00:24:16] Marshall Shepherd: I

[00:24:17] Michael Hawk: Okay. Before Dr. Shepherd answers that question since it’s been a few weeks now, I figured I should give a little more context about hurricane Otis. So hurricane Otis had actually was a tropical storm and it was forecast to make landfall on Mexico as a tropical storm. And the definition of a tropical storm is a maximum sustained wind between 39 and 73 miles per hour. And once you hit 74 miles per hour, you become a hurricane and that’s a category one hurricane. If a storm system sustains wins, that’s not gusts, but just constant wind of 157 miles per hour.

[00:24:56] It’s a category five storm. So this storm. I was predicted to form and it did, and it was predicted to make landfall as a tropical storm. Again, that’s less than 73 miles per hour. But it had some of the most rapid intensification ever observed and it became a category five storm, 165 mile per hour storm.

[00:25:18] By the time it made landfall. There’s actually a whole Wikipedia page on the storm because it’s just so much of an outlier. That, people were quick to say, okay, this is special. We should document it. So I’ll be sure to link to that in the show notes.

[00:25:32] Marshall Shepherd: think it does in many ways. I think we’ve seen several examples in recent years of what we call rapid intensification where we gain 35 miles per hour of wind or more in less than 24 hours. And I think in the case of Otis, it was even a much larger gain. I think it went from a. tropical storm essentially to a cat 5 hurricane in about 13 or hours or less, which is hyper intensification, not rapid intensification.

[00:25:57] And we’ve seen that. That’s been a hallmark. This was a forecast bust on intensity. Our weather forecast models are pretty good, particularly with track, but any way you look at it, the intensity forecast for Otis were underestimated severely and one of the reasons why, and I was going over this in one of my classes at the University of Georgia, is we were looking at the sea surface temperatures just off the coast of Poco, Mexico, and they were quite warm.

[00:26:27] There was a pocket of very warm water. And so. Otis moved over that rapidly intensified all of all of the other conditions were favorable in that moment. There wasn’t very much wind shear, which is changing wind speed and direction as you go up in altitude and so boom, it moved from a what I’d like to say it’s a patch of 89 octane fuel to 93 or 94 octane fuel warm ocean content.

[00:26:52] Is the fuel supply for hurricanes. And so, uh, we’ve known for some time now that our models lag behind skill on intensity forecast. They get track forecast pretty good because the track is really determined more by large scale circulation patterns that our models can handle. Intensity changes with hurricanes are driven more by internal mechanisms, by ocean water, by the sort of big heat engine processes that are going on in the eye wall of the system that oftentimes we may not have the best observations for to feed the models.

[00:27:27] And so, uh, this was a case where we will go back to the drawing board and understand. Some things we, to help us improve intensity forecast, because again, as a scholarly community, we know that that’s where we lag.

[00:27:40] Michael Hawk: we’ll talk about models, I think a little bit more here in a moment, but this might be a good lead in to talk about the tools that you have at your disposal for weather prediction in the first place. I know there’s lots of different ways that you could, you could go, you could probably like someone skilled, like yourself, you could probably go outside and look at the, look at the sky and look at the current temperature and have a rough idea of what’s going to happen for the next few hours.

[00:28:02] But tell me more about what goes into, , the short term weather forecast, what tools are at your disposal today.

[00:28:10] Marshall Shepherd: Yeah, no, we, we, it’s a question I often ask in public lectures of how do we make a weather forecast? How do I know that it’s going to be certain, certain, certain things? Three days out or five days out. And people say things like, oh, you just look at how things move from west to east. Or, oh, you look at the farmer’s almanac of the groundhog or, or radar or satellites.

[00:28:28] And I mean, it’s, it’s, it’s computer models. The atmosphere is a fluid. It’s a fluid on a rotating body and they’re a set of boundary conditions. And you’ll certainly understand that as an engineer. And so. You know, the Navier Stokes equations, other types of sort of thermodynamic equations and so forth are solved on a system of grids representing this fluid on a rotating body.

