Loss of key US satellite data could send hurricane forecasting back 'decades'
Researchers say the satellites themselves are operating normally and do not appear to have suffered any errors that would physically prevent the data from continuing to be collected and distributed, so the abrupt data halt might have been an intentional decision.
Wait, the U.S. aren’t even going to try selling the satellites? We’re just scrapping them?
How large systems with exposure to these places (insurance, capital markets) respond is what you should look to next. What do you do when you don't have the data to accurately price risk?
Relevant comments:
https://news.ycombinator.com/item?id=43366311
https://news.ycombinator.com/item?id=42450680
https://news.ycombinator.com/item?id=41664750 (top comment of this thread aggressively relevant)
SCOTUS, which has demonstrated that it is ideologically aligned with the regime
If you read SCOTUS's opinions this is obviously false. Alito and Thomas are bought. But the others have their own quirks and agendas.
For ordinary people it can feel reasonable to keep your head down and hope that somehow this blows over. But for SCOTUS it's entirely within their power to draw a line, and it seems like at best their idea has become "Maybe if we give him what he wants he'll go away?" which is dumb, Kipling wrote his famous poem "Dane-geld" about this, it's well over a century old and it's about a mistake England (or rather one of its Kings) made last millennium (when he wrote it, ie now over 1000 years ago).
could probably imagine that ACB is just very stupid I guess? She's made choices which only make sense if they're out of blind loyalty to the man who gave her a job
Barrett has sided with the liberals on various decisions. SCOTUS has a problem. But its problem isn't blind loyalty to Trump. It's that there is a deeper conviction about the way the world should work that sometimes aligns with Trump in ways that are deeply damaging to our society.
If you want to see a judge who's blindly deferential to Trump, that's Aileen Cannon.
It's not like we're asking for SCOTUS to accept constitutional slights from the left side of the aisle, its about consistency of reasoning regardless of which party is involved.
As you've noted, the conservatives of SCOTUS are working backwards from their desired goals rather than pursuing justice for all.
I haven't seen evidence that the Austin robotaxi launch was unreasonably bad. There were a couple viral incidents of undesirable behavior, though no collisions as far as I've heard, which is significantly better performance than one expects from typical human drivers.
What do you do when you don't have the data to accurately price risk?
Insurance companies will just be sending up their own satellites, and that is the true goal. Force people to pay money to private entities for a service that used to be provided by the government for free.
Functionally, in such a system there is no difference between that and regular taxes, just in a private system there's opportunities for those in power (because you gotta have a lot of money to send up a powerful satellite) to make even more money.
With the current US administration, always look at the grifting opportunities, that will explain virtually all policy decisions.
Which is why the government running satellites it would need to run anyway is much more efficient.
There's no reason to expensively launch your own forecasting system when you can instead just wait for someone else to do it and then use their insurance rates to do your own forecasting.
Indeed but who's going to do that? The US government will, more likely than not, have lost the ability entirely, and Europe... good luck waiting on us.
Which is why the government running satellites it would need to run anyway is much more efficient
Indeed. But there is no opportunity for continuous recurring grift revenue in that, and that is all that drives the current administration.
On the other hand, in the first Trump admin the AccuWeather spam site guy was trying to restrict NWS data to private companies:
https://en.wikipedia.org/wiki/Barry_Lee_Myers
I think AccuWeather opposed the Project 2025 plan to remove weather tracking frothe government though, they just wanted it to be tax payer paid but exclusively provided to corporations for sale to make competitive upstart weather sites harder to establish (you can bid more if you already have lots of users, without them you have to build something so great and potentially profitable that you can get VC to fund your purchases of the data).
https://www.masslive.com/news/2024/07/accuweather-rejects-pr...
When you control the propaganda that lies between the data and the public, the underlying availability of the data is irrelevant. The propaganda already overwrites the data.
Honestly I would suspect that limiting the data is a strategic asset. The US can use its knowledge of weather events as leverage to cow other countries, or a weapon against countries it dislikes.
Hello <other_country>, are you worried about the impact of weather on your population? Lower your tariff rate on us and we will be glad to help give you advanced warnings so you evacuate your people
And likewise they would completely withhold such data from "enemy" countries.
