 Good evening everyone. Thank you for joining us today for this NCAR Explorer series lecture called Turning Down the Thermostat Climate Intervention Using Stratospheric Aerosols with Dr. Yaga Richter and Dr. Doug McMartin. My name is Dr. Lorena Medina Luna and I am an education designer and a lead organizer for the NCAR Explorer series. The National Center for Atmospheric Research or NCAR which is a world-leading organization dedicated to understanding the earth system science including our atmosphere, weather, climate, the sun and the importance of all these systems to our society. I'm really glad to be with you all today to learn more about the computational modeling work that Yaga and Doug will be presenting about. For this lecture we'll be taking questions at the end but please definitely submit any questions that you might have during the talk using the Slido platform. If you scroll down this web page you can see the Slido window just below where you're seeing the live stream the live stream. If you haven't already done so go ahead and click on the green join event button and then you can ask questions on the Q&A tab and answer poll questions on the poll tabs both of which are found in the blue bar across the top and definitely be sure to join the Slido to add your thoughts on our word cloud question which is what do you think of when you hear climate intervention or geoengineering? We'll get to that after I present our speakers and before they start their lecture. This lecture is being recorded and will be available on our NCAR Explorer series web page in addition to any past lectures that you might be interested in checking out. With us again we have Dr. Yaga Richter who is a scientist in the climate change research section of the Climate and Global Dynamics Laboratory at NCAR or CGD. Her areas of expertise and scientific interests include gravity waves and their parametrisations in global climate models, middle atmospheric dynamics, quasi-by-annual oscillations or QBO, whole atmosphere climate modeling, sub-seasonal to sub-seasonal forecasting and geoengineering and if you're not aware in 2016 Dr. Richter also is co-founded these NCAR Explorer series the whole series in an effort to work with scientists to speak about the work that is relevant for the community so I'm really honored to be able to work with her in this capacity today. And Dr. Douglas McMartin is a senior research fellow in the Sibley School of Mechanical and Aerospace Engineering at Cornell University. His research focuses on climate engineering also known as solar engineering or climate intervention with the aim of helping to develop the knowledge based necessary to support informed future societal decisions in this challenging and controversial field. Yaga and Doug I welcome you to turn on your cameras to give a quick hello before we check out the word cloud. Great to see you both thank you for joining us. We're still virtual so I appreciate everybody joining us from wherever you might be and let's go ahead and Paul or Brett would you be able to share the word cloud with us and we do have surveys again that you're welcome to to fill out and we will be going into those surveys during the lecture as well. So thank you so much for a lot of these some phrases and words. So yeah it is scary to be to be thinking about what's happening but fortunately we have computer models to try to help us learn before we do too much implementation and we have a lot of scientists that are working in different areas so I'm really excited to learn more about this field as I'm sure our audience is. And with that thank you so much Paul and Brett I'll hand it over to Yaga and Doug to start us off on our lecture and I'll see you both at the at the Q&A set portion of this talk. Hello everybody my name is Yaga Richter and it's my sincere pleasure to talk to you today with one of my close collaborators Doug McMartin from Cornell University. As Lorena mentioned I'm one of the co-founders of the NCAR Explorer series and I believe that it's really important for NCAR to communicate the science that we do here to a broad audience and in particular the topic of climate intervention that we'll discuss here is something that would potentially affect every single person on the planet and hence I think it's especially important to share this research with you so thank you all for tuning in today. I will start by describing to you the motivation for carrying out this research which is fundamentally actually really simple and it boils down to this the earth is getting hotter and our climate is changing. So what I have here is a picture of the global temperature from 1880 to present day relative to what we call pre-industrial average in particular case 1881 to 1910 average and what you see in the late 1800s and in the early 1900s that sometimes we would have a warmer year sometimes a cooler year sometimes the warmer years would last a few years but then we have a period of cooling however since about the 1920s the global mean temperature has been drastically increasing and this increase has been really strong since about 1970 so 2021 was the planet's sixth hottest year on record and the top 10 hottest years on record have all occurred in the last 12 years and along with climate change and global mean temperature changes come changes in temperature and other extremes so in july 15th 2011 this idling coal lake in china hit 122.4 degrees Fahrenheit in july and august 2011 these two places one in iraq and iraq hit 127.4 degrees Fahrenheit in this little town in southeast colorado las animas experience a high of 114 degrees Fahrenheit on june 23rd 2012 and those are just a few examples and along with changes in temperature we're also seeing changes in floods and an increase of these events here are some pictures from pakistan in august 2010 india june 2013 manhattan in october of 2012 and not very far from encar here a flood in lions colorado in september 2013 climate change also has brought about increases in drought and fires so here's a picture of a morse reservoir in noblesville indiana in july 2012 completely empty and everybody who in the bolder area remembers the marshal fire in superior and we know that wildfires have been increasing year after year in our area there's also changes in the polar regions and very large ones so what i'm showing you here on this top left plot is the average monthly arctic sea ice extent from 1979 to present day and you see that the sea ice extent has declined from a little bit over seven millions of square kilometers to about four and a half and if we look at year 2022 in the second plot in this blue line it is way below the average over this time period which is the gray area and there were years in which the sea ice was very low for example 2012 the sea ice extent was only four millions of square kilometers and that affects not only people but obviously the animals living in those parts of the world so why are we here well the reason is increases in carbon dioxide and other greenhouse gases so for the last 650 000 years the atmosphere concentrations of carbon dioxide or co2 have always been under 300 parts per million there was a lot of variability so it was warmer it was colder but these numbers never went over a line of 300 parts per million currently we just checked yesterday we're at 417 parts per million so there's been a really really sharp increase and we're zooming in on that increase in the figure on the left and where you see that the changes in carbon dioxide are very well correlated with the changes in the global mean temperature and you see a lot of more wiggles in the in the white line here the global mean temperature that's internal variability that's something we can't control that's just something part of the natural system but there's a general trend for these temperatures to increase and they're going to keep increasing as long as we're putting more greenhouse gases into the atmosphere so here's a quick lesson on the earth's energy budget there's the sun and in its short wave radiation or solar radiation and most of this radiation is absorbed by the earth and it warms it however the portion of it is rereadeted back to space