 We'll start our next lecture is by Lance Bosart. Lance is a distinguished professor in the atmospheric and environmental sciences department at SUNY Albany. Lance has broad research interests in planetary scale, synoptic scale and mesoscale meteorology. Lance works on variety of multi-scale research problems that relate to weather and climate of higher and mid-latitude regions as well as tropics. Research problems that Lance works on involve winter storms, hurricanes, organized convective systems, and the predictability of individual flow regimes that are especially attractive to him. His current research projects focus on observation and modeling studies of synoptic and mesoscale phenomena from a multi-scale perspective. Yeah, Lance really enjoys working with students and from what I've heard from his students, his students really enjoy working with him as well. Just on a personal note, Lance organized the ASP colloquium on weather and climate intersection or interactions about a decade ago. And Lance was one of the main organizers and that ASP colloquium was a huge success. And a lot of what we are doing now is also modeled on that ASP and it had a huge impact on my career as a student participant then. So thanks, Lance, for being a long-term ASP colloquium organizer and contributing to a community. You're welcome, Anisha. It was a great having you in that colloquium back in 2012. So today I want to talk to you about some synoptic dynamic meteorology concepts, concepts that are relevant to when you think about the S2S time scale problem. And I want to acknowledge my two colloquium students, Tyler Light and Alex Mitchell, because they're very good at keeping me on my toes and making sure that I toe the line, so to speak. All right, so start out with a little overview of forecasting skills, because there's some issues that tend to get lost in all the uncertainty or all the arguing about whether it's a good forecast or a bad forecast. I'll also talk about a couple of interesting West Coast forecast uncertainty events from January 2017 and February 2019. We need to talk a little bit about potential vorticity and potential vorticity streamers. And then in one of the big elephant, one of the elephants in the room and predictability issues is we're curving and transitioning Western Pacific tropical cyclones, create all kinds of downstream forecast mayhem, and then conclude with some discussion of Q vectors and jet streams and to try and point out why that matters for some S2S forecasting. So let's start out with an overview of forecasting skill. This is taken from the European Center. It's the anomaly correlation coefficient for over a 41 year period at 500 millibars from 1981 to 2021. And it's day three, five, seven and 10. All the colors you see on the left, the gap that's between the Northern Hemisphere and the Southern Hemisphere. So there's more skill than Northern Hemisphere, but the advent of excellent satellite coverage in polar orbiters in particular here in the period of 2000, 2004 basically closed the gap and it also increased the skill. But you see here, it's think the curves are leaning over and starting to flatten. So that's a sign that we're starting to saturate our skill. And if you look at the continuous probability skill score that Tim Palmer really likes and I think it's the best way to verify probabilistic forecast on 850 millibar temperature, reaching a threshold of 25%, you can see the steady increase into about 2010. And then we really plateaued and now there's been a gradual slow increase to basically getting out to about day nine and 10 on the information. So we're kind of reaching some of the limits. So, and then we look at, say we look at the most recent CPC forecast, the most recent winner, the only way to summarize it is Houston, we have a problem. If you look at the top forecast, the surface temperature anomaly forecast for December, January, February, above normal pretty much everywhere. The only question is how above normal the verification shows this big cold gap down through here because there happened to be the great cold outbreak in February 21, 2021, for which there was the forecast were clueless. And also I should point out that we're getting a free ride on skill because if you don't know, you should always forecast above warm with an upper trend in the data, you can show skill, but it's all artificial. Forecast reliability is an important process. So if we look at some of the TIGI, this is Thorpec's Interactive Grand Global Ensemble. I can't remember what the acronym Thorpec stands for. I look it up and I forget it after five seconds in there, but basically, and then the tertiary one is the MCGE help in through here. But nevertheless, when you do this and you look at forecast reliability, let's just think about this. Reliability diagram, observed probability and forecast probably ought to match. If they don't match, there's a problem. So perfect reliability would be right on the XX, Y on the 45 degree slope line. Poor reliability, in this case, overconfident, you're flattened on the curve this way. Underconfident would be in the other way. Okay. So here's the National Weather Service, over 2 million forecasts on the National Weather Service between 1994, 2004 and the Eastern region, showing