U.S. Hurricanes History and Forecasts, hosted by Inland Marine Underwriters Association. I'm Eileen Monreale, Education Training Specialist at IMUA. Thank you for joining us today. We have over 220 attendees registered, and we look forward to a great presentation on a topic that affects us all personally and professionally. I want to also thank my colleagues, Lillian Colson and Ellie Monreale, for working behind the scenes to plan and coordinate this webinar. You'll find your attendee control panel on your screen that it will minimize, perhaps, by default. If that happens, you can right-click, you can right-click on it to bring it back and keep it permanently opened by going to the options and unchecking the auto-five option. There is a question box in that control panel that you can type your questions and click submit. Our presenter, Phil Klotzbeck, will be sure to leave time at the end to answer all of your questions. To encourage your questions and participation, Phil will select a winner, and IMUA will email a $25 Amazon gift card for the most-slash-best-engaged attendee, which is a combo of those two criteria. We just want to keep it fun and connect with you in any way we can, so let's have a lot of engagement. Phil also shared a copy of his recently released, actually just released this week, extended-range forecast of Atlantic seasonal hurricane activity and landfall strike probability for 2023 as a PDF, and it's available in handouts along with the PDF of the presentation. Allie is also popping a link in to take you directly there, so that's a great report, a great resource. So please be aware of the IMUA statement, and now I am honored to introduce today's presenter, Phil Klotzbeck. He is a research scientist in the Department of Atmospheric Science at Colorado State University. He received his PhD in atmospheric science from Colorado State University in 2007. He has been with the Department of Atmospheric Science for the last 22 years and has co-authored on the Atlantic basin hurricane forecast with Dr. William Gray throughout 2025. Phil became the first author on the seasonal forecasts in 2006. He developed the two-week forecast currently being issued during the peak months of the hurricane season between August and October. Phil has authored over 100 articles in peer-reviewed journals, such as Journal of Climate and Weather and Forecasting. He graduated from Bridgewater State College with a BS degree in geography in 1999. After receiving his master's degree from CSU in 2002, Phil hiked through the Appalachian Trail from Georgia to Maine over 2,100 miles. I am not sure how many pairs of hiking boots he went through, but maybe we can have a polling question with that. Phil has also climbed all 54 14,000-foot peaks in Colorado and has completed nine marathons and six ultramarathons. And fortunately for us, he's going to share his expertise with us today. Please welcome Dr. Phil Potsback. Well, thanks so much, Eileen, and thanks so much to all of you for attending today. So today we're going to talk a lot about the 2023 hurricane season, but before we get there, I do want to start by talking a little bit about how these seasonal hurricane forecasts are done. We'll then spend just a bit of time going through what happened last year. Excuse me, spend the bulk of the time talking about the outlook for 2023, and then finish with a brief discussion of hurricanes and climate change, I think that's a hot topic, so to speak, that's been discussed a lot in the literature in recent years and want to kind of dive into that and unpack it in a little bit more detail. But when it comes to seasonal hurricane forecasts, a lot of people think we're like my meteorological namesake, the groundhog, Punxsutawney Phil, waking up to see if we can see our shadow, but that's not how we do these seasonal forecasts. We use a lot of historical data and a lot of statistics to come up with these outlooks. And so the seasonal hurricane forecasts were pioneered by Dr. Bill Gray all the way back in 1984. So Dr. Gray was the founder of the seasonal hurricane forecast. In addition to founding the forecast, he made fundamental contributions to tropical cyclone genesis, structure, intensity change, just a giant in the field of tropical meteorology. And I was very fortunate to have the great opportunity to work with him for over 15 years. One of the things that Dr. Gray used a lot of was he was basically an encyclopedia of knowledge. So before we had Wikipedia, we had Bill Gray. And so he knew which years in the past were active hurricane seasons and he knew which years in the past were El Nino years. And he noted that when you have El Nino conditions in the Eastern and Central Tropical Pacific Ocean, you tend to get fewer hurricanes in the Atlantic Ocean. And so then it was a matter of trying to figure out the physics as to how that goes about working. So in El Nino years, so we're gonna talk a lot about El Nino today. In El Nino years, you have warmer than normal water in the Central and Eastern Tropical Pacific. La Nina is pretty close to the opposite of El Nino, colder than normal waters in that same region. And so the question is, why would something going on in the Tropical Pacific Ocean impact hurricanes in the Atlantic Ocean thousands of miles away? And through research that Bill Gray and others did, they found that it was primarily through alterations in what's known as vertical wind shear, which is the change in wind direction and speed with height in the atmosphere. If we think about a vertical cross section of winds in the Tropical Atlantic, your winds at low levels blow out of the East. These are the trade winds. At upper levels, so you're talking 25, 30,000 feet, they blow out of the West. So in a vertical cross section, we have a shear. Low level winds blowing one way, upper level winds blowing another way, and too much shear is detrimental for hurricanes. It tends to tilt the hurricane circulation, disrupt it, and you're less likely to get tropical storms and especially hurricanes. Too much shear also tends to encourage mixing in of dry air, which squelches the thunderstorms that are the building blocks of hurricanes. Alternatively, when we have La Nina conditions in the Eastern and Central Tropical Pacific Ocean, conditions tend to be more conducive for Atlantic hurricanes, and you have overall lower levels of vertical wind shear. So that's one of the big predictors that ourselves at CSU and most other groups issuing seasonal hurricane predictions use, but there's a lot of other things as well. And so what I want to try to convince you of on this next slide is that hurricane seasons are not random. If we look at this plot here, we're looking at 10 busiest seasons since 1950, difference from the 10 quietest seasons since 1950, and we're looking at August through October sea surface temperatures, otherwise known as SSTs. Sea surface temperatures are a critical factor driving hurricane activity. So if seasonal hurricane activity had no relationship with climate, we would expect this to just be kind of a few random blotches. But because we do see coherent spatial patterns, that is also indicative of large-scale climate does drive hurricane seasons. So in addition to busier seasons having colder than normal water, which is indicative of La Nina, they also tend to have warmer than normal water in the tropical Atlantic. Makes sense since hurricanes live off of warm ocean water. More warm ocean water provides more fuel for storms. It also tends to create a more unstable atmosphere, which is more conducive for storm formation and intensification. So the question that you might be asking is, how good are these seasonal hurricane forecasts? And a couple of things I want to take you to, have you note on this plot, we've been issuing these forecasts all the way back since 1984. So these are the real-time scale of the forecasts. And a couple of things, you'll note that the scale improves from April to June to August. Makes sense since we're closer to the