in the form of infrared radiation or infrared waves and then we have a layer of greenhouse gases of carbon co2 water vapor etc and some of this radiation is rereadied back down to the earth and that's how the earth warms up some more so sometimes people say then the radiation is strapped or that the co2 is a blanket over the earth however we say it the more co2 you have in the atmosphere the more infrared radiation is going back towards earth and the more the earth warms so the physics behind this are very straightforward and we're very confident in them there's really a very little question that the more co2 we put into the atmosphere the warmer the earth's going to get and these changes associated with increasing carbon dioxide vary with location so what i'm showing you here is the anomalies of temperature in year 2021 compared to a pre-industrial average and what you notice that the polar regions the northern polar regions warm the most and then the other feature that you notice is that the land areas are warming more than the oceans and in general the northern hemisphere is warming more than the southern hemisphere and the reasons for this have to do with differences in land masses and distributions of oceans etc and you can see this that the change is not the same across the united states so the western united states is warming a lot more than central and east except for florida and the very north eastern states all right so what will the future look like so what i'm showing you here is a figure from a report by an intergovernmental panel on climate change or the IPCC and these reports take a very long time to put together by hundreds of scientists then and then look at projections of what might happen in the future and so what we look at is these shared socioeconomic pathways and those are assumptions that social scientists make about what our emissions might be and then we look at under those different scenarios what the temperature the global mean temperature might be if we follow one of these paths so these ssp 1.26 and 1.19 are the optimistic scenarios so for example for ssp 1.9 we assume that we'll have net zero co2 global emissions by around 2050 and this is the only scenario that keeps to the Paris agreement of keeping the global mean temperature under one and a half degree Celsius above pre-industrial this is at this point not very likely so hence we look at other scenarios like 2.45 3.70 and 5.85 so the 2.45 is a moderate scenario in which we assume that we'll keep co2 emissions remaining at current levels until 2050 and then they're going to start declining after that so this is in some ways a hopeful a reasonable scenario and then ssp 5.85 would be the more extreme scenario business as usual the society does not change we just keep emitting more and more co2 okay so right now we're in 2022 we're about 1.2 degrees Celsius above the pre-industrial following this moderate scenario we would be at one and a half degrees Celsius over pre-industrial somewhere between 2025 to 2035 and the exact number is not known because these projections are made using different models and there is uncertainty in our projections but we know will sometime happen between 2025 and 2035 and will it reach two degrees sometime between 2035 and 2060 and along with that the Arctic will become seasonally ice free around 2060 to 2080 and these are the projected patterns of change if we were to reach one and a half two degrees or even four degrees warming and these patterns look very much like the pattern I showed you in 2021 so there would be a lot more warming in the polar regions and the continents would be the land masses would be getting warmer than the oceans so the number of people affected by climate change would be about eight billion however you see this would not be equal across the world so along with changes in the temperatures there will be changes in the water cycle which manifests themselves in changes in rainfall which is shown in this bottom figure so you would see for example in Africa we would increase and increase in precipitation and in other areas we would increase expect decrease but there would be over the majority of land areas there would be an increase and you may say isn't that good don't we want more rain in those countries well the problem is it's not going to come in a form that's particularly useful so most of this will come as heavy intense precipitation and it will be alternate with heat waves and drought so it actually would not be really good for farming so basically any extreme weather will increase which is not what our crops and our populations are used to so the recent report the APC6 assessment report it came out recently it said it will be impossible to limit warming to one and a half degrees C with no or limited overshoot without stronger 2030 climate action so one of the key takeaways from this talk that you should take away that eliminate elimination of emissions needs to be our top priority to solve the climate crisis we need to either get this CO2 out of the atmosphere and our technology is being made to remove the carbon out of the atmosphere and primarily reduce the emissions so we would consider any climate intervention it could be part of a strategy to reduce the worst consequences of climate change but it's not a substitute for and so the reason we look for strategy to be a part of the the strategy is because the carbon dioxide removal technologies are not really available in its scale and it's not looking like the countries are keeping their commitments to reducing CO2 emissions and we're on a path at least of a moderate scenario that I showed earlier meaning that more than likely we will reach one and a half degree Celsius or two degree Celsius and not in the next decade or two so if the idea of climate intervention sounds scary I would fully agree with you so I started this research seven years ago and I still remember the day when I sat across the desk from my lab director who asked me to work on this topic I have great respect for him so I didn't say anything at the time but my initial reaction was that the idea was really crazy and I was not really thrilled to be a part of the research going forward however I've been in it now for seven years and the reason I still stick with it is that it's looking less and less likely that we'll be able to avoid doing this so hopefully we will do all this research and not have to implement any climate intervention hopefully society will make choices to keep us underneath one and a half or two degrees Celsius but if that doesn't happen I believe that we really want to do our research thoroughly and research is very slow so it's in my opinion will probably take about 10 to 20 years to just really get basic understanding over the consequences might be but let's hear now from you what you would like to what you think about that so let's bring up the poll and then I'll hand it over to Doug so I think there's no wrong answer here I think the distribution of answers here is probably fairly typical and at some level the honest answer is we actually don't know because we don't know what the future looks like I think when I want to move on to the next question when I first started working on climate intervention which was back in 2006 I was introduced to it a few years before that if I talked about it nobody had ever heard of it and so if you had asked this question 15 20 years ago people would have looked at you and said what are you talking about and now I would say most people that I've talked with have at least some exposure to the idea so I started like I said about 16 years ago this was just a just a fascinating idea just a curiosity didn't really take it seriously as a possibility I've been working in I was an aerospace engineer originally and the more that I started looking at the subject the more basically concerned about climate change realizing sometimes we need an insurance policy and then started working with Yaga actually I guess about when you were just started to work on the on the topic yourself moving more or less from looking at sort of very simple climate models to how would you go simulate this in a much much more realistic