the observed frequency of probably a precipitation in the reliability in the 24 hour forecast. Remarkably, reliable forecast, a little overconfident at the higher end, for example, and down near zero, down and through here. When I show this to economists in the economics department at Albany, they just go, you've got to be kidding me. They just assumed the forecast were jokes. The fact that they're reliable comes as a lot of astonishment to users of these forecasts. So that's a very important part of the process to how reliable the forecasts are. Now, the basis for ensemble prediction. I love this figure from the European Center, which I call lumpy and bumpy PDFs because in say the temperature, here's the PDF, it's sideways, the initial condition, forecast time, forecast. The tendency is when talking about extended forecasts is to show PDFs that broaden and flatten. But in reality, PDFs may develop multiple maxima. This is an example of the three of them and through here. So you have all this uncertainty of how the solutions can diverge with time, but they're not going to diverge in nice uniform ways. They're going to diverge in very unpredictable ways. So how reliable then are some of these forecasts? This is for extreme precipitation from the ticky, for this is from the year 2007 to 2013, three days, five days, nine days and 15 days. The grand global ensemble of everything on the left side, European Center in this column, Japanese meteorological agency in that column, NSEP in this column, UK Met Office in this column. Let's just go down to 15 days. Note that the European Center, this is the skill no skill line, basically goes, it crashes into the zero axis here at the other range. So basically nobody has any real skill and no surprise there, but the global ensemble of averaging everything together does a little bit better than all the individual models. And it really shows up at nine days, for example, out and through here, pretty reliable even at nine days, even though all the individual members are overconfident in through here. And, but clearly the European Center during this particular six year period is the model of choice overall, but the ultimate model of choice is the grand global ensemble. This is for high temperature, similar kind of record for high temperature shows up again. Issues with the overall, but overall the grand global ensemble is there, relative to everybody else. But after about nine days, there's a rapid decay in scale between nine days and 15 days, which is this limit of two weeks of predictability that we have seen so many times and Ed Lorenz told us about long time ago. You can also look at things like reliability and tropical cyclone intensity forecasts in through here. And you can see anything to the left of the 45 degree slope line. They're not like, for example here in the Bureau of Meteorology in Australia, it's very, very tough to the model to predict anything below 980 hectopascal central pressure. Whereas the European Center over here has both sides of the line, more on this side of the line of the under forecasting than on the over forecasting side and through here. But you can certainly see the differences between the individual centers. For example, the German Weather Service and through here is mostly under predicting the intensity of the stronger cyclones and through here, same with Medio France down and through here. So you can look at this reliability of all kinds of forecast. This happens to be intensity and it's a useful thing to do. Now something we don't do very often is why I call animal farm meteorology. All forecasts are equal but some forecasts are more equal than other forecasts. We lump everything together. But what happens at Ron McGarrett Cowan of Environment Canada sent me this. This is a month of, this is in February, a whole month, this is February of 2019 and through here, the day-to-day anomaly correlation coefficient. And this is the time when all the real forecast problems on the West Coast of the United States, this is for the whole Northern Hemisphere. But you can see there were two major periods for the Northern whole Northern Hemisphere where predictability was much lower than if you just took a simple mean predictability here and didn't really consider what the day-to-day variability is. These are the issues that basically contribute and these are the important kinds of weather regimes when these happen. And these are the things we need to pay attention to but they get lost in the sauce when you do all the averaging. So all forecasts are local like politics and Alan Pearson who used to be the director of the National Severe Storms Forecast Center which was the predecessor to the Storm Protection Center always told this hilarious story of a woman who would all, and that was based at her base in Kansas City at the time, woman out near Topeka, Kansas who would call them up after every forecast for a severe thunderstorm and say, you guys don't know what you're doing. Nothing happened here. This went on for years. And then finally one day she called and say, all an excited voice say, now we're getting somewhere. I lost the roof of my barn last night. You guys are finally figuring it out. So that really illustrates the whole issue of forecast predictability, forecast