Reducing harmful air pollution has led to a surprising effect — more hurricanes in the North Atlantic
Theo Severud
CNN Article
Scientific Literature
Aerosols are solid particles or liquid droplets suspended in the gas phase. Some common aerosols include: Sea spray, which can contain salt, organics, and biological species (Schiffer et al.); Mineral dust, an aerosol formed from wind blowing over deserts and drylands to lift small particles into the atmosphere (Goodman et al.); and Smoke, which can contain carbon-based particles from biomass burning (Schill et al.). Atmospheric aerosols, also called particulate matter, create a cooling effect on Earth because they are able to reflect light back into space, rather than absorbing it like greenhouse gases. However, these aerosols, especially fine particulate matter (<2.5 um in diameter), also have a large effect on human health because they are able to deeply penetrate the lungs, with prolonged exposure leading to long-term health effects, such as cardiovascular and respiratory issues (Pinkerton et al., Pope and Dockery).
A 2022 article in Science Advance by Hiroyuki Murakami aims to investigate a link between changes in atmospheric aerosol concentrations and the abundance of tropical cyclones (TC), also known as hurricanes. This study uses a combination of observed data, models, and statistical analysis to analyze a decrease in TC frequency over the western North Pacific and Southern Hemisphere and an increase in the North Atlantic based on the respective increase and decrease of anthropogenic aerosols (Murakami sometimes specifically mentions simulated Sulfate aerosols based upon an early (CNTL) and late (ALL21) decade experiment measuring mean aerosol emissions). The article generally links the reduction in anthropogenic aerosols in Europe and North America to Northern Hemisphere heating relative to the Southern Hemisphere, and draws a large comparison to increased aerosols from China and India which have helped cool these areas. This is connected with a discussion of wind speeds and direction, both locally (Relative to an ocean) and globally which ultimately gives a strong message that aerosol concentration has some effect on TC frequency, even if we cannot determine the exact link due to the number of compounding variables.
This article is very well written, however, for someone like myself who does not have really any background in modeling methods it is a difficult read (Not the authors fault). The article makes many, in my opinion, strong arguments for the effects of decreased and increased aerosol concentrations on localized wind direction, wind speeds, global wind direction, and hemisphere temperatures which would influence the frequency of TCs in a given region. At the time of this blog post, this paper has been cited 38 times, which would indicate that it provides a new perspective on the global differences in TC frequency, whereas previous work has mainly focused on more localized regions. I think the most interesting part of this article is the last point the author makes in their discussion, where they point out how any changes to emissions can have widespread effects on our planet, which based on this article I would agree with. If you did not understand how bad aerosols are overall, you might think the solution to this problem is to increase aerosol emissions, which would increase cooling, and take a reverse effect when compared to the long-term studies used in this papers simulations. Increasing aerosol emissions is obviously not a solution to this problem, but it does put into perspective how important political changes, related to emissions, are because of any pollutants ability to have an impact on various aspects of the environment.
CNN reporter Rachel Ramirez published an article shortly after the release of this paper titled: Reducing harmful air pollution has led to a surprising effect — more hurricanes in the North Atlantic. In this article, Ramirez summarizes the key findings of the paper, namely the inverse relationship between aerosol concentration and TC frequency in both the North Atlantic and the western North Pacific. This article also seems to include an interlaced interview with several scientists in the field, including the paper's author, Hiroyuki Murakami. The interview with Murakami does a great job in summarizing the body of the paper for a general audience, without getting bogged down in the modeling and statistical methods of what is admittedly a very methods heavy paper. This overall creates a strong narrative to stay on target with the papers main finding and allow a general reader to stay involved. The interviews with other field-related scientists are actually very valuable to this interview because they help show how important this work actually is, rather than what some readers might take away without them, which is that this is the "expected" result.
Overall, I was extremely happy with this article and its summary of Murakami's paper, but I first want to discuss what the article did not do so well. First, and arguable biggest, was their definition of aerosols. Given that there is zero discussion of the paper's methods (More on this later), I think a slightly more substantial definition of what an aerosol is would have been great. Whether that was in the form of specifically stating that these are solid particles or liquid droplets suspended in gases, rather than saying they are "tiny particles of pollution that float in the air," or giving a couple additional sources of aerosols, such as sea spray or mineral dust, rather than just the main 20th century sources. As previously stated, there is no discussion of the methods used in this paper, but I am on the fence about whether or not this is a bad thing. In all honesty, I understood very little about the modeling methods used in this study, but could still generally understand the paper. I believe that if there was discussion of the methods in the CNN article, it would likely only distract and discourage readers. Not understanding the methods does not limit ones ability to understand the importance of the study, so I think that Ramirez' choice to not include the methods is justified. Again, I was extremely happy with the CNN article overall because it gives the reader the key takeaways of the study, briefly explains what these results are based on, and gives additional information (Expert interviews) that puts the work in perspective for the reader. Given this, I would give the article a 9.5/10!
