Wikipedia talk:WikiProject Weather

I would like to announce that a new task force has been created to re-examine the status of every GA in the project. Many good articles have not been reviewed in quite a while (15+ years for some) and notability requirements have changed quite a bit over the years. The goal of this task force is to save as many articles as possible. Anyone not reviewing an article may jump in to help get it up to par if it does not meet the GA requirements. The process will start officially on February 1 and will continue until every article has been checked and either kept or delisted. The task force may be found at Wikipedia:WikiProject Weather/2024–25 Good Article Reassessment. Noah, AA^Talk 15:22, 26 January 2024 (UTC)Reply

Articles under review

• 1949 Texas hurricane (review)
• Windsor Locks, Connecticut, tornado (GAR)
• Wind shear (GAR)

Should weak and unimpactful tornadoes be included in list articles? Departure– (talk) 14:20, 11 December 2024 (UTC)Reply

Opening comments: This all began because of an above discussion, where an editor placed a tag on List of tornadoes in the tornado outbreak of March 31 – April 1, 2023 article for "excessive examples", and upon discussion stated that weak tornadoes with little effects were getting too much prose in the lists given their impact and shouldn't be listed in the same manner as other tornadoes. This goes against the status quo of the "List of tornadoes in the XY outbreak" and "List of United States tornadoes from X to Y, YYYY" list articles which have remained largely untouched in policy and unquestioned on notability since their origins. I personally believe that, since other tornadoes in the list are practical, all tornadoes that can be reliably sourced to be included should be listed with a brief summary. Another potential solution which I personally oppose but could be implemented is prose in the articles for EFU/0/1 tornadoes, stating that "X weak tornadoes producing little impact were also observed". I'll also state that this statement will make tallying tornadoes harder, given the lack of specificity that can lead to under or overcounting. Whatever the outcome of this RFC, I merely hope the solution will prevent this issue from producing policy-based stalemate with maintenance tags having no clear and easy solutions as we have at the first article I mentioned. Departure– (talk) 14:27, 11 December 2024 (UTC)Reply

• WP:CSC: "Lists are commonly written to satisfy one of the following sets of objective criteria:" When we apply this to e.g. List of United States tornadoes from July to August 2023, we see that it certainly doesn't meet #1 (not all these tornadoes are independently notable), it doesn't meet #2 (some of them are notable (e.g. Tornado outbreak sequence of August 4–8, 2023), and it doesn't meet #3 ("reasonably short (less than 32 KB)": the example article is more than 200K, and is already a random subdivision of US tornadoes of 2023). Fram (talk) 14:58, 11 December 2024 (UTC)Reply
• Cherry picking certain tornadoes to include here would be overly subjective and impossible, so it’s really an all or nothing scenario. I support including all tornadoes as is done currently, with no changes needed to the current core status of the lists. The only way these lists can be totally objective as Fram mentioned above is to include all of them. United States Man (talk) 19:22, 11 December 2024 (UTC)Reply

• That´s the same kind of argument used for years for sports, everyone who played one game is the only objective measure. Didn´t fly there, doesn´t fly here. Including e.g only tornadoes of, say, EF2 and above is equally objective. Or all rornadoes which have at least one non-local reliable source apart from the curre t database. Or probably other rules, these are just some first thoughts. Fram (talk) 19:40, 11 December 2024 (UTC)Reply

  Actually no, there are numerous cases where EF1 and even EF0 tornadoes include more damage description and even media coverage than some EF2s, so then again, it is subjective. You can frame this anyway you want, but your argument here is actually not an improvement and is detrimental to the Wikiproject and the flow of information of Wikipedia as a whole. There are actual issues afoot here in this wikiproject, such as mass creation of tornado articles with bad grammar, multiple factual errors, and content-fork creation. The list pages are not a hill to die on for you imo. United States Man (talk) 19:53, 11 December 2024 (UTC)Reply

• I've been following this discussion without really chiming in, so I will offer a possible solution. Noting, if I was actually choosing, I choose to not alter anything. However, this is a possible compromise to the dispute:

1. Monthly U.S. tornadoes articles remain stand-alone list articles (merges to combine additional months open to case-by-case basis).
2. Any tornado that has one non-NOAA source is automatically notable for summary details (i.e. summary details as the lists have now).
3. The leader is altered slightly from the current lead versions to denote this includes notable tornadoes (i.e. at least one non-NOAA source)
4. In the lead, any weaker tornadoes are noted without full summaries. For a hypothetical example: "In the month of July, 20 tornadoes occurred across the U.S., with 3 rated EF2, 10 rated EF1, and 22 rated EF0."
5. The hypothetical example above would be cited by the NOAA database set just to the monthly tornadoes, which is a reliable source.

As mentioned, I don't necessarily support this at this moment, but I wanted to throw a possible solution into the water. If consensus/compromise would be falling towards allowing this type of solution, I would be for it. The Weather Event Writer (Talk Page) 19:54, 11 December 2024 (UTC)Reply

That´s more or less what I intended with my second suggestion, and seems like a good basis for discussion. Fram (talk) 20:26, 11 December 2024 (UTC)Reply

Almost every single tornado will have a non-NOAA source if you look for them (i.e. local news). So we’d end up excluding maybe 2-5%, so why not just include all of them and be done? United States Man (talk) 21:09, 11 December 2024 (UTC)Reply

If almost every single tornado has local news coverage... what's even the point of being selective? Wouldn't they all be considered notable? Departure– (talk) 21:14, 11 December 2024 (UTC)Reply

That’s my point. United States Man (talk) 21:18, 11 December 2024 (UTC)Reply

Playing devils advocate for a moment: Then the local news coverage source should also be listed with the NOAA primary source. While those of us (y’all and myself included) generally understand that fact, I’ll be honest, in the example article listed by Fram above, List of United States tornadoes from July to August 2023, there is 0 non-NOAA sources outside of the lead. Out of the entire list article, which has 262 sources, 260 comes from NOAA and 2 come from non-NOAA sources. Part of the overall issue is that WikiProject Weather got in the habit of citing NOAA and then not anyone else since the info was already cited. The topic of “Is NOAA a primary source” has come up multiple times and the answer is yes it is (WP:VNTIA). So technically, if we look at Wikipedia policy to the letter, that article is basically cited entirely by WP:PRIMARY sources, which is actually cautioned against, not secondary reliable sources, which is preferred over primary sources. Basically, a possible solution to not even change the list is to just add a secondary reliable sources to the tornadoes. Then, see where it goes from there. Anyone else think that may be a good idea? Actually see how many tornadoes do/do not have secondary sources?