[00:28:52] So we have to account for things like Coriolis force and so forth. And we make a prediction for how that fluid changes. The analogy that I like to give is that, let’s say I was up in St. Paul, Minnesota, Minneapolis, St. Paul. That’s where the headwaters of the Mississippi, or at least the beginning points, I could put a beach ball in that river, and because I know some things about the flow of the river, its temperature, its depth, and where I put the ball in, I could solve a series of fluid dynamics equations to predict where that beach ball would be three days later downstream.

[00:29:23] Um, that’s kind of what we do with the atmosphere. We’re solving the system of equations to predict that fluid change one day out, three days out, seven days out. Uh, we know due to some things that Ed Lorenz and chaos theory told us that we probably, yeah, it’s an initial value problem if you go back to calculus class.

[00:29:41] Uh, but we know that there’s probably a limit of predictability out to about 10 days or so based on what some studies have shown us. So models, weather models, very high, high end computational models are the primary weather forecast tool, but you’ve got to have data to initialize those models and even assimilate into the models as the models are running.

[00:29:59] And so weather satellites and buoys and Doppler and dual polarimetric radar and when the profilers, which, are a sort of a special radar that points straight up into the atmosphere can give us things among wind and solometers, which are lasers that give us things on cloud height. all of these tools are important.

[00:30:18] Now there’s an emerging tool. That you’re hearing quite a bit about, which is artificial intelligence, machine learning, we produce so much data and some of weather forecasting is pattern recognition. So I think artificial intelligence and machine learning is going to help quite a bit and parsing through the Tremendous volume of data that we have from not just the atmosphere, but from the oceans.

[00:30:40] As you said earlier, the oceans a couple system. So we ocean surface temperature and water. These, these things are very important to us as well. Land surface temperature, but it’s a ton of data. And so we, we need innovative ways to manage it. And so that’s why I think AI techniques are going to be quite helpful going forward.

[00:30:58] Michael Hawk: It really does blow my mind to think about an earth scale, model, you know, as you talked about, you break it down into these little quadrants and all the variables in each of those quadrants. And, you know, the challenge in, in crunching that, and it, it does seem like AI for all of its flaws is, uh, has a lot of potential in data crunching and pattern recognition.

[00:31:20] So it’s, it’s interesting that you characterized it that way.

[00:31:22] Marshall Shepherd: Yeah, no, it’s very, very important to what we do.

[00:31:26] Michael Hawk: you mentioned that with chaos theory and some of the limiting bounds that exist, uh, that today about 10 days out seems to be the limit. Is that, do you see that as something that even with AI is going to be hard to break through a 10 day window?

[00:31:42] Marshall Shepherd: Yeah, I do. I think it will be tough to break through it through. I mean, Lorenzo actually predicted this. You know, many, many years ago, and even, you know, more recent studies, there was a study out of Penn State a couple of years ago that sort of confirmed what Lorenz knew several years ago. I think there will always be some bound.

[00:31:59] We might be able to push it a little bit. I mean, there are some people that sort of, you’ll, you’ll see some sort of longer term deterministic type forecast out well beyond. 10 to 20 days out there, but in terms of what most of us sort of believe is the limit of predictability, I think 10 to 12 days.

[00:32:17] There are certainly some anomalous cases that kind of go beyond that. Uh, where the models do, but most people that I talk to in this field believe that we probably won’t get too much beyond that on a day to day basis. Now, there are some seasonal predictability capabilities that are emerging that can help us with things like in the one month, three month, two, two to three month timescale.

[00:32:41] We signals in the El Nino and La Nina patterns and some of these other alphabet soup patterns that I mentioned earlier in the podcast, They can help us, you know, there is a real interest in this sort of seasonal, interseasonal, seasonal predictability problem, but I mean, you’re really not talking about predictability in terms of a day, what’s happening on a given day, three months out, but sort of, sort of seasonal predictability, I think that’s kind of a next, a new frontier that people are really thinking quite a bit about now.