events that we're trying to predict. And while August is two months into the hurricane season, over 95% of your major hurricane activity, your category three, four, and five hurricanes, winds of 111 miles per hour or greater, occur after the 1st of August. So it gives us one last shot to kind of fine tune the numbers. The other thing I want to draw your attention to is the red columns are higher than the blue columns, indicative of improved skill in more recent years. I think a lot of that is due to both better historical data. So we now have a better idea of what was going on in the past, as well as what's going on now. Obviously that's helped for calibrating our models. And we also now use climate models to be able to forecast what might be coming during the season. And if you go back to when Dr. Gray started issuing these forecasts in 1984, we just didn't have climate models that were able to forecast things like El Nino, things like water temperatures, wind patterns. So that really does help us improve the skill of our forecasts as well. And when we dive into 2023 in detail, I'll spend some more time going through exactly how we come up with the outlooks that we issue. But now before we move in and talk about 2023, I did want to look back and talk about 2022. The way that I termed the 2022 hurricane season is that it was the most abnormal normal hurricane season on record. And so why do we call it normal? Well, the numbers are pretty close to average. An average season has 14 storms. We had 14. An average season has seven hurricanes. We had eight. An average season has three major hurricanes. That's category three, fours, and fives, winds up 111 miles per hour or greater. We had two. So overall, the numbers were pretty close to an average season. Another metric that I do want to draw your attention to that we're going to talk about occasionally throughout the rest of this presentation is accumulated cyclone energy. That's an integrated metric accounting for storm frequency, intensity, and duration. So long-lived major hurricanes like Irma in 2017 or Dorian in 2019 generate tremendous levels of ACE, whereas weak, short-lived storms like Arlene that formed just last week generate very low levels of ACE. So just keep that in mind when we talk about accumulated cyclone energy. It's kind of just a geeky metric that accounts for storm longevity and intensity. But the question is, how good were CSU's forecast last year? And I would say it turned out better than I thought they would at the end of August. And we'll talk more about that in just a minute. But overall, the numbers that we forecast were too high. We anticipated an above-normal season and we got a normal season. The number of hurricanes forecast was not great, or it was good, but the other metrics were way too high as a general rule. And one of the big reasons why I think that was the case was that we had an extremely quiet August, and I'll talk more about August here in just a moment. This is the track map showing the storms that do form in 2022. A couple of things I want to draw your attention to here. First of all, the Eastern Atlantic was pretty much empty. There weren't a lot of storms forming right off of Africa. We think part of the reason why that was the case was that there was a lot of dry air in the Eastern Atlantic last year. Also, I want to draw your attention, there were a lot of storms in the Western Atlantic, and unfortunately, that's not great because that's obviously where most of the land is. Two purple lines represent the major hurricanes. And on average, major hurricanes cause about 85% of the damage, even though they only make up about 20% of the storms that you get in an average year. But in 2022, the two major hurricanes that formed, Fiona, which obviously caused devastating flooding in Puerto Rico, parts of the Dominican Republic, and then caused a lot of damage up in the Atlantic provinces of Canada as a post-tropical cyclone, or a system that was a hurricane, but was now transitioned to a mid-latitude storm. And then obviously, by far, the biggest story of the year, Hurricane Ian, which made landfall in Southwest Florida as a category four hurricane, causing over $100 billion in damage, and unfortunately, over 150 fatalities as well. And so just to highlight the landfalls briefly, Ian was by far the biggest story of the year. It made its first landfall in Southwest Florida as a category four. It made a secondary landfall as a category one hurricane in South Carolina. We had Colin really early in the season as a tropical storm. Almost no one remembers that because it was a weak storm, didn't really do any damage. And then Nicole, late in the season, in early November, made landfalls, a big, sprawling category one hurricane, and did cause a fair amount of erosion. It had some pretty significant storm surge damage, especially as it followed fairly closely on the heels of Hurricane Ian just a few weeks prior. So here's a satellite image of Hurricane Ian as it was approaching the Southwest Florida coastline. It was upgraded after the hurricane season to a category five while it was in the Gulf of Mexico, but when it made landfall, it was a category four hurricane. And you can see it was a well-developed, very well-developed storm, classic eye, and unfortunately caused a tremendous amount of storm surge damage. And that was by far the biggest driver of the damage. It was the first category five Atlantic hurricane since Lorenzo in 2019. It's landfall intensity of 150 miles an hour. Tied it with five other hurricanes for the fifth strongest on record. Tied it with Hurricane Laura in 2020, as well as Hurricane Ida in 2021. $113 billion in damage from this storm. 66 US fatalities directly caused by this storm, with 90 indirect fatalities, which could be due to electrocutions, carbon monoxide poisoning, all sorts of other things as well. But unfortunately, over 150 fatalities directly or indirectly tied to Hurricane Ian, and a huge price tag, well over $100 billion in economic damage. So kind of what drove that season is, remember, as I mentioned before, we do like to focus a lot on sea surface temperature, what we call anomalies, which are differences from average. So in these plots, blues are colder than normal water. Oranges and reds and yellows are warmer than normal water. So in the tropical Pacific, we had a La Nina, a very strong La Nina, which as you remember from our earlier discussion, tends to favor above normal Atlantic hurricane activity. While in the tropical Atlantic, we have water temperatures that were a little bit above normal, and that would tend to also favor above normal hurricane activity. However, we also had colder than normal water in the subtropical Eastern Atlantic. And I think that played an important role, especially quieting down August, compared with what you would normally expect during the month of August. So here, we're going to focus on the 2022 season and kind of how it played out. So the red columns are the average values, and the blue columns are 2022. So you can see, you know, very little activity in the early season. We went from July 3rd to September 1st with no storm activity in the Atlantic Basin. The last time that we had gone that period with no storms was all the way back in 1941. So it had been a long time. But then we had a very busy September, a pretty quiet October, and then finished off the season with a busy November. We had three hurricanes from late October into early November. And so what I wanted to talk about next is why August was so dead. During the month of August, we had below normal water temperatures in the subtropical Atlantic, and near normal to even above normal temperatures in the tropical Atlantic. And when you have that strong kind of gradient or difference in temperature between the tropics and subtropics, the tropics generally are warmer than the subtropics. So when you have warmer temperatures in the tropics and colder temperatures in the subtropics, that tends to encourage more frontal