If one or two do not, then the list, bluntly, is fine (once non-NOAA is actually added). If 20+ do not in a monthly list, then we may have a true problem. As I see it, the problem is that primary is being used and secondary is basically being ignored, leading to Fram’s conclusion that most of the tornado may not be notable enough for the list. The Weather Event Writer (Talk Page) 21:54, 11 December 2024 (UTC)Reply

Yes, it should be mentioned in the list as its omission would be misleading (showing less tornadoes than there actually was), less accurate, and less comprehensive. I'm fine with a brief summary, mention, or omission of some of the events outside of the list only if certain details of the tornado would be inappropriate or rule-breaking. _ZZ^Z'S 05:00, 14 December 2024 (UTC)Reply

• (Summoned by bot) - Yes, if I understand the question. All tornadoes are sufficiently eventful and concerning that, if there is a list, they should be included in a list. Tornado warnings are disruptive. People who have been disturbed by tornado warnings and have headed for cover when there was no damage would still like to see that event in a list. Robert McClenon (talk) 00:06, 12 December 2024 (UTC)Reply
  • Tornado warnings are general, not for a specific tornado. "All tornadoes are sufficiently eventful"? Many tornadoes are likely to remain undetected as they are very minor and shortlived and if no camera or storm chaser is nearby and they happen on unpopulated land they will likely not be noted. Even among the ones listed. Look at e.g. List of United States tornadoes from July to August 2023#July 7 event; none of these 4 tornadoes were, as far as we know, eventful; we have no idea if tornado warnings were given, and if so where and when. As an aside, I have no idea why this is called the "July 7 event", these were not one event but can perhaps be considered two events (the ND ones and the Texas one have nothing to do with each other). Just labeling it with the date (so here "July 7") would be at least better. Fram (talk) 12:34, 12 December 2024 (UTC)Reply

There are times when it's useful to list (or attempt to list) every tornado (like List of Australia tornadoes). I guess the bigger questions comes down to the effort to document every tornado in the United States each year, and how best to do that. The way we do it now, we have the yearly Tornadoes of 2024, plus monthly lists in the US such as List of United States tornadoes in May 2024, as well as individual outbreak articles, such as Tornado outbreak sequence of May 19–27, 2024, and sometimes those outbreak articles have individual lists, such as the List of tornadoes in the tornado outbreak sequence of May 19–27, 2024. While that might seem like a lot of overlap, any single tornado has the potential to be notable. Take the EF2 tornadoes for example: none of the EF2 in the List of tornadoes in the tornado outbreak sequence of May 19–27, 2024 get a mention in the main outbreak article. But given the current length of the outbreak article (7,500 articles), it would be too much to include every single EF2. Now most of them weren't that significant, but an EF2 can still destroy a building, so they still deserve mention. Even EF0 and EF1's have the potential to cause significant impacts - the most recent tornadic death in New Jersey was caused by an F0. In the interest in being inclusive, I don't think it makes sense to be unnecessarily restrictive. At the same time, requiring non-NOAA sources could be tricky, since a lot of news sources just regurgitate NWS reports. I realized that while working on List of California tornadoes. I think the way that the severe weather project has been handling tornadoes is honestly pretty impressive. I should also note the importance of digging into each tornado directly, rather than just relying on random NCDC links, as there can be multiple reports for the same tornado if it crossed state/county lines, or if the tornado touched down multiple times. In short, I don't think much needs to change, other than maybe summarizing more here and there, and trying to include non-NOAA sources (when the info doesn't just repeat what's in the NOAA sources). ♫ Hurricanehink (talk) 18:11, 12 December 2024 (UTC)Reply

Here’s my opinion: if it’s a list of tornadoes in a specific outbreak: I believe that ALL tornadoes that occurred in that outbreak, even if their impacts were trivial; need to be included. Hurricane Clyde 🌀^{my talk page!} 20:01, 13 December 2024 (UTC)Reply

If we’re referring to the Tornadoes of YYYY articles and the like, then I think WP:TornadoCriteria should be followed. And only list the more notable ones. But it really depends on the case. But if it’s concerning individual outbreaks; then every tornado needs to be listed. Hurricane Clyde 🌀^{my talk page!} 20:03, 13 December 2024 (UTC)Reply

If we're only listing tornadoes notable enough to be mentioned in the yearly article anyway, what's the point of even having the list? In any case, if we make such a move, there will still need to be a list of all tornadoes in project space so we can keep an accurate tally. TornadoLGS (talk) 05:38, 14 December 2024 (UTC)Reply

Tornadoes of YYYY isn't the same as the "List of tornadoes from M to M YYYY" lists that get created every year. Departure– (talk) 12:54, 14 December 2024 (UTC)Reply

What may be more prudent here would be to split those M to M lists into monthly lists rather than omit tornadoes. Hurricane Clyde 🌀^{my talk page!} 18:48, 14 December 2024 (UTC)Reply

So, in all honesty yes even weak and unimpactful tornadoes should be included in these lists. If the lists grow too long; we should instead split the lists. Hurricane Clyde 🌀^{my talk page!} 18:50, 14 December 2024 (UTC)Reply

List of United States tornadoes in May 2024 is a list focusing on one month and already reaches 118kb in size. This is why I in particular beg for a {{cite pns}} or {{cite storm events database}} template. I'd wager there's a non-zero chance that the Storm Events Database is the single most used citation across all of Wikipedia, and these lists are a big part of that. Cutting down the size in bytes can also be done by cutting summaries of tornadoes from outbreaks and simply including a main article tag with the small table, rather than the excerpt format used today. Departure– (talk) 19:31, 14 December 2024 (UTC)Reply