[00:33:08] Climate prediction uses some of the same techniques. We use climate models. We use some of the same equations. Okay. But with a climate model, we’re not trying to predict what a day in the year 2050 looks like on Wednesday in San Francisco. We’re trying to look at sort of climate states and sort of the broader picture of climate in those timeframes.

[00:33:27] So I think people who often say, well, the climate models aren’t reliable, the climate models can’t. So, that climate models aren’t designed to predict the weather in the same way that a weather model is. It’s predicting a climate state. And by the way, we can run these models in such a way that we run them from past starting points to predict our climate today.

[00:33:48] And they do pretty well. So, that gives us confidence in their ability to predict going forward.

[00:33:52] Michael Hawk: I think that’s a really good point to make, uh, because I, I have heard that stated before, where, you know, if they can’t predict what the weather is going to be X days from now, how can they predict the climate? And I think that the El Nino predictions are a good case example. and I tend to point people towards that because when I see these El Nino predictions, or even just the seasonal predictions, it’s an opportunity to say what they’re talking about here is a , probability.

[00:34:20] Over a large area of an, you know, of an average or above average or below average, whatever the condition, you know, is so you’ve, you’ve kind of dampened that you’re not looking at a specific day in a specific city. Instead, you’re saying this entire area over this period of time. Is likely or unlikely to see the following conditions.

[00:34:44] And I think that’s a good stair step or like initial step into what climate prediction is like. Uh, so tell me more, I think about the climate prediction models that you use, that climatologists use and, how they differ from these, weather prediction models.

[00:35:03] Marshall Shepherd: Well, I mean, they’re more earth system models. They’re looking at much larger spatial scales, even at the weather model scale, sometimes we can’t resolve certain things because of the grid size of the computer model. So for example, uh, some of our best weather models are now getting down to a couple of kilometers in grid size to kind of give you an analogy.

[00:35:23] If think about your cell phone and your, your Apple phone or your Android phone. The number of megapixels in that camera means a much more detailed picture. Well, the smaller the grid cells are in our weather models, the more things we can resolve. We can resolve that thunderstorm cloud or we can solve that microburst and so forth.

[00:35:42] Well, climate models, one of the big sort of challenges has been that they’re still fairly coarse. One grid cell may be. You know, several tens of kilometers or 100 kilometers. So, uh, you’re not necessarily representing individual hurricanes or cloud systems. You’re representing large parameterized sets of data.

[00:36:02] But the good news is with computing power and technology, we are now starting to get some of those climate model scales down. The other really interesting thing about climate models that may differ from, say, a weather model is that the climate Then this used to be a criticism of climate models. Like, oh yeah, they don’t really represent the Earth system very well.

[00:36:21] And some of the early generation climate models were just either atmospheric models or maybe an atmospheric model coupled to an ocean model. But now these models are sophisticated. They have cryospheric processes, which is the ice parts of the world. They have biospheric processes that can represent vegetation, things like the Amazon region very well.

[00:36:40] They have scenarios that represent, human activity and population and force things associated with that, like land use change or emissions. So they’re very sophisticated, complex models. With that having said, there still are some things that we don’t, that are, you know, how to, the scenario estimation of sociodemographic.

[00:37:00] Challenge futures are still very challenging, you know, population and so forth. We still know, for example, that the models may struggle a little bit with things like aerosols, like the pollutants and particulates in the atmosphere and how we best represent them and what their radiative effect is so that we, but we know we understand where the challenges are.

[00:37:20] But I think there are more opportunities with the models than there are challenges.

[00:37:24] Michael Hawk: So just cherry picking like one little. System effect that has been in the news here and there, and I’ve read that as certain as oceans warm and even like, uh, like Pete soils and things like that, that methane can be released at a higher rate are those sorts of details incorporated into the climate models?