Additional references:
Schiffer et al., Sea Spray Aerosol: Where Marine Biology Meets Atmospheric Chemistry, 2018 https://pubs.acs.org/doi/10.1021/acscentsci.8b00674
Goodman et al., Trace element chemistry of atmospheric deposition along the Wasatch Front (Utah, USA) reflects regional playa dust and local urban aerosols, 2019 https://www.sciencedirect.com/science/article/abs/pii/S0009254119304243
Schill et al., Widespread biomass burning smoke throughout the remote troposphere, 2020 https://www.nature.com/articles/s41561-020-0586-1
Pinkerton et al., Distribution of particulate matter and tissue remodeling in the human lung, 2000 https://pubmed.ncbi.nlm.nih.gov/11102298/
Pope and Dockery, Health effects of fine particulate air pollution: lines that connect, 2006 https://pubmed.ncbi.nlm.nih.gov/16805397/
Great analysis Theo, I agree with your high article rating. The article clearly gets to the paper's main points without wandering too deep into the methodological weeds. It is in these weeds that I find myself wandering and pondering. Something I found interesting was that the study only used TC data over the period 1980–2020 (40 years). The period is relatively short, and I wish that more observational data could have been used. Realistically, how many hurricanes even occurred in these 40 years? Further, I would have found it more interesting if sampling included the 1940s-1970s, years when there was global cooling associated with high levels of aerosol (Nasa: https://earthobservatory.nasa.gov/features/GISSTemperature/giss_temperature4.php). I would have found the author's argument more convincing if they showed a decrease in hurricanes during the years of global cooling. Another issue I have with the methods is the global grid sizes. At a 5º x 5º grid size, modeling a hurricane (about 350 miles in diameter or about 5º x 5º grid size) would surely be challenging. I understand that the modelers smoothed over the grids, so this is less of an issue, but I am still not convinced that at such as scale the models would accurately reproduce hurricanes. Obviously, all science has its limitations, but I find these especially problematic. What are your thoughts?
ReplyDeleteHi Joe! I am not hurricane expert, however, I would assume there were a fair number of hurricanes over that 40 year period, partly because there are a lot of hurricanes even today that occur over the ocean and never reach main land North America. I agree that comparative data under global cooling would be extremely interesting, but I was not able to find any sort of publication related to that and hurricanes at a brief search.
DeleteAs for the grid sizes, I think that is a very reasonable concern. As I previously stated, I know very little about modeling, but based on this authors previous work in modeling related to tropical cyclones, I am inclined to believe in their work. I also believe that making large grids could be a product of how infrequent hurricanes actually are- if you make grids 1/25 the size (1 degree square) you might run into a scenario where you have a lot of 0s and 1s for hurricanes, meaning you get so few in some areas and can't make predictions based on that. I think somewhere in the middle would be reasonable to me (Perhaps 3 degree square, if possible) to give a middle ground of large enough to have frequent hurricanes, small enough to supply a strong data map.
I agree that Ramirez did a great job summarizing a complex modeling paper into a digestible news article. I would also agree that the lack of a further definition of what aerosols are was a bit disappointing. Do you think this stems from Murakami's lack of definition of anthropogenic aerosols in his paper? Do you think he could have further investigated if there are differences in the contribution of PM10 and PM2.5 to hurricanes, and if that would have made a difference in the results?
ReplyDeleteI think that is a very large part of it- because the paper as a whole does not discover differences between the type of aerosol and its contribution to tropical cyclones, Murakami does not really need to provide a background on aerosols. If he say differences between the type (Or size) of aerosol and the frequency of TCs, it would be extremely important to talk more about aerosols in the introduction (Which would be reflected in the news article). Really, there wasn't a lot of dislike about this article, but I think everything can be improved, so maybe the point about aerosols is undeserved.
DeleteYour analysis is very thorough, and I agree with most of your main points. The article was concise and to the point, but still very thorough on the points that the general public is likely concerned with.