I have no more hope of changing anything at this project, which is probably as bad as the former roads project when it comes to closing ranks and not seeing how completely inappropriate their efforts are to duplicate a database in all its excessive detail. List of United States tornadoes in May 2024 has more than 500 tornadoes, most of them very ephemeral, and the suggestion to deal with this is ... creating a new cite template to reduce the size of the sources. Try to imagine some other weather phenomenon, say a hurricane or a winter storm or whetever, and having a place by place list describing in place X "damaged some vegetation and fencing", in place Y "damaged a small shed and utility trailer", in place Z "caused no known damage", elsewhere "no damage could be found.", and in many places strong winds were observed but nothing more. And buried among this endless list were the serious, noteworthy instances with deaths or truly massive damage. But hey, congratulations all around, we have repeated every instance from the weather database, good job people! Fram (talk) 14:45, 10 January 2025 (UTC)Reply

Per the above comment by User:Departure– above, we should probably have template for referencing storm events, since, as the user said, "there's a non-zero chance that the Storm Events Database is the single most used citation across all of Wikipedia." Every single URL has the same beginning, so such a template might also need something like Template:NHC TCR url, which shortens the URL for TCR's released by the NHC.

There is a little bit of inconsistency over the publisher and author, but since we don't know the people who actually write the event reports (other than the local NWS office), I think the default publisher should be "National Climatic Data Center". Does anybody with template knowledge think they could work on this? I can try tackling it after the new year if no one does it. ♫ Hurricanehink (talk) 20:28, 14 December 2024 (UTC)Reply

On the database I've been experimenting with a citation that displays as "Storm Events Database (LWX survey BALTIMORE MARTIN ST, 2024-06-05 20:27 EST-5). I think the WFO and ID are all that are needed, but I'm definitely in support of the begin location or timestamp being alongside the WFO. Departure– (talk) 20:38, 14 December 2024 (UTC)Reply

I have previously wondered about setting up a template similar to the NHC TCR URL one before now, however, I'm loathed to as I have previously been informed that the URL ID changes from time to time.Jason Rees (talk) 22:09, 14 December 2024 (UTC)Reply

As long as the time between when the URL is put into the article, and there is an archives of that URL, then it shouldn't matter too much for when it changes. Linkrot is a problem that's avoidable. Data compression is also helpful for articles loading faster, so a template would be useful. As for what User:Departure– made, I think it should have the "National Climatic Data Center" as the publisher, but "Storm Events Database" should be the series, if that's possible. The details about the exact time and location is good, but that is ultimately extra coding being added to one of the most common citations. Perhaps a title of just "[Weather type] event report"? The weather type would be whatever is the first entry. That way the NCDC URL could be used all across the weather project. For example - "High wind event report" or "Hurricane event report" or "Tornado event report". If we wanted to be more specific, maybe add location, so you could have "California high wind event report", or even "Monroe County, Florida tornado event report". There are options, but seeing how often the NCDC reports are used, there should be some discussion on it. ♫ Hurricanehink (talk) 01:05, 15 December 2024 (UTC)Reply

While you're at it, might as well knock down "Cite storm data publication" as well. Access links are temporary and only for 24 hours on the NCDC site. Wildfireupdateman :) (talk) 05:22, 28 January 2025 (UTC)Reply

Should we have notability standards for individual tornado articles? We already have informal inclusion criteria for "Tornadoes of YYYY" articles. Below is a preliminary proposal for such criteria, with the hope that it can evolve into a formal guideline that can possibly be referenced in future AfD discussions.

  Previous discussions: New tornado articles and the news, Proposal - Criteria for inclusion on Tornadoes of XXXX articles

This has been nagging at me for a while now, and since another editor has talked to me about this issue, I think we bring this up. Since we have a sort of "inclusion criteria" for "Tornadoes of YYYY" articles, I suggest we come up with notability criteria for individual tornadoes as well. See User:EF5/My tornado criteria for what this may look like.

This is my very primitive way of determining the notability of several tornado articles I've written, and am hoping that it could be integrated into a refined set-in-stone WPW policy that could be used in actual AfDs. I'd assume that the table will be gotten rid of and turned into a list. This has been discussed in the past, but never really came to anything. Maybe it could be... WP:NTORNADO (with it's own project page)? Starting an RfC, since obviously community input is needed. Also pinging @Departure–:, who suggested this. :) E_F⁵ 18:56, 16 December 2024 (UTC)Reply

I support these guidelines, but please see my suggestions on the talk page - the wording around fail-if-pass criteria make this much more difficult to read than it needs to be. Perhaps putting them in their own section separated from the other criteria would resolve this. Departure– (talk) 19:01, 16 December 2024 (UTC)Reply

Support waddie96 ★ (talk) 11:50, 18 December 2024 (UTC)Reply

I see WP:CONCENSUS (3-0-0; lack of continued participation after over a month), @Waddie96: and @Departure–:; shall something be drafted up? Would be nice to have multiple people work on this. E_F⁵ 17:26, 1 February 2025 (UTC)Reply

Precipitation has been nominated for a good article reassessment. If you are interested in the discussion, please participate by adding your comments to the reassessment page. If concerns are not addressed during the review period, the good article status may be removed from the article. Z1720 (talk) 02:27, 5 January 2025 (UTC)Reply

tl;dr: climate change is already increasing the number of weather catastrophes, and I don't believe WP:Disambiguation provides adequate guidance to name articles when e.g. significant fires share a name.

I was directed here from the teahouse as I'm relatively new. There's currently an going conversation on the talk page for the Palisades Fire about whether the 2021 Palisades fire or the 2025 Palisades fire is the WP:PRIMARYTOPIC.

Without getting into that specific conversation, it occurs to me that as climate change fuels more and more weather catastrophes, it will be helpful to have guidance and/or policy on naming articles. I provided a table below that shows California fire names since 2013 that have been used 10 or more times. I'm, of course, aware that not all named fires meet the WP:N guidelines. But as more fires occur, there will simply be more notable fires that share the same name.