climate model and that's basically what Yaga and I are going to talk about today so let me just share my screen here carry on so formal definition of climate intervention or geoengineering is typically a deliberate large-scale intervention in the climate to manage climate change the this is something that we only study using climate models it's not something there's no no intention right now to actually think about doing this doing anything like this in the real world and any decision on that would require some discussion among pretty much everybody on the planet the picture here is from a report that I was involved with last year from the US National Academies like I said 15 years ago pretty much nobody had heard of this this is getting more attention now so we had a report from the US National Academies there was actually one in 2015 and then another one last year there was actually a report released this morning from Council on Foreign Relations um and whoops went the wrong way so let me just give you a really qualitative big picture overview first and then we'll talk more about the methods for how to intervene in the climate um nothing that we say here changes the fact that we have to cut our emissions and we have to cut those aggressively and I deliberately made this plot I took any numbers out of this plot just to sort of get the basic ideas the problem is that the lifetime of carbon dioxide in the atmosphere is so long that when we get to zero emissions globally when we're not using fossil fuels anymore at all that's not when we've solved climate change that's when we've stopped making it worse this black line basically levels off Yaga mentioned the potential to remove carbon dioxide from the atmosphere there's a lot of ideas in that space right now nothing is remotely close to the scale that you would actually need to implement it so we certainly hope that those technologies will be ready but we don't really know for sure when they will be ready um so I mentioned uncertainty even if I tell you exactly what the future emissions are there's uncertainty in the climate response there is of course policy uncertainty we don't know when future when people will ultimately get to zero emissions there's technology uncertainty as well and so you add all of this up we're sitting somewhere over here we've already experiencing significant climate impacts and we can certainly hope that these tools basically cutting emissions and removing co2 from the atmosphere will be sufficient to manage the risks of climate change but the honest answer is we don't know and so that's sort of the context for thinking about solar geoengineering or climate intervention is a potential way of reducing some of those climate impacts and of course it comes with its own risks these points that I've labeled one and two here it's clearly colder you will reduce temperatures but you don't recover the same climate that you would have if you had neither the greenhouse gases nor the geoengineering so so in that sense point two and point three are different so this is a picture from the national academy report again just talking about the methods the idea that's best understood and the one that we're going to talk about is putting aerosols into the stratosphere so an aerosol is a small droplet in this case a fraction of a micron and this and then the stratosphere is a higher layer of the earth's atmosphere where things are stable so if you put pollution or something like that out of a smokestack near the ground there's turbulence there's rain all the weather and so forth and that all leaves the atmosphere within a week or so but if you go high enough in the atmosphere there's a layer that is stable and if you put material there it can persist in up there for a border a year so nature gives us this example every now and then this picture is from the eruption of Mount Pinatubo in the Philippines in 1991 that put something of order 10 to 20 million tons of sulfur dioxide into the stratosphere and that cooled the planet by something of order point three to point five degrees Celsius for the following year or two it's a little hard to see the signal on this plot but it's a robust effect after every large volcanic eruption there's a there's a dip in the global temperatures so we know that the idea works if you essentially mimic this by flying aircraft up to the stratosphere and releasing material you know we know you will cool the planet the other idea we're not going to talk about much it's much less well understood the other idea that gets some attention is marine cloud brightening so this is a satellite picture of clouds off the coast of europe and each of these lines you see in the picture is what's called a ship track it's basically formed from the pollution from the smokestack of the ship can basically create a cloud where there wasn't one before and that cloud can last for a week and you can replicate that not with pollution but simply by spraying saltwater into the clouds but the physics of that as much as i said is much less understood so the idea with stratospheric aerosols is this is the same picture as before but now you put an additional layer of aerosols into the stratosphere that reflects a little bit of the sunlight back to space by a little bit if you were to reflect of order one percent of all of the sunlight back to space you would cool the planet all the way back to pre-industrial temperatures so you're not talking about very large changes in the total sunlight in order to maintain radiation balance of the planet and as mentioned we you know we have this natural analog in the form of volcanic eruptions we can use those basically to help us understand how to model the climate system and so if our models match the observations after eruptions then we have greater confidence that those we can use those climate models for projecting what the impacts of a deliberate intervention would be so just two quick movies here the one on the left is from an injection from a volcanic eruption in the northern hemisphere Tessitocchi and you can see the aerosols primarily stay in the northern hemisphere and if you visit from the eruption of Mount Pinatubo in the Philippines again in the northern hemisphere but much closer to the tropics the aerosols spread around the planet and they spread into both hemispheres this eruption was slightly in the northern hemisphere so you do see more aerosols in the northern hemisphere so we're going to come back to those ideas in a minute so let me pass that back to Yago who's going to talk about the climate model yeah so what do we use to study climate intervention so we use what we call earth system models or sometimes climate models and they have evolved from what we call in the past general circulation models basically all of these take the atmosphere and they divided into little cubes and then on these cubes we solve lots of equations then then show us where the air is blowing and then also have lots of equations to solve chemistry and other reactions in the atmosphere and then we take this set of equations and we solve them on a supercomputer and I'll get to it on the next slide how long that takes and then we can produce simulations of the past the present and the future so if you go to the next slide Doug at Encore in particular we are developing the community earth system model and with the whole atmosphere community climate model is what I'm going to talk about today that's our model that goes really up high into the atmosphere so Doug talked about the troposphere and he talked about the stratosphere and these models this model even goes higher than that so it goes into the mesosphere and the lower thermosphere so about 140 kilometers about the earth surface and I'm in some of the studies we're using we're using the version one of the model and for some of the recent studies we're using version two or CSN two but they fundamentally have a similar structure so there is the atmosphere and then there's the ocean model and then there's the land model and they're all coupled together so there are all the interactions there's interactions between snow cover and precipitation evaporation ocean currents clouds convictions