activity because fronts tend to form along strong differences in temperature. And when you have those fronts, it tends to encourage a lot of mixing in of dry air, more vertical wind shear, both of which tend to suppress storm activity. And certainly we could see that during the month of August, we had high shear throughout the Western part of the Atlantic and in the Eastern part of the Atlantic where the shear was lower, we had a lot of dry air. And so a lot of times the dry air is attributed mostly to coming from the Sahara Desert. And some of that did, but a lot of it actually came along the backside of a low pressure system, which was situated to the West of that strong high pressure system that was over Western Europe that was causing the significant heat wave. So we had a lot of air coming out of the North in the Eastern part of the basin. And when you have air coming out of the North, it tends to be drier, more stable, and that tends to squelch thunderstorm activity. But then we moved in and had a very busy September with vertical wind shear well below the long-term average. And we ended up obviously with a very busy month, most notable for Fiona and then obviously Hurricane Ian. But now I want to kind of switch gears and jump from 2022 and then talk about 2023. So we did put out an updated forecast as Eileen noted on June the 1st. We're calling for a total of 15 mean storms. Of those 15, seven becoming hurricanes. And of those seven, three becoming major, category three, four, five hurricanes. And so again, our forecast is pretty much dead on average. And the numbers that we're forecasting do include both Tropical Storm Arlene that formed just a few days ago, as well as a Subtropical Storm that formed in January. At the time in January, the National Hurricane Center did not name it, but they did note in May that, hey, a Subtropical Storm did form in January. So that actually does officially, actually count as a named storm. It's an unnamed named storm. It's kind of confusing, but basically anything that's in the Hurricane Center, either names in real time, or identifies in post-storm analysis, does actually count as a named storm. We also forecast this ACE Index, but also especially west of 60 degrees west. And the reason why we focus on that region is because when you start getting west of 60 degrees west, that's where people live. So when storms are strong in that area, that generally generates a lot of angst. Basically because obviously people are very nervous about that storm because it could obviously cause very significant impacts. So we forecast ACE west of 60 separately, and I'll talk a little bit about how we disentangle that from base and wide ACE here a little later in this presentation. But as I mentioned before, we focus a lot on sea surface temperature anomalies or differences from average. And the thing I want to draw your attention to first is there's a heck of a lot of red. Basically everywhere right now is much warmer than normal. And so the last few years, we tend to have colder than normal water in the Eastern and Central Tropical Pacific, as La Nina conditions tended to dominate. But now those water temperatures have been replaced by warmer than normal water. So we're not officially up to the El Nino threshold yet, but we're likely to get to that El Nino threshold here very shortly. We also, if we look at that area denoted by the red rectangle, that's known as the Atlantic Main Development Region. As you would expect, the main development region is where most of your hurricanes form in the Atlantic Basin. So we kind of have these two pretty strong factors where, El Nino is looking extremely likely and El Nino tends to cause quieter Atlantic hurricane seasons via those increases in vertical wind shear that we talked about. Whereas the Atlantic is also extremely warm and the warm Atlantic tends to favor an above normal Atlantic hurricane season. So now shifting gears and focusing solely on the Tropical Pacific Ocean. If we look at this figure here, we're looking at sea surface temperature anomalies. That blue rectangle denotes the Nino 3.4 region. That's what happens when you have scientists naming stuff. They don't always come up with the most exciting names, but that region is what NOAA uses to monitor the strength of El Nino or La Nina. And these temperature changes in that box are fairly small. If the box is half a degree Celsius or more warmer than normal, or about one degree Fahrenheit or more warmer than normal, NOAA will declare an El Nino event, provided those warm waters persist for a few weeks. Alternatively, when the water temperatures in that box are colder than normal, half a degree Celsius or more colder than normal, NOAA will declare a La Nina. So this week, the water temperatures in that box are actually eight tenths of a degree Celsius, warmer than normal. So it's pretty likely we're going to see El Nino declared in the next few weeks. NOAA does tend to be fairly conservative in its declarations. And also they do want to make sure that those warm waters persist before declaring an El Nino event. You don't want to declare an El Nino event and retract it several weeks later. But one of the things that we monitor very closely when we're looking for El Nino or La Nina is what's going on across the tropical Pacific Ocean. So here is winds averaged across the, basically averaged across the entire tropics from five south to five north. And here, the center of the plot is an international dateline. So you can see basically reds here are westerly wind anomalies, or differences from average, and blues are easterly anomalies. And if we look at this figure here, we can see a very strong westerly wind event that occurred during the middle to latter part of May, partially associated with Typhoon Maywar, which was, which actually became a super typhoon, extremely powerful storm, as it tracked northwest across the western North Pacific. But why do we care about these winds? And the reason why we do care about the strength of the winds blowing across the tropical Pacific is because if they're strong out of the east, that tends to basically prevent El Nino development, whereas when they develop when they're stronger out of the west, that tends to enhance El Nino development, because on average, your waters in the western tropical Pacific are much, much warmer than in the eastern tropical Pacific. So when you get westerly winds, it basically transports that really warm water eastward. So it just had a strong westerly wind event that started to transport that warm water eastward. And we can actually see it in this next schematic here, where if you look at the temperatures beneath the ocean surface, you can see those warm colors going from west to east across the tropical Pacific. And that's a sign that we are seeing, again, continued transition over to El Nino. And again, probably in the next few weeks, NOAA will officially declare an El Nino event. So the big question is, is El Nino coming? Hopefully some of you remember this classic Chris Farley sketch from Saturday Night Live about 25 years ago. Are we going to see El Nino this summer and fall? And I think the odds are quite good. Here is the latest forecast from the European Center for Medium-Range Weather Forecasts or ECMWF. We like our acronyms here in meteorology. And you can see that all the model members are calling for El Nino this summer and fall. Most of them calling for a greater than one degree, or all of them calling for a greater than one degree Celsius warm anomaly, which would be at least a moderate strength, if not a strong El Nino event. The way that these ensembles are done is you basically take your initial observations, which you know have some uncertainties associated with them. And, okay, say for example, you have a buoy and it says it's 83 degrees Fahrenheit. Well, you know, it could be 82.8, 83.2, there's some margin of error there. But what we can do is use those margins of error to basically put in uncertainties into the forecast