[00:37:47] Marshall Shepherd: I think so. I mean, I think, you know, methane is, you know, much more potent of a greenhouse gas than carbon dioxide is, uh, but the good news is it’s not nearly as robust in the atmosphere as CO2, so that’s why you mostly hear us talking about CO2, but yes, we, are we concerned about sort of methane, uh, permafrost thaw, uh, methane from hydrates and so forth.

[00:38:09] I think all of that is of concern, and I think they are captured, or to some degree, in our models.

[00:38:14] Michael Hawk: And another, I think, challenge that people sometimes have is you just walk through how climate is. it’s more of an earth system model on a longer scale, longer timeline. So when I think about back to the weather models, you mentioned with hurricane Otis being able to go back and, you can see that the, that the models didn’t verify it didn’t, it didn’t pan out the way it was expected.

[00:38:38] And you can, uh, understand some of that and probably update the models pretty quickly on a quick turnaround. Uh, but with climate, it

[00:38:46] takes longer to play out.

[00:38:47] Marshall Shepherd: Yeah, but I would say even with Otis, I think we know what the problem was. I don’t think it was a problem with the model. I think it was a problem with what we can input into the model. So, in other words, I think that we have some high resolution hurricane models, but if we don’t get the details of that patch of warm water corrector, if we don’t get the details of that latent heat release that’s going on as condensed condensations occurring, um.

[00:39:10] I think we’ll still miss with on some of these intensity forecast. So that’s what I was saying earlier when I talked about that the track forecast have improved dramatically because we can get the large scale stuff that’s pushing or driving the storm along. But a lot of the internal mechanisms that are associated with intensification or things that are requiring much more fine scale observations that in some cases we don’t always have.

[00:39:34] Michael Hawk: great distinction. So it’s, it’s some of the fine scale inputs into the model that are necessary on the weather side. what are the gaps that you see today in on the climate models?

[00:39:48] Marshall Shepherd: Yeah, I think the big gaps in the climate models, as I mentioned, are aerosols. I think they’re getting the models down to a resolution where we can resolve individual hurricanes and some of the fine scale, processes like, for example, many climate models in the last 10 years or so couldn’t resolve a hurricane.

[00:40:06] Resolve the blob or resolve some kind of a convective wet system there, but, you know, we’re now getting to a point, but we still have some ways to go where we get resolutions of computing power that essentially can be at weather model scale, but running essentially as a climate model. So there’s a process that some, some scholars have been working on at Colorado State guys like Dave Randall, even my colleague at the University of Georgia, Gabe Cooperman.

[00:40:33] It’s something called super parameterization, where essentially they’re running cloud models on each grid point of a climate model. And so the, the thinking there is that you can resolve some of these smaller scale processes, an embed essentially individual cloud models within the larger. Cell red cells of the climate model.

[00:40:52] Now that can, you can’t imagine on your technology guys. So you can imagine this computing power that something like that takes to put a, essentially run a cloud model at every grid point around the planet. So, but those types of things are there because we understand that we need to understand the cloud, cloud impacts in macro physically and radiatively in the climate models, the aerosol effects and the resolution effects, but I think we’ve made significant progress in at least coupling all of the systems together.

[00:41:20] Michael Hawk: And when some people hear aerosol, they may be thinking about, ,

[00:41:23] Marshall Shepherd: spray

[00:41:24] Michael Hawk: spray can. So can you elaborate on what you mean by aerosol?

[00:41:28] Marshall Shepherd: Yeah, aerosols are just particles or particulates or droplets in the atmosphere. I mean, so it could be dust, or it could be soot or smoke from wildfires, uh, pollens. Those are all examples of aerosols. Those, they’re very important, as in some cases as seedlings for cloud formation. Cloud clouds need some type of nuclei to form on our, In our atmosphere, they also have impacts on the radiative impacts on the amount of radiation from the sun that are coming in.

