Based on the photo from the NASA website ([1]), they use a Thales Satellite modem ([2]). My best guess it's VesseLINK 700 ([3]) that uses Iridium Certus constellation ([4]) and costs around $8K ([5]).

Key Features:

* Robust, Light-Weight Communications for at Sea Operations

* Certus 700 Services (352 kbps Up/704 kbps Down & 256 kbps Streaming Capable)

*100% Global Satellite Coverage and Low Latency for Critical Data and Voice Communications

1. https://blogs.nasa.gov/earthexpeditions/wp-content/uploads/s...

2. https://www.thalesgroup.com/en/markets/market-specific-solut...

3. https://www.thalesgroup.com/sites/default/files/database/doc...

4. https://en.wikipedia.org/wiki/Iridium_satellite_constellatio...

5. https://seatech.systems/product/thales-vesselink-700-for-iri...

I would bet that modem is using a phased antenna array [1] (I'm guessing this is what "solid state, no moving parts" means on the product page). With the right sensors (gyros, etc), a solid-state system like that should be able to keep a pretty tight lock on the satellite even in the roughest conditions.

1. https://en.wikipedia.org/wiki/Phased_array

Look at the Terminal Equipment tab of this page linked below. It shows the Cobham ( different manufacturer from Thales) antenna for the same satellite service. It looks like it is a set of six or so patch antennas. It's not clear if it is switching between patches or combining the signals to/from the patches. If the latter, it is indeed a phased array. That seems likely because the other manufacturer, Intellian, describes their antenna as a 12-element phased array. I'm guessing the Thales also uses a phased array.

https://www.otesat-maritel.com/article/2128/iridium-certus

It's great tech. Same next generation Iridium network is used for offshore maritime, aviation, land mobile data, etc. Lots of places where a traditional two way VSAT is much too large. Its main market competition is the INMARSAT I-4 and I-5 series satellites and BGAN network.

The main problem with it is the very high dollars per megabyte cost. If you're a billionaire or a nation state with a $30 million Gulfstream jet and an Iridium terminal on it you probably don't care. But it can be cost prohibitive for any appreciable amount of data transfer from remote scientific systems.

Jeez. When I work with Iridium I'm constrained to 300 byte messages (for budgetary reasons).

Yes, me too. But Iridium recently (2017-2018) launched the new constellation that they call Iridium NEXT which supports L-Band and offers significantly more bandwidth. So, it's now possible to get a 1GB plan for ~$1300/month ([2]). This obvisouly means that one needs to use a different modem. RockBLOCK modems are great but limited to the Iridium "Classic" with these tight limits and insane prices (~$1/KB).

1. https://en.wikipedia.org/wiki/Iridium_satellite_constellatio...

2. https://www.satphonestore.com/tech-browsing/satellite-intern...

This is using the next-generation Iridium network which is capable and marketed for offshore maritime, for aviation purposes, land mobile data, land based portable terminals (where people would previously need an INMARSAT BGAN), etc.

It is still very costly on a dollars per MB of data transferred basis.

Per this little video, the Thales Vessel link does some 350kbs uplink and downlink.

https://www.youtube.com/watch?v=OoVbH7gFrVY

Damnit, this is why love HN.

If someone comes along with a plan and budget for 10K drones does the system have spare capacity ?!

at 8000 each for hardware, that's 80 Million, plus the data plan charges for each line.

With that amount of money I'm sure the system would be able to grow to accommodate (including sending up more satellites) if only to make sure the money doesn't go to another competitor.

Just stick a Starlink antenna on it, dummies. /s

Noting the satire marker, I’ll just say for completeness: Starlink consumes 100W continuously to operate its phased array beam and the computation requires to drive it. That’s too much power for a sail drone.

It's really not. Assuming your power source has the required amperage, short, infrequent bursts are much more effective than continuous operation. In embedded systems, it's all about duty cycle, and by that metric, Starlink blows the Thales modem out of the water.

The Thales VesseLink modem they used consumes 65W nominal/120W maximum. It offers a connection speed of a couple hundred kbps, so sending up a video file of a fixed size will require it to be on for quite a while - Assuming 200 kbps average, and a 360 MB video, that's 4 hours of uploading or 260 Watt-hours. Also, it's 12x9x2", and weighs 7.5 lbs; this is a boat not a hobby quadcopter. 260 Watt-hours is a lot; that's like 3 laptop batteries, but that's still smaller than the modem itself.

Starlink does consume 100W, but offers a connection speed of about 200 Mbps. The 360 MB video upload could complete in 14.4 seconds