ReplyDeleteI thought it was vital that the author of the CNN article clearly stated that although the conclusions drawn by the paper do not mean that we should stop making attempts to control air pollution. I feel that at face value, this may seem like a logical conclusion for someone without background information to come to. Ramirez paraphrases one of the authors, "When a person stops smoking, they improve their health and can avoid cancer. But in some cases, quitting also comes with side effects, including gaining weight and feeling stressed." This is a very tangible and digestible analogy because most people understand that one should stop smoking, even if it causes negative side effects because the alternatives are presumably worse.
Because of the somewhat contradictory nature of these findings, I thought it was a good idea that Ramirez added additional context about the immense complexity of atmospheric science and even hyperlinked another CNN article about how research reports on heatwaves in the Northwestern US. This makes it clear that this particular paper is studying one relationship in a very complex system with numerous other interrelated factors. The quote from Gabriel Vecchi furthers this idea, explaining how the study of aerosols, in particular, involves many uncertainties.
I agree completely with your points on the quitting smoking quote. I was even partially conflicted on my first read because some people won't see the benefit in quitting smoking, but those people may not be very inclined to listen to this type of article anyway. Like all atmospheric and environmental research, there are too many factors to say that one thing will always be solely beneficial, and this is a perfect case- we know that an increase in aerosol concentration is generally bad for air quality, but on the other hand, we could reduce tropical cyclone frequency if we put more aerosols into the air. I even fear that people might read this and think "What if we just put aerosols into the atmosphere above the oceans" thinking this will reduce hurricanes and the aerosols will stay above the ocean. Ultimately, there will always be pros and cons related to specific pollutants and the atmosphere, and I think it's important to keep that perspective because in this case the cons of increased aerosol concentration would heavily outweigh the benefits.
DeleteThe article was quite good at summarizing the paper, but didn't mention much about the methods, which may leave some readers (like me) to question how the scientists behind the paper got to such a conclusion (the paper never mentions computer modelling). Though its true that most readers won't understand the systems of aerosols anyways, I still believe that CNN should not have given readers so much work just to find out more about how aerosol pollutions affect tropical storms besides saying "the ocean could absorb more sunlight." This kind of writing makes the paper sound more like a bunch of guesses and hypothesis rather than an actual analysis of empirical data. Perhaps the CNN author should have also considered adding a few graphs/correlations rather than some blanket statement, or maybe add more emphasis on the complexities of atmospheric science.
ReplyDeleteThe analogy to smoking was quite helpful though, as it helped me understand the complex nuances behind increasing or reducing atmospheric aerosols. The article could definitely have explained more on the aerosols part though, such as pinpointing what type of aerosols, and what size (the specific elements/components of aerosols were not noted in the article).
I think the points you make about modeling are extremely important. There definitely should have been mention because simply not saying anything is definitely worse than saying something the reader might not understand. When going into this assignment, I decided to first read the news article and then read the scientific paper to see how much was left uncovered in the news, and this was a glaring miss. My main thoughts on my first read were "How did they get to these conclusions" because while the news article gives the reasoning, it does not give how they got to those reasons (Modeling). I do think that not adding figures was the correct decision because the figures are complicated if there is no background of modeling, and most figures build upon others, so one stand alone figure would likely not be super helpful. Overall though, I completely agree that there needed to be at least some mention of the modeling process.
DeleteI think one of the major challenges with science communication is finding a way to express findings in a way that is difficult to misconstrue, particularly when there's a controversial topic involved (in this case, pollution and climate change). I think Ramirez did an acceptable job with this in her CNN article, despite the somewhat click-bait-y title. Her 5th sentence in the article establishes that the answer to the increase in hurricanes is not increasing aerosol pollution, which is nice. There have been a few articles this semester that required you to read most of the article before the "catch" to the title is addressed. Ramirez specializes in articles about climate change and this likely informs her decisions on which scientific information to include in each article she writes -- while I wish she included a more rigorous definition for aerosols, I can understand her not giving any attention to Murakami's methods. This article isn't written for scientists and I'm not sure that including information about the computer modeling will make much of a difference to her target audience.