[00:41:56] For example, we know that Sahelian dust, this dust, the layer that comes off the Sahel in Africa can dry out the atmosphere and choke off the environment for hurricane development. So, uh, these aerosols play, uh, one of the things that we know about. Smoke, wildfire smoke is, they have a radiative impact and a meteorological impact, but they also are a hazard for airline engines when airplanes are flying around.

[00:42:21] Uh, so there are these practical aspects of, aerosols as well. I know that, uh, some colleagues of mine have been studying the Sahelian dust in Caribbean, uh, Caribbean regions. We get these big dust outbreaks and that dust settles. In the Caribbean, and there are some significant health and upper respiratory issues associated with that.

[00:42:38] So, but from a climate standpoint, we know that aerosols can have an impact on sort of cloud microphysics, the process of making clouds, and there can be radiative impacts in terms of the energy exchanges that are happening in the atmosphere. And right now, sometimes those climate models aren’t There’s still some uncertainty in sort of the aerosols.

[00:42:56] Part of that, which is because we haven’t really been able to robustly measure them. But now there are satellites up there that are measuring aerosols and their distributions and so forth. And so that’s helping us.

[00:43:07] Michael Hawk: And for what it’s worth, there was, we had an episode on nature’s archive last year about snowflake Genesis. And there’s a little bit of overlap in that episode with what you’re talking about here.

[00:43:18] Marshall Shepherd: Yes, because snowflakes are, you know, well, one of the things that when you, you asked me earlier about things that I tell people that may sort of surprise them or blow their mind. One of those things that I often tell them is that no matter where you are, uh, what time of the year it is, the rain that you’re experiencing probably started out as a snowflake in the cloud.

[00:43:38] So you could be in. , St. Louis on a hot summer day, in August and it starts raining, but that rain likely started out in the cloud as a snowflake because the processes that lead to rainfall, it’s called the ice crystal process or the Bergeron process. ice crystals grow much more efficiently in the tops of these deep cold clouds.

[00:43:58] Remember in the upper atmosphere, even though it’s 80 degrees surface, it’s very cold in the upper atmosphere. And so these things form ice crystals and they need an ice nucleus. to form. And so that crystal, ice crystal forms, it grows at the expense of water vapor that’s there and gets larger. It starts to fall and it clumps on the other ice crystals or collides with super cooled water, which is water that’s below freezing.

[00:44:23] But when you collide with something, it freezes immediately and then it falls. And if the temperature is below freezing, it stays as snow. But if it’s above freezing, it melts and then we see it at the surface as rain. So these things are all very much connected.

[00:44:35] Michael Hawk: Yeah. That’s, that’s a fun one. I kind of have. Taking it for granted in recent years, but yeah, it’s, it’s good to remind yourself of, of this comp complexity. back to this concept of verifying a model, maybe you can define what that means just to start off and you can pick like a weather model or a climate model, whatever is easier to explain what, you know, what’s meant.

[00:44:55] Marshall Shepherd: Yeah. Well, like I said, you know, models are just computational entities. They’re solving equations and spitting out numbers at the end of the day, which we can then codify as rainfall or temperature or what, what clouds and whatnot, but they’re models. So how do you know they’re, they’re right? Well, you have observations.

[00:45:14] So the model is just say you’re using a weather model. It’s predicting a line of thunderstorms to move through Atlanta, Georgia. Well. I mean, how do you verify that that line of thunderstorms came through? Well, I look at radar and I look at the rain gauges to see, uh, that, and, and like I said, in most cases today, Um, Within, you know, 0 to 10 days and particularly within 0 to 5 days, our, our weather models are quite good.

[00:45:39] That’s us. That’s another myth that’s out there that weather forecasting is bad or that we’re not very good. I think that myth exists because people only remember the days. where there was the occasional miss and it impacted them in some way.

[00:45:53] They don’t remember the 95 percent of the days where the weather forecast was right because it didn’t affect them in any way. And it was right. So you don’t tend to focus on it. So I’m a big football fan. So if you think about a field goal kicker in a football game, they can make every field goal all year long in every single game.