to try to kind of give you a better idea of the uncertainty associated with these forecasts. So the European Center model, you know, is very aggressive at calling for El Nino this summer and fall. Likewise, the formal forecasts from NOAA is also very aggressive at calling for El Nino and 94% chance for August through October. And the reason why I've got the arrow over August through October is because even though it's half the hurricane season, it's over 90% of your major hurricane activity that occurs during those three months. So the question is, you know, what the heck does El Nino and El Nino mean for hurricanes? Well, El Nino tends to mean more hurricanes and also stronger hurricanes compared with El Nino. And again, a lot of that is due to lower levels of vertical wind shear in La Nina compared with El Nino. It's about a two and a half to one difference. You get about twice, two and a half times as many major hurricanes as well as major hurricane days, which just means the number of days that these storms spend at major hurricane strength. Whereas in El Nino, you tend to get a lot fewer major hurricanes as well as major hurricane days. One of the things you will note is that you especially get a reduction in storms, making landfall along the Florida Peninsula and East Coast in El Nino years. And that's because El Nino tends to really knock down storm activity, both in the Caribbean, but also farther East in the Atlantic. Whereas in the Gulf of Mexico, it does reduce your storm activity, but not as much because shear isn't as strongly related to El Nino in that region. But also when you get storms forming in the Gulf of Mexico, sometimes they actually form off of things like fronts and they're not necessarily going to have a strong relationship with La Nina versus El Nino and some of these other phenomenon. On average, for a continental US landfall, you get about twice as many hurricanes in a La Nina year as an El Nino year, and about three times as many major hurricanes. Now I want to shift gears and talk about what's going on in the North Atlantic. And you can see it's really, really warm right there. I believe that was quoted in one of the, I think it may have been, you've been on CNN, and it's really stinking hot. And it is, it's extremely warm. And so these temperature anomalies of half a degree to one degree Celsius don't seem like a lot, but one degree Celsius temperature difference in the tropics is huge and makes a big difference in how the atmosphere then responds. You can see certainly warm waters almost everywhere right now. We like to focus during the springtime on this box denoted here. When waters in that box are warmer than normal, that is very critical for the strength of the subtropical high region. When the waters are warmer than normal, you tend to associate that with lower pressures. And where you have lower pressures in the subtropics, the subtropics normally have very high pressure, the tropics have low pressure, and that's effectively what drives your trade winds. Whereas when the subtropics have lower pressure, you tend to get weaker winds blowing across the tropical Atlantic. And weaker winds mean less evaporation, less churning up of the ocean surface, and you're more likely to get warming faster than normal. Obviously, always North Atlantic warms from spring to summer to fall, but when you have weaker winds, it tends to warm even faster. But you can see if you toggle back and forth that the current sea surface temperature pattern and the optimal pattern for a busy hurricane season are pretty close. It's not a perfect match, but it is quite close. And especially that water in the subtropical Eastern Atlantic which is extremely warm, is again, typically a harbor driver well above normal hurricane season. So you can kind of see why there's a lot of kind of angst and stuff going on about this season. And now I want to kind of just briefly highlight a few of the parameters that we use to come up with our forecast. One of the tools that we use is a statistical model, and this would be very similar to what Bill Gray used all the way back in the early 1980s. A statistical model again uses historical data available up through the time of the outlook. So since this outlook was put out June 1st, we use data through the end of May. And so we use two predictors, one in the Eastern Atlantic, again, showing those water temperatures when they're warmer than normal that favors an above normal season. And we also look at winds blowing across the tropical Central and Western Pacific. When those winds are blowing stronger out of the West, that indicates an El Nino is likely. Whereas when those winds are stronger out of the East, that tends to indicate that La Nina is likely to occur. And you can see a statistical model has reasonable levels of skill. So this is hindcast skill. So basically how well would the forecast have worked? Have we had it in each of these years? And you can see it does a reasonably good job. There are some years where the forecast busts, and that's to be expected because you're using, there's uncertainty in the observations, but also even if you have perfect knowledge of the large scale conditions, hurricanes at the end of the day are weather events. And so while climate conditions will dictate overall how much storm activity approximately forms, you could, for example, get a strong storm that forms in just one small pocket of conducive conditions. In an otherwise harsh year. We also use a statistical dynamical model. And this is a hybrid model. It basically takes forecasts from various climate models and says, okay, assuming those forecasts are perfect, how much do you, or what are you forecasting for water temperatures both in the tropical and subtropical Eastern and Central Atlantic, as well as in the Eastern and Central Tropical Pacific where La Nina tends to develop? So this is something Bill Gray did not have at his disposal because we just didn't have climate models that could forecast water temperatures. And the nice thing with climate models is that they're built on hindcast data. So you can basically see using today's climate model, how well would it have forecast years in the past? And so with dynamical models, in general past performance is a pretty good indicator of future results. And so when we combine the forecast of these two models in combination, we can get reasonable levels of skill. So we use a variety of different models, including the European Center for Medium Range Weather Forecasts. And this model is calling for a very strong El Nino. So our El Nino box is very warm, but also the Atlantic is warm. So it's these two predictors kind of in combination again, push pulling, tugging each other one way or the other. I mean, you can see again, this model does show reasonable levels of skill going back to the early 1980s, which is as far back as the European Center model has its hindcast skill documented. We also use analogs, which involves going back into the past and looking for years in the past that had conditions most similar to what we currently see and what we expect to see for the peak of the Atlantic season. Now we know like snowflakes, no two hurricane seasons are exactly alike. And there's not a lot of great analogs for this year because I think the odds of a pretty strong El Nino are quite high, but also the Atlantic is extremely warm and that's not a typical pattern. When we look back at some of the strong El Ninos we've had in recent years, such as 1997 or 2015, the Atlantic was maybe a bit warmer than normal, but not nearly as warm as what we're seeing right now. So it's not a great analogs for this current situation. However, the next time we get a really strong potential El Nino developing at the same time, the Atlantic is extremely warm, we'll have 2023 as a good analog year. Just to show you the activity that was experienced in those five analog years, we have 1951, 1957, 1969, 2004, and 2006. Those five years, the activity spread was huge between a little