2016 failure of DMSP 19 without replacement[edit] On 11 February 2016, a power failure left both the command-and-control subsystem and its backup without the ability to reach the satellite's processor, according to the U.S. Air Force Space Command investigation released in July 2016 that also announced that DMSP 5D-3/F19 was considered to be 'lost'. The satellite's data can still be used, until it ceases pointing the sensors towards the Earth. The satellite was the most recent on-orbit, having been launched on 3 April 2014.[15]The failure only left F16, F17 and F18 – all significantly past their expected 3–5 year lifespan – operational. F19's planned replacement was not carried out because Congress ordered the destruction of the already constructed F20 probe to save money by not having to pay its storage costs. It is unlikely that a new DMSP satellite would be launched before 2023; by then the three remaining satellites should no longer be operational.[16]
To anyone acting as if this is a surprise or they're suddenly caught out and have to switch to another provider, I have to wonder, with the writing on the wall for 8 years now, how have you not already updated your plans?
That's the guardian for you. Remove context. Generate hyperbole. Beg for money.
To anyone acting as if this is a surprise or they're suddenly caught out and have to switch to another provider, I have to wonder, with the writing on the wall for 8 years now, how have you not already updated your plans?
That doesn't accurately capture the reason why there's outrage here. In the weather community, we're constantly thinking through contingencies because a great deal of things are out of our control - and we rely on aging infrastructure, much of which is already flaky to begin with.
Data outages and data loss happens. But there's no reason to allow a _preventable_ data loss to occur. The DMSP data is still being collected, it's just not being distributed downstream. And the decision to make this policy change was seemingly done rapidly and with no input or feedback from the user community of this data - both inside and outside the federal government.
There's no reason to turn off the spigot of this data. And there certainly is no reason to do so abruptly and with virtually no notice. As a consequence, the community is limited in its ability to adapt. For instance, it would take time (and money) to spin up more hurricane hunting resources to replace the overpass data that the SSMI/S instrument captures. Some private companies operate PMW satellite constellations and we could accelerate the acquisition of these data, but there are limited (read: none) federal mechanisms to do this and due to vertical integration in the weather industry, the operators of these constellations may not actually be inclined to do so - and certainly won't do so on the cheap, especially for the federal government.
So this isn't hyperbole. This is a really big deal. It might not be visible to you, but there is a panic and scramble occurring in the weather community to figure out what to do from here.
And for the record - yes, the same panic would happen if the DMSP satellites failed suddenly due to natural causes. But this current situation could've - and should've - been prevented.
Some private companies operate PMW satellite constellations and we could accelerate the acquisition of these data, but there are limited (read: none) federal mechanisms to do this and due to vertical integration in the weather industry, the operators of these constellations may not actually be inclined to do so - and certainly won't do so on the cheap, especially for the federal government.
That's the goal, actually. You can be sure someone in the admin owns stock of these companies and pushed for this policy for this very reason.
But Hanlon's razor ought to apply until shown otherwise.
But Hanlon's razor ought to apply until shown otherwise
I'm no longer willing to grant this administration this privilege. The last few months were an utter clownshow of corruption.
At some point you take your hand off the burning stove, even if it means amputating your arm. Some folks should prepare for that contingency while those of us who can still stomach it pursue reason.
At some point, the consequences of actions are felt.
“The market can stay irrational longer than you can stay solvent,” is a truism from before the Trump era, but it still rings true.
That the administration might eventually realize that one of their policies is hurting small business owners, well, that’s cold comfort to someone whose business is struggling or failing now due to unpredictable tariff rates.
It just so happens that the communities most likely to be adversely and quickly impacted by the loss of these data are deep Republican bastions in the South / Gulf Coast.
At some point, the consequences of actions are felt.
yup, and that's when a Democrat comes in, fixes the worst of the mess, and then a Republican comes in whining about soooo much change. And fiscal stability. And god knows what else. And then, they cut taxes for the rich again and seriously hike the debt.