[00:46:10] He’s a really good kicker. But if he misses the one field goal in the Superbowl that could have won the Superbowl, people will remember that and be talking about it. And some may even say he was a bad kicker because he missed that one field goal. So that, that doesn’t actually line up with the statistics of his, of his skill.

[00:46:26] It’s just perception.

[00:46:27] Michael Hawk: Yeah. I think understanding statistics probabilities, it’s, it’s not real intuitive to the way our brains work anyway.

[00:46:35] Marshall Shepherd: Right.

[00:46:36] Yeah, and people also don’t assess risk very well. I mean, that’s something that the part of weather warnings, you know, you, had significant flooding in New York City a couple of weeks ago, just, just tremendous amounts of rain falling over short periods of time causing flooding.

[00:46:51] Yet, I still saw people driving through flooded roadways and we tell people not to drive through flooded roadways, but people have these sort of skewed sort of normalcy biases. Well, I’ve driven through a flooded road before a while. Got to really get to my kid that they care so they will sort of

[00:47:05] use flawed logic in their risk assessment

[00:47:05] Michael Hawk: if a change is made to a model, are you able to then reinitiate the updated model? like, say, go back in time to to that instance or other instances 10 years ago, using the initial data that would have been available at that time and run that model and see how it plays out as a way to verify.

[00:47:33] Marshall Shepherd: sure, no, I think that type of thing happens quite often in our field. You’re, you, you have all kinds of verification projects with modeling groups or reanalysis projects and so forth. But one of the things that really I should mention is really the way weather forecasting from a modeling standpoint is done today.

[00:47:49] We use what’s called ensembles. you might actually run a same model with slightly different initial conditions, you might run the same model with 10 different slightly perturbed initial conditions. So you get 10 different outcomes. But if they’re close to the same outcome, you have this more confidence that that’s actually what’s going to happen.

[00:48:09] But oftentimes what you have is you’ll have. different outcomes out of those different ensembles, and by the way, we, we use multiple models and a lot of people hear about the European model of the American model, but there’s several models that we use and good forecasters do use all of them.

[00:48:24] They don’t just anchor in the European or the American, and oftentimes the, the ensemble of the European and the ensemble of the American are exactly. The same or very similar. Many times they’re very different. And so that’s where the meteorologist skill and looking at other things has to come into play.

[00:48:40] Things like similar scenarios they’ve seen like this before or trends in the model or things. What data is not there. So it’s a very challenging but interesting problem.

[00:48:52] Michael Hawk: I love to look at on the National Weather Service websites. They have the forecast discussion and sometimes they get into some of the. Details that you’re talking about and you can get a feel for the confidence of the forecast. That’s currently on the table with that.

[00:49:05] Marshall Shepherd: No, and that’s a, that’s a really good point because I think you’ll see some people that are a little more amateur to this or maybe just sort of for whatever reason, just like 1 mile over the other that don’t seem to realize that the National Weather Service forecasters, for example, the Hurricane Center forecasters, they, they, if you dig into those discussions, they’re very clear about the nuances of the forecast and.

[00:49:25] Why they are trusting one particular model over the other or why they’re going in a certain direction even though Many of the sort of twitter meteorologists may be saying something else. Uh, so it’s really the nuances in that discussion.

[00:49:38] Michael Hawk: So you were talking a little bit about reanalysis projects on the models. Yeah. I’m going to maybe pull a couple of different things together here that I think makes sense, but call me out if it doesn’t. And that’s with, uh, I’ve always been really interested in paleoclimatology and for climate models.

[00:49:59] I’m curious how much of, how much ability exists to reanalysis based on what we know through paleoclimatology.