My understanding is that this is largely due to how fires are named: often by dispatchers trying to simplify radio traffic for firefighters. I believe that the NWS/NOAA World Meteorological Organization retires a storm name once a named storm becomes significant.

So, some questions (of course feel free to propose your own):

• Does the Wikipedia:Disambiguation policy adequately address article naming for natural disasters?
• If yes, please elaborate

• If no, which catastrophes need clarification? Fires only? Hurricanes? Snowstorms? Tornadoes? Derechos? Fire Whirls? Other?
• Are you aware of naming schemes for weather catastrophes in countries outside the US that could cause confusion? What are they? It would be helpful to ensure this is not a US-only discussion.
• Does the timing of article creation/title selection affect your decision? e.g. there was a 1981 Hurricane Katrina. At what point did the 2005 Hurricane become the WP:PRIMARYTOPIC Hurricane Katrina?

Thanks for taking the time to read this, and I look forward to hearing your thoughts!

California fire names since 2013 that have been used 10 or more times. From this dataset(scroll down to Incident Data) .

Lastly, as an aside, there was also a 2019 Palisades Fire in CA. Good thing we stopped burning fossil fuels! We really need to stop burning fossil fuels!!! Delectopierre (talk) 05:42, 10 January 2025 (UTC)Reply

Yes. See Lake Fire and Lake Fire (2024). No matter the size of the second fire, the first fire with that name will always take the “main” name and the second one will have the “(YYYY)” after it. E_F⁵ 13:06, 10 January 2025 (UTC)Reply

Is this spelled out somewhere? The current conversation about renaming the Palisades fire is looking like it will not be following that convention. Delectopierre (talk) 18:35, 10 January 2025 (UTC)Reply

I'm not seeing how current article title policy is inadequate. If two or more distinct events are similarly named, we disambiguate by year, then by month if necessary (e.g. Hurricane Alice (June 1954) and Hurricane Alice (December 1954)), and a set index article is set up (following the same example, List of storms named Alice). If one event is much more significant than others of the same name, we drop the disambiguator, in accordance with WP:PRIMARYTOPIC.

Determining the primary topic is usually a subjective process, though this can be backed up by objective statistics like pageviews, death tolls, or damage totals. Katrina is probably not a good case study since it existed in a time where content policies were less strict - Hurricane Katrina (2005) was moved to Hurricane Katrina on 27 August 2005, two days before landfall in New Orleans. These days, people prefer to wait for things to settle before making an assessment, like at Talk:Typhoon Doksuri#Requested move 15 August 2023. Faster moves do happen though, like with Talk:Hurricane Beryl/Archive 1#Requested move 1 July 2024 where the gap in usage and long-term significance became quickly apparent, making it easy to reach an early decision (and this is also what's happening with the ongoing Palisades Fire). ~ KN2731 {talk · contribs} 14:45, 10 January 2025 (UTC)Reply

I think the example you gave is exactly why it could be helpful to provide guidance. My instinct is to leave the year on both and give it a few months before changing the name.

That wasn’t a popular viewpoint, and of course we make decisions by consensus so I don’t pretend that it has to be my way. But it seems to be pretty subjective, at least to me.

Unfortunately, I’d say it’s fairly likely there’s another significant fire (although hopefully not as significant as the current event) in the palisades in the next ten years. What will we do then? Delectopierre (talk) 18:40, 10 January 2025 (UTC)Reply

The guidance across Wikipedia, per WP:COMMONNAME, is that the most significant event of a name should get the primary topic. Significance should have an objective truth to it, like number of fatalities, or acreage. ♫ Hurricanehink (talk) 19:06, 10 January 2025 (UTC)Reply

I strongly support development of a guideline that spells out how to determine that objective truth. like you said fatalities and acreage are important. I think including other metrics would be important as well. For example:

- people evacuated days (eg # of people * number of days evacuated)

- structures destroyed

- housing units destroyed

- economic damage

- firefighters deployed

- firefighting equipment deployed

- economic damage

- changes in how future calamities are handled because of learnings

All that said, I haven’t a clue how to make all of that objective. Delectopierre (talk) 20:01, 10 January 2025 (UTC)Reply

I would support a metric based on some these guidelines (structures destroyed, damage $$), but not others -

• People evacuated days is a good metric but hard to calculate, because evacuation orders are not issued or withdrawn uniformly and good reporting on # of people affected is often hard to find.
• Number of personnel or amount of equipment is also tricky because some major fires will have relatively few resources assigned to them if there are a large number of fires burning concurrently (as with this SoCal event).

Penitentes (talk) 20:21, 10 January 2025 (UTC)Reply

Yes, agreed on both points.

I tried to calculate the people-evacuated-days for a few fires in 2020 during the lockdowns...because I had the time. After 3 or 4 days of maddeningly refreshing facebook posts from rural sheriffs and tracking down nixle alerts...I gave up.

Michael Wara expressed some interest in studying this in 2020, which is what inspired my effort, although I'm unsure if he ever did so. Delectopierre (talk) 20:44, 10 January 2025 (UTC)Reply

Commendable effort, I'm a fan of Wara's work. Figuring out evacuation timelines and stats is always my very least favorite part of writing wildfire articles. Penitentes (talk) 20:47, 10 January 2025 (UTC)Reply

Thanks!

Yeah, he's a very smart guy, and it helps that he's so friendly and available, even to the public.

Evacuation timelines are such a mess -- frankly as are evacuation notifications in the real world. Its too bad that the companies that have 'tried' to solve that...haven't. Emergency alerts have gotten better, and some communities have improved their evacuation policies and procedures. But as a whole, its still a giant mess. Delectopierre (talk) 22:34, 10 January 2025 (UTC)Reply

I recently worked on at least starting the process of getting List of tornadoes in the outbreak of March 31 – April 1, 2023 to featured list status. As part of this, I changed the layout of the page (organizing tornadoes by state), however that change was reverted for not being consistent with the standard layout of tornado list articles.

