Embedded - 37: Surf's Up

Episode Date: January 31, 2014

Dr. Karen Shell and Elecia talk about modelling vs. building models, ocean albedo vs. ice, climate vs. weather, and science vs. policy. They gloat about being on vacation only intermittently. National Oceanic and Atmospheric Administration (NOAA) NASA's climate change home  Help run climate models on your home computer at climateprediction.net Karen's class will be looking at data from NOAA's Climate at a Glance  

You are listening to Making Embedded Systems, the show for people who love gadgets. I'm Elysia White, here with Dr. Karen Schell, to talk about the earth, the environment, and education. This is a bit of a departure from our usual engineering-centric podcast. It's a bit more on the science side. It's also a departure from our usual studio sound since Karen and I are on vacation. And we're recording from our beach house,

enjoying the pelicans in the ocean and the sight of little sailboats. Karen, thank you so much for joining me. Thank you for having me. We went to college together at a small college where the fight song was sung to the Mickey Mouse theme song. And you majored in physics. And then you did technology work a little bit. And then you left the Bay Area to get a PhD. I think that was totally the wrong choice. But you apparently have been having a good time. What did you do after you

left our little technology world? Well, I actually went to San Diego, which given that's where we are now, I think you probably think that's a reasonable choice. And I went to the Scripps Institution of Oceanography and studied, as you might expect, atmospheric science. I was sure you were going to say oceanography there. And then I went to the National Center for Atmospheric Research for a postdoc and studied atmospheric science, as you might expect.

And now? And then I, about seven years ago, I joined the faculty at Oregon State University in Corvallis, Oregon. And I've been there ever since. And you do mostly research with your job. Right. I do mostly research. I work with computer models of the Earth's climate. And the point is to improve the models to be better able to predict future changes in climate that will be occurring over the next century. So what's a day in the life of Karen Schell like?

Or a day in the office life? We don't need to hear too much about the cats. More email than anyone would like. But so I split my time. I will often be running a model on a Linux cluster that we have or spending more time usually analyzing output from the model and trying to understand why we're getting a certain temperature change in one version of the model versus another version of the model.

Why would you? I mean, is this like setting different reflectivity rates? It's more complicated. Sometimes it's code bugs. But normally it is the way that these different models take into account processes that occur on smaller scales and can really be resolved by the models so a like the butterfly thing that we hear about how it causes hurricanes all over the world those darn butterflies yeah yeah um so so that that actually is is right um the global climate models that i normally use have a resolution of 120 kilometers or so about clouds, butterflies, convection,

those processes all can occur on meters, tens of meters, kilometers, scale, and we don't have the computational power to resolve that sort of scale in the model. No, here in San Diego, I mean, you go from the beach and it's one temperature out 100 meters into the ocean, it's a different temperature even in the air,

and then a mile inland, it could be 100 degrees. Well, not today, but it could be. But it's never going to get to be 100 degrees here on the beach. It's always 80 plus or minus 5. So that makes the weather forecasting easy for here. But it makes your models more difficult. It makes it really difficult for the models. The boundary conditions have got to be strange.

Well, the boundary conditions, we typically, what we call boundary conditions would be like your land surface. And so the land surface, as you know, you're talking about the variation in air temperature, just think about all the different land surfaces you have, or you have ocean or desert or forests and so those also add complexity to the model and also are not you know you don't have a 100 kilometer by 100 kilometer patch of forest here's here's a square forest here's a square sand here's a square farm pretty soon we're going to be in minecraft and everything will be made of squares so so we do everything in terms of squares or or cubes but they're really big and so one of the ways a lot of models differ is that they have different ways of trying to incorporate the effects of these small scale features on

the larger scale circulation. So we're not trying to predict what the climate will be like here in San Diego in 20 years, but if you look at the some broader region what's going to happen to the overall temperature or for here more importantly maybe the amount of rain and so it's you still need to account for the smaller scale things going on in order to figure out how the the climate is going to change even on the broader scale. Yeah, that makes a lot of sense. Yeah. So these models that you run, do you, I mean, do you program them? Or do you push the button? Do you put the data in?