matter at the local level. So some realities is weather forecasting much better in 2021 than 50 years ago. Yes, much better. However, our customers expect more from us and they have different views of success than we do. All forecasts are local across all timescales but backyard meteorology, so to speak, matters. And sometimes when we really talk about like S2S timescales for predictability being very, very small a week, two or three in there we're arguing, we're really arguing to ourselves in the weeks because it's not actionable information as far as most users would be concerned. So one of the challenges is how to get to the point where we can have actionable information on S2S timescales. So here's an example of a significant GFS forecast error from January, 2017. And this was when there was a California deluge on California on the eighth and ninth in January and through here. And so based on deterministic forecast verifying on 12Z on the eighth of January, 2017, here's a 210 hour forecast and the solid contours of sea level pressure, the 1,005 thickness is the dash lines going from red to blue at the 540 decometer thickness and all the shaded represent the strength of the jet starting at 40 meters per second and through here. So you see, you have one jet up to the north and it's the chronically curve suggesting there's a block up there and then the Pacific jet coming off the Asian coast extending eastward and another jet coming in over a northern northwestern Mexico. This is the 210 hour forecast and this is the time when all things were going rapidly south in California but you'd never know it from the 210 forecast. Here's the 180 forecast verifying the same time. Let me toggle back and forth and you can see how the changes of the jet structure along on the West coast and also look at the cyclone off the coast of Japan there in the, excuse me, off the coast in the 210 forecast is now a big hole in the Western Pacific out and through here the jets move north, now you go to 90 hours and the jet moves further north and it's into Northern California in Oregon. So that big difference in where the jet was located suggests a much more amplified flow pattern and that made a one heck of a difference to the forecast. And you can see it in the corresponding precipitable water forecast, the 210 forecast the first orange is 24 millimeters. There's no real atmospheric river activity in the 210 forecast. 180 forecast verifying the same time you see the atmospheric river now you have one to the east here coming into extreme Southern California, Northwestern Mexico and one out in the East Central Pacific then you look at the 90 hour forecast when it locked in this one becomes the dominant figure and this is where all the issues occurred in California. So I'll show you a couple of loops on this quickly the 500 millibar heights, winds and vertical motion ascent only in through here and then thicknesses in through here. This is the 500 millibar, the vorticity is shaded so you can see how the ridge keeps changing its configuration as disturbances crawl up the backside of the ridge and then come down the east side of the ridge and cut off lows. The blue is the area where there's strong asset. So the evolution of the anticyclone upstream it's continually evolving and how that works and how disturbances go up the west side and the east side matter greatly. And you can now look at it in terms of sea level pressures the same thing with the jets. And again, you see the disturbances crawling up the backside of the ridge, rebuild the ridge and then they eventually break underneath the ridge. Well, the transition from going up and over the ridge to breaking under the ridge is very crucial for how much rain falls on the western coast. So some of the takeaways on that would be like blocking and cyclonic way breaking and cut off cyclones prevail. You have anticyclonic way breaking that's common across the North Pacific and all of that when it allows the deep moisture to reach the west coast and there's a California deluge in a Portland snowstorm that result. If we look at two years later, February 2019 biological mayhem on the west coast. Tyler Light made this diagram shows the mean the time mean 250 millibar heights and winds for the first 15 days in February. And you can see you basically have like what it looks like a Rex block this way and kind of an Omega block that way but basically have a blocking pattern in the East Central Pacific along 150 West but how did you get there? And I point out that time mean maps users beware time averaging may hide importance and optic dynamic flow signals. So you can look at the over a seven day period until here 500 millibar heights on the left standardized height anomies are shaded so blue, the cold colors are negative standard anomalies standardized and the warm colors are warm standardized anomalies. Same thing on the right, 500 millibar heights standard price, precipitable water anomalies aren't shaded. So what you see is the high precipitable water anomalies that extend well-pulvered on the west side of the ridge ahead of the trough all the way up to high latitudes. And there's also a weak cone of low here which is going to allow for moisture to come in out of the tropical Pacific into mid