Former GOES engineer here. At this point I'd almost be surprised if 19 didn't have something go wrong. We had issues on almost every other satellite. GOES-17 had the loop heat pipe anomaly(Supposedly from someone stepping on it in the cleanroom...), GOES-15 (IIRC) had a micrometeorite strike, and GOES-13 had a fuel tank anomaly right before deorbit.
GOES-16 and GOES-17 are on-orbit spares, so in the extremely unlikely event of a total failure there's at least another spacecraft on-orbit ready to take up station.
That said, I have every faith in the GOES team to get to the bottom of this. They're the best, and I often wish I was back there working with them.
"Anomaly" is one of those words that, in the space industry, can mean anything from "Our payload isn't behaving as well as we're expecting" to "Someone stepped on it" to "A rock hit it"
You can use it to describe literally anything that's off nominal. It's fantastic.
“Measurements taking of the person: foot and shoe weight, total weight including clothing, height and leg length, estimated velocity of travel at time, estimated duration of load”
> The main NOAA satellite for tracking Atlantic, Gulf Coast hurricanes is out until further notice
> GOES-19 is the main instrument used to identify tropical waves as they strengthen and move over the Atlantic, Caribbean and Gulf of Mexico, providing real-time tracking for forecasting.
I love how "safe mode" for a satellite is basically: "extend solar panels, turn self towards sun, don't do anything unnecessary, wait for further instructions".
They should rebrand it as "Praise the Sun" mode. We are sorry, GOES-19 is temporarily unavailable during a planned solar worship break of indefinite length.
The cool thing about geostationary orbits is that they're far enough out that they get 24/7 sun (Except around the equinoxes). We could easily fit a solar worship break in the schedule in between imaging and momentum dumps.
Sure, GEO sats are almost always in full sunlight. But still, if you loose orientation, you could end up with the solar panel wings pointing in the wrong direction or overheating parts of the satellite not meant to be in full sunlight for long time.
It'd be easy to optimize with some additional religions too, since from that distance pretty much all possible Earthly [0] holy sites are in the same direction.
[0] I just stumbled onto a fun grammar question! Is it "earthly" (non-supernatural) or is it Earthly (proper-adjective related to the planet), or both? I submit that it's possible to have an "earthly Martian holy-site", and therefore only the capitalized version works here. :p
Interestingly, I noticed this in aproximately real time. I had been checking up on the visible-light geocolor composite images every hour or so to look at the massive plume of Canadian wildfire smoke that was turning the skies in the northeast dark orange yesterday.
I haven't interacted with the GOES site or cared too much about the image output until the last 2 days, and the it immediately broke. Somewhat humorous to me.
Looks like they're making progress toward getting things restarted: "Update #2: The GOES-19 Safehold has been resolved and engineers are working to prepare for restart of the onboard instruments. More information on the recovery timeline will be provided when known." [0]
Anyone interested in accessing GOES data at scale will find this interesting - I created a Zarr index over the 7 billion chunks of data in the GOES-16 archive.
Yes! But if you want to share it with anyone else that would be great, since we're advocating for fairly radical changes within a big bureaucracy here, as I'm sure you will appreciate :)
```
Update #3: DCS and SAR have returned to service as of 1630Z. Engineers will now work to restore ABI and expect imaging to resume by 1900Z. Image navigation may be slightly degraded for the first hour after imaging starts. The GOES-19 instruments will be restored in the following order:
ABI
GLM
SUVI
CCOR-1/EXIS/MAG/SEISS
The recovery process to return all GOES-19 instruments to normal operations is projected to take approximately 8 hours.
Update #2: The GOES-19 Safehold has been resolved and engineers are working to prepare for restart of the onboard instruments. More information on the recovery timeline will be provided when known.
```
Looks like its started to publish new imagery again! I'm sure those fellows are relieved to have it working... I can't imagine the stress of trying to debug something that far away and unreachable!
As an aside, I'm always surprised how US Gov websites look like they've been made in Dreamweaver in about 2006. Not even seemingly with a emphasis on usability either.
While it may not be flashy, I personally find the GOES sites extremely useful. Things are often simply placed at obvious and expected URLs, so scraping or monitoring is extremely easy.
I wrote the script that provides the GOES NavSum [1] and it pretty much just builds a standardized text file and drops it in the folder. The neat thing is that this makes it really easy to programmatically scrape and parse the data.
I wrote a personal script at one point that would download the GOES-EAST CONUS image and both EAST and WEST full disk images and composite them into a wallpaper. At one point my server had 500GB of archived GOES imagery. I liked to joke with my former coworkers that I could report image anomalies before they notice because my desktop wallpaper would change every 10 minutes.