You don't go out and measure temperature all over the world. No, I don't. I don't measure temperature, but other people do. And that information is incorporated into the models. Personally, I do a little bit of model development. And you're probably going to laugh, but it's usually in Fortran. I like Fortran. It's a great language. And certainly there are some models where I wouldn't call them push the button, but there are models that you can download,

and if you read through the instructions, you could run them on your own desktop. And in fact, there's one of those, climateprediction.net, which is like some of the astronomy programs or things where you actually download a model onto your computer and it runs on the background and then it sends the data back to a center where it's used for climate prediction oh that's like the old study at home yeah you're basically using your extra cycles to help somebody else. It's pretty cool. And so you feed in different parameters and then run the model and then write papers on which parameters led to less world destruction.

We write papers on which parameter, which parameters lead to differences in, in the model. If we can understand what sort of feedbacks are different. So on sort of a higher level, viewing it not just on the base programming, what is the exact code level, if you look at it at a more emergent level, these models differ because the climate sensitivity in them is different. So that means if you double the

carbon dioxide concentration in 10 different models, and you look at what happens to the temperature in the year 2050, you'll get a dozen different answers. And if you look at sort of the first level of why they differ, you find that the feedbacks within the model are different. So one example of a feedback is the ice albedo feedback. So that means that if you have a surface that's covered with sea ice, so you're high latitude, well, it gets warmer. You have less sea ice because now some of the sea ice, so you're at high latitude. Well, it gets warmer. You have less sea ice because now some of the sea ice is melted. And the sea ice actually reflects a lot of sunlight. When it melts, you have this ocean, which is very dark, and so it absorbs even more sunlight.

So that's going to warm up. And it gets warm, and then the ice melts, and then it gets warmer, and then the ice doesn't reflect and then it gets, all right. So some models think that's really important and some models don't. Is that? Right. They all included less as opposed to unimportant. Right.

And do you have to, I mean, how do you decide which model to use? Or do you just try all of them with the same parameters and average the results? That can't be right. Well, first of all, you can look at observations. So you use satellite data, data from meteorological stations, in situ measurements of water vapor or you measure the ice and try to figure out which model does best. So which model has the best current temperature or has the best temperature for the last 10 years or so. So do you ever pretend that it's 1950 and you put in all the data that you have from then,

run the model and see if you can get to year 2000? Yes. And is that one of the ways you check models? That's one of the ways you check the models. Makes sense. Use the data you have the results for. But the interesting thing that we found, though, is that if you take a bunch of different models, so you have two dozen different models,

you actually get the best answer if you just average them all together. I was kidding about that. I really was. So you were on the right track that it actually turns out that just averaging all the models together works the best, which is pretty crazy and counterintuitive. I guess as an Uber model, the sum of all models it cancels out errors or something so so going back to fortran uh-huh um you you do a little bit of model development and do you have to worry about computer science topics or do you just have you just kind of learned what you need to learn and gone on uh i probably should do more about computer science topics uh it's my profession and i feel that way too so yeah um hesitation with that

there's always more to learn we are limited in the number of years you can simulate you know the and the more years you can simulate the more model years you can simulate. And the more years you can simulate, the more model years you can simulate, the more you can learn about the system, the more models you have to average together, for example. And you're limited by cycle time, by how much time you get on your Linux cluster? Cycle time is a limit. These days, actually, storage space is getting to be a real problem. The models are just so huge and generate so much data that some places they've actually decided to not save as much of the data because it's actually cheaper to rerun the model and save data if they need it than to store, you know, terabytes of data from one model run and then have hundreds of model runs and, you know, each of a dozen or two dozen different models.