below normal in a year like 57 all the way to a hyperactive season like 2004. One thing to draw your attention to is just because the year is selected as an analog, say 1969, not saying we expect to see per se another Hurricane Camille, excuse me, like we saw in 1969. Excuse me, with these forecasts, we also have uncertainty. And so the numbers on the left are the forecast, but obviously we know that these forecasts aren't perfect. So the uncertainty is listed on the right hand side in that column. One thing I do want to point out is the reason why we give you kind of the midpoint or the best estimate is because a lot of it is just for general public interpretation. The general public would take, or the media will take a forecast say of seven hurricanes, or if you give the uncertainty range and then say five to nine, the media will generally report that it was up to nine, which then means nine to the general public. So we thought by kind of giving people the midpoint, but also publishing the range helps kind of clarify the uncertainty associated with these forecasts. And certainly you can see there is a fair amount of uncertainty associated with these forecasts based on historical data, just because again, even if you get the climate situation even perfect, there's not a guarantee you're going to get a specific amount of hurricanes with that climate. The climate certainly loads the dice, but again, hurricanes are weather events. So with that, I kind of want to open it up to the general public or to the attendees here to see how many hurricanes you think the Atlantic will get this year. So if you could please launch the poll, I'd be very interested to see what people are saying. Okay I can't see the poll results. So less than six we have 12%. Six to seven, 22%. Eight to nine, 39%. Ten to 11, 14%. And more than 11 is 12%. So the dominant answer was that eight to nine. All right okay so going a little bit even more aggressive than what we did at CSU. That's good to see. So yeah we'll see what happens. There definitely is you know a potential for a lot more activity than what we predicted. If the Atlantic remains as warm as it is right now relative to normal and continues to warm at that rate. But certainly also too if El Nino comes in and is extremely strong that could really knock down the storm activity. So a lot remains to be seen. This is a forecast with a lot of uncertainty and we will update the forecast a couple of times. We'll talk more about the updates here in a few minutes. But certainly stay tuned because there is the potential the numbers could go up or down a fair amount with the July and early August updates. So now moving ahead. One thing I did want to draw your attention to is we do forecast accumulated cyclone energy west of 60 degrees west. And so the reason that we forecast that is because that's where your land masses are. And so you know it's really hard to forecast number of landfalling storms. But what we can do with some skills forecast accumulated cyclone energy in various regions. If you have a lot of accumulated cyclone energy west of 60 degrees it tends to correlate pretty well with landfalling hurricanes. And this is landfalling hurricanes anywhere in the North Atlantic. So I'm not just saying the United States. This could include Mexico, the Caribbean, Central America. And while these places may not get overall as much financial damage because they don't have as much infrastructure. You can actually see more loss of life and obviously a lot higher percentage of GDP than you would see with hurricanes hitting the United States. But you can see with Basin Wide Ace there is a decent relationship. But if we use west of 60 west and North Atlantic landfalling hurricanes, we do see an improved overall relationship there. And so you know you can basically take weights west of 60 is just a percentage of Basin Wide Ace. But you tend to see a little less Ace west of 60 relative to the Basin Wide Ace in El Nino years. And the reason why that's the case is because storms in El Nino years tend to be a little more likely to recur. But also the Caribbean tends to be extremely harsh in El Nino years. You get a lot of shear there. And since Ace is a function of intensity and duration, the Caribbean is generally extremely warm. So when storms get into that region, they have the potential to intensify to quite high thresholds. So in El Nino years, the Caribbean tends to have a lot more shear. You're less likely to get rapid intensification in the Caribbean. And consequently, overall, you get more Ace in the eastern part of the basin in El Nino years, which is good. That's where we want to see the Ace where no people are living. Obviously, when you get eight storms west of 60, people get nervous because likely someone is going to be impacted, potentially very significantly. Now, we also do probabilities of landfall. And so these are probabilities for the 2023 season. And these probabilities are generally adjusted by the forecast, with the idea being that more active seasons have more landfall in hurricanes. It doesn't work out every year. You can have a very busy season like 2010, where we wrote a great streak of luck and had 12 hurricanes in the Atlantic Basin and no landfalls. Or like 1992, where we forecast one major hurricane at CSU. We got only one, but that was Hurricane Andrew. But in general, busier seasons do have more landfalls. This year, our probabilities are pretty much dead on average because we are forecasting pretty much a dead on average hurricane season. So the numbers on the right are our climatology or the long term average, while the numbers to the left are for the 2023 season. We also provide probabilities at finer scales. So here we're looking at storms tracking within 50 miles of each county or parish if you're in Louisiana. And the reason why we use a 50 mile buffer is because you don't have to have a landfalling storm in your county for it to cause significant impacts. Obviously, if a storm makes landfall in an adjacent county, you could still be very heavily impacted. So certainly if there's a hurricane within 50 miles of your county, you're looking to see at least some damage there. And so here we're looking as an example of Plymouth County. We're looking at all hurricanes within 50 miles of Plymouth County. The reason I selected Plymouth County is because that's the county I grew up in. But if you look, we have 10 hurricanes in those within 50 miles of the county. So you can basically take those 10 hurricanes over 141 years from 1880 to 2020 and calculate probabilities based off of that. So again, we have this for every coastal state, every coastal county, coastal parish, all the Atlantic provinces of Canada, coastal states of Mexico and various countries in Central America and the Caribbean that are hurricane prone. So just to give you an example, for 2023, the probabilities on the right hand side are climatology, while the probabilities to the left are for 2023, again, dead on average for all these quantities. One thing, one kind of good example of why having a 50 mile buffer is, for example, Hurricane Sally in 2020, which made landfall in Alabama, but it still brought very significant impacts to the western part of the Florida Panhandle because it was very close to the border. So here by using this buffer that kind of helps to smooth out some of that noise. So if you go to our website, you can get more and more information about kind of how these probabilities change and what the probabilities are for various places along the U.S. coastline. These forecasts are not stagnant, however, we will issue an update on the 6th of July and a final update on the 3rd of August. And while the 3rd of August, again, is two months into the season, you still have over 90% of your major hurricane activity to go, more like 95%. And it gives us one last shot at a mulligan if we realize we really messed up up to that point. But it really more seriously does give us a better chance to kind of fine tune