You can see my take on the style here and the original here. I want to get more opinions on whether the new style really is more readable, and whether or not keeping consistency with other project-space articles is more important than readability. Pinging @TornadoInformation12 as they were the one to revert my edit; bringing this to WPWX because this could easily affect other articles. Cheers! Departure– (talk) 20:52, 11 January 2025 (UTC)Reply

If nobody objects, I'm going to restart working on the article under the new style per WP:SILENTCONSENSUS. Departure– (talk) 15:47, 28 January 2025 (UTC)Reply

We need a unified format for the "Weather of XXXX" articles. For example, Weather of 2008 and Weather of 2009 lists a blurb for each significant weather event (although very incomplete, missing tons of stuff), while Weather of 2024 simply lists Wikilinks, with some info on each type of disaster at the beginning. Thoughts? Wildfireupdateman :) (talk) 18:15, 15 January 2025 (UTC)Reply

Also note that I support the 2008 and 2009 format. Wildfireupdateman :) (talk) 18:15, 15 January 2025 (UTC)Reply

Weather of 2021 & Weather of 2022 is the best format in my opinion. The Weather Event Writer (Talk Page) 18:20, 15 January 2025 (UTC)Reply

Those seem like the intermediate between the 2008 format and the 2024 format. I could work with that! Wildfireupdateman :) (talk) 18:22, 15 January 2025 (UTC)Reply

I've begun a rewrite in userspace. Wildfireupdateman :) (talk) 21:06, 15 January 2025 (UTC)Reply

Yea, when the Weather of 2008 was originally written, it was called "Global storm activity of 2008", which was simplified to "Weather of 2008". The overarching articles should include a summary of all of the different weather types, and mention the deadliest events, I think that's a good way of making the article useful. ♫ Hurricanehink (talk) 21:15, 15 January 2025 (UTC)Reply

Elijah and I like the 2021 format like you mentioned, so I'm rewriting it. It looks like it's gonna be a lot of work, as 2008 isn't the only year that has a different format/issues(e.g. somehow Elie 2007 is not mentioned). Wildfireupdateman :) (talk) 21:19, 15 January 2025 (UTC)Reply

Most "Weather of" articles need a lot of work, so I appreciate you doing that. ♫ Hurricanehink (talk) 21:21, 15 January 2025 (UTC)Reply

In connection with this discussion, I'd like to know whether there is any guideline to distinguish what can be called cold from what is just cool. I assumed that a Spanish town where 4 months per year have mean daily minima below freezing point could be described as having cold winters (rather than very cold or freezingly cold), but there is no consensus. For comparison, I have checked a couple of random articles; for example the one about Paris states: In winter, ... days are cool, and nights are cold but generally above freezing... Furthermore, it seems that it is correct to define winters in the town as cold for Spain, which I would understand if it was a low-latitude country such as Malaysia or Panama, but not for Spain; this makes me think that there is some kind of Euro-centric or Northern-centric bias in how climates in the world are described. Jotamar (talk) 22:32, 15 January 2025 (UTC)Reply

With a historic winter storm about to impact the Gulf Coast, I would highly recommend anyone who is able to contribute and expand the newly created article I started for this at January 2025 Gulf Coast winter storm. There is going to be February 2021/Uri/Viola-level disruptions here in these regions, and seeing how much stuff we got to add there, I would hope to see this page get there as well. I will contribute as well, but I wanted to send this message here to hopefully spread the word and get as much people on board. Cheers! MarioProtIV (^talk/_contribs) 06:20, 21 January 2025 (UTC)Reply

I see this get brought up on quite a few DYK reviews/GANs. Would be good if we establish a community consensus on this topic. Wildfireupdateman :) (talk) 20:03, 24 January 2025 (UTC)Reply

• I say generally reliable up to GA, with no prejudice for/against FA/A/FL. They cite their summaries (I can give proof of that if needed) and I'd even go as far as calling them experts in the field of tornado history, although that can be debated. I see it as no different to citing Grazulis's "big books". I can directly contact the runner of the project (Jen Narramore) if y'all have questions for the person who heads the project. E_F⁵ 20:06, 24 January 2025 (UTC)Reply

  If you're going to contact the team, can you tell them to either add a button to cite their page or get rid of the annoying "Copying text is disabled on this page" plugin? Not a dealbreaker for use on Wikipedia but it seems they really don't want anyone using anything from their site, even for simply citing information from there. Departure– (talk) 18:48, 25 January 2025 (UTC)Reply

  Sure, although I do understand why copying is disabled, obviously to prevent against unattributed plagiarism. E_F⁵ 19:01, 25 January 2025 (UTC)Reply

IMO Generally reliable for anything up to GA. I'm not sure about FAC/FL. Wildfireupdateman :) (talk) 20:11, 24 January 2025 (UTC)Reply

I may have confused it with other sites, but Tornado Talk's forums in particular brought me to the conclusion that it was unreliable. However, I see now that it's run by a select group of people instead of the quorum of faceless internet usernames. However, before giving my support to reliability, have any of their claims failed a verifiability challenge, what kind of claims do they typically back up, how often do they back up their sources, and what credentials do the editorial team have? Departure– (talk) 20:13, 24 January 2025 (UTC)Reply

I believe you're confusing it with https://stormtrack.org/. Wildfireupdateman :) (talk) 20:14, 24 January 2025 (UTC)Reply

Here is a list of researchers, the vast majority of whom seem to be professionals with credentials. Funny enough, Lon Curtis, photographer of the 1997 Jarrell tornado, is on their team. E_F⁵ 20:16, 24 January 2025 (UTC)Reply

Alright then.

My current view is that, as a tertiary source apparently run by freelancers (and with an annoying extension blocking me from copying content out of the page), it should be treated as such. Most claims should be attributed. Where possible, cite other sources, however, they seem reliable enough for many claims.

I just checked the "about" section and saw that Grazulis is getting on in age. Before the inevitable event of his passing, I hope to get his article to GA or FA, because he's contributed so much to the contemporary weather and tornado scene. Above all, I hope either him or another member of the Tornado Project is able to publish the next edition of "Significant tornadoes" within my lifetime, assuming one of the monsters in the book doesn't take me first.