etc and one of the things there is unique to CSM Wacom is that it has a really comprehensive chemistry module and in particular CSM two Wacom six their latest version of the model is really comprehensive chemistry not only in the stratosphere where you need it for where the injections on sulfur dioxide would go but also in the troposphere so we can look at impacts on ozone and other things and on the right here I have a picture of what the resolution of the CSM is and it's about 100 kilometers so if you look at the United States and particularly in the state of Colorado uh state of Colorado is about 610 kilometers by 450 kilometers so there would be about 24 model grid boxes and there's a lot more grid boxes across the world and don't forget that they go in three dimensions so they go all the way up to 140 kilometers and we're solving lots of equations in each one of this grid box so one simulation year takes about six to ten hours using over 1000 CPUs so that's why we need a super supercomputer so if we want to run a 30-year simulation with CSM Wacom that takes us about two weeks all right next slide so how do we know that these models work so the best way we know that they work is we run simulations that start in a past so in particular here on the left you see a plot that started in 1850 and the black line is the observations and the other colored lines are different ensemble members of simulations with CSM Wacom and the different ensemble members are not alike different ensemble members of the weather forecast where you make a small change in initial conditions and then by the effect of the butterfly effect that they're the forecast keeps spreading however the forced response you see it's very consistent in all of the simulations and it follows the global mean temperature that we see in observations and I should mention in the models we can play so if we remove the increase in carbon dioxide from our models we do not get the increase in temperature after in the recent 50 or 70 years so that's the nice thing about models we can learn about a planet without actually changing anything in the real world and on the right side it's just a plot of ozone so we verify multiple parts of the system and again the black is observations and the blue is our model just to show you that we're really doing well with this modeling system representing not only the processes on at the bottom of the surface but also in the stratosphere and particularly ozone here and back to you Doug. I forgot that I was muted. A lot of the early simulations that were done took a climate model and just like let's see what happens if you turn down the sun without even simulating stratospheric aerosols or they simply specified the aerosols in the stratosphere without simulating all of the processes the plot on the left here is a typical example of what what one found so then so the process is like let's just try something and let's see what happens and I'll point point out here this is effectively comparing if you look to my my previous diagram it's effectively comparing point two and point three so it's how is the climate system different than if you had neither the greenhouse gases or the climate intervention and this is a fairly typical result from a lot of the early simulations is for example if you put the aerosols in near the equator you will over cool in the tropics and you'll under cool at high latitudes it's still colder than it would be if you hadn't done anything but you haven't managed to restore the climate exactly so that was sort of the older approach and then the work that I started doing with yoga a number of years ago was trying to sort of change the the way of thinking about this problem and saying rather than just trying something and see what happens let's actually say what we want to happen let's set deliberate goals and see if we can design a strategy to actually meet those goals so there's a couple of parts to that one is what's what goals do you pick the answer some levels we don't know we picked three so global temperature is sort of a fairly obvious one a measure of the equator to pull um temperature gradient um or sorry the the inter hemispheric temperature gradient so that's making sure that the northern hemisphere and the southern hemisphere both are cooled appropriately um and otherwise you wind up shifting tropical precipitation and then this pole to pole temperature gradient to make sure that the high latitudes uh you don't have this over cooling in the low latitudes and under cooling at the high latitudes so that's the first thing is setting the goals um and as I said a lot of the early simulations might have just injected the material at the equator what we did instead was to say we get you know if you're actually ever want to go do this you get to choose where you're actually going to put the material um as I showed you in the movies with the volcanic eruptions if you put the material in in the northern hemisphere it stays mostly in the northern hemisphere in fact if you put it further towards the poles it stays further towards the poles and you can use that knowledge to actually do better uh than equatorial injection and you can use it to maintain those degrees of freedom those those goals that I said in the previous slide the second piece of this is a question of how much do you inject um and this is originally my background as an aerospace engineer is what I did was was control theory um and the idea is not fundamentally different from what happens when you take a shower in the morning you check the temperature of the water if it's a little too warm you turn the knob down if it's a little too cold you turn the knob up um and in principle we just did more or less the same thing in the climate model so we run the climate model for a year if it's a little bit too cold we inject a little bit less aerosols if it's a little bit too warm we inject a little bit more and so this is just the result from one of those first sets of simulations from a number of years ago demonstrating that the idea worked and this is in the rcp 8.5 this is the the business as usual or very high emissions scenario so it's not really a realistic projection of what we at least hope won't happen in the future but demonstrates the ability to maintain global mean temperature on the top the inter hemispheric temperature gradient in the middle um and then we didn't quite meet the equator to pull um temperature gradient in the bottom um so how did we do this we started by just saying all right let's pick a number of latitudes and see what happens if we put material in at each of those as the first step and then we can use those to sort of mix and match and say how do how do we actually design that strategy so this is just the plot here um is the the bottom is the surface going up in altitude and this and then on the left side is the south pole the right side is the north pole and this dashed line is showing the height of the tropopause and the black and yellow points are the places that we tried putting material in so either injecting close to the tropopause just barely into the stratosphere or significantly above that um and for reference when you go fly in a commercial aircraft um you at high latitudes you are in the stratosphere just slightly but you're just barely in the stratosphere um and at lower latitudes where you have much more influence on the climate system um your sub you know the current aircraft just cannot do that you need you need to be quite a ways you need to be above the the tropopause you need to be much much higher than current aircraft can fly i'll come back to that point later on so here's where the sulfate aerosols wind up when you do that if you inject the aerosols at the equator you wind up with a peak um in the aerosol concentrations at the equator so each of these plots again is altitude um on the vertical axis and south pole to the north pole on the horizontal axis if you inject in the northern hemisphere the aerosols would stay more in the northern hemisphere um and if you look at how that affects the temperature um the the red i didn't show you which corresponds to which but unsurprisingly these red and purple