latitudes. So that's one day later, two days later and the streams coming towards the coast now but note the ridge building in the Pacific. See how the ridge, the leading edge of the ridge is replaced by upstream ridging to the west continually to the west. And then what happens is as the ridge becomes very elongated in through here at 500 millibars the trough coming down the east side merges with a cone of low and that allows high precipitable water there it's now merged right in through here. Now you've got plus six sigma precipitable water values off the coast of Baja, which are coming inland and that led to the third brainiest day ever in Palm Springs, California and that and other widespread flooding in parts of Arizona and Southern California. So a little dynamics here talk about potential vorticity potential vorticity streamers and Ross v way breaking. First of all, PV is basically on a theta surface the product of the absolute vorticity times the stability in through here one PV unit because the units would use PV units because who's gonna write down for all these units in through here but one PV unit basically calculate a temperature change of 10 degrees Kelvin over a hundred hectopascal layer and a vorticity value of 10 to the minus four that's the definition of one PV unit. So some potential vorticity concepts that you all know from dynamics classes first order advantages, conservation and invertibility conserved for frictional salivatic motion invertibility allows recovery of heights, winds and temperatures and invertibility requires a balance condition, a reference state for example, potential temperature and top and bottom boundary conditions invert the PV field globally. However, to me the most interesting aspect of PV is non-conservation because that's where the most exciting and interesting weather events occur. I won't put you well on this, but these are the Bible the Bible is Hoskins, HMR, people know yesterday HMR you know what that is, who did that and then follow up by Hoskins and Beresford in 1988 trying to get some people to use it and to show for the weather people and the weather and then Brian Hoskins looking at a potential vorticity viewpoint of synoptic development in 1997. Okay, anisoclonic way breaking downstream trough development and PV streamer formation. First of all, what is the PV streamer? It's an elongated filament of high potential vorticity air that has a high length to width aspect ratio. PV streamers correspond to cold upper level troughs and can be identified by following a PV contour on a theta surface that marks the dynamic tropopause. Why should you care? This is what a PV streamer could look like. This is one from 2008. The wind bar, wind arrow, the scale is 20 meters, whoop, 20 meters per second down there. Here's the East Coast of the United States and here's over here the West Coast of Africa. So note the anisoclonic way breaking is going on upstream and you get this trough that extends equated with a 20 degree South in through here and PV unit about four. So this is the deep trough. These matter greatly because these can be a source of exporting tropical moisture to higher latitudes and also big severe rainstorms in the subtropics. So why should you care about these things? I said they can be linked upstream, raw speed, weight breaking and downstream baroclinic development. And it's so shared with a variety of extreme weather vents that usually centered around tropical moisture and export on the East side. And the PV's thin and elongated. When you were, I don't know, when I was a kid you had this rubber cement. I don't know if I had a rubber cement anymore or if you took the cap off the rubber cement and pulled it upward, the stream basically thin and then eventually broke. And that's the analogy. And when you do that when the PV streamers thin and they break, you get cut off cyclone formation. But once you have a slow movie cut off cyclone and lower latitudes, that's a recipe for heavy rainfall related extreme weather events and big predictability problems. So the science question here, so shall be curving and transitioning West Pacific tropical cyclones and downstream impacts. How can a reconfiguration of the North Pacific flow induced by curving and transitioning Western Pacific TCs trigger downstream baroclinic development and the current extreme weather events? All right, Heather Archambault and a student of Dan Kaiser and me we wrote this paper on looking at events when the 272 events and looking at these, gonna show you the top quintile 54 events of how they impact the North Pacific jet stream then impact downstream predictability. So here's a diagram for timing of 54 events. The blue lines are potential vorticity units two, four, six and eight. The arrows are the irrotational wind in meters per second scale that way. The shaded and gray is precipitable water starting at 40 millimeters. The red symbol here is the tropical cyclone. So when the tropical cyclone is beginning to interact with the mid-latitude flow and through here and the green showing the 500 millibar assent in through here. So what does this do to the strength of the jet? Well, if you actually quantify the negative PV advection by the irrotational wind, you get the red dashed lines and through here. If you