everything. What happens in June, but especially what happens in July, is critical. The atmosphere in the tropics tends to kind of get locked into whatever it's going to do during July. So there's a lot of persistence basically between July and, say, August. If the shear is low in July, it tends to be low in August and vice versa. So we'll certainly be very closely monitoring things as we move ahead throughout the rest of June and throughout July. But if you think, hey, CSU stinks, we want to look at other people, I invite you to check out seasonalhurricanepredictions.org. We have over 25 different groups submitting seasonal forecasts. These groups include other universities, private sector weather companies, as well as government agencies. And on the website, there's links to all the different groups, the methodologies, how they use. But also we have all the different forecasts displayed. So here they're color-coded. Blues are government agencies, purples are private sector weather companies, and greens are universities. One thing I'll draw your attention to is there's a lot of spread this year. For example, the UK Met Office, Skyfora, have gone for extremely busy seasons, while other institutes like a couple of the institutes in Mexico have gone for very low seasons. So basically you can see the ones that are going for the busy seasons are more weighing the warm Atlantic strongly, whereas the ones going for the quieter seasons are weighing the El Nino strongly. CSU is the second forecast from the left, so we're kind of near the middle, but it's kind of nice to show just how big the spread is this year. There is a lot of uncertainty associated with the 2023 hurricane season and the various outlooks issued by these groups. But now I want to shift gears and just in the last few minutes, talk a little bit about some longer-term trends that we see in hurricane activity. And so to start with, we're going to look at some observed trends since 1990. And the reason we're starting in 1990 is because data prior to 1990, especially globally, isn't great. For the Atlantic, you can go back further, but globally, we're very dependent on satellite imagery. And satellite imagery prior to like 1990 really wasn't very good. It was kind of coarse resolution. So you have a hard time detecting storms, but especially hurricanes or major hurricanes, rather, super typhoons. Hard to tell exactly how strong these storms are. But since 1990, the data is reasonably good. So we published a paper looking at these trends last year, and you can see that for the number of storms globally, there's a very slight increasing trend, but not significant. And there's a lot of uncertainty in the scientific literature exactly how that's going to play out in the future, how many storms we're going to get. But one thing that we have observed in recent years is that we're getting more of these short-lived named storms, storms lasting two days or fewer, and it's really jumped out in the Atlantic Basin. We have this really large increasing trend from about two of these per year in 1990, averaging now close to six per year in 2022. So the question is, you know, why are we seeing more of those storms? Is it actually climate signals, or is it more due to just improvements in observations? And so, you know, going back to 1990, we could observe a hurricane pretty well, but these weaker storms, we just didn't necessarily have the capabilities to observe. And there's a couple of big technological advances which have really helped at observing these weaker storms. So this is a nice figure that Chris Lancy put together a few years ago, highlighting how observations have changed with time. Certainly flying planes in the storm, starting in the mid-40s, better satellites, or having satellites and then better satellites. But really two things have really helped with the weak short-lived storms, both scatterometers, which help us to be able to assess surface circulations. We can measure the winds quite accurately, and we can also tell if the circulation is closed, which is critical for determining whether it's a hurricane. Also, we have microwave data and the cyclone phase-space diagram, which helps us to be able to determine whether the storm truly has a warm core. And the warm core is critical for whether it's a tropical cyclone or a mid-latitude cyclone. So all these technology is likely one of the big reasons why we're observing increases in the short-lived storms. Now, if we step ahead and look at the global number of hurricanes slash typhoons, so not just the Atlantic, but globally, we're actually seeing a significant decrease. So while the Atlantic has gone up since 1990, the rest of the globe has gone down, especially the Western North Pacific. And that's critical because the Western North Pacific is a much larger ocean basin. You tend to get a lot more storms there. So when that basin goes down, it tends to knock down, overall, your global frequency of hurricanes slash typhoons. If we look at the next slide, we can see, though, at the same time, the Category 4 or 5 hurricane percentage. So basically, the odds of any Cat 4 or 5 or the odds of any hurricane reaching Cat 4 or 5 have gone up. So kind of like we have with El Nino versus the really warm Atlantic, here we have a tug-of-war between the decreasing number of typhoons and hurricanes globally versus the increase in the Category 4 or 5 percentage. And obviously, this is critical in terms of judging potential future threats from storms. If the number of typhoons goes down a tiny bit and the Cat 4 or 5 percentage goes up a lot, we're going to see more overall number of Category 4 or 5s. But if they kind of cancel each other out, your hurricane threat, at least the wind threat, would actually go down. Because even though the Category 4 or 5 percentage would stay the same, the Category 1 to 3 hurricane typhoon percentage would actually go down. So that's a big question that we still don't really quite know the answer to. And one of the big factors that we think is leading to the decrease in global hurricanes slash typhoons is this trend towards a more La Nina-like environment. And so as I mentioned earlier, La Nina tends to increase your Atlantic storms, but it tends to decrease your Pacific storms. The cooler water tends to kind of suppress storm formation, and it really tends to shift storm formation in the western North Pacific farther to the north and west. And if you shift formations to the west and north and northwest Pacific, those storms just have shorter times over water, so they're less likely to be able to become typhoons, super typhoons, and they also just generate lower levels of ACE. Because one of the big drivers of ACE is duration. So if we look at the trend in the last 30 years since 1990, sea surface temperatures, we can see, you know, it's been warming almost everywhere, and those slashes highlight statistical significance. But if we look in the eastern and central tropical Pacific, we see no trend or even a little bit of cooling. So overall, the environment has trended more towards La Nina. And if we think about El Nino as measured by a longitude index, typically in El Nino years, it tends to shift thunderstorms farther to the east, and that increases westerly winds over the Atlantic. So when this longitude index is shifted westward, that's more typical of La Nina, and when it's shifted eastward, it's more typical of El Nino, and we can see that that trend has been significantly trending westward. So this year, notwithstanding, overall, the trend has been towards a more La Nina-like environment. So now I want to shift gears and talk a little bit about hurricane losses. Just have a couple more minutes to go. If we look at hurricane losses adjusted for inflation, we see this huge increasing trend. So the question is, you know, what's driving this increasing trend? Is this climate change, natural variability, changes in demographics? And a lot of it, we think, is just changes in demographics. We've seen a huge increase