And so it adds up quickly. But if you can run a model faster, then you can have either run it for longer, run it at a higher resolution, include more processes or do more different experiments with it. And so there is a certain amount of, you know, algorithms and computational efficiency that that some people do pay attention to in terms of model development. Are there ever computer science PhDs who come through and try to optimize things for you? Or is

it the atmospheric scientists coming in and saying, Oh, well, now I'm going to learn some proper computer science and optimize this puppy. There are a few people who have computer science training, but honestly, they normally go on and take jobs where they make lots of money. And so... I like that you've gone the other way, though. You know, get out of technology

and into something that pays less. But you need to save the world, so that's nice. Maybe. So are there gadgets in your line of work? You mentioned a little bit about data collection. Yeah, there are a lot of gadgets. It's the show for people who love gadgets. Right, right.

But if you think about it, the more data you have, the better. The better we can test the models, the better we can understand the climate system. And you get your data from, you mentioned satellites. Right. And probably temperature buoys in the ocean. Mm-hmm. Yeah. What else? Well, there is some people are starting to take advantage of the fact that you can have tiny sensors and put up a whole bunch of tiny sensors all over the place. So I have a friend who studies snow in mountains. she goes and she hangs tiny little sensors under trees all over the place in the sierra nevadas and cascades and is able to get a lot of data so she has you know a high

a very high resolution and that they're cheap sensors she can just go put out a bunch and if she doesn't get them the next year that's fine because and so so some people are really starting to take advantage of the fact that that it's not you don't have to have some big weather station to get all your data you can just put out little sensors other things like cell phones that are starting to have air pressure or temperature sensors, that's something the community hasn't really taken advantage of. I did a show not too long ago from the MEMS conference where we talked about humidity sensors in cell phones. I'm still curious about that, but that would be a data point.

Yes. still curious about that but that would be a data point yes i guess as long as i'm not showering with my cell phone which realistically is well i did recently get a new cell phone because it took a swim so yeah don't shower with your cell phone in fact don't do most things with your cell phone that involve water okay um back to the humidity sensors, which clearly if I had one, it would be broken now, like the rest of the phone. What else? What other sensors? I mean, can you use sensors that people have? Do you ever use home-based weather stations? I know Wonderground has a whole list of them. you can sign up to be one of the to have an official NOAA weather station at your house and there's some people who manage these you know for decades they've recorded the temperature and sent in the data to NOAA

but you can't just do it with a regular weather station that you've bought at the store and installed at your at your house but um i mean that may be changing i mean maybe nest will start sending information to google and and soon google will have a great database of temperatures and heating degree days and things were used with climate models. Do things have changed even in the seven years you've been working at Oregon State? Mm hmm. How? I mean, has there just been a huge explosion of data? I think we haven't... There's more data available, but we still don't know how to use it.

So the big explosion of data was really when we started getting satellite data because then we could have worldwide coverage of all sorts of things like land surface reflectivity, cloudiness, you know, just even places where you don't have a weather station, you don't have lots of people in the Southern Ocean, for example. Wait a minute, cloudiness. I didn't think about that before, but your 120 mile or 120 kilometer blocks, do you have like this block is cloudy and this block is sunny?

I mean, clouds aren't, well, it used to be that way, but it's a little bit more sophisticated than that. But clouds are actually the most complicated climate feedback. So if you take those couple dozen models and try to figure out, when I was talking earlier about the climate sensitivity and the different feedbacks, the uncertainty in the cloud feedback

is the largest uncertainty in all of those models. And that's because clouds are so difficult to model. Well, yeah, how do you get the one with the little turtle that's swimming across and then the sheep and then you know the dragon it does seem really hard it's a it's a fluid mechanics problem but it's more than that really yes it's a fluid mechanics problem but it's also a thermodynamics problem because you have condensation. Oh, that's known as rain. Yes. Well, it condenses and then it forms cloud droplets, which sometimes rain out, which make it frozen.

