Oil Fields Release Far More Methane Than Thought, Study Finds
Baker Angstman
News article: https://www.nytimes.com/2024/03/13/climate/oil-gas-industry-methane-release.html?searchResultPosition=1
Journal article: https://www.nature.com/articles/s41586-024-07117-5
Methane is the second most abundant greenhouse gas in Earth’s atmosphere, trailing only carbon dioxide. While it has a shorter lifetime in the atmosphere than carbon dioxide, methane is roughly 80 times more effective at trapping heat than carbon dioxide, making it one of the most potent and destructive greenhouse gasses. A major source of methane emissions is the production of energy, particularly onshore oil and gas production sites. This methane can be released into the atmosphere through leaks at well sites, processing plants, pipelines, or other energy facilities. When released unburnt into the atmosphere, methane accelerates global warming. This environmental problem is complicated by technological limitations that have made it difficult to accurately quantify methane emissions at oil and gas production sites.
The New York Times article written by Hiroko Tabuchi reports on a paper published in Nature that found US oil and gas systems emit more methane than previously thought. Tabuchi begins the article with a couple of standout statistics from the Nature paper then gives a succinct summary of methane as a greenhouse gas and its connection to climate change. A very brief overview of the methods used in the scientific study and discussion of the economic implications of the methane emissions follows. The remainder of the article discussed methane emissions more broadly, featuring an analysis released by the International Energy Agency and methane’s connection to meeting the Paris climate agreement’s goal of limiting global warming to 1.5 ℃ above pre-industrial times. The article ends on a hopeful tone, citing nearly 200 governments' agreement to reduce methane emissions by 2030 and the Biden administration’s implementation of rules requiring oil and gas producers to increase monitoring for methane leaks.
The journal article estimated methane emissions across six major oil and gas-producing regions in the US, which together account for 52% of onshore oil and 29% of natural gas production in the country, as seen in Figure 1 below. The researchers merged the data from approximately one million aerial site measurements collected across 15 aerial campaigns with an empirically grounded statistical model for smaller emissions to estimate the methane emissions in each region. All aerial campaigns used hyperspectral infrared spectroscopy to detect and quantify methane. The statistical model was based on an EPA dataset that contained component-level measurements of methane emission frequency and magnitude, combined with counts of the number of each relevant component (like valves, connectors, open-ended lines) at each site. The researchers combined these two methane emission datasets, using the aerially measured emissions for well sites with emissions large enough to detect and simulated emissions for all other surveyed sites, to yield accurate methane emission estimates.
Figure 1a: Surveyed well sites. Remaining active US well sites in black.
The study found that there was a high degree of variation across the six regions in terms of the fraction of methane lost to the atmosphere, ranging from 0.75% in Pennsylvania to 9.63% in the oil-focused region of New Mexico. The overall weighted average methane loss was 2.95%, about three times higher than official US government estimates. A key finding was that a small fraction (0.05–1.66%) of well sites contribute to the majority (50–79%) of emissions, with midstream facilities like pipelines also playing a significant role. Economically, the emissions represent a loss of $1 billion in commercial natural gas value and $9.3 billion in social costs. The study highlights the significant underestimation of methane emissions from oil and gas systems and introduces a more accurate method for quantifying these emissions.
While the New York Times article was successful in conveying the results of the journal article, it glossed over the paper’s methods. The article spent only two sentences total explaining what was a methods-heavy research paper. Granted, the methods were rather dense and confusing, but slightly more explanation of where the methane emissions statistics came from would have strengthened the article. Additionally, the article omitted a few critical statistics, including that 0.05–1.66% of well sites produce the majority of methane emissions and that 52% of onshore oil and 29% of natural gas production in the country were surveyed in the study. However, it is understandable that not all of the important statistics can be mentioned in an 850 word article.
Overall, this was a very strong article that successfully explained the issue at hand, accurately referenced statistics from peer reviewed research, and expanded upon the research to report on methane emissions more broadly. All of this was accomplished while remaining a relatively short, digestible piece of writing. I appreciated that the research paper was conveniently linked in the second paragraph of the article for interested readers. Quotes from the lead author of the paper were also incorporated well into the writing, highlighting a silver lining that the methane emissions were concentrated in a small fraction of sites, which makes the methane problem simpler to solve. This was a very well written article, I rate it a 9/10.