[00:50:11] Marshall Shepherd: Yeah, I don’t, I don’t, I’m not as much of an expert in that question. I do think that people sort of try to blend sort of past climates based on paleoclimate records, things like tree rings and ice cores and things. So we do have some understanding of like past climates in a very broad sense. And so I do think that there are people that sort of then try to take that knowledge and reconstruct climates and then sort of see what our model or our models able to sort of assess sort of in sort of hindcast mode, if you will.

[00:50:41] Those climate scenarios based on more pristine environments where you didn’t have as much CO2 in the atmosphere and so forth. So I do believe those things have that go on. I’m just not as versed in them.

[00:50:52] Michael Hawk: Okay. I want to talk a little bit about communication of these concepts. I mean, you’ve been exhibiting communication here, uh, so far today, but, you know, there, there does seem to be a disconnect between climate scientists and people. Sometimes, maybe it’s a. A willful disconnect, sometimes not,, why do you think there is such a disconnect between what you see as a weather scientist and, what a lot of people tend to believe?

[00:51:24] Marshall Shepherd: So what I would say, and I’ve got a couple of different answers to that. So bear with me. One, one of the things that I don’t think there’s as much disconnect as we perceive there to be. I just think the people who are disconnected tend to be very loud in their perspective of the Yale. Climate communication group surveys America every year and they do something called the six America study.

[00:51:44] And in their six America study, they find that about 10 percent of America’s just dismissive or on climate. Then there’s a group that’s just kind of not as aware, but they want to know more. And so when they’re just different levels, uh, up all the way up to alarmed. So I think some of that is just. We have a, have a quite a bit of noise out there from, from people that have just very disingenuous intent perhaps because you know, you the whole bite the hand that feeds you mentality and, and or other reasons that you might wanna sew discord.

[00:52:15] But I have a Ted talk out there if you have a chance to catch it. It’s a Ted X talk at UGA, talking about what shapes people’s perspectives on science. And in that talk, I discussed the various biases that are out there, the confirmation biases where people consume information that supports what they already believe because of their political or religious or other biases, or other beliefs, I should say.

[00:52:38] There are things like the Dunning Kruger effect out there where people feel like they know more about things than they actually do know, and they underestimate what they don’t know. and then I, I often talk about marinades. We all grew up in these political, cultural, geographic, and other types of marinades that it shaped what we believe.

[00:52:56] I am a scientist of faith, but I will often have people that come up to me and say, I think it’s blasphemous to think that humans or mankind can change the weather in any way. And I say, well, we do it all the time. Have you seen pollution or have you touched the, The surface of a parking lot in a city versus a rural landscape.

[00:53:12] There’s the urban heat island. So there’s certainly many ways that humans affect weather and climate. We see it clearly. but I think it’s just biases and perceptions and marinades, and in some cases, just more devious intent because people have a vested interest in seeing things not change.

[00:53:30] Michael Hawk: Yeah, I will definitely link to that Ted talk. And I think was that the Ted talk where you, you mentioned that. Like if 97 percent of doctors told you that you need heart surgery, but the public disagrees, like who do you believe was that the same one?

[00:53:44] Marshall Shepherd: No, I think that’s an earlier one from 2012. My saying the climate zombies, uh, Ted talk, but there’s a more recent one from about 2019, 2018 ish. , but yeah, that one’s certainly out there. I was because of again, the vast majority of climate scientists that publish and study this certainly are on the same page for the most part, but there is a small percentage of very credible scientists.

[00:54:07] Three or so percent, or maybe more depending on which survey you look at. So the analogy often used in addition to one you just said as well, if 97 percent of people, of the engineers told you not to drive across the golden bridge, it’s going to collapse and 3 percent of the engineers said, Oh, it’s fine to go across, which, what are you going to do?

[00:54:27] Michael Hawk: Yeah. That, uh, that makes it real. How do you respond when people ask, was this weather event climate induced or caused by climate change?

[00:54:37] Marshall Shepherd: I say that, you know, we have, that’s, that’s, by the way, a part of our field called attribution science, this idea of can we link climate to climate change to today’s weather? I never sort of say that it was caused by climate change, but I’ll give you a good example. There was a recent study that came out that said the heat wave in the Pacific Northwest from 2021 was 150 times likely to be that intense because of climate change.