The "run deficits in my 4 years to pay for nice things, to be paid for by taxes once I'm out of office" shit has to stop.
companies I'm referring to are (generally) not publicly traded
Stock doesn't have to be publicly traded to be traded.
Generally, you use space hardware until it dies, which is hopefully well beyond the design life.
Can we ask dang to change the title to something like "Blocking of key US satellite data could...."?
There is one operating satellite in this constellation, and congress voted to shut down the program in 2015.
The DMSP program was discontinued in 2015 by a vote in congress[1]. Virtually every working stallelite in this program has failed. As best as I can tell there's just a single working one specifically NOAA-19[2].
Instead the program has switched to JPSS[3] which is part of GEOSS[4].
[1] https://en.wikipedia.org/wiki/Defense_Meteorological_Satelli... (scroll up slightly)
[2] https://en.wikipedia.org/wiki/NOAA-19
[3] https://en.wikipedia.org/wiki/Joint_Polar_Satellite_System
[4] https://en.wikipedia.org/wiki/Global_Earth_Observation_Syste...
So to be frank, the only thing that's "NOT TRUE" is nearly all of your post.
- DMSP satellites are up and measuring data - These data will continue to be measured after Monday - the government is discontinuing processing and public access to the data - This will impact our capacity to predict hurricanes and monitor sea ice.
Which of the above are “not true”?
>The loss of DMSP comes as Noaa’s weather and climate monitoring services have become critically understaffed this year as Donald Trump’s so-called “department of government efficiency” (Doge) initiative has instilled draconian cuts to federal environmental programs.
We can't actually say this was DOGE, so we're going to imply it using emotionally charged words, and 90% of folks with bad media literacy will come away thinking it was DOGE (just check the reddit comments).
This in-vogue method of "lying without lying" is shockingly common nowadays, but apparently it's okay for media to lie because Bad Man Bad.
We can't actually say this was DOGE
The article is saying it was DOGE. DOGE directly attacked our hurricane-forecasting capacity[1]. OMB, i.e. Vought, continues that attack[2].
Given the top three states by hurricane risk voted for Trump in ‘24[3][4] this should make for an entertaining hurricane season. (Particularly if both a red and blue state get hit and request federal assistance.)
[1] https://apnews.com/article/national-weather-service-layoffs-...
[2] https://en.m.wikipedia.org/wiki/NOAA_under_the_second_presid...
[3] https://www.realtor.com/news/trends/states-most-at-risk-for-...
[4] https://en.m.wikipedia.org/wiki/2024_United_States_president...
The article is saying it was DOGE
Yah, but it's the guardian. They aren't exactly reliable.
For it to be DOGE would require a time machine, because this project was shut down in 2015.
but it's the guardian. They aren't exactly reliable
This is valid and I'm open to someone calling out the reporting as non-factual with evidence.
Pretending The Guardian is trying to imply this was DOGE when it straight out says as much, on the other hand, is closer to a reading-comprehension issue.
Now media gets a free pass on certain lies because Bad Man Bad, and (evidently) people aren't even allowed to point out the lie.
Hint: when the media can make up whatever they want about someone, they can quickly twist perception to make anyone into the Bad Man.
And now the current funding request enacts a ~30% funding cut.
I'm not sure the factual issue you're seeing. Is it that the statement wasn't definitive enough in saying that DOGE apparently was a large part of instituting these cuts?
(Yes, I know OPM implemented many of these programs, but they're apparently at DOGE's request, named after the "Fork in the Road" initiative at Twitter, using data gathered by DOGE IT staffers, &c. If we give credit for any cuts, we have to give them credit for significant cuts at NOAA.)
Yes, the DMSP program was aging and slated to wind down as replacements - both federal and commercial - came online in the second half of the 2020's. But in general, if valid and useful data continues to stream from these types of satellites, you use it and monitor for disruption.
As someone who uses the DMSP data every single day, let me be very clear: there was no warning or expectation that such an abrupt change was going to happen. Yes, we all have contingency plans for if a satellite fails or a data link goes down. But to be given basically 5 days notice that a significant, mission-critical asset would be taken offline? That doesn't - and shouldn't - happen.