I would definitely agree with the 9/10 score you give. The overall structure of the new article is very nicely laid out so that the reader can take in the information bit by bit. For example, it introduces methane and its effect on global warming first, before taking a more detailed look at the study. It also incorporates data presentation well, not overflowing the readers with numbers, but at the same time presenting all the important values. They also credit the institutions who conducted the studies, putting the big names out there for reliability. I'm also a big fan of the "One take way" paragraph they have in there, which allowed me to quickly understand the "conclusion" within minutes. Overall, I think it's a well-constructed article targeting audiences at different stages of education, making it understandable and reliable. It makes sense for them to discard the method section, which is the last thing many people will care about (those who are not in the field). Although I would agree that they should at least add a few more sentences to the methods.
ReplyDeleteI totally agree that the structure of this article was very well thought out and easy to follow. I was also impressed how Tabuchi was able to sprinkle in statistics throughout the article without bogging down the reader, especially since there were a lot of important findings from the study that needed to be mentioned. You mentioned how the article credited the institutions responsible for the research to gain credibility, which is a great point. As Lauren mentioned in her comment below, the article further heightened this credibility by also referencing a report recently released by the IEA, which I thought was a nice touch.
DeleteI agree with your 9/10 scoring. Something that I particularly found refreshing in the paper's and article's analysis was the economic focus. Many (if not most) environmentally-focused papers/articles exclusively frame their analysis around climate change and ecosystem losses. I think that framing can be powerful, but I imagine it mostly resonates with those who care about our environment (as the saying goes, preaching to the choir). Money is a great incentive to start some meaningful change. A loss of $1 billion could incentivize oil/gas companies to fix the minority of well sites and stop methane release. Even if this would lead to the burning of more gas, the methane would be oxidized to be a less potent greenhouse gas. In today's polarized political system, I believe people on both sides of the aisle can agree with the science and the need for change.
ReplyDeleteYou make a great point that regardless of political affiliation or environmental beliefs, everybody agrees that losing $1 billion from methane leaks is inherently an issue. Another way the author drove home just how much methane is leaking into the atmosphere is equate the methane emissions from oil/gas wells to the annual greenhouse gas emissions from the energy used by 20 million(!) homes. I thought the combination of these two relatable statistics was powerful.
DeleteI agree with your 9/10 rating; it does a great job at succinctly summarizing the takeways of the study. By noting that concentrated emissions occur at a comparably smaller fraction of the sites studied, it leads to an implicit call to action that is understandable from the general public. One aspect that the article adapts from the publication that I particularly like is the inclusion of the monetary value of methane gas released since, from a policy point of view, is a concrete number that policymakers can point to and take action on. I think such an impact could be further emphasized with a mention of the monetary value of subsequent environmental damage and other associated costs but the point still stands.
ReplyDeleteDiscussing the monetary value of the environmental damage from methane leaks is a fantastic idea! In the research paper, they actually did report a finding related to your comment - they estimated the "social cost" of the methane leaks to be $9.3 billion annually. The term social cost was not explained in the paper, but they referenced the Inflation Reduction Act of 2022. I looked a little further into this and found that this bill is in part related to the redirection of funds to reduce greenhouse gas emissions. So I think the "social cost" reported in the paper is exactly what you are talking about with the monetary value of other costs associated with methane emissions.
DeleteI think the NYT article was very successful at conveying the information from the Nature paper to a general audience. I particularly appreciated the citation of a separate report from the IEA to establish multi-source credibility. I appreciated the article's mention of the number and location of measurements used, and that they just said "computer modeling" instead of getting bogged down by unnecessary details. I agree with you that the article should have mentioned the fact that 0.05-1.66% of well sites contribute to the majority of emissions, as it would strengthen their point that emissions are very concentrated at a small number of sites. I think the numbers they included, of emissions rates varying from 0.75% in Pennsylvania to 9% in parts of New Mexico, don't do as well at conveying this point: 9% doesn't seem very large at all, so readers may not care. Additionally, as you said, getting quotes from the paper's authors speaks to the accuracy of the article. I agree with your grade of 9/10 for this article.