cases with more cooling in the northern hemisphere than the southern hemisphere that's what happens when you inject in the northern hemisphere when you inject in the southern hemisphere you get the reverse because the the hemispheres are not symmetric because there's more land mass in the northern hemisphere so you can use this to basically tailor how you design your your uh injections to reduce some of the side effects so knowing that where we used that knowledge to compose the first gene engineering large ensemble and we use an algorithm that specified where the injections were going to go to try to keep to the goals and we used our older version of the model csm1 whackham here and we went against the most extreme scenario and the part of the project we were looking at here was to see whether we could keep the temperature at about 2020 mean so about present day levels it was not a realistic implementation uh in the modeling framework even we're trying to see if we could do it and how much cooling could we offset how much cooling could we produce using stratospheric aerosols so um in the simulation you know as we have to keep injecting more and more as under the rcp8.5 scenario it's getting warmer and warmer and in order to offset the warming from the co2 emissions we need to put in about one pin one pin a tuba or about 20 terrariums of so2 uh by about year 2050 and if you wanted to keep going business as usual and use climate intervention just to cool the temperatures down which is not what we're advocating here but if you um we would just want to see what happens by 2080 you would need to be putting in two mount pinot tubos every single year so that's a lot of so2 going into this stratosphere however uh we showed and if you go to the next slide that this works actually really well so the black line here is the global mean temperature without intervention and the purple here is with intervention relative to about 2020 and you see that we can keep the temperatures pretty much uh at a very constant level and what does this look like in latitude so i'm looking here at this period from 2075 to 2095 this end of the century versus this base period it's around centered about 2020 and again a reminder if you do not do anything then the temperature changes are going to be very large mostly in the polar regions but if we do apply intervention then most of the changes disappear except for a little warming here in the atlantic and in northern europe and similarly for precipitation if you don't do anything under the rcp 0.5 scenario there'd be tremendous changes in precipitation a lot of wetting but with intervention most of those will be reduced to very few changes so if we go to the next slide a lot of people have been analyzing glens or changing large ensemble for all sorts of impact so a recent study from a person here marie tyatt anchor has looked at extreme weather and extreme events so this is showing changes in warmest nights again with no intervention more nights are getting warmer and warmer however with intervention we're pretty close to back to present day climate except for a few areas like the united states where we still have slightly warmer nights than we would have otherwise but much cooler than if we didn't apply intervention unfortunately other parts of the system do not benefit for climate intervention and in particular the ozone hole so the black here shows you um october total column ozone in the rcp 8.5 scenario and you see the dip down in the 1980s and 90s and we're now on a recovery path and we're on the path to recover by around 2040 if we apply climate intervention unfortunately there would be delayed probably i would say 2090 or 2100 and that's because when we're putting aerosols in the stratosphere then you're causing a lot of heating which changes the temperature in the stratosphere and water vapor and there's lots of chemical reactions that occur then cause depletion of ozone to just last longer and then go on to the next slide however other parts of the earth system have a tremendous benefit so this is arctic sea ice and the black line is the rcp 8.5 scenario in our model it shows that there would be hardly any sea ice left by 2050 and if we do apply intervention we can restore that to present levels or even above that and you similarly you see that pictures on the right for 2080 you see that there would be no sea ice in the arctic by that time and if we applied intervention you could get it back next slide so recently we repeated the set of experiments in our version newer version of the model CSM 2 Wacom 6 and we didn't exactly repeat them we tried to make the setup more realistic and also consider a more moderate emission scenario so we would be able to look at the responsible climate intervention to make the impacts what might be more expected rather than the drastic scenario so this new set of simulations is called arise SAI or assessing responses and impacts of solar climate intervention on the earth system or stratospheric aerosol injection okay that's a mouthful so we'll use arise SAI moving forward and in particular arise SAI 1.5 because we decided to cool to about one and a half degrees above the pre-industrial value and as with CSM 1 and CSM 2 we can perfectly well not perfectly but we can do a really good job keeping the global mean temperature to our target level below what the SSP 245 projection is but if you go to the next slide their regional impacts are a little bit different so what I'm comparing here is the residual changes in temperature on the top and on the bottom and precipitation on the right from glens so there was a run with our earlier model and with arise with our newest model and what you see is that in arise we show a little cooling in the Atlantic and in glens we're seeing worming over the Atlantic and in northern Europe well you may wonder which one is right which one would happen and similarly for precipitation in this pink box I'm highlighting that for example Australia was predicted to be wetter and then it's no longer predicted to be what it was our newest model so this is where the largest uncertainty is in this research and that's their regional impacts so if you go on to the next slide all of the research we have done has shown that using stratospheric aerosols can absolutely reduce global mean temperature the more you put in the more you can reduce all that's been very very straightforward and in agreement with basic physics and our model all the models are getting the similar answers however we are not getting much confidence in what their regional residual changes would be from present day so we know that the changes will be lower much lower than if we don't do anything however we don't exactly know in the given region of the world whether it may be a little wetter a little drier relative to present day the changes will definitely be lower relative to an unclimate without any intervention and why is that well it goes back to what their climate models can and cannot do so I described a few slides back that our grid boxes are about 100 by 100 kilometers in the horizontal and then about 500 meters stick in the vertical but if you think of releasing any particles from the back of an airplane or releasing a plume those interactions happen on much smaller scales so we make assumptions on what happens inside of those boxes similarly if you think of clouds and cloud droplets and rain droplets there are much smaller than the grid boxes of our models so everything in those boxes that's grid scale what we call needs to be parametrized so we use approximations to to assume what's going to happen in there and we do our best to verify that with existing observations but there are uncertainties and that's what causes differences between our this friend versions of the models and the uncertainties are in every part of the system in this stratosphere in the ozone chemistry in the ocean in particular with regard to the Atlantic meridional returning circulation or amok and also in the representation of key modes of variability for example El Nino etc so we know what climate intervention can do in many ways however the tiny details this is where a lot of research is still