now superimpose the wind speed starting the light oranges at 40 meters per second, you see all this is happening in the equator with jet entrance region of this system. So when we curving tritropical cyclones interact with equator with jet entrance regions, all kinds of downstream mayhem usually results. And here's a composite of that for the 54 cases. And you can see where the TC is relative to the jet core and the 1,005 thickness. So ahead of the TC, as it's beginning to interact with the jet, you have loads of warm air advection which is forcing strong asset in the equator with entrance region to the jet. And then there's going to be a predictable downstream impacts. To show you an example, super typhoon Nuri here in 2014 induces downstream Barrick Clinic development. This was not too shabby of a TC. It got down to about nine, 10 central hectopascals, central pressure. And that was the track, but this is all source of digital typhoon, the greatest best website ever for tropical cyclones that the Japanese run. And then they give up on the storm when it's no longer a TC. But then that's when things really start to get interesting because the resulting extra tropical cyclone after the extra tropical transition is sitting out here about the deepest cyclone, extra tropical cyclone tine ever observed in the Pacific out and through here with broad cyclonic circulation. Well, what are the impacts? So we can look at this from induced downstream flow evolution in schematics to avoid showing you the maps. This is the 4th of November at zero Z. Here's where Nuri is and the jets are the colors in through here. And the flow pattern, and you can see there's a moderately amplified flow pattern. Now super typhoon Nuri. Now watch what happens as we go forward seven days. The remnants of Nuri are up and through here, but now you have downstream this big building a mega block and the first cold surge coming into the U.S. H1 is there gonna be three anticyclones. So building a mega block, four days later, hard not to recognize an a mega block. The first cold surge is dissipating in the whole Ohio Valley, but here comes the third second cold surge and the third cold surge is waiting in the wings. There's the 18th of November. The blocking ridge is still in place. The ex-Nuri cyclone is still sitting out here. And now you have these monumental cold surges coming down in November, 2014. You can see it in a Havmahler format here 35 to 55, 55 to 75 over here, 29 October to 28 November. So time runs down 120 East to 60 West. You can see where the height anomalies as Nuri, when you see the colors here, it's in the plane of this cross-section. Big height rises occur downstream of Nuri, Nuri's inducing downstream ridging and downstream of there, you get downstream troughing. Ridging becomes troughing for the downstream. That's the downstream baroclinic development. You look at it in the V-wind anomaly and through here. So you've got anomalous southerly flow here, anomalous northerly flow here as big anisoclonic circulation develops ahead of the recurring Nuri. And then if you look at it at a 50 millibar temperature at higher latitudes in here, 55, anonymously warm downstream, anonymously cold and anonymously cold underneath the block as the block over sits over below. And so that was the result. 2677 minimum temperature records broken over the conus during this particular period, one of the 16th to the 22nd of November. Impacts, this guy in Buffalo had a bit of tunnel vision for lack of a better term. This guy had quite the load on his mind and pretty lazy about clearing off the snow. If he hits a bump and you're following him, look out. But the CPC November conus temperature forecast is completely derailed. Here's the forecast for November, 2014. That's what was observed. These recurring Western Pacific typhoons they have huge impacts when they're able to interact with the Westerlies as you're deep into the autumn season in through here. And we just basically don't have any predictability on longer timescales of those kinds of events. So to finish up, we'll quickly a dynamical overview on some vertical motion applications. I should point out that such that Kevin Trenberth, and I wanna make a point here, Kevin Trenberth is an internationally renowned climate scientist for very good reasons. We went to school together, but what people probably don't appreciate is that like Ed Lorenz, Kevin was grounded in weather. He was forecasting and we worked for the New Zealand Met Service for a while. And after he came to the U.S., he eventually wound up at the University of Illinois and then spent a very productive career at NCAR. Trenberth built upon the famous Sutcliff Development Equation of 1938 to show that vertical motion was mostly forced by the Borticity advection by the thermal wind. And you could get that by knowing just the sea level isobars, which converted to 1,000 millibar heights and then calculating the 500 millibar analysis and doing a graphical subtraction to get the thickness. This is what was done in the Southern Hemisphere because there was so little data and people did this manually and then with simple computer forms. At the same time, in the same year, Brian Hoskins did his famous Q-vector paper at all in through here and the most Q-vector