ReplyDeleteYou make a good point about multi-source credibility - it's a lot easier to trust an article when two credible sources are reporting similar findings! I'm glad you also thought the 0.05-1.66% of well sites contributing over 50% of total methane emissions was worthy of being mentioned in the article. While the quote from the lead author did make clear that a "small fraction" of sites were contributing the majority of emissions, it leaves the reader wondering how small that fraction is. "Small fraction" is quite an ambiguous term, and I think it would have been pretty easy to include the actual percentage range in parenthesis somewhere.
DeleteBaker, I agree with your overall conclusion regarding the news article’s ability to communicate the main idea of the Nature’s article to a widespread audience [Oil and gas systems are releasing amounts of methane emission larger than predicted (328)]. However, I would rate the news article sightly lower than a 9/10. I think the large absence of details regarding the methodology keeps the audience “in the dark” and that the inclusion of some of the limitations to the methods could’ve made Tabuchi’s argument stronger.
ReplyDeleteThe Nature article does a great job highlighting the shortcomings and limitations to their study. For instance, they mention in lower detection ranges for aerial systems, they can only detect smaller emission particles for a “part of the time” (329). I also liked that they admitted to partial bias in the Monte Carlo interaction by randomly selecting one emission value to measure at all sites. Lastly, some small particles were not included in their combined distribution due to the limitations across different methane quantifying instruments regarding detection limit due to the sensitivity of the sensor, flight altitude, local environmental conditions (wind), and more. I think these details could’ve been included in the news article in a sentence or two emphasizing (1) the increased need to continue research in their area/develop better technology and (2) the amounts of methane emissions might even be larger than predicted in this study (“This suggests that our simulated estimates for Pennsylvania may be conservative” (333).
Overall, the news article includes the important and most eye-catching statistics discovered in the Nature article. I understand the targeted audience are American/International voters across the world; however, I believe the inclusion of the limitations of the study could’ve created a more urgent voice and call to action. With the recent election coming up, it’s imperative to write with a commanding voice regarding climate change and our world.
What do you think? Do you think these methodology details could’ve been confusing to readers? Do you think it was confusing to the journalist?
Thank you for your in-depth comments and questions, Mia! You bring up a lot of important aspects of the journal article. I completely agree that a main takeaway from the research is the increased need for developing better technology for monitoring methane emissions. This isn't a focus of the NYT article, but it is discussed briefly: "Scientists are increasingly turning their attention to getting better measurements of human-caused methane emissions"... then goes on to talk about MethaneSAT, a new satellite launched by the Environmental Defense Fund. I don't have an issue with the author not going more in depth on this topic, simply because there was a lot of content to cover in only 850 words.
DeleteAs to your questions related to the methodology - yes, I do think including too many details would have confused readers. I think the majority of readers care more about the results than the spectroscopy used to measure methane, the altitude of the flights, and the details behind the simulations. Plus, most people are willing to trust the methods of peer-reviewed research. I could also imagine readers getting bored/confused by the methodological details and deciding not to finish the article.
I do think there is a chance the journalist was confused by the paper's methods, because after reading the paper multiple times, I'm still trying to wrap my head around some of the details. I am guessing the decision to mostly gloss over the methods was influenced by the scope and length of the news article, audience interest, and partly the author's comfort level with reporting on the methods.
While I agree that the NYT article does a decent job of not conveying too much information in lieu of general comprehensibility, I felt that it left out some details that could have strengthened the articles impact. For example, the "9% of natural gas escaping above New Mexico" metric is important, but 9% relative to what? What is a "normal" level of emission? Similar to what Mia and Lauren said, what sort of "computer modeling" was done? I think even one sentence mentioning that a Monte Carlo simulation was performed would be sufficient for readers who become skeptical upon hearing such a buzzword. With enough hand wavy statements left to the reader to verify, a reader less well versed in the sciences may feel as manipulated as by a true misinformation source, leading to a further distrust of scientific authority.