Hah originally making an animation was my plan, but as so often happens it fell on the backburner and then I ended up with a massive archive. I just deleted it once I realized that A) Better archives exist elsewhere and B) I wasn't going to do anything with it.
I still have the script somewhere. I should throw an LLM at it and see if I can't sand off a few rough edges.
It would be great if an LLM could be trained to generate interpolated images between the 10m intervals of the full disk geocolor product. The animations would be fantastic.
I've got about 2.5 years worth of that imagery if someone knows a good way to do this on a budget.
I doubt you need an LLM for that, a diffusion model or perhaps even a deep CNN could probably do a passable job. You could train it by taking consecutive triplets of images [A, B, C] and providing [A, C] as inputs with B as the expected output.
The ones that look old are old. The USG has newer design systems that you'll see used on many of the websites that have been redesigned more recently: https://designsystem.digital.gov/
This admin gutted both NOAAs budget and workforce so a website redesign is probably low priority at the moment.
Sites like NASA's APOD have not changed by design. So many third parties have been built up around sites that any change [w|c]ould break so much for no effective gain. Same holds true when people ask why things like NOTAMs and even NOAA's alerts are formatted the way they are.
The link OP submitted appears to be a webpage displaying a screenshot of another web page, and the image aspect ratio has been altered. It's so comically bad it had to be on purpose, or someone is doing their web dev in MS Word.
Edit: I think actually it's a screenshot of a screenshot even, and this appears to be the entire design of spaceweather.gov. What in the holy heck is going on there? This has to be a top 10 worst website designs of all time.
The counter to this is the UK.Gov website. Very simple and basic, modern (maybe) but very accessible. Government websites should put accessibility over everything else.
You can thank AccuWeather for nerfing any funding for site modernization. I'm surprised the tiled radar map hasn't had the Biden performance fixes reverted.
A safehold is like maintenance mode, shutting down all non-essential systems, after it detects something is wrong. Doesn't necessarily mean it is gone for good, but not a good sign.
What is it that the space aliens don't want us to see? The obvious conclusion is that they are hiding their invasion fleet arrivals inside of hurricanes. The proof will be when the system comes back online and only permits us to see ordinary weather.
> Please note: This status information on this website is generally updated on a monthly basis. Recent outages and anomalies on data flow are highlighted at the top of the page.
> Update #2: The GOES-19 Safehold has been resolved and engineers are working to prepare for restart of the onboard instruments. More information on the recovery timeline will be provided when known.
No, you can’t see ships with this satellite. Too small. Besides, Russia and China have way better satellites that they’ll happily share intel from with Cuba.
Former GOES engineer here. At this point I'd almost be surprised if 19 didn't have something go wrong. We had issues on almost every other satellite. GOES-17 had the loop heat pipe anomaly(Supposedly from someone stepping on it in the cleanroom...), GOES-15 (IIRC) had a micrometeorite strike, and GOES-13 had a fuel tank anomaly right before deorbit.
GOES-16 and GOES-17 are on-orbit spares, so in the extremely unlikely event of a total failure there's at least another spacecraft on-orbit ready to take up station.
That said, I have every faith in the GOES team to get to the bottom of this. They're the best, and I often wish I was back there working with them.
From now on, every time I see the word "anomaly" I will assume it is an euphemism for "someone stepped on it".
"Anomaly" is one of those words that, in the space industry, can mean anything from "Our payload isn't behaving as well as we're expecting" to "Someone stepped on it" to "A rock hit it"
You can use it to describe literally anything that's off nominal. It's fantastic.
> Supposedly from someone stepping on it in the cleanroom
I would be too embarrassed to return to work if I did that.
How do they track that? Is there a log book where someone has to write to?
"Observed EMP555 step on loop heat pipe. Conducted visual inspection of the affected area; no damage found, pipe remains nominal."
I'm sure they take detailed close out photos and likely video of the entire handling process.
“Measurements taking of the person: foot and shoe weight, total weight including clothing, height and leg length, estimated velocity of travel at time, estimated duration of load”
> GOES-17 had the loop heat pipe anomaly(Supposedly from someone stepping on it in the cleanroom...)
Was this reported anywhere in the news? Sounds like one of those "not even once" kind of mistakes.
https://www.nola.com/news/hurricane/weather-satellite-goes-1... explains a bit more what this is, and what this means.
> The main NOAA satellite for tracking Atlantic, Gulf Coast hurricanes is out until further notice
> GOES-19 is the main instrument used to identify tropical waves as they strengthen and move over the Atlantic, Caribbean and Gulf of Mexico, providing real-time tracking for forecasting.