And you've got all of these phase transitions and they just make atmospheric modeling a pain. A fun pain. I'm... Fun. Fun, but difficult. So what papers have you written? Actually, while you're thinking, you tweeted about how pleased you were

when you saw a paper title that you were kind of excited about it you thought it was interesting and then realized it was yours do you recall which paper that one was i think that one was one where we were trying to figure out how to relate the climate feedbacks that we observe in models trying to figure out how to relate the climate feedbacks that we observe in models, trying to figure out if there's some way we can use observations to figure out how big those feedbacks are. And so this was a paper where we were looking at historical runs of climate models. So looking at 20th century runs of climate models, which are trying to replicate the last century, and looking to see if the water vapor feedback, for example, so

water vapor is a greenhouse gas. And as it gets warmer, you get more water vapor in the air and so that water vapor then is going to strengthen the greenhouse effect which then makes it warmer which then means you have more water vapor in the air which then strengthens the greenhouse effect etc etc and so we were looking at how strong that feedback was in a variety of different models and trying to relate that to the feedbacks for those different models for sort of future runs. Now, you might think a model would just have a strong water vapor feedback. It would always have a strong water vapor feedback, but that's not actually the case. Because it depends on so many things. Yes. I mean, this is, you mentioned that the amazing amount of data that comes out of these models, there's a fair amount that goes into them, isn't there? Yes, there is. And so you tweak one little thing whether it's a parameter or the temperature

in bangkok a year ago those count they do um one thing when you when you look at climate versus weather some of those things sort of wash out what's the difference? So weather is what the temperature is right now. So today we have a temperature. It's cloudy. It's not cloudy. It's raining. It's not raining. And that's what we're experiencing right now.

And you can forecast what that will be in a few days. So you can forecast what the weather is going to be like on Friday. And then, and there's a certain skill to that, um, you know, depending on, on the particular time of year and where you live, it may be more or less skillful, but, um, but once you get out to about two weeks, um, there's actually a theoretical fluid dynamics limit that you don't really expect to have skillful forecasts beyond that because of the butterfly effect that these small errors in the current climate, in the current weather, I'm sorry, when you run those models forward in time, those little small errors magnify. And so after about two weeks,

you can't really do a skilled weather forecast. There's just too much uncertainty. Right. Okay. For climate, you're not really interested in what the particular temperature is going to be on a certain day in 2020. You're interested in more, well, how much warmer the January is going to be on average, how much more rain or less rain would we expect to get on average? Would you expect to get rain a different month than you get it now? Seems like there's a big distance between weather in the next two weeks and climate in the next 10 years. Is there something in between that talks about drought in California and ice storms

throughout the rest of the country, including New Orleans. Ice storms in New Orleans. Sorry. Yeah, yeah. You can do seasonal forecasts. And so, again, there you're not able to predict a particular day. But you can say, well, based on the ocean patterns that we see. That's like the hurricane forecasts. Right.

They try to predict how many hurricanes there will be this year. And what happens is with weather, you're really dealing with just the atmosphere. And the atmosphere doesn't have a long memory. So after a couple weeks, it's just chaotic. But when you deal with things like ocean or ice or changing atmospheric concentrations of carbon dioxide, those are longer-term processes. The ocean has a lot of heat in it. It's a huge thermal mass.

And so it's not going to change its temperature that rapidly and it's also going to be having a huge influence on what the atmosphere is doing and so if you know what the ocean temperature is going into a hurricane season for example you don't know what the weather is going to be doing two weeks from now but you're you might be able to tell that the weather is going to be doing two weeks from now, but you might be able to tell that the atmosphere is going to be a lot warmer because the ocean is a lot warmer. And so you can't predict, oh, we're going to have a hurricane on this particular day, but you can say the conditions are more favorable for having a hurricane this season than they were the previous season.

Do you think that better software, more computer time, more disk space will improve weather forecasts, or is that really about climate? Weather forecasts are still being improved. But you don't really care because you're all about climate right all right i'm going to quit the weather angle here what has been your favorite line of research i think one of the most interesting things that I've done in the past couple years was some collaborators at University of Michigan, Mark Flanner and some other people, looked at the albedo feedback in the Arctic for the length of the satellite record. So looking at the past 30 years, and we tried to figure out how strong the actual feedback was. So how much did the albedo of the sea ice, the ocean, the land, the snow, how much did that albedo change if you look at the last 30 years?