I would bet you, but that money's too easy. :)
Again, this exact conversation is the genius behind 'lying _without lying_.' You can always claim in high-literacy communities like HN that no, nobody would ever be silly enough misread it like that, all while watching your misinformation spread across the low-literacy communities like facebook and reddit.
The Guardian et al has done this too often for plausible deniability. Even I can pick up on the pattern, and that's without access to the big boy's social media engagement and sentiment tracking tools.
I see this sentiment a lot lately, and I see your HN join date is similar to mine. HN is more mainstream than it used to be, for better or worse. There is a lot more overlap between commenters on HN and Reddit nowadays, especially in certain categories of subreddits.
Personally, I lament the web being a high-literacy community.
But a couple of things should be considered here:
* Intention
* Degree
* Impact
Degree is likewise to be noted, where it can range from lost nuance to outright lies.
Impact is also a concern if it is a concern. A news article that compels people to randomly attack their neighbors is more of an issue than one that tempts you to buy a new snack.
And most importantly of all: "the media" is not a singular entity and they vary strongly in their veracity and scope, as well as their agendas. Some are at their core intending to serve the public, others are a business to sell advertising, and others are literally propaganda outfits to serve vested interests (e.g., Fox News was created to be the PR arm of the GOP -- this is a fact and not conjecture).
So yes, the NYT can get things wrong (like the lead up to the Iraq invasion), I trust them more than Fox News (which destroyed a community by spreading lies about their new immigrant neighbors eating people's pets).
Hope this helps!
A lot of these important projects have a single point of failure - who is the president every four years. I wonder how we build institutions and resources resilient to that?
I realize privatization is an ugly word, but could some of this stuff be provided by the private sector?
Can we make it possible to fund initiatives in a multinational manner where countries contribute to these efforts, but if one country blinks out, then you still have it go along?
A lot of these important projects have a single point of failure - who is the president every four years. I wonder how we build institutions and resources resilient to that?
We already did. The legislative branch allocates funds for stuff that the people deem worthy. That budget becomes law. The Constitution says the "President shall take Care that the Laws be faithfully executed." There's even a specific law that prevents the President from withholding Congressionally-approved funds.
What you are seeing here is not a lack of designed resilience, it's the wilful removal of that system.
who is the president every four years...could some of this stuff be provided by the private sector?
A president cares about election every four years. Private sector cares about it every quarter. I doubt privatization is improvement if you want to focus on long term.
could some of this stuff be provided by the private sector?
Yes, but the key man problem still exists. For instance, SpaceX could build/operate a network of weather satellites for various nations but the instability of the founder leads to similar issues.
If you know your car's engine is going to need replaced after exactly 100,000 miles, you know to save up for a new engine or a new car - and you know how long you have to save, so you can precisely set aside an appropriate figure every month.
If you know your car's engine will die sometime within the next 15,000 miles, you know you need to start saving up immediately, but b/c you don't know when in the next 15,000 miles you have to rush your saving.
If you have no idea when your car's engine is going to die, you are likely to end up dead engine and little to no savings.
The real reason insurance is high is because of fraudulent claim risk. Hurricanes themselves are more or less a solved problem in Florida. That data is useless.
Hurricane risk has been grossly exaggerated for years
Year-over-year, economic impacts and disruptions due to tropical cyclones are dramatically rising. Most of this is an exposure issue. But long-tail events - like Andrew's utter devastation of Homestead in 1992 or Katrina's unique confluence of storm surge in urban/suburban parishes in LA - can and do happen.
One day, there will be another Galveston or Homestead.
There won’t be another Andrew because the building codes were changed so that all new construction must withstand category 5 storm force
I sincerely hope you're right, but there is plenty of evidence suggesting that this will not be the case, owing to a multitude of factors:
- not all housing stock is <30 years old and has been properly retrofitted to meet state specs
- the climates around the Gulf, which tend to be more humid, can lead to premature degradation of things like strengthened anchor bolts and roof attachments
- there continue to be immense factors related to cost and time-to-build which provide significant negative pressure towards cutting corners and minimum-compliance which may mitigate some of the attendant benefits of strengthened building codes
An event like Andrew _is the selection event_ that you're referring to.