Additional considerations apply, but generally reliable. Departure– (talk) 20:23, 24 January 2025 (UTC)Reply

•   Comment: This question has come up several times in the last few years (including in GANs and possibly FACs). I wonder if we should take this to WP:RSN with an RFC to help establish reliability. An RFC over at the Reliable Source Noticeboard would allow us to formally add TornadoTalk to the big list at WP:RSP and help limit any future discussions for weather GANs and FACs. I can actually personally attest that TornadoTalk was questioned and removed during the GAN for Tornado outbreak of February 12, 1945 after its reliability (secondary-source confirmation of reliability) couldn't be verified. Grazulis' book(s) and TornadoProject were considered reliable due to Grazulis'-well documented RS and usage by the National Weather Service. However, TornadoTalk, at least during that GAN, didn't have clear RS-establishing their reliability, so it was removed as a technical self-published source from a group that had (at least at the time of that GAN) no sources confirming they met the qualifications as a subject-matter expert.

So with that, would y'all be ok if I start a formal RSN, since even GANs have questioned TornadoTalk's reliability? Courtesy Pings: Departure–, EF5, Wildfireupdateman. The Weather Event Writer (Talk Page) 19:00, 25 January 2025 (UTC)Reply

I mean, I think a project-specific discussion like this is fine as-is, from here to RSN is an unnecessarily large step. E_F⁵ 19:02, 25 January 2025 (UTC)Reply

This is partisan group so a RSN would be appropriate Noah, BSBA^Talk 20:32, 25 January 2025 (UTC)Reply

• Discussion opened at RSN: Wikipedia:Reliable sources/Noticeboard#RFC: Tornado Talk. @Hurricane Noah:, Departure–, EF5, Wildfireupdateman. The Weather Event Writer (Talk Page) 21:48, 25 January 2025 (UTC)Reply

There is a requested move discussion at Talk:Tri-State tornado outbreak#Requested move 18 December 2024 that may be of interest to members of this WikiProject. Aviationwikiflight (talk) 13:24, 26 January 2025 (UTC)Reply

This page hasn't been updated or properly sourced in 15 years. Please, rescue it or go to WP:AfD. 2025 is a year of decisive action. Bearian (talk) 17:09, 26 January 2025 (UTC)Reply

To further collaboration, I've assembled a list of tornado articles in draftspace as of 17:39, 30 January 2025 (UTC).

• Draft:1931 Birmingham tornado
• Draft:1954 London tornado
• Draft:1967 Belvidere tornado

One of mine. Definitely interested in bringing this up to quality once I get some time. Departure– (talk) 17:55, 30 January 2025 (UTC)Reply

• Draft:1973 Columbus tornado

Abandoned, not enough sources. E_F⁵ 13:32, 31 January 2025 (UTC)Reply

• Draft:1974 DePauw tornado

I have abandoned this one, not enough sourcing. E_F⁵ 17:47, 30 January 2025 (UTC)Reply

• Draft:1976 Jordan tornado

One of mine. One of the most interesting tornadoes I've written about. Departure– (talk) 17:55, 30 January 2025 (UTC)Reply

• Draft:1979 Cheyenne tornado

One of mine. As with Jordan, there's a lot to love about Cheyenne from what few sources exist. Departure– (talk) 17:55, 30 January 2025 (UTC)Reply

• Draft:1995 Pampa tornado

I have abandoned this one too, not enough sourcing. E_F⁵ 17:47, 30 January 2025 (UTC)Reply

• Draft:2002 Indianapolis tornado

I'd work on this more today, but I can't get my hands on the Storm Data publication for September 2002 because the server's offline! Argh! Departure– (talk) 17:55, 30 January 2025 (UTC)Reply

• Draft:2003 Moore-Choctaw tornado

I have abandoned this one, not enough sourcing (although it could very well be notable). E_F⁵ 17:47, 30 January 2025 (UTC)Reply

• Draft:2006 Millsfield tornado
• Draft:2010 Millbury tornado
• Draft:2011 Askewville tornado (draftified from mainspace)
• Draft:2011 Enterprise tornado

Still a work-in-progress. E_F⁵ 17:47, 30 January 2025 (UTC)Reply

• Draft:2014 Noxapater tornado

I have abandoned this one too, not enough sourcing. E_F⁵ 17:47, 30 January 2025 (UTC)Reply

• Draft:2015 Holly Springs tornado

See 2015 Holly Springs–Ashland tornado, we both started it at around the same time, funny enough. E_F⁵ 17:47, 30 January 2025 (UTC)Reply

• Draft:2021 Newnan tornado

Sources supporting sustained and significant coverage, along with lasting impacts that can be detailed in the Aftermath section, do exist, I have just been busy IRL lately and haven't had much time for substantial article work. Will continue when I have a bit more time on my hands. Chris ☁️^{(talk - contribs)} 04:07, 5 February 2025 (UTC)Reply

• Draft:2018 Alonsa tornado
• Draft:2023 Covington tornado

Just a fork now, but I was interested in getting this to mainspace in the not-too-distant future. Departure– (talk) 17:55, 30 January 2025 (UTC)Reply

• Draft:2024 Tallahassee tornadoes

This is the article I have the highest confidence in getting to mainspace. Departure– (talk) 17:55, 30 January 2025 (UTC)Reply

• Draft:Kapuskasing tornado (unlikely to go anywhere)
• Draft:7/14/2000 annville ky tornado (definitely not going anywhere)

See also:

• Draft:List of US tornado records by state

Not abandoned yet. Departure– (talk) 17:55, 30 January 2025 (UTC)Reply

• Draft:Tornado outbreak sequence of June 11–14, 1976

Partially abandoned. I'll resume work once Jordan's in mainspace. Departure– (talk) 17:55, 30 January 2025 (UTC)Reply

• Draft:Tornadoes in Iowa
• Draft:List of places hit multiple times by significant tornadoes

I'm slowly chipping away at this one. Departure– (talk) 17:55, 30 January 2025 (UTC)Reply