in population, about 50 million more people along the east coast. So east coast basically meaning from Texas to Maine. And so obviously these 50 million more people along the coast, that increases the exposure or the threat to damage from storms. And obviously these people that are moving there are buying houses, buying condos, townhouses, and so we're seeing a big increase in housing units as well. And so these big increases that we've observed since 1950 are really causing, you know, a really big bump up in the exposure that these hurricanes cause. So what we want to do next is talk about normalized damage, which is basically adjusting damage that was caused at the time to how much damage the storm would likely cause if it were to make landfall today. And as an example, we're going to talk about the 1926 Great Miami Hurricane, a large in size Category 4 hurricane slammed into downtown Miami. Because it was large, its winds would have also really buffeted Broward County, all the way up probably into Palm Beach County. But at the time, the storm caused about $80 million in damage. And obviously now we get tropical storms that cause billions of dollars in damage. But, you know, Miami-Dade County was a lot different in 1926 than it is today. For example, in 1926, Miami-Dade County's population at that point, just Dade County, was about 100,000. Now it's about 2.7 million. Obviously you can see there's a lot more stuff on Miami Beach now than there was in 1926. So if we take the observed population today, and so adjust the population in 1926 to the population today, as well as adjust by various economic metrics, we estimate that storm would cause about $200 billion in damage. And that is the most damaging hurricane since 1900, if it were to reoccur today, at least estimated. And so certainly, you know, if we were to get these storms today, we could see damage even higher than what we saw, say, with Hurricane Ian, Hurricane Katrina, or even Hurricane Harvey. So if we look at the normalized losses going back to 1900, we see basically a flat line, a very slight increasing trend. Doing a lot of losses in the past several years, but the normalized losses match up a lot better with the actual, looking at the actual number of landfalling hurricanes. Going back to 1900, we basically see a flat line in the number of landfalling hurricanes since 1900, as well as the landfalling major hurricanes. Basically no trend in those two quantities. So we would expect that the storms are getting stronger, and we're not seeing more landfalling major hurricanes. The damage trends should be a fairly flat line. I always advocate if you're looking for climate change trends, you want to look at the actual physical phenomenon themselves, as opposed to looking at things like damage, because obviously damage becomes a lot more demographics and other things, which are, you know, mostly the demographic changes, which aren't necessarily due to climate change, but can cause significant changes in the signal. But we do know, you know, damage is going to likely continue just because more and more people are moving to Florida, about two to three million more per decade. And so obviously with that increase in population, again, you're increasing the exposure. Obviously we hope with the improved building codes, that that will reduce some of this increase in future damage. Also, we look at sea level rise. We do know the sea levels are rising, and while sea level rise of a few inches, you know, we've gone up about eight, nine inches since 1880. It doesn't seem like much to me in Colorado, but obviously when you live right along the coast, small increases in sea level can cause big differences in storm surge inundation, which basically is how far inland the storm surge penetrates. And obviously you don't need six feet of salt water in a house to condemn it. Just six inches or a foot of salt water can be enough to cause massive problems and potentially even condemn a house. So that sea level increase is something we know very well. I think that's one of the big things when people ask me about climate change and hurricanes, is one of the big things that I emphasize. I also emphasize protected increases in heavy rainfall, just because with climate change, you warm the entire atmosphere, not just the ocean surface, but also throughout the atmosphere. And that warming of the atmospheric column means that the hurricane will likely produce more rainfall because a warmer atmosphere can hold more moisture. Storms may also go slower in the future. There's more uncertainty on that point, but obviously if storms go slower in the future over land, then that could potentially increase the rainfall threat as well. So thank you so much for putting up with my blather for so long. I'd like to always finish with an admonition from Arago, who was the head of the Paris Observatory in the mid-19th century, who stated that, never no matter what may be the progress of science, well, honest scientific men, keep in mind this is the mid-19th century, so we can now say honest scientific men or women who have regard for their reputations venture to predict the weather. And so with that, I will take any questions that you may have. And thanks again so much for your time today. Excellent. Thank you so much, Phil. That was fascinating. And you certainly go through it at a good clip. So a couple of questions. One, Michelle asks, I think on the last few slides when you were going through them, what about Hawaii? Yeah, so here we're focusing on the Atlantic basin. Hawaii, you know, obviously most years it's not impacted by storms. In El Nino years, you actually are more likely to get storm potential threats towards Hawaii. And that's because East or North Pacific storm tracks tend to shift a little farther to the West. And in the Central Pacific, which typically is fairly non-conducive for storm formation, you can get a lot of shear near Hawaii, that tends to relax somewhat. The threat to Hawaii is typically increased in El Nino years. Obviously we don't see, you know, imminent threats to Hawaii right now, but when it comes to Hawaii, you know, really significant landfalls, you're likely to have to get a storm coming more out of the South than out of the East. That's because if you start to get storms out of the East, the waters temperatures tend to be cooler as they come in. And also you tend to get a lot of shear just to the East of Hawaii. Whereas if it comes from the South, the waters are warmer and the shear tends to be lower. So you end up with like a Hurricane Iniki, which obviously caused tremendous devastation on Hawaii back in 1992. But obviously in general, Hawaii thankfully is kind of a really small target and a huge ocean. So thankfully most years, Hawaii is not significantly impacted, but in El Nino years, the threat to Hawaii does increase. Excellent. Truxton asked early in the presentation, so you covered it to some degree, but has the average or normal temperature ranges been increasing over the last couple of decades or are those normal in quotes static? Okay. So NOAA, which is what we use for our normals as well, the National Oceanic and Atmospheric Administration uses 30 year averages that change every 10 years. So for example, from 2011 to 2020, they use the 1981 to 2010 average. Then starting in 2021, we use the 1991 to 2020 average. And since going back in time, the last 50, 60 years, we've seen a fair amount of warming. Those quote unquote normals are warming with time. So if you were to plot the sea surface temperatures right now relative to a 1971 to 2030 year normal, the water temperatures would look even warmer. So that's one of the reasons why one of the things that we focus on a lot is gradients or differences in temperature. Because obviously with climate change, we generally are seeing warming everywhere. But what really matters for hurricanes isn't necessarily the overall temperature, but it's how the temperature say in the Atlantic is relative to the Pacific. So basically when