I love how "safe mode" for a satellite is basically: "extend solar panels, turn self towards sun, don't do anything unnecessary, wait for further instructions".
My restorative mode is also "turn self towards sun, don't do anything unnecessary".
"Turn your face to the sun and the shadows will fall behind you." Māori Proverb
They should rebrand it as "Praise the Sun" mode. We are sorry, GOES-19 is temporarily unavailable during a planned solar worship break of indefinite length.
The cool thing about geostationary orbits is that they're far enough out that they get 24/7 sun (Except around the equinoxes). We could easily fit a solar worship break in the schedule in between imaging and momentum dumps.
Sure, GEO sats are almost always in full sunlight. But still, if you loose orientation, you could end up with the solar panel wings pointing in the wrong direction or overheating parts of the satellite not meant to be in full sunlight for long time.
> a solar worship break
It'd be easy to optimize with some additional religions too, since from that distance pretty much all possible Earthly [0] holy sites are in the same direction.
[0] I just stumbled onto a fun grammar question! Is it "earthly" (non-supernatural) or is it Earthly (proper-adjective related to the planet), or both? I submit that it's possible to have an "earthly Martian holy-site", and therefore only the capitalized version works here. :p
Earthian, I believe? If you adhere to the same rules as Martian, Venusian, et al. I’d me more inclined to use Earthen, capitalized, though.
Terran
Just don’t point the thing at any maggots.
Initiate Sol Invictus mode
Interestingly, I noticed this in aproximately real time. I had been checking up on the visible-light geocolor composite images every hour or so to look at the massive plume of Canadian wildfire smoke that was turning the skies in the northeast dark orange yesterday.
I haven't interacted with the GOES site or cared too much about the image output until the last 2 days, and the it immediately broke. Somewhat humorous to me.
Is this what you're talking about? https://www.star.nesdis.noaa.gov/goes/conus_band.php?sat=G16...
Looks like they're making progress toward getting things restarted: "Update #2: The GOES-19 Safehold has been resolved and engineers are working to prepare for restart of the onboard instruments. More information on the recovery timeline will be provided when known." [0]
[0] https://www.ospo.noaa.gov/data/messages/2026/07/MSG_20260716...
Anyone interested in accessing GOES data at scale will find this interesting - I created a Zarr index over the 7 billion chunks of data in the GOES-16 archive.
https://www.earthmover.io/blog/virtual-zarr
I used to work for NOAA. I think the scientists would flip out over this. Did you share it?
Yes! But if you want to share it with anyone else that would be great, since we're advocating for fairly radical changes within a big bureaucracy here, as I'm sure you will appreciate :)
From the latest update (https://www.ospo.noaa.gov/operations/goes/status.html#datafi...), it looks like they're already restoring systems.
``` Update #3: DCS and SAR have returned to service as of 1630Z. Engineers will now work to restore ABI and expect imaging to resume by 1900Z. Image navigation may be slightly degraded for the first hour after imaging starts. The GOES-19 instruments will be restored in the following order:
The recovery process to return all GOES-19 instruments to normal operations is projected to take approximately 8 hours.Update #2: The GOES-19 Safehold has been resolved and engineers are working to prepare for restart of the onboard instruments. More information on the recovery timeline will be provided when known. ```
Looks like its started to publish new imagery again! I'm sure those fellows are relieved to have it working... I can't imagine the stress of trying to debug something that far away and unreachable!
https://www.star.nesdis.noaa.gov/GOES/sector_band.php?sat=G1...
As an aside, I'm always surprised how US Gov websites look like they've been made in Dreamweaver in about 2006. Not even seemingly with a emphasis on usability either.
While it may not be flashy, I personally find the GOES sites extremely useful. Things are often simply placed at obvious and expected URLs, so scraping or monitoring is extremely easy.
I wrote the script that provides the GOES NavSum [1] and it pretty much just builds a standardized text file and drops it in the folder. The neat thing is that this makes it really easy to programmatically scrape and parse the data.
I wrote a personal script at one point that would download the GOES-EAST CONUS image and both EAST and WEST full disk images and composite them into a wallpaper. At one point my server had 500GB of archived GOES imagery. I liked to joke with my former coworkers that I could report image anomalies before they notice because my desktop wallpaper would change every 10 minutes.
[1] https://www.ospo.noaa.gov/resources/cemscs/navsum.txt
Hey, I have a script for updating my background too! I'm not archiving the old images though, but I've thought about it to make some cool animations
Hah originally making an animation was my plan, but as so often happens it fell on the backburner and then I ended up with a massive archive. I just deleted it once I realized that A) Better archives exist elsewhere and B) I wasn't going to do anything with it.
