This headline could really use the word "hurricane" in it. I had no idea what Otis or Acapulco were, and my initial guess was wildly off: some sort of space probe experienced a catastrophic failure on the way to an asteroid.
Otis is the name of one of the Fat Bear Week audience favorites so that was where my first thoughts went but I was very surprised to imagine an “Acapulco, Alaska” near where he lives.
If I heard that Otis exploded after visiting Acapulco, I wouldn't really consider that unexpected, but admittedly that's only based on personal anecdata.
I guessed the correct Acapulco then hypothesized that Otis was something being transported by truck to there. I even thought that Cape Canaveral or a lab like CERN would be a more likely destination for dangerous things that people like us would care about.
I think we'll see more of this sort of thing. Weather models have been made in a climate somewhat different than the current one and those models are relatively accurate when modeling the past. As more data rolls in about anomalies like this the models will get updated and they'll become more accurate again in specific cases that are outliers right now, but they may end up slightly less accurate overall (due to the increased bandwidth of possible outcomes). This happens in almost every field where modeling is used to predict complex systems. You can't guarantee both great performance on past data and great performance on data that you haven't seen yet, there is always the risk that you've lost generality as you increased accuracy.
> So how did every reliable model we use miss this? That’s for graduate students and researchers to answer in the coming years, because I have no formal idea.
Would love to see followup on this. Wish there were some way to like bookmark or something and come back in a few years.
I'm one of the researchers on a new disease forecasting project, and I'm bookmarking this because we often use hurricane forecasting as an example of a level of accuracy we aspire to, so this is a really interesting breakdown.
The big miss here seems to be that not enough data had been seen yet to even know the envelope of what's possible. Without that there will always be inputs where a model breaks down. Disease modeling is also always going to have to contend with the fact that our historical data was collected when things weren't quite what they are today, for instance population numbers and densities are still entering uncharted territory compared to the past.
Detailed summary from interview:
• only 1 hurricane hunter flight was done into this storm.
• mexican hurricanes just don’t get as much reporting coverage (historically). For example for atlantic hurricanes, NOAA hurricane hunter airplanes have tail doppler radar, which provides good data to feed into the model.
• without detailed doppler radar, they had to fallback to satellite measurements.
• Mexico doesn’t have ground based radar coverage at all, like the US coast does.
This is the most likely answer. Since the atmosphere is a chaotic system weather models are extremely sensitive to initial conditions. A lot of the improvements in weather forecasts have come from better data assimilation for better initial conditions.
too late to edit my comment, but after more reading it seems like mexico does have land radar along the coast, but it’s a much smaller network compared to the US coast. So for this particular hurricane, there was lack of coverage.
> Otis intensified so quickly that it basically outran the ability to measure how intense it actually was.
We're working on this problem at Care Weather (www.careweather.com). Current wind sensing satellites only measure ocean surface winds once per day. They're too expensive for governments to build additional satellites for more frequent refresh. We're lowering their cost by three orders of magnitude and launching a constellation to get hourly updates.
For me it is completely relevant to the article because it make it impossible for me to read for no legitimate reason.
Even if the site was down for everyone I would comment. Lots of people comment about “hug of death”, for example, and I find it very useful. I know it is not me and I can find an alternative.
And since it’s not down for everyone, only a few, for a new, odd, technical reason, I find it even more relevant to comment about it.
Just like it is useful when people comment when a site is not accessible from a certain country, like BBC sometimes or some YouTube video.
I see these comments often on HN and I always find them useful. They also usually are among the top comments, so likely not downvoted.
That said, I was premature in editing the comment to show my surprise in being downvoting. That, indeed, is against HN etiquete and also it was just some temporary state, since I got many more upvoted since then.
If my ISP was hosting the page, sure. The article is the thing. And if you can't read the article due to false positives around how the article is hosted, how in the hell is that irrelevant? "I got mine, screw you." is a disappointing lifestyle choice.
But great news, we're on HN, so maybe jgrahamc or eastdakota can show up and explain why it's totally normal and fine that their product prevents random people from reading about the causes of natural disasters.
I've always found the relative unreliability of meteorological predictions fascinating. A lot of the time they're spot on (even the notorious "probability of precipitation") and the way we're able to track and predict storm paths is so impressive.
But from time to time we get an event like this, and it really underscores how little we really know. The sheer complexity of the phenomena / the theory we use to contain and understand them, it just seems to misfire spectacularly on occasion.
In a fantasy world, I would like to see weather forecasts that reach several months out, with high reliability. Mapping winds, precipitation, etc, with high spatial and temporal accuracy.
Perhaps I'm just misunderstanding what you're saying but when it says "30% chance" what it is actually saying is "there's a 100% chance of rain in 30% of the coverage area".
“It feels like this was a combination of bad luck, bad timing, and bad placement. And it just so happened that a metropolitan area with over 1 million people was in the way.”
Readit News
https://www.imdb.com/title/tt0093617/
Deleted Comment
Would love to see followup on this. Wish there were some way to like bookmark or something and come back in a few years.
tldr: garbage data into the model, garbage out.
Detailed summary from interview: • only 1 hurricane hunter flight was done into this storm. • mexican hurricanes just don’t get as much reporting coverage (historically). For example for atlantic hurricanes, NOAA hurricane hunter airplanes have tail doppler radar, which provides good data to feed into the model. • without detailed doppler radar, they had to fallback to satellite measurements. • Mexico doesn’t have ground based radar coverage at all, like the US coast does.
https://news.ycombinator.com/item?id=38038558
https://www.reddit.com/r/RemindMeBot/comments/e1bko7/remindm...
Or put it on your calendar with an email alert ;)
Isn't this the expected behavior once it made landfall?
I found this analysis more comprehensive https://yaleclimateconnections.org/2023/10/nightmare-scenari...
We're working on this problem at Care Weather (www.careweather.com). Current wind sensing satellites only measure ocean surface winds once per day. They're too expensive for governments to build additional satellites for more frequent refresh. We're lowering their cost by three orders of magnitude and launching a constellation to get hourly updates.
edit: wow, that was some unexpected downvotes
https://arstechnica.com/science/2023/10/trying-to-make-sense...
Can anyone recommend a good one-click firefox extension to open an archive copy of the current page?
I'm using it on my phone. It gave me this https://archive.ph/20231026233758/https://theeyewall.com/try... with a few taps.
Even if the site was down for everyone I would comment. Lots of people comment about “hug of death”, for example, and I find it very useful. I know it is not me and I can find an alternative.
And since it’s not down for everyone, only a few, for a new, odd, technical reason, I find it even more relevant to comment about it.
Just like it is useful when people comment when a site is not accessible from a certain country, like BBC sometimes or some YouTube video.
I see these comments often on HN and I always find them useful. They also usually are among the top comments, so likely not downvoted.
That said, I was premature in editing the comment to show my surprise in being downvoting. That, indeed, is against HN etiquete and also it was just some temporary state, since I got many more upvoted since then.
But from time to time we get an event like this, and it really underscores how little we really know. The sheer complexity of the phenomena / the theory we use to contain and understand them, it just seems to misfire spectacularly on occasion.
In a fantasy world, I would like to see weather forecasts that reach several months out, with high reliability. Mapping winds, precipitation, etc, with high spatial and temporal accuracy.
Perhaps I'm just misunderstanding what you're saying but when it says "30% chance" what it is actually saying is "there's a 100% chance of rain in 30% of the coverage area".
Not sure if that makes it more or less accurate.