ReplyDeleteI think your last sentence is the best takeaway for us analyzing the article- the article isn't meant scientists. Unfortunately, with the way popular news works in todays society (Relating clicks to pay) it is difficult to avoid the title being somewhat clickbait no matter what. I also agree that in terms of "flaws," a lack of rigorous aerosol definition is more pressing than the lack of discussing the methods, though I do think modeling should have been mentioned (Even if extremely broad). I was overall very happy with the article, and to your point about not misconstruing the findings, I think she did a great job because she gives the "clickbait" of aerosols equals hurricanes, and then gives the full explanation. I think it would be hard to avoid any sort of clickbait unless the structure of this part of the article was reversed, but I totally see your side to this.
DeleteUpon opening up both the CNN article and scientific paper, I was pretty concerned about the validity that was going to be presented by the CNN article. As the paper was quite methods heavy with numerous figures, I was not sure about how accurately the article was going to explain the results, especially based on the article length. Upon reading both, I was pleasantly surprised and impressed, leading me to agree with your high rating of the article. Based off the title and the caption to the image, I was very scared that she was going to suggested that adding more aerosols into the atmosphere was going to be the solution to the increase in hurricanes. Not only did she avoid this, she also drew attention to the complexity of aerosols and air pollution. I really appreciate how concisely Ramirez was able to explain the results, and that she was able to clearly communicate the importance and relevance of this study to the broader study of climate science. She also brought in some good credibility with clear quotes from the many scientists in the field, including the author of the study. It it hard for me to critique her decision to not include any description of the methods in the article, as it was pretty difficult to follow, but I think she could have said that the study used a mixture of modeling, statistical analysis, and observation data (as you did in your blog post) to get the results. I think this would have further added credibility to the results, as I as a reader, would have been very curious as to how one gathers data at this large of a scale accurately. I also would have liked for her to mention that this study only accounts from data within the last 40 years, using 1980 aerosol concentration as the baseline, because if it was much longer time than 40 years, 33% could mean a much less significant increase. My last critique is her very lose definition of aerosols, like you mentioned, especially because her definition makes it harder to visualize how they are different form gas emissions. Overall though, I think the article did a great job conveying this information to a more general audience and agree with your great analysis!
ReplyDeleteI absolutely agree with your opinion that Ramirez did not do a great job with defining what aerosols are. The definition for aerosols being just “tiny particles of air pollution” does a disservice to her audience, especially people who are just learning what an aerosol is. This definition gives the unfamiliar reader the idea that aerosols are solely just air pollution particles, which as you pointed out is not the case. This is especially true in the sentence of “there are natural aerosols, but much of the pollution in the early to mid-20th century came from sources like industrial smokestacks and car exhaust”, where it comes off as Ramirez implying that the natural aerosols are also pollution (which is not always the case). While this simplified definition does not impact this article that much, I believe that it could cause readers confusion in different articles if they were to read about aerosols in a non-pollution context like for disease transmission (biological aerosols), medical treatments or even natural fog/mist.
ReplyDeleteBesides that, I think the discussion about the impact of the study is incredibly well written in the CNN article. I really liked the context Ramirez included about aerosol studies and the emphasis that the solution to the increase in tropical cyclones is not to increase aerosol production. I also think the distinction between greenhouse gases and aerosol impacts on climate systems was also incredibly helpful as it makes it easier for someone not knowledgable on climate science to grasp the idea that aerosol pollutants are not the same thing as greenhouse gases (something that people tend to solely associate air pollution with more often than not).
This comment has been removed by the author.
ReplyDeleteI agree that the definition of aerosols is not clear. The CNN articles did poorly on that by saying "They are tiny particles of pollution that float in the air". The article could have a proper definition with more categories like PM 2.5, PM 10 and so on. Also, the position of that definition is not good. I feel like it would be better to bring it up at the very beginning so that the public reader would have some background to that they will not get confused what the aerosols are at the first three paragraphs.
ReplyDeleteBut besides this, the article made a good summary of the paper. They point out that although aerosols could reduce temperature, it does not mean it is a solution to global warming because it could also cause health problems. Treating aerosols is complicated because reducing aerosol pollution would cause more hurricanes. And the simile with quitting smoking is actually great. I am quite satisfy in reading this article.
I notice many people commenting that CNN did not discuss methods of the paper. However, I wonder what percentage of the audience would be lost as soon as methods are discussed. I would assume the audience is reading the article because they want the methods. The paper is linked in the words "new study," for the engaged reader to find said methods themselves. I do wish, however that any of the data from the paper was displayed in some sort of figure. I would have loved to see one of the maps discussed in a little more detail.
ReplyDelete