needed so I'm going to I'll wrap up in a few minutes here but so I'll say just a little bit about the technologies that one would need to do this so as I had mentioned earlier there are no aircraft that exist today that can both get to a sufficient altitude and do that while lofting a payload aircraft by the way look like the by far the most likely way that you could get material to the stratosphere and principle there's there's there are other ways to do that as well so the picture shown here this is from a colleague Wake Smith at Yale he has a background in aircraft he works with a number of aircraft designers and they basically did this thought experiment it's like well could we actually go design an airplane to do this and their answer more or less is yes you could use fairly straightforward technology that we understand you could go design a new airplane it would probably cost you something like five billion dollars to design it and it would take about five years to develop and they figured that it would be relatively straightforward to design something that could get to about 70,000 feet or 21 kilometers this by the way is one of the reasons that we changed in the if you look back at Glenn's and some of those original simulations that I had shown we were injecting material as high as 25 kilometers in the stratosphere and we concluded that that's just not realistic and so understanding what's actually possible from an engineering perspective then affects well what we actually go simulate in the climate models so it's not that we're trying to it's definitely premature to actually be developing deployment technology but it's important to understand what's possible just so that one knows so that one's simulating useful things and then the actual cost of deployment this would not be a small effort when talking about putting millions and millions of tons of sulfur dioxide into the stratosphere every year it'll be really really easy to see that from satellites it would require quite a few aircraft with constant flying up to the stratosphere coming back down reloading going back up to the stratosphere the number sort of depends on how many how much you're trying to cool but it would basically be like a small airline or the operations of a fairly decent sized airport so you wouldn't be able to hide these operations at all they're not cheap by my perspective the cost would be measured in tens of billions of dollars per year but when you compare that with what the impacts of climate change are its cost is not going to be the reason to avoid doing this there's lots of other concerns with doing it but cost isn't ultimately going to be the barrier on talking about implementation this is the slide that I showed earlier on or the the figure I showed earlier on the top left showing the tropopause height the tropopause is lower at high latitudes and so there is a valid question of saying well could you do this with existing aircraft if you went to higher latitudes so that's another strategy that we've been looking at just to understand what the implications would be again not a recommendation that one does it but to use climate models to understand what could be done and there's definitely advantages to that as well so the plots I've shown here on the left is September sea ice mid-century without any intervention at all the middle plot is coming from a rise so these is a global strategy using injection at four different latitudes to try to overall balance the impacts on the climate change if you took roughly the same injection rates and you put them all in at high latitude you would actually do a much better job at restoring CS of course you're doing much worse job at some other things so there's clearly going to be some trade-offs there to understand what are the range of choices that are available and what are the implications from this so the and as I said one of the reasons for considering that strategy is that that actually is much more implementable with relatively easier aircraft technology and then in from a research perspective I'm not going to dwell on this but there's a number of one could basically skies the limit right one could no pun intended look at look at multiple different strategies and understand what the impacts of each of them would be so when you're thinking about what would stratospheric aerosol injection do to the climate system the answer is not just that we don't know because of the uncertainties but also that it depends on how you do it so each of these will have a different set of impacts at a different set of risks associated with it to some extent in the interest of time I'm going to skip some of those slides and talk a little bit more about what we need to know I just pose this a bit more as a thought experiment back to you so imagine now transport yourself forward to 2030 let's assume for the sake of argument that we've managed to that emissions have now peaked in their declining but we've already passed one and a half degrees of warming and if you project where we're going to wind up being we're going to look more like two and a half degrees and keep in mind that these pictures on the right those come from a climate system that is only warmed by 1.2 degrees Celsius so if we're going to warm by twice as much things are going to be at least twice as bad so one can reasonably project more and stronger heat waves forest fires storms and so forth arctic summer sea ice might be pretty a lot smaller in 2030 than it is now there are risks coming from places like Antarctic ice shelves we don't know when we how much warming it takes to destabilize those permafrost is already showing signs of fine we don't know what the rate of future thought will be so all of these statements are plausible now imagine we're in this future world and you hear a talk from Yaga and I that looks kind of like the one we gave today and says well we know we can cool the planet but we don't know the details what do you do and in some sense more importantly what do you wish you knew because that's what we need to spend the next eight years or 10 years or whatever doing research to better understand so we need to do a better job of understanding what we think will happen how does that depend on the choices that we can make what are the uncertainties so being able to say not just here's our best guess but here's sort of the range of possible outcomes that could happen and you need all of that information to be able to ultimately support some sort of informed decisions about these technologies so that's also the basis for recommendations from the US National Academy report that was released last year I'm not going to read through all of these it emphasizes some of the same things that we just said about not being a substitute for reducing greenhouse gas emissions that our current knowledge is not enough to support informed decisions I mean in particular it recommended that the US go forward with creating a research program and coordinate that with other countries and that that research program shouldn't just be addressing the physical sciences but should also be addressing social sciences and that there ought to be adequate research governance associated with that as well so I'm going to wrap up there we of course should be reducing our CO2 emissions much more aggressively than we are I think a lot a lot of us would would support that but even if we do hopefully that will be adequate but we may need to also in the future consider additional options research is pretty early stages right now we need a large if we really want to support informed decisions we need to much more aggressive research program and climate model simulations are a core piece of that it's the only way that you ever project the future is using a climate model and you can use that to help understand how would you actually do this the way that maximizes benefits minimizes risks what do you think would happen and how confident are we so I will wrap up there and Yaga and I are happy to take your questions right well thank you so much Doug and Yaga for the wonderful presentation giving us some insight on what it does take to to create these models and it's it