I still have the script somewhere. I should throw an LLM at it and see if I can't sand off a few rough edges.
It would be great if an LLM could be trained to generate interpolated images between the 10m intervals of the full disk geocolor product. The animations would be fantastic.
I've got about 2.5 years worth of that imagery if someone knows a good way to do this on a budget.
I doubt you need an LLM for that, a diffusion model or perhaps even a deep CNN could probably do a passable job. You could train it by taking consecutive triplets of images [A, B, C] and providing [A, C] as inputs with B as the expected output.
There's an app that updates the desktop https://downlinkapp.com/
make torrent of it
Maybe if the UX was nicer, you wouldn't need to write scrapers and parsers and could just use their site.
We don't need a bloated React framework to show a plaintext file with the fuel tank levels. It's NOAA, not Microsoft.
> We don't need a bloated React framework
Could have stopped there for 99% of websites
They're scraping to automatically update the wallpaper on their desktop. That's not something a website can do, even with fantastic UX.
"That's not something a website can do, even with fantastic UX."
Actually, I used to have a live-updating website AS MY BACKGROUND. Windows 98 and Me, website used AJAX and Comet to make it happen.
You used to be able to set websites as your desktop background.
I'm sure that was 100% secure, too. /s (I did the same thing back then.)
The ones that look old are old. The USG has newer design systems that you'll see used on many of the websites that have been redesigned more recently: https://designsystem.digital.gov/
This admin gutted both NOAAs budget and workforce so a website redesign is probably low priority at the moment.
Sites like NASA's APOD have not changed by design. So many third parties have been built up around sites that any change [w|c]ould break so much for no effective gain. Same holds true when people ask why things like NOTAMs and even NOAA's alerts are formatted the way they are.
The link OP submitted appears to be a webpage displaying a screenshot of another web page, and the image aspect ratio has been altered. It's so comically bad it had to be on purpose, or someone is doing their web dev in MS Word.
Edit: I think actually it's a screenshot of a screenshot even, and this appears to be the entire design of spaceweather.gov. What in the holy heck is going on there? This has to be a top 10 worst website designs of all time.
Lots of the web still looks like this when you step outside the comfort zone of big tech search engines, content streaming sites, and social media.
Why must everything look and be modern
The counter to this is the UK.Gov website. Very simple and basic, modern (maybe) but very accessible. Government websites should put accessibility over everything else.
They even have their own design system and CSS framework: https://design-system.service.gov.uk/get-started/
You can thank AccuWeather for nerfing any funding for site modernization. I'm surprised the tiled radar map hasn't had the Biden performance fixes reverted.
A safehold is like maintenance mode, shutting down all non-essential systems, after it detects something is wrong. Doesn't necessarily mean it is gone for good, but not a good sign.
What is it that the space aliens don't want us to see? The obvious conclusion is that they are hiding their invasion fleet arrivals inside of hurricanes. The proof will be when the system comes back online and only permits us to see ordinary weather.
They're concerned because GOES-19 deployed with the newest Rothschild Space Weather Laser, which could disrupt their atmospheric camouflage
What would a space hurricane look like?
https://memory-alpha.fandom.com/wiki/Lightning_storm
We really need to finish the Gollop Chamber.
Very unfortunate timing given the ongoing wildfires and associated smoke spreading across eastern North America in recent days.
Better link: https://www.ospo.noaa.gov/operations/goes/status.html
https://www.ospo.noaa.gov/#datafile-3
I disagree. That just shows GOES-19 as "green", whatever that means. The OP link is also not very informative, but this link is even less so.
The outage list at the top is up to date, but the main status page is nearly three months old - the last updated date at the end is April 20, 2026.
> Please note: This status information on this website is generally updated on a monthly basis. Recent outages and anomalies on data flow are highlighted at the top of the page.
status page says
> Update #2: The GOES-19 Safehold has been resolved and engineers are working to prepare for restart of the onboard instruments. More information on the recovery timeline will be provided when known.* https://www.ospo.noaa.gov/operations/goes/status.html
* https://www.ospo.noaa.gov/operations/goes/status.html#datafi...
Invasion of Cuba incoming?
No, you can’t see ships with this satellite. Too small. Besides, Russia and China have way better satellites that they’ll happily share intel from with Cuba.
is this old cache?
* https://cdn.star.nesdis.noaa.gov/GOES19/ABI/CONUS/GEOCOLOR/2...
* https://cdn.star.nesdis.noaa.gov/GOES19/ABI/CONUS/GEOCOLOR/G...
Looking at the timestamp, that's from yesterday. Nominal product delivery happens ~every 10 minutes.