ReplyDeleteHowever I do feel that this was a good brief overview of an otherwise slightly overwhelming Nature article to anyone unfamiliar with environmental jargon. The comparison to economic loss was especially captivating as someone who might not be used to thinking in environmental figures (ex: I know $1 billion is a lot of money, but is 1000 kg a lot of methane?). Overall, I think this was a decent article with a few, easily fixable flaws, and so I would slightly disagree with your rating and give this article a 7/10.
I agree with your assessment of the article for the most part. It is extremely well-written and includes exactly what it needs to from the research paper to get the point across without getting lost in the details that most of the population would gloss over. I do disagree with your statement about including more of the methods because in my personal opinion while methods are important for more scientifically minded folks that would want to reproduce the work to confirm it, it is not needed for the average person reading this article.
ReplyDeleteI do think that the NYT article should have included the map showing where the well-sites and their emissions were. I feel it is an extremely impactful image that it portrays, and people tend to do better with visuals than reading. I also know this is a bit off the main topic of the paper, but I like that they included a correction at the bottom of the article, it shows that they find being factual important and are even willing to admit that they may have been wrong with a correction (even if it was just a minor one.) It raises the reputation of the news source and makes it more trustworthy.
I'm torn as to whether I agree or disagree with the comments that the methodology of the study should have been discussed in greater detail in the New York Times article. A thorough explanation of the Monte Carlo method would be out of place for the intended audience of the New York Times, but a simple generalized statement like "Researchers took repeated random samples of measurements" would have sufficed. Tabuchi wrote that the data was processed with computer modeling. Is "computer modeling" too vague? I think that it is an accurate statement. I am not sure that diving deeply into the modeling parameters or the simulations would be worthwhile here. However, I think it is important to discuss that there are limitations to every study, especially a study that relies on data collected from a third-party source. I think when the methodology isn't described in detail, there should be information provided about the shortcomings of the study along with future research. In the case of the research by Sherwin et al, what really struck me was that the data used only paints a partial picture of the methane emissions in the US:
ReplyDelete"Given the regional variability in emission rates we show in this work, an accurate estimate of US emissions will require surveying a large fraction of the 48% of onshore oil and 71% of natural gas production not covered here."
Overall, I think there is a healthy balance between explaining the methodology and communicating its limitations. Additionally, I think there is a great desire today to have conclusive, irrefutable data on chemical processes related to climate change. There is a lot of suspicion surrounding climate change, and any scientific data to prove its effects can be helpful to educate. But no data has 0 error associated with it, and I think it is important to be transparent about all aspects of a study, including shortcomings. Overall, Tabuchi leaves it up to the reader to decide whether or not they are satisfied with the nebulous explanation of the Nature paper's methodology. The Nature publication is linked in the New York Time article, so if someone was curious they would be able to access the details.
As an aside, Tabuchi makes a connection between the study and the MethaneSAT, a satellite that is being implemented to track methane emissions from outer space. The article displayed a graphic in the middle of the article which I really enjoyed. The graphic displayed crucial components of the satellite to make methane emissions measurements. Another article about MethaneSAT followed. Even though the picture of the MethaneSAT wasn't what was used in the study, I think showing scientific equipment used in research can be very informative. Having visual aids are also helpful to readers. As scientists, we are keen on including graphs in our work, but those can be confusing to readers. I would have appreciated if Tabuchi included a picture like in Figure 1C of the Nature paper.
This is a well-rounded and thoughtful analysis of both the New York Times article and the original journal paper from Nature. You clearly outline the strengths of the article, particularly in how it simplifies complex scientific data on methane emissions into a digestible format for the general public. I agree that the omission of some key statistics and the brief treatment of the methods could be a drawback, especially since the research paper is methods-heavy and those details contribute to a better understanding of the study’s findings. However, as you pointed out, the constraints of a news article likely made it necessary to streamline certain aspects, especially given the 850-word limit.
ReplyDeleteI also like your insight about the silver lining the article emphasized—that methane emissions are concentrated in a small fraction of well sites—giving a hopeful tone to the solution. This kind of nuance is important in climate reporting, where it’s easy to feel overwhelmed by the magnitude of the problem.