And we earlier had the example of the albedo goes down as the ice melts because it doesn't reflect as much. Are there areas where you get higher levels of albedo? I mean, it isn't all going down. What goes up? There are some places where the, so if you have deserts that are expanding those have a higher albedo they're shinier and brighter right and and cities if you put in more reflective uh you know if you cut down a forest, which is dark, and put a city there, that also might change the albedo.

If you paint the top of your house white, that can change the albedo. hemisphere was that the climate models that were simulating the last 30 years were only getting about half of the change in albedo so they were they were only getting about half of the effect in the climate models as opposed to what we were really seeing which would suggest that the climate models were underestimating the this ice albedo feedback they were underestimating the warming this extra warming that you get due to melting snow and ice and there wasn't anything that was replacing it right so there wasn't an offset there wasn't another thing that was being misestimated in offsetting that error so this this would indicate that the models are not indicating as much warming as is happening or as happened right over the period there could be some compensating error somewhere else that you know is resulting in you know something some compensating error in clouds, for example.

But in terms of just looking at the albedo for snow and ice, the models were underestimating that change. Even if you averaged them together? Even if you averaged them together, yes. And so you end up looking at a lot of historical data. Yeah, yeah. I didn't realize that. That's all we have.

That's how you check your answer. Right. And does it horrify you looking forward? I mean, when you run these models going ahead, do you just cringe? Well, it's when you get some results and see that the models are underestimating some warming effect, that is concerning. And there have been a lot of studies that have come out recently that

suggest to me that maybe the warming that we should be expecting is on the higher end of what some of the sort of standard estimates of climate change are. And so that, you know, it's one of those things where I really hope I'm wrong. And I think a lot of climate scientists really hope they're wrong. We would love to find some compensating effect that is going to cool the planet down. Do you think there are some climate scientists out there who are planning to blow up volcanoes in order to put some ash in the air

in order to cool down the environment? Have you joined the Evil Mad Scientist League of Climate Scientists? You can tell me. There are people who are studying that well i mean there was there were a couple of big volcanoes uh early in the previous century that seemed to cause quite a bit of cooling they do but they only last for a couple of years yeah so you've got to keep keep with your mad scientist machine making the volcanoes erupt every couple years if you want to do that.

And they did kill a lot of people. Yeah. Unrelated to keeping Europe nice and cool that summer. Yeah. And if you stop making the volcanoes erupt, it warms up really quickly. Because now you have a bunch of ash in the air. Well, actually, once the volcanic,

once the aerosols that the volcano emits, once those settle out of the atmosphere, now you have all that warming that would have happened that you prevented by having the volcano erupt. It comes back. It just turns on all of a sudden. Rebept. Yeah.

Do you talk to people getting away from blowing up volcanoes and not moving to blowing up other things despite the fact that that's what everybody's expecting i'm sure do you talk to people who don't believe in climate change sometimes i i talk with people who have an open mind so if someone wants to know what the science is then I'm happy to talk with them and I'll go and give presentations and just try to let people know we have all of this evidence

we have all of this physics that tells us what happens when you add more greenhouse gases to the atmosphere, when you add more CO2. And when I'm giving a public talk, I stay away from policy. So I don't want to tell you what to do. I just want to say if you put this much CO2 in the atmosphere, then this is what the warming will be. And these are some of the consequences of the warming in terms of, you know, drought or storms and things like that. Um, but what you want to do about it is, you know, I can't, I can't tell you as a scientist what to do about it I

have my personal views but I think I won't communicate the science and so but if people some people aren't open to the science at all and in that point in that case it's it really isn't effective because they're not interested in the actual science. Have people gotten more interested in talking to you or more interested in talking about the science since the weather's gotten a bit strange? What with ice storms in New Orleans and drought? It's definitely come up. I think it's, and it's been in the media more. TV shows are talking about it more and you see more news articles about it. can happen naturally. You didn't need that to motivate the conversation, but it's certainly good to make people more aware that the climate is changing, I think.