[00:55:03] So, in other words, 100 years ago, that exact week in time, you might have gotten a heat wave, but because of climate change, it was 150 times more likely to be as intense as it was, which killed so many people. So, we can say conclusively now that some events have a likely DNA of climate change within the weather today.

[00:55:25] the strongest signals are in heat waves, intensity of rain, drought. And some aspects of hurricanes, likely the intensity aspect of hurricane, I can’t say as conclusively anything about tornadic storms right now. So if you ask me, are we seeing an uptick in tornadic activity because of climate change, I can’t be as conclusive with that answer as I could with heat or intensity of rainstorms.

[00:55:50] Although there are certainly studies that are sort of moving in that direction. So again, I was an author on a report. That was produced by the National Academies of Science about in 2016, talking about this attribution science and what we can say about certain stores. Now, it’s about 7 years old. Now, there are some updates coming.

[00:56:07] So I think a lot of even some of the things we said in that report. We likely can say with a little bit more strength, but now what I would emphasize, though, is the things that we can say are a lot more aggressive than what climate scientists may have said 10 years ago, when you asked about that heat wave in Portland, Oregon, or in the Northwest Pacific Northwest, and we’re like, well, maybe, or there would have been a little bit, a lot more uncertainty, but I think increasingly, there’s a much more conclusiveness in our ability to say something about attribution, but I still never say it was caused by something.

[00:56:39] Michael Hawk: Yeah, that makes sense. And I think that I can wrap my head around the hurricane scenario. So, uh, I think. much more cleanly than other scenarios, because you still need the right base conditions for the hurricane to form in the first place. But then if you’re giving it

[00:56:52] more

[00:56:52] fuel,

[00:56:52] Marshall Shepherd: Yeah, you might, yeah, you might get the same hurricane even in a pre industrialized environment where we didn’t have CO2 fossil fuel burning. Would its behavior be the same now? Or would it’s the way it evolved? It probably would be different now. our best thinking on hurricanes right now, although I often hear this misstated.

[00:57:10] Is that there might be actually at least in terms of hurricanes, fewer hurricanes on average as we go forward. But when they happen on average, they’ll be stronger. So you’ll often hear people sloppily say, Oh, we’re going to have more hurricanes because of climate change. That’s not necessarily the case.

[00:57:25] But when we do have them, they will likely be stronger. They may rapidly intensify. Uh, in some cases, there are some literature that suggests that they may produce more rainfall or be slower moving. So, it’s important to sort of be very precise about the things we know and don’t know.

[00:57:41] Michael Hawk: What do you think is the most important thing that listeners to this podcast can do to, to help. Mitigate or live with or whatever the proper reaction would be, to the climate crisis to climate change.

[00:57:54] Marshall Shepherd: yeah, and it’s not what a lot of people say. I mean, I mean, some people when I get that question are looking for me to say, well, buy an electric vehicle or eat less beef and all those things are fine or compost your food. Those are, those are fine, but those are still incremental solutions going to take transformational, large scale action policy action at the state, local, national, international level.

[00:58:15] So. The thing that I say is just become very climate cognizant voters

[00:58:20] Michael Hawk: Great advice. And, where do you find hope in, in the climate situation that we have?

[00:58:27] Marshall Shepherd: because we know what we need to do. I mean, we’re not, I go back to the vaccine, but when COVID hit, I mean, COVID hit and it disrupted us significantly, but we knew what to do. We knew we had to develop a vaccine or we know what we, we’re not sitting here as far as saying we’re scratching our heads saying, what do we need to do to get out of this?

[00:58:44] We know. We just have to take the action to do it. We’ve got to reduce carbon emissions. In some cases, we have to adapt to certain things that are already happening. And we have the technology to do those things quite effectively.