was nice for me to hear that you're being strategic about placing goals for what you want in the model rather than just trial in error as I can always do trial and error but it's good to have goals in mind and we do have some questions from the audience so thank you everybody for submitting those and you're also welcome to submit them on Slido if you haven't already joined definitely scroll down the page and join Slido great so let's go ahead and see the first question from David or the yeah let's just see what the first question is is it possible to design a technology that can counter the stratospheric aerosol injection to accelerate the fading process of the particles? I think my answer to that would be that I don't think it's absolutely necessary so I was not specific I don't think either of us were specific the aerosol lifetime in the stratosphere is of order a year so if you you have to constantly replenish to keep the aerosol there there and if you stop basically the aerosols will go away after about a year and then of course this is one of the risks if anybody were relying on this and they never stopped then there's a termination shock where the temperatures sort of grant go back up to where they would have been if you were never doing it at all so you could in principle I guess I haven't thought of ways that one could accelerate that but the you know you can you can view the this the reversibility as a good thing right if you if you decide that you don't want to do it the effect goes away great thank you so much because I did see that that was part of the model input is the injection rate and how often and where in that earth model you actually injected so that's awesome um let's go ahead to see the next question this says if carbon dioxide emissions go to zero won't sequestration by for example trees plants um help reduce what's left monologue carbon dioxide data shows seasonal variation due to this um so maybe I'll take this one as well um yes uh is the short answer I think I was not uh I wasn't precise when I sort of uh went through that um if you zero emissions today the co2 levels in the atmosphere will fall because some of that co2 will continue to be taken up by the biosphere and by the ocean it is also true that the earth is not yet in equilibrium with the amount of co2 that's in the atmosphere so if you help the co2 levels constant temperature would continue to rise if you stop emissions the co2 levels actually fall they don't stay constant and those effects roughly sort of to first order balance each other so that when you go to zero emissions co2 levels fall the temperature is roughly remain constant thank you let's go ahead and see the next one says are there other choices besides so too or so for dioxide to inject are there better choices on what volcanoes do naturally I can take that one so the reason we do with volcanoes do naturally is because we have observations and to verify our models so if we consider it's something that hasn't happened in nature we would first have to study in the lab then you would perhaps do field experiments and then putting that in a stratosphere and having a measurement system around it is not a simple issue so there is an idea of a calcium carbonate that a group at Harvard is studying and that would not have the issue of stratosphere stratosphere heating however again that does not occur in nature therefore we would need to repeat a lot more studies in order to have any confidence in the results thank you how will this intervention impact large systems like the polar vortex which depends on the temperature gradient between the poles and the equator or the golf string all right so I'll take that one as well it follows from the previous one that if you put in sulfur dioxide in this stratosphere which you will actually create a layer in the lower stratosphere of heating and that the will by thermal wind balance change the circulations and in particular strength and the polar stratospheric vortex and indeed they would have influences down at the surface and it's one of the reasons why we do carry out these large ensembles with our models because there's already lots of variability in the polar vortex and so we need to really have a lot of ensemble members a lot of simulations to see which direction that would change and I think on a similar note we have looked at the amock and in CSM one the amock recovers and in CSM two it does not so there is a lot we don't understand but we do know for sure there will be stratosphere heating there will be changes in the polar vortex and by coupling down there will be residual changes and that's where a lot of our research is going beyond the basic physics and examining different aspects of the earth system so I think about 30 papers have already come out of Glenn's the dataset that we ran in 2000 early 2017 and there's already lots of researchers starting to look at a rise great it's it's good to hear that it's an active point of research currently so there's definitely opportunities for people that are interested in in doing this type of work the next question asks does the problem to be solved seem futile or are you inspired by your work I don't know maybe that's for both of us but personally I you know I think it's easy when people have not heard of this topic much before talking about this sounds a little bit like we failed right but at the same time you know if I get into a car I might hope that I'm not getting into a car accident but I'm very happy to have airbags and seatbelts I don't you know I don't smoke but you know if I'm at risk of cancer I'm happy to know that there's chemotherapy out there so when I'm originally getting involved in this it's it's recognizing that climate change is going to cause an awful lot of suffering if there's a technology out there that has the potential to reduce that suffering then to me it seems really really worthwhile to try to understand it same time I you know like I said I would rather be in a world where we did not have to rely on this and I still hope that we can all work towards that world yeah what I find hardest about the work is that we know how much time it takes to do good research and we're up against a ticking clock so I think a lot of the people who work in the field they spend a lot of time on their weekends their nights some of the students and postdoc the Doug has working I think they are always online you can always get an email back from them because they realize the urgency of the problem and there's just simply not enough people doing the research that we will have the answers to the questions that people want to know if people if we reach the point that we want to consider this so hopefully the society of all the countries in the world will get it together and we don't have to go there but it's again the problem is that it's increasingly less and less likely that some sort of intervention will be needed hopefully maybe that's in the form of carbon dioxide removal maybe those technologies will be available but I think the hardest part is that we're really up against the clock and the resources for this research have been limited yeah but it's good that at least we have like Doug you were saying the opportunity to do these computational models to kind of see what would be the impact of doing this type of work does the aircraft have to be manned or could we use satellite or space tourism technologies is a question that was asked as well my personal view would be I don't see what the purpose of putting a person in the cockpit of the aircraft would be right now quite frankly if you're ever behind in an airliner and the pilot disappears for some reason the airplane's going to go to its destination and land because it's programmed to do that so you know we use pilots and aircraft for the extreme cases where you know for the unexpected I would assume myself that if you were ever going to do this you would not be operating out of a commercial aircraft you would not be in commercial airspace and you would not have a pilot but I know wake smith who actually did the studies that I was referring to he actually thinks otherwise so we'll see thank you let's see how confident are