Yeah, and it's hard because the climate is a 30, 50-year process, and we only remember about a week at a time. Right. And one week is not climate, that's weather. Exactly. But a lot of the things I do see in the news are highly politicized. And how do I, how, where do you get your news from? So there is an international panel on climate change, IPCC. And every six or seven years, they put out a huge report and summarize the state of climate science. They don't do any new research, but they read, and summarize the state of climate science. They don't do any new research, but they read,

take all of these peer-reviewed articles and try to make a synthesis report of what we know. And there was one that just came out in 2013. And that's a great starting place in terms of what the science is and what we know. It's a little bit difficult to read. So there are various places that talk about this report. So if you find some news article and it's quoting something from the IPCC reporter or something related to it, unless it's saying, you know, the IPCC report is politicized, then, you know, it's probably they're getting their science from some reasonable source. NASA and NOAA, the National Oceanographic and Atmospheric Association.

Don't look at me. I'm not the one who's worked for them. That's it. I just say NOAA. They are both a really good source of information. And they have, if you go to their websites, they have a lot of educational materials available. And part of their mission is to communicate about weather and climate to the public.

And so if you see a press release or something from a NOAA or NASA website, I would tend to trust those. And then generally the further away from the sources that you get, the more, you know, the more careful you have to be. Well, I don't really want to go read peer-reviewed journal articles. And if I did, I probably wouldn't be in climate science. And the NASA and NOAA websites have lots of things that are not for scientists.

They're for public consumption. Right, right. And do you, you said you don't want to talk about policy. But as a scientist, isn't it, do you feel some responsibility to make sure people are hearing what's coming out of the science community? I feel the responsibility. I don't think that every scientist needs to be communicating to the public. In fact, some scientists shouldn't be communicating to the public. Oh, name names. Don't name names.

But I do try to make it clear when I'm talking about science and things that are very well established versus if we start to veer off into policy. And so for, for, uh, normally if I'm giving a talk to the public, I just try to avoid policy altogether. But if I'm talking to, to, you know, a friend or something, then as you know, I, I don't have a problem talking about policy. So, uh, but I do want to make clear that distinction so that, uh, you know, this is the science and this is, you know, the area of expertise. And then when we get beyond that to, you know, we really need to do something about this, that sort of thing, though, is something that necessarily know more than, than, uh, any other person about how the whole political machine functions and how decisions are made. So.

And so you mostly are interested in outreach and education and making sure people understand, and then hopefully they can make wiser decisions sounds like a reasonable summary okay so so let's say i am that person well now i i'm i i i agree i don't want to be a burden on the planet i want to leave the planet in a state that future generations can then build rockets and go explore the rest of the universe, because I'm not sure that that's going to happen. I'm not sure I'm making it past Jupiter. And I do want the humankind to. What can I do? I have an electric car.

We have solar power. I'm mostly a vegetarian. So I'm doing all the things I can think of. But what advice would you, and I'm not even sure those are the right things. What advice would you say? So, and again, this is now beyond the science in terms of what I think might be most effective. So all of the things you list would decrease your carbon footprint.

Certainly, you know, buying energy-efficient appliances and energy-efficient cars and things like that do decrease the amount of carbon dioxide you produce. But relative to the amount of carbon dioxide that everyone produces, it's pretty small. So really, you know, this is a global problem. We have to get people all over the world to agree to make reducing carbon emissions a priority. And that's, in one sense, that's beyond what an individual consumer can do.