• Draft:List of Canadian tornadoes and tornado outbreaks in the 2020s
• Draft:Rope tornado
• Draft:Tornadoes in the United Kingdom
• Draft:List of the most active tornado seasons
• Draft:List of airports struck by tornadoes

It's... probably notable? Departure– (talk) 17:55, 30 January 2025 (UTC)Reply

Yes, this is all from a search, but there are a lot more draftspace articles than I expected there to be. Departure– (talk) 17:39, 30 January 2025 (UTC)Reply

Wow, a suprising about of those are me starting things and not finishing them. Maybe I need to commit to an article, and finish it before moving on to something else. :) E_F⁵ 17:47, 30 January 2025 (UTC)Reply

There is an ongoing discussion/request in progress for a citation template to be created specifically for the Damage Assessment Toolkit (DAT). You can see the discussion here: Wikipedia:Village pump (technical)#Creation of new citation template for the U.S. Gov Damage Assessment Toolkit (DAT). The Weather Event Writer (Talk Page) 18:48, 30 January 2025 (UTC)Reply

How should articles deal with damage estimates for weather events? Departure– (talk) 14:46, 5 February 2025 (UTC)Reply

Opening comments: This was brought to the forefront of my attention by this damage estimate at 2023 Selma tornado. An inline comment there says Per an RfC, NOAA-damage totals MUST be supported by a non-NOAA reliable secondary source. The NOAA damage totals are not supported by a secondary source, therefore, per the RfC, the secondary source MUST be used. I have never seen this RFC before and this is the first I'm hearing of it because these NOAA estimates are being used all over Wikipedia unbounded and I didn't see a problem with that. Let me lay out a few points that might help discussion:

• NOAA:
  • NOAA often fails to provide an estimate for certain damage events. These are listed as $0.00k in the crop and property damage sections of the Storm Events Database.
  • The methodology for getting NOAA estimates isn't often discussed, but from the event summary of the Tallahassee tornadoes of 10 May 2024, we can get this much:
    • Damage costs to the city of Tallahassee were extensive. The city accrued at least $50 million in damages, not even including residential damages. Residential damage was significant. There were a total of 174 structures deemed destroyed, 742 with major damage, 780 with minor damage, and 417 that were deemed affected. The median home price in Tallahassee as of July 2024 is roughly $286,000. Thus, a rough estimate for residential damage is an additional $50 million for the destroyed structures (assuming $286K damage per structure), $74.2 million for the structures with major damage (assuming $100K damage per structure), $7.8 million for the structures with minor damage (assuming $10K damage per structure), and $2.1 million for the structures that were deemed affected (assuming $5,000 damage per structure). This brings the estimated grand total to $184.1 million, which will be divided equally between the two tornadoes since they merged together over the city.
  • NOAA damage costs are often an acceptable WP:CALC estimate of multiple Storm Events Database entries. For instance, the cost of $90 million for 2023 Little Rock tornado came from two database entries.
  • NOAA also operates the Billion Dollar Disaster report, a database of events that cost $1 billion or more in damage. This appears reliable and is used in many articles, especially for hurricanes.
• AccuWeather:
  • AccuWeather was a previous point of friction (see Wikipedia:Reliable sources/Noticeboard/Archive 452#AccuWeather for damage estimates) - this discussion entailed a preliminary figure of about $95 to 110 billion for Hurricane Helene. I believe the current AccuWeather estimate for that storm is $225 billion. The NCEI, a division of NOAA, states the median estimate for Helene is around $78.7 billion.
  • In addition, they also appear to have different figures for total economic loss in addition to or in lieu of property or insurance toll figures.
• CoreLogic:
  • CoreLogic appears to be an insurance appraiser used at a professional level to determine whether or not a property was affected by a storm. Their work appears to be in determination of the total value or exposure of properties potentially impacted by severe weather events. Take their March 31 summary - the title is CoreLogic Identifies Approximately 358,000 Homes Worth an Estimated $83.2B Potentially Within Tornado Paths and Hail Boundaries. This disagrees with the Billion Dollar Disaster listing for March 31, with their estimate being at $5.9 billion.
• Local sources:
  • If I'm not mistaken, local sources echo what insurance appraisers and NOAA relay. However, as is the case at 2023 Little Rock tornado, they can have citations for damage tolls known to be paid instead of just what could be paid.

In my opinion, sources should be cited inline attributed with what they're valuing the cost off of if relevant, and for AccuWeather in particular, attributed in-text. For instance:

• $59.1 million (property damage)^[1]
• $68.2 million (total insurance payout)^[1]
• $50 billion (AccuWeather initial estimate)^[1]
• $88.7 billion (AccuWeather total economic lost estimate)^[1]
• $7.2 billion (CoreLogic potential property exposure)^[1]

Let me know of what you think of this proposal. If you have another suggestion, feel free to discuss it. Departure– (talk) 14:46, 5 February 2025 (UTC)Reply

@Departure–: The RFC you mentioned at the beginning is why "Verifiability, not truth in action" was created and why Tornadoes of 2022#Costliest United States tornadoes has confirmed, factually incorrect information, but verifiable information. I'll do a larger comment in a little bit later, I just wanted to let you know on that RFC. The Weather Event Writer (Talk Page) 15:41, 5 February 2025 (UTC)Reply

Courtesy link: Special:Diff/1152938833#RfC on clarification of WP:CALC for costliest tornadoes Departure– (talk) 15:53, 5 February 2025 (UTC)Reply

Alright, so let me give my input on this RFC from a bygone age. This appears to be solely based on the use of NCEI sources in Costliest tornadoes of XYZ year articles and really shouldn't have any bearing beyond that context. Really, I don't know how much we even need those "costliest tornado" indexes on this encyclopedia, and I am sure as hell against these damage estimates being discounted outside of that context. Departure– (talk) 15:58, 5 February 2025 (UTC)Reply

Just a quick comment that List of costliest tornadoes in the Americas also exists as a stand-alone list. The Weather Event Writer (Talk Page) 16:27, 5 February 2025 (UTC)Reply

• Ok, here is my whole thought on the process. I have attempted to implement it in the past, with it almost always getting reverted.
  1. If NOAA is available, use NOAA as the damage estimate, as these, for U.S.-based weather events at least, are always regarded as the "official" damage totals.
  2. If NOAA is not available, (i.e. no-NOAA damage total in final report), then use an RS-based range with a note of "unofficial". For example, if NBC News said $1 million and AccuWeather said $100 million, then the infobox should reflect the RS range.
  3. If pre-NOAA finalized reports (i.e. within like 2-4 months of the weather event), then use an RS-based range, with a note of "unofficial".