Your rating of 9/10 seems well justified, given how effectively the article talks about complex information while still managing to cover broader implications and policy solutions. Overall, I found your analysis comprehensive and fair in its critique of both the article and the underlying study.
Tabuchi does a great job of making this paper digestible for the public and showcasing methane emission's importance by focusing on the economic impacts of this study. While science as a whole may be difficult for the general public to understand how it relates to their daily life, the economy and economic impacts are something that everyone can loosely understand. Money talks and in this case could motivate people, and companies, to change their minds about how important this issue is and remediations. One part of the article overall that I did not like was the way that the data is displayed. The average American can't really fathom these values in terms of "millions of tons of methane" or percentage values of total emissions. Even to someone in science these values aren't super helpful. There is mention of this methane release being the energy equivalent of 20 million homes but technically does not specify that this is per year. Regardless, hearing that this is as much as 20 million homes (~50 million people at the 2.51 people per household average) puts this value into perspective. This is the type of reporting needed for the non-scientific community because a raw value (Like 6.2 million tons of methane) does not tell the ready anything- although they report percentages as well, the reader may think, "Do we really care about 6.2 million tons of methane (3% of total gas produced)?" I would generally agree that this article does a great job (And I think 8+/10 is reasonable), but keeping the data/information digestible is the most important part in my opinion.
ReplyDeleteI feel that you are overemphasizing the necessity for news articles to go into the nitty-gritty of the papers they are referencing. The paper would have been fine without ever referencing the methodology on how to data was collected as that is something that the reader of the news article does not need to know to become informed about the topic. The fact that they did reference the aerial surveys and computer modeling, along with talking about other methane emission monitoring methods like the MethaneSAT, is a nice bonus. Also, while they do not reference many exact data points, such as the 0.05-1.66% of well sites contribute 50-79% of the methane emissions, the paper explicitly mentioned that “just how concentrated emissions are in a very small fraction of sites,” while quoting the articles author. Scientific data is often imprecise, so making more broad statements are often better than specific statements that might overburden or confuse the reader. I do however agree with you that the news article should have included the 52% of onshore oil and 29% onshore natural gas figure, as this emphasizes that the results of the paper were not an anomaly based on a small data size or cherrypicked numbers.
ReplyDeleteI think the article was written very well and includes all the important concepts from the research paper, this is great for getting straight to the point for the viewers so that they don't get confused. Though I agree with you on the aspect that this article was written very well, I disagree when it comes to your opinion on the articles lack of method representation. For researchers and others genuinely in the pursuit of in depth knowledge , it can be a great idea, however, for me and the average audience which I believe is the biggest viewers the amount of methods produced in the article I think is perfect for keeping the audience engaged.
ReplyDeleteI do think that the NYT article is quite strong at showing the data from the Nature paper to the average audience. I agree with your assessment of the article should have showcased the 0.05-1.66% of well sites contribute to the majority of emissions, I think this would ultimately strengthen their opinion that emissions are highly concentrated at a smaller number of sites. I do think there parts of the article that maybe ramble on a little too much, which is unnecessary, which after looking at a number of our classmates responses, they also seem to agree that a one sentence response on a topic would keep the audience who may not know much on the topic engaged. I also agree that the lack of the map showing the well-sites and their emissions where, was a minus for the article since a lot of the general audience, in my opinion, do well with visual learning, so for them to add this would have really made this article shine!
I agree with both your evaluation of the article, as well as the other commenters here in that this article not only does a pretty excellent job of extracting much of the most vital information from the Nature paper, but also is able to advance beyond it in its analytical scope. The original article, though necessarily wide-reaching, acknowledges its own points of limitation within the bounds of an academic journal. The simplification of the methods section, specifically the 'game' of data acquisition the researchers had to undertake with the aerial spectroscopic and simulated well-site diagrams, feels understandable given that the NYT article is not intended as a critical review of any sorts.