the Atlantic is super warm and the Pacific is relatively cool, it tends to force more air going up in the Atlantic, which tends to support strong thunderstorms, tends to support hurricanes, reduces the shear. Whereas in a year like this year, there's a lot of question because the Atlantic is super hot, but the Pacific is also likely to be super hot. So it really depends on exactly how hot each one is, is what's going to determine if the storm activity tends to pile on in the Pacific Ocean or if it tends to pile on in the Atlantic Ocean or perhaps maybe even more closer to average everywhere. So that's really the big uncertainty that we have. Well, I guess there'll be an interesting year for sure. So Peter asks, what impact do you see from the new tropics satellite? Okay, so, you know, basically what we do is, you know, all the observations that we can get the better, because basically what happens is when you're looking for forecasting an individual hurricane, the more observations that we have, especially what the hurricane looks like right now is critical. So for example, when we fly into hurricanes, so hurricanes, when they're out over the open Atlantic, not near land, we generally just monitor via satellite, but when they get closer to land, we actually fly airplanes into them to get a better idea of exactly how strong the storm is, the structure of the storm, and when we put that into the models, that helps. We also do actually launch weather balloons. We launch more weather balloons when storms are close to land to try to kind of better help understand exactly the kind of the large scale kind of steering flow. So basically the atmospheric wind patterns that are going to push the storms one way or the other. So basically, you know, any one individual satellite, you're probably going to see, you know, pretty small impacts, but basically higher resolution satellites, more observations is better for individual storms. And also the better that we can kind of know what's going on right now, that also is critical for when we do seasonal hurricane predictions in the future, because if say, you know, we thought the winds were 10 miles an hour and they were five miles an hour, that makes a big difference when we're going back and trying to calibrate our models using historical data. All right, do we have any other questions? Sorry, I think maybe we've lost them. I'm sorry, I'm sitting here reading the question. To myself. To myself. I've noticed a predictive, Rick asks, I've noticed a predictive forecast on actual landfill location versus cone forecast. Are there plans to return to a cone forecast as we've been hearing that preparations slash evacuation plans are based on landfill prediction rather than wider cone? So I think, so in this case, we're forecasting for a season. So the National Hurricane Center does still have like the cone of uncertainty that goes out with its forecasts. The challenge with the forecast cone that the Hurricane Center puts out is that is based on historical errors associated with storms. So it's based on the last two thirds of the last five years worth of forecast would fall in that cone. So over time, because the forecasts get better, that cone is shrinking. But we know each storm is different, kind of like each snowflake is different, each storm is different. Some storms, the uncertainty is larger, such as for example, Hurricane Ian last year, there was a lot more uncertainty associated with that particular storm. And so that makes it more challenging. So basically right now, the cones are static, but hopefully in the future, maybe we'll get to more dynamically based cones because we do know in general, we don't necessarily know where the storm is going to go, but we kind of, you know, kind of, I guess, like what was it, Donald Rumsfeld, they talk about the known unknowns and the unknown unknowns. That's more of a known unknown. We can at least know that the forecast is less certain, and that's valuable information. If we know, if we're super confident in a forecast, that's also valuable. So there's a lot of pretty good guidance that the Hurricane Center does provide in addition to the cone, where to provide ideas of the impacts, which is obviously what's most critical, because obviously, you know, the hurricane is forecast as a point, but obviously we know the impacts go well beyond the center of the storm. And so obviously larger storms are going to have impacts spread out fairly far apart. Here, we're just looking overall more at seasonal stuff. This isn't talking about individual storm landfalls. All right, opening my mic this time. So we will end with a question that is at home for you. What is, and this is from Marina, what is triggering the more than usual rainfall in Colorado almost every day in the past few weeks? Well, that's near and dear to my heart, where I've had eight and a half inches at my house in the last 30 days. It's extremely green. I'm mowing my lawn almost every day, which is great. It's been certainly great for the drought, but you know, we've tended to have kind of an area of low pressure kind of lingering in the Southwest United States, and the circulation around that area of low pressure being counterclockwise, has basically been kind of bringing in flow out of the South and Southeast, which is basically tapping moisture, actually from the Gulf of Mexico. But yeah, it's been quite remarkable, just, you know, day after day after day of rain. We've had a few breaks here and there, but it's been a pretty juicy setup for Colorado, and it looks to continue for about the next week, and then potentially looking more towards summer light weather, as we move more towards the later part of June. Excellent, excellent. Okay, well, we are at time, and I want to thank you so much for sharing your expertise. Everything was so fascinating. I, you know, look at the potential hurricanes for 2023, and the trends, definitely a lot of great resources and a springboard for further learning. I love the resources you gave us, so thank you so much. Ali just posted a survey link into the chat, so our attendees, please take a minute to complete the survey. Important feedback, because we base our future webinars off of it. So, Bill, now you have to tell us who has the most slash festing days. Well, it's kind of hard, because I can't see the questions, so I don't know who all asked the questions. I know you mentioned names. Did anyone ask more than one, or did we just have a bunch of people with one? I think it was all just one. It's kind of tough, because I can't see the questions, so it's hard to remember what you said. Any good questions stand out in your mind? We'll look back on it, and that's totally fine. Sorry, I can't see the questions at all, so it's kind of hard. Actually, I could probably make you a presenter, and then you would see them. Hold on one second. Make organizer, yes. So now you should have that Q&A pod in your control panel. But I bet it's not going to show you the questions, because it came in late. Yeah, sorry. So we'll send them to you, and then you can pick it. How about that? Yes, that would be good. That sounds good. Excellent. So everybody will stay tuned, and there'll be a lucky winner that will receive an email. So thank you again, Phil. It was really fascinating, wonderful. Maybe we can have you back next year to do next year's forecast, because every June you offer the updated forecast, right? That's correct. We'll hopefully do it next June as well. That would be wonderful. So that ends today's webinar. Thank you again, and thank you for our interesting questions to our attendees. This ends today's webinar. If you want to just pop through the last two slides, Phil. There you go. So we have your contact information again. It's all on the slide deck that you received. So thank you so much, everybody. Have a great rest of the day.

hurricanes

US

2023 hurricane season

forecasting

El Nino

La Nina

climate phenomena

historical data

forecast uncertainty

climate change