The worst is some flooding and damage at coastal areas, but it’s always anti-climactic.
The residents of what used to be Ft. Meyers Beach would probably disagree with you.
Hurricanes themselves are more or less a solved problem in Florida.
I have been in Florida for nearly a decade now. I'd say that the above statement is at best, disingenuous. It's just not true. MAYBE Cat1 hurricanes are a solved problem, but nothing above that. The busiest economic center in Florida (Miami's Brickell area) is 6 feet above sea level. Any major storm locks that part of town down for days. My own building's parking lot is 5 feet above sea level, and yes, it's flooded every time we have a storm.
In 2022, Hurricane Ian caused extreme flooding in the Orlando-region, including in areas that have never suffered from hurricane flooding before. For me personally, all 3 cars parked at my house were total losses b/c of the flood damage.
The extreme and extensive damages in the Appalachian region last fall is another great example of hurricane risk not being "grossly exaggerated".
Hurricanes themselves are more or less a solved problem in Florida.
I'm going to go with less, though I suppose you could call "experience widespread destruction, get bailed out by the federal government, rebuild in the same spot" to be a permanent solution.
Florida has maybe solved cat 1-2 hurricanes.
They all voted for this with extreme skew towards the current policies. What is the point of trying to save this satellite data if the very people most affected dont care for it?
There may come a day when they have saved up enough grievance against the Republicans to look for an alternative. But right now they have a solid foundation of anti-woke grievance and they can be counted on to keep voting the same way.
It's part of the Administration's war on ... Florida?
The administration of Florida has a war on the idea of climate change:
* "Ron DeSantis signs bill scrubbing ‘climate change’ from Florida state laws": https://www.theguardian.com/us-news/article/2024/may/16/desa...
* "Florida Officials Barred from Referencing “Climate Change”: https://climate.law.columbia.edu/content/florida-officials-b...
This allows (certain) Florida politicians to put their head in the sand even more than they already have.
There is no war on anyone, and this has nothing to do with Trump, DOGS, or Climate change. Rather there were too many satellite failures, leaving just a single operating one in orbit.
Also from NOAA: “Noaa said they would not affect the quality of forecasting.”
Decommissioning old sensors?
The ATMS sounder that remains is far inferior to hurricane forecasters than the SSMIS instrument the Department of Defense is discontinuing. Unlike the SSMIS which scans at a continuous resolution, the quality of the ATMS degrades considerably on its edges, rendering the sounder useless for most operational hurricane forecasts. The example below shows the difference between the scans from both instruments for Hurricane Erick last Wednesday, June 18th.
There is also a useful figure included.
The damage to weather prediction and climate research will go far beyond "decommissioning old sensors." We're talking about an abrupt end to nearly all science related to climate and weather in the US. Even if a future administration decides to turn funding back on, it will take a generation to rebuild the research community.
There is an embarrassing trend at the US EPA for example, to get interns and volunteers to do some official work, with no real way to transition to full time with job security and health insurance. This is partly due to the unending stream of applicants, despite few positions.
So from this context, rebuilding research means .. what? Both institutional knowledge store and authoritative titled positions. After some advances in data management by Google and some other commercial players, it is not clear what role Federal science had anyway, from this desk.
https://www.pbs.org/newshour/politics/fact-checking-what-pro...
I'm actually surprised that the successors to DMSP don't meet the same needs. Or is the problem that they do and the government just doesn't share that data?
HAFS is often the best
https://www.noaa.gov/news/new-noaa-system-ushers-in-next-gen...
European models assimilate in data from US satellites and vice versa
The GOES-R satellites seem to have equal or better resolution: https://journals.ametsoc.org/view/journals/atot/4/4/1520-042... https://en.wikipedia.org/wiki/GOES-16
DMSP resolves to 600m, while GOES-R resolves to 500m (don't confuse it with the older GOES satellites mentioned in the article).
* Ok that's an oversimplification. They actually turn encryption on while the satellite is over certain areas. But if you're in the Continental US I think it's in the clear