ReplyDeleteOne of the most interesting aspects of the article was the highlighting of the regional differences of gas emissions and how they come about. I think that's a great example of the different bounds in which both sources worked and the different motives behind each one. The NYT article by no means pushed forth any completely radical claims, but the point about what certain drilling plants 'looked for' and what they considered waste brought up an interesting thought about the wastefulness immanent in the very production process of these companies. Whether or not that complicates the 'silver lining' brought up in the quote from the original paper's author, or if such a complication was even the intention of Tabuchi, I'm not sure.
I agree with your rating of the article as a 9/10. The paper clearly communicates the implications of methane emissions, and it also concisely illustrates the fact that methane emissions are higher than previously reported. I do think the paper could have included a little bit more information in the methods section, such as describing a bit more about how the data was collected and analyzed rather than simply stating that the authors used "computer modeling" on data collected from "aerial surveys"; however, I do not believe that the paper needed to include significantly more details about the methods, as the paper utilized complicated methods that would likely overwhelm or confuse the target audience/general public. Moreover, while I agree that some more statistics could have been included to convey how a small fraction of sites are emitting a disproportionate amount of methane, I do believe that the article did a sufficient job explaining this to the reader, using quotes by one of the authors of the study in order to do so: "'That’s the silver lining. If we can figure out what’s happening at these small fraction of sites, we’re halfway toward solving the methane problem in oil and gas...'"
ReplyDeleteIn addition, I appreciated that the article addressed the issue of methane emissions from both an environmental and an economic lens. By both highlighting that methane "can warm the planet more than 80 times as much as the same amount of carbon dioxide over a 20-year period" and that methane emissions result in the wasting of "about a billion dollars’ worth of gas," the article alerts the reader that lots of economic resources are being wasted, and in the process a more potent greenhouse gas than CO2, the product of combustion, is being released into the atmosphere. This economic perspective may also help appeal to those who are not as concerned about environmental issues, but are concerned about economic consequences such as an increase in oil/gas prices due to loss of materials involved in their production. Overall, the article does a great job summarizing a complex paper and making it digestible for the general public; moreover, it excels in explaining the implications of the paper's findings to increase public awareness and encourage climate action.
I also agree with what you and many others have said about the article being very well written and getting the main ideas of the study across. I really liked how this article was written so that a wider audience can understand it, and the explanations of important takeaway and data from the article get the main point across. I slightly disagree with the idea of adding a bit more explanations about the methodologies, as the average person is not really looking to know exactly how the data was produced, but rather the results and impacts (especially if it may impact their life). Going into more depth about the methodologies behind gathering the data reported I feel like would actually distract from the main point of this article. Instead of adding more clarification about the methodologies or more pieces of data, I feel like if the article described what the data points it already presented in a way that the average person could understand why the 9% for methane emissions is noteworthy (As to the average person, 9% does not seem like a percentage that is newsworthy)., I feel could have strengthened the article a lot more.
ReplyDeleteI can see why you and others rate this article so high but I think it could be rated only a small point lower. I really appreciated the article and the ease of accessibility in term of digesting information. One problem I had that has already been mentioned is the lack of telling the weakness of the paper is you will call it that. While collecting data at this scale is difficult the paper itself acknowledges the limits of data collection that I felt are important to incorporate into the news article. At the same time I can see why if you were writing an article like this you might want to leave some of that information out to make your point come across more efficiently. Either way it was a pretty impressive article I rate 8.4/10
ReplyDeleteI am a bit conflicted with your opinion regarding the need to emphasize the methods. Sure it's true that there is no need for articles to delve that deep into the scientific paper to the regular American audience (as mentioned by a few), but you are also right in arguing the importance for papers to give a scope on how big the actual affects of oil fields are on the environment, along with how many places did the scientific paper took measurements at. Though some critics may argue that readers of the news article do not need to be well informed on the topic, others could claim that it adds dimensions to the accuracy of the data compared to the rest of the nation as a whole (notably the 52% of onshore oil and 29% onshore natural gas statement). However, since they already included references to MethaneSAT, the "empirically grounded statistical model," and collaborations with the International Energy Agency, I would already assume that this is enough info regarding the methods section.
ReplyDelete