in through here and the forcing. So when del dot Q is negative, the right-hand side is positive, which means omega is negative when you invert the Laplacian in through here. But note, I wanna point out what the Q-vector is. It's the rate of change of the geostrophic wind along the flow dotted with the horizontal temperature grade. So that suggests that things where there's shear along the flow and in temperature gradient regions, the Q-vectors will be important. In 1990, Fred Sanders and Brian Hoskins wrote a reader's guide to how to use Q-vectors, estimate them from weather maps. And this came about because a few years before this paper was published, Brian and I were talking, he was lamenting why a lot of people were not really using Q-vectors in forecast office and otherwise. And I remember saying, Brian, what you need to do is team up with somebody like Fred Sanders and write a paper, which is a user's guide to tell people how to use these things in languages that the forecasters will resonate with. So that's what they did. And to make a long story short, look, the Q-vector then depends upon the magnitude of the temperature gradient and the shear of the wind along the flow. So where does that Q-vector going to be large in jet entrance and jet exit regions where partial V sub G, partial X is large. But if you're dealing with a jet, you're going to have a horizontal temperature gradient. So where the horizontal temperature gradient is maximized in the North-South direction in this case, and when you have variation wind speed along the X axis, which say the thickness contours are oriented east-west. So you get more bang for your Q-vector bucks in jet entrance and jet exit regions, particularly for short stubby jets where the change with X of the wind speed is large. So here's one of the figures from the Sanders and Hoskin diagram that tells it. This is the classic four corners way of deformation zone for surface-front or genesis. When you have a pattern like this and your isotherms are the dashed lines and your flow is like this, that's going to tighten the temperature gradient. Now imagine taking just the right-hand side of this diagram and putting it down here and looking upstairs. So now you have a confluent jet entrance region and through here, and think of these as thickness contours. So along here, partial V sub G partial X is increasing to the east. The temperature gradient partial T partial Y is increasing to the south. K cross partial D V sub G partial X is a vector going to the right. So the Q-vectors point towards rising warm air, low-level rising warm air or away from cold air in through here. And the strength depends upon the gradient of temperature and the rate of change in wind speed along that temperature gradient along that thermal boundary. So there are going to be large in jet entrance and jet exit regions. And if you have a jet core here and the standard straight jet model in through here, divergence convergence, divergence in through here, jet entrance region is right in through here. There's going to be a minimum value of vorticity here or a maximum value of vorticity here. So you have weak cyclonic vorticity advection on the right entrance region, strong cyclonic vorticity advection in the left exit region. And the Q-vectors will be pointing in this direction towards rising low-level warm air, rising warm air here and the flow you're looking at the arrows of the return flow. So the circulations in jet entrance and jet exit regions are going to be proportional to depends upon the strength and rate of change of the wind speed along the flow. And the Q-vector measures that. So jet streaks and S2S predictability, jet entrance and jet exit regions, I call the traffic cops of the atmosphere. Transverse agiastrophic verticirculations can be more vigorous in short stubby jets. Error growth is maximized in equator jet entrance regions where you typically have access to tropical moisture. Think back to what we just discussed about recurving tropical cyclones interacting with the equator or jet entrance region of the North Pacific jet stream. Error growth is also maximized in the poleward left exit regions where areas when you have vigorous cyclogenesis. And again, if you have uncertainty in the strength of the wind flow relative to the temperature gradient in the jet exit region, that's going to lead to large error growth with mid-latitude disturbances. So the last slide here would be key takeaways. Synoptic dynamic thinking and weather analysis and interpretation skills must have a place at the S2S table if we're going to make sense of what is going on. Forecast reliability remains a challenge on time scales beyond one week. Negative PV infection by the irritational wind ahead of recurving and transitioning West PAC TCs or West Atlantic TCs poses downstream predictability challenges on S2 time scales because one or two recurving TCs that interact with the jet can totally rearrange the downstream circulation. And the dynamics of vertical circulations and jet entrance and jet exit regions is a key governor of S2S predictability. And so it's time for me to shut up. Thanks a lot, Lance. That was a great talk and a great bridge between the weather and how it connects to the climate through the S2S time scale.