So in terms of effectiveness, and here I'm just, I'm going out on a limb here. I don't actually know what the most effective thing is. But first is, you know, letting your policymakers, letting your elected officials know that this is a priority and that you will vote based on what they do. And hopefully if you get enough people who are telling politicians that this is a priority, things will happen. And here in California, it seems like there's a little bit, it's a little bit easier of a, yeah, yeah. And so to some degree, in the US, I think at the state level, you're going to see more of the success in terms of carbon emissions and maybe carbon trading or support for alternative energy, it probably is going to take longer for the national, for U.S. national policy to catch up. The other thing that I think does make an effect is, so you buy an energy efficient washer and it is going to be you know first of all it may you know reduce your electricity or water bill or something and reduce your emissions but it also

sends a message to the the manufacturers that this is something you value. So by buying energy efficient appliances or cars. By paying 5% more, you're saying give me more like this. Yes. This is important to me. This is a feature that I look for in these different products. And so that's sort of the consumer side, the market- based approach that, you know, to try to encourage manufacturers to make energy efficiency a higher priority. And so, in some sense, I think that is more effective than the actual carbon reduction you get from your more efficient car. Because if you can get, you know, one person doing it, the amount of difference is pretty small. But if you use peer pressure to convince all of your friends and all of your neighbors to buy a fuel efficient car,

then you can start to have an effect. So you need to do it and you need to talk about it. Yes. And help people realize there is science that is out there and it's not hard. I mean, NASA and NOAA have pretty good ways of talking to you that don't involve, here's the fluid mechanics equation that you need to understand

in order to even begin to comprehend the intricacies of butterflies and climate change. Is there anything else you'd like to say everybody should, I mean, recycle? Wash their hands after sneezing. Especially if you're going to be around me. Yes, please wash your hands after sneezing.

I'm on vacation in a house full of germaphobes. That's great. I'm not one of the germaphobes. You just keep saying that. You were, this weekend, you were grading quizzes for a climate change class that you're teaching.

Is that right? Okay. What did you ask and what was the best answer or worst? Put your students on the spot. Or were you actually going to assign them to listen to this? I'm hoping they don't find this. Then they'll be like, why weren't you grading our homework

rather than recording a podcast? But did one of them really say that aliens came to... No, none of them said that aliens are responsible for global warming. So that was good. I would totally have given extra credit for that. Actually, I read over your shoulder for next week's quizzes. What was the website with all the data? So that was a NOAA website.

That was the National Climatic Data Center, the NCDC, which is a division in NOAA. And sometimes it's a little bit too difficult to figure out where on the NASA or NOAA websites to find the nice interfaces, but they do actually have a lot of websites where you can go and play around with some of the data yourself. precipitation, northern hemisphere, southern hemisphere, and what years and what area, the west, the southwest, Europe. And that was really neat because you could see the average and the trend lines and the year to year. And it was very satisfying to my data nerd soul. Yeah, so there's a lot of data out there. And there are a number of different websites which have nice interfaces so that you don't have to download the data

and put it in your Excel spreadsheet and plot it. Although they don't allow that because I was kind of into that. Yeah, so if you want to do that, there is even more data that's available as text files or Excel spreadsheets that you can download. But there has been a push recently to make a lot of data available so that people who know something about science and can look at a plot and understand what it means can actually go and look at the temperature record, go and look at the carbon dioxide record. And so the data is out there. And so for people who are skeptical and don't really,

you know, want to actually see, see the data before they decide what to do about climate change, or just want to understand better what's going on there there are a lot of places that you can do that and i can certainly give you the link to the website if you want to get the link of the show notes yeah so if you want to play with that data you can go play with it and i have no problem with the skeptics as long as they're willing to spend a little time playing with the data. And I teased you last night that I found one that was perfectly flat that indicated no climate change over the last hundred years. And how many did you have to look at first before you found the one that was flat? A few.

Maybe more than a few. Maybe I found the only one. Well, I think that's about it for the show. I think we should get back to vacation. Are there any last thoughts you'd like to leave us with? It's nice and sunny outside, so there's not really an incentive to keep this going any longer. All right, then.

My guest has been Dr. Karen Schell, Associate Professor of Atmospheric Science at Oregon State University. Thank you for being on the show. Thank you, Alicia. As usual, I'd like to thank you listeners for listening, and to Christopher White for making the audio sound better than it does at this very second. I'm sure he's got some filters planned. And now I'm getting back to my regularly scheduled vacation. So your, you know, final thought of the week is surf's up.