For example, prior to NOAA releasing their official damage total for Hurricane Helene, I attempted to do an infobox with a similar format to how the 2013 El Reno tornado's infobox is. I will admit, my format was a little bad on the Hurricane Helen infobox, so I am not saying to go right back into it. However, the idea behind it still stands in principle. Hurricane Helene had an impact section chart listing various damage totals. I'm not saying every article should have a chart with damage totals, but maybe a bullet-point list reference or enf note listing the various sources for the "range" would be good. The Weather Event Writer (Talk Page) 16:42, 5 February 2025 (UTC)Reply

• Now, on the note above, I am strongly against the use of any AccuWeather-related damage "estimate". They have been known for a long time to be way off on their estimates and forecasts. AccuWeather claimed Hurricane Helene was going to be costlier than Hurricane Katrina and Hurricane Ian combined! Obviously, NOAA highly disagreed with their estimate and officially said Helene caused less damage than either Katrina or Ian by themselves. Honestly, another renewed RSP on AccuWeather might be in order. Anyway, that was just a small P.S. I wanted to say since AccuWeather got brought up. The Weather Event Writer (Talk Page) 16:42, 5 February 2025 (UTC)Reply

  All predictions are iffy, and AccuWeather failing at a prediction tells us nothing about accuracy of their post-storm damage assessments.  — SMcCandlish ☏ ¢ 😼  19:21, 6 February 2025 (UTC)Reply

  @SMcCandlish: They are really off on every prediction. I just gave a single example. At the next RSP (which I will probably open sometime in the next few weeks), I will give dozens of examples of them, in short, always being the highest-possible damage estimate, to the point where they easily cross into the “generally unreliable” territory on that category, but also their forecasts are often challenged/laughed at by other meteorologists as basically just trying to cause public panic. AccuWeather is a big sensationalist weather source, which a very detailed RSP in the future will show. Anyway, that was just a single example and for this discussion, you can just picture that type of damage estimate as their norm…i.e. always the upper-end of damage estimates for any weather event. The Weather Event Writer (Talk Page) 20:57, 6 February 2025 (UTC)Reply

  Fine, that's a case to make at WP:RSP then. It doesn't change my position at all with regard to applying WP:DUE policy properly, including no longer treating US government source as "super-reliable", which is something we should not have been doing in the first place.  — SMcCandlish ☏ ¢ 😼  21:24, 7 February 2025 (UTC)Reply

• I agree with "sources should be cited inline attributed with what they're valuing the cost off of if relevant, and for AccuWeather in particular, attributed in-text", but would go further and require in-text attribution for NOAA and another source. I do not agree with WeatherWriter's take that NOAA is more reliable because it is "official". WP is not in the habit of giving more weight to governmental sources, and given what is happening to the US government right now, no such source can any longer be taken as prima facie reliable, and Trump and company's outright hostility to emergency management agencies of all kinds in particular calls into question whether NOAA will remain reliable enough to use at all. In the interim, any time WP has ostensibly reliable sources that provide conflicting numbers, with regard to anything, our job is to provide a range (when summarizing), with sources, and to provide more specific numeric details claims (again with sources) in the main body of the article. This general principle is not magically voided just because the subject happens to be weather.  — SMcCandlish ☏ ¢ 😼  19:21, 6 February 2025 (UTC)Reply

Well, your opinion on whether Trump being President will influence how the National Weather Service conducts a tornado damage survey is noted and entirely irrelevant for this discussion…sorry to break it to you. Anyway, what do you think of the proposal below? The Weather Event Writer (Talk Page) 21:23, 6 February 2025 (UTC)Reply

You're clearly not sorry, and I'm not interested in any apologetics from your or anyone else's direction. It really has nothing to do with "Trump being president" intrinsically; this is not a popularity contest. It has everything to do with the Trump administration's specific policies of a) hostility towards emergency management agencies both within and without the US government and clear intent to interfere with them for propagandistic and worse purposes, and b) an agenda to hobble or even dismantle federal agencies of all kinds to the maximum extent possible, even beyond the limits imposed by the US constitution. Whatever effects this will have off-site, the obvious effect it has with regard to Wikipedia is that US federal agencies are, or soon will be, less reliable sources, both because their mandates are being interfered with to serve political purposes and because their financial and human resources, and other factors of their ability to operate effectively at all, are being slashed. You can praise this approach, as a political move, all you like, but it will do nothing to change the facts of what it means for WP relying on them as sources.  — SMcCandlish ☏ ¢ 😼  21:24, 7 February 2025 (UTC)Reply

PS: Here's [1] just a hint of where this is all headed, in a directly pertinent area of government data, and we're only part way through the third week of this administration.  — SMcCandlish ☏ ¢ 😼  21:45, 7 February 2025 (UTC)Reply

No comment on any other sources. But AccuWeather should NOT be used for damage estimates under any circumstances. They majorly inflate the totals to unrealistic numbers. Hurricane Clyde 🌀^{my talk page!} 20:36, 7 February 2025 (UTC)Reply

For my proposal, I am thinking of a damage total range & collapsable list inside the infobox listing the damage totals from various sources. Below is a copy/pasted version of the 2023 Rolling Fork–Silver City tornado’s infobox, with the new damage total list. These sources can and should also be listed an explained in the article’s aftermath section. Note, source 1 and 2 are NOAA sources and source 3 is a secondary RS source.

• Thoughts on this type of proposal? The Weather Event Writer (Talk Page) 21:16, 6 February 2025 (UTC)Reply