you that you sufficiently understand the physics of climate to define the parameters for the climate models I'm not sure which parameters okay I'm not sure which parameters of the climate model there so there are parameters that we make assumptions about and we try to validate them on the present in a historical record so that's the plots that I've showed we are certain that with our models that we can reproduce the historical record and it gives us confidence for looking into the future but we are yes there is an uncertainty that some things and especially due to the rapid change in the earth system that some of the relationships that we assume that they will hold may no longer hold and then there's a amount of uncertainty and luckily the approach that we've been working with Doug again because of this background in aerospace engineering and control theory that the algorithm is designed to change right so if we're not meeting our goal then you can adjust for that you can inject more or you can inject less I might just sort of add in broad context we're never going to reduce the uncertainty to zero right there's uncertainty in the climate what will happen if we don't do geoengineering and there's always going to be uncertainty in what happens if we do hopefully we can reduce the uncertainty and we can at least bound it to some extent enough to be able to support an informed decision that says let you know either let's not go forward or let's go forward in moderation and monitor what goes on and learn more and hopefully find a way that if one is going to deploy it that one can do that in a responsible way that minimizes the risks but you don't have to you don't wind up needing perfect knowledge to do that hopefully better than what we have right now thank you and what's the expected impact of enhanced sulfate deposition is is that something that you have studied as well yeah my when I'm post-doc working with me then Visione actually wrote a paper on that last year the amount of sulfate pollution that we put in the form of industrial pollution right now is truly enormous it's of order 100 million tons a year of sulfur dioxide that comes down in the form of things like acid rain mostly relatively close to where you put it up if you were to put material up into the sulfur dioxide into the stratosphere of order say 10 million tons would cool the planet by an entire degree Celsius so that's like a 10 increase on what we're currently doing because the industrial pollution comes down close to where people put it up those places aren't going to be affected at all but places that are currently more pristine could see an increase in acid rain that's a risk the one really needs to understand and and weigh relative to the other impacts yeah so again future areas of study and it's good that you're a graduate student was able to to to write about that next question says has there been progress on research governance since the nas 2021 report what is NCAR doing to advance research governance and oversight i gotta let you take the first part of the question i think the answer is no that right now that 2021 report is sort of the de facto best rec set of recommendations nothing in particular has been implemented beyond sort of the basic governance that would apply to anything right so you can't go do an outdoor experiment without satisfying environmental oversight but there aren't right now any specific recommendations in addition to that if somebody were to consider doing outdoor research i do i am aware that there's language congressional language so recommending more effort be done to understand research governance and on the NCAR side so our focus is the physical science the modeling of the earth system but we do have an effort it's called the community climate intervention strategies effort led by simona tilms and trying to bring together different people from different fields including governance and so i don't foresee that ever being our key our primary area of expertise but we certainly want to be working with those individuals and communicating with them at various venues thank you let's see the question asks how well do the current models fit the known historical volcanic eruptions were any changes needed to be made to the models to better fit the cooling scene so i'll take that one so actually we have a scientist called michael mills in our chemistry laboratory and he specializes with that and i think he even gave an exploratory series lecture so he's done a lot of work on this with csm wacom and i think our model is one of the best models to be capturing the effects of volcanic eruptions and as duc said for example mount pinotubo observations show us that it emitted about 10 to 20 teragrams of so2 and in order for our model to get the right temperature cooling and the right distribution of aerosol optical depth i know mike has to use 10 teragrams so on the lower end of things so which other models may need to use a little bit more so that's where that uncertainty comes in and you know that comes back to the different aerosol parametrizations how the size distributions that we make again it comes down to the assumptions that we make about the things that are not resolved in the models thank you um the next question asks given a small change in impinging light makes a significant difference are there trends or forecasts on solar output for any part of the future so there is a we can go ahead there is an 11 year solar cycle where the solar variance varies by about 0.1 per cent per year so a lot of that is already captured in our models and it's an 11 year solar cycle that's a pretty slow varying so that's something we can put in the models but uh those yes so so we try our best to do to capture long-term changes yeah and those those changes are even smaller than the changes that we're talking about thank you um does SO2 or would SO2 affect air quality at the lower levels and I think you briefly talked about this uh yes to some extent so again there's the comparative relative to what our current pollution is of course our current pollution is a big problem right we we should be cleaning that up regardless um the other thing is the size distribution because they've been in the stratosphere for a little bit longer the particles tend to be a little bit larger when they come to the surface they wind up almost entirely coming out as wet deposition so in rain rather than dry um and so uh I don't I think my best understanding is that it does not a significant contributor directly in terms of uh to air quality thank you and I know um you kind of also mentioned this um the alternative to the SO2 um just in case people didn't um were unable to capture that first question and response yes um so I think I think Yaga mentioned earlier the calcium carbonate or calcite uh there are other there's lots of other things that could in principle work but that's probably the only other one that's sort of seriously considered it's just that we don't have the we don't even have the data really to put that into the climate model with any confidence right now well thank you so much um and thank you everybody um thanks for a highly informative talk um and for those of you watching we will have a survey that goes out if you registered through a VEMPRI um we'd love to hear your feedback on how you um how this talk went and how we can continue to improve our explorer series lecture um we don't see any more questions that have come in so I think um Doug and Yaga for being able to answer those questions and for a great talk as well again we will have this lecture up on our explorer series website um for future reference um and with that I don't know Doug and Yaga if you have any closing thoughts or anything you wanted to anything else you'd like to leave us with but just thank you for your time and if you have any questions don't hesitate to reach out okay likewise well thank you so much both and um thank you everybody for joining and tuning in for this explorer series we look forward to um see you in the future for our next talk um and with that I'll just say thank you and um reach out if you need anything we'll talk soon. Thank you. Bye. Thanks. Bye.