4 ms·
So I'm not buying it from the articles. They are unconvincing and not really saying too much. They keep point back to satellite data and that surface temperatur
by godelski 3y ago
So I'm not buying it from the articles. They are unconvincing and not really saying too much. They keep point back to satellite data and that surface temperature stations are biased. I actually addressed some of this in this comment[0]. The arstechnica article actually discusses a lot of this. We know that airport data is biased. No one has claimed otherwise. It's also why the data is pre-processed before being used. It doesn't make it garbage, it just makes it biased. If you know the direction it is biased, you can unbias it. I'm not sure what Watts is talking about and I can't read his report. Even using the wayback machine I can't load it. But if he did what he's saying he did then it looks like he just unbiased the debiased data, not the raw. The arstechnica post even goes through this btw.
Processing data is pretty common. I mean we aren't fucking idiots. If you can figure out an issue in 2 seconds, you bet it has been addressed. Other sources are just other blog posts so forgive me for not taking them as good value given that I can read articles from scientists and understand all their methods and models (which are open sourced[2]). I mean even the blog you linked me has a graph of arctic sea ice[1] (also addressed in other comment) where they didn't even bother to remove the 1981-2010 mean line. Which both years are below... But if you look at my comment[0] you'll also see why their selection (especially of the time range in months) is deceptive. I walk you through how to verify yourself. It is fine if you don't believe the NOAA data, but if you don't then you also can't believe dailysceptic since they are using that same data to dispute NOAA's claims. Both are dependent on the data being accurate. Which a big irony given our above discussion.
To be more direct with satellite data (because I'm guessing you aren't going to read the arstechnica article), satellites don't measure temperature, they measure brightness. That is then turned into temperature. But they also don't measure ground temperature. So you're biased to that. But this can of course be adjusted and corrected for. But again, if we can do that for satellites why can't we do it for airport thermometers? The reasoning just doesn't line up in a consistent fashion.
I can tell you're really passionate about the subject. I congratulate you on that. We need passionate people and in no way do I want to get rid of your passion. You're trying to seek truth, and that is honorable. But an important part in science and truth seeking is to become your own adversary. Once you feel that an idea is good you have to attack it pretty fucking hard. You can't be sad when things get knocked down, it is just part of the process. You take what's left and rebuild from there and repeat. You need to challenge your own viewpoints. If there is nothing to convince you that would cause you to change viewpoints (even in the hypothetical!) then you aren't actually seeking truth, you're seeking validation.
So if you're seeking truth and you are acting in good faith, tell me just one precise thing that would get you to change your mind if the data suggested. I'll go first: I would doubt the weather models if I saw a constant or increasing trend of arctic sea ice that was outside variance levels.
[0] https://news.ycombinator.com/item?id=36564370 https://news.ycombinator.com/item?id=36564370
[1] https://dailysceptic.org/wp-content/uploads/2022/07/image-69.png https://dailysceptic.org/wp-content/uploads/2022/07/image-69...
[2] https://berkeleyearth.org/archive/about-data-set/ https://berkeleyearth.org/archive/about-data-set/ (btw, their skeptics guide also addresses all these issues. I hate to break it to you, but these are the same talking points from skeptics over the last decade. So it tends to be fairly easy to refute because noting additional has to be done. Just reapply the previous method to the newer data)
- revelio 3y agoYour posts are quite confusing, do you realize that? I started a reply asking if you had mixed up different locations in the thread, because it seemed like none of the stuff you were writing was connected with the topic or people being discussed. Now I look carefully you seem to have gone off on a tangent and started replying, without any setup or intro, to random other parts of the linked articles, instead of treating them as a citation for the claims being made and sticking to that. I was like, Anthony Watts? Huh? Why is he suddenly appearing? What report? What are you talking about? Going back it seems you've picked a fight with paragraph 9 in the second link when the relevant information for answering your query is in paragraph 1?! We can discuss other unrelated stuff, that's fine. But please accede first to these three requests: 1. Let's resolve the RAF Coningsby discussion before moving on. You appear to have mis-read the initial claim and then decided it didn't make sense and asked for sources. You now have the evidence. As you don't mention RAF Coningsby or the record breaking temperature in your reply, is it OK to assume you accept that evidence and thus that this reporting problem is real and did in fact occur as stated? 2. Please make it clear what the heck you're actually replying to! Quotes would avoid a lot of confusion here. Remember that some of these articles came out a year ago and I dug them up to satisfy your request for sources, I don't remember every single sentence in them by heart. 3. Finally, please cut out the attempts at philosophical lessons. You seem to be a climatologist, from your posts saying "we" when talking about that group? If so then that's a small community and one with a poor track record of accepting when its theories and evidence have been successfully knocked down by others. Before lecturing others about epistemic humility and truth seeking you may wish to take on Mann and other famous grifters that clutter up the scientific landscape, before random internet commenters of no note or impact. Moving on. > If you know the direction it is biased, you can unbias it. Processing data is pretty common. I mean we aren't fucking idiots. If you can figure out an issue in 2 seconds, you bet it has been addressed. If you know the direction and magnitude of the bias, which the Met Office clearly do not, and if the bias is constant and well characterized, etc. That's important! As you seem to have accepted by choosing to debate something different, climatological agencies are happy to report momentary blasts of jet exhaust as evidence of climate change. That's an issue that literally anyone can figure out in two seconds the moment they discover the thermometer isn't in the village as claimed, but rather right next to a place jet fighters take off, and yet, unambiguously, the Met Office had not figured this out. They even claimed they'd used a rigorous verification process! So please forgive us if we assume that you guys are in fact not addressing issues that any "fucking idiot" can spot in two seconds, because this would seem to be one of them. More generally, your field's attitude to data quality is guaranteed to create train wrecks like this. If your data is poor quality you often won't know how to characterize the corruption, which is exactly why good scientists obsess over how to collect data with high quality instruments and setups in the first place. They don't just scrape whatever data they find on the internet and then assume with 100% confidence that 100% of the measurement error both can be well characterized and actually will be. That's an absurd methodology that is guaranteed to lead to continuous disasters like this, as well as the entirely reasonable lack of trust when people discover that you constantly rewrite the historical record. You guys can't even change the data once, you keep doing it, meaning that the vast majority of the climate literature is based on measurements you later decided were wrong! Most of the climate literature should have been retracted by now! This is inevitable when you can't reliably characterize measurement error. Thermometers aren't advanced tech and it's easy to build networks to a standard that satisfies even critics so why do you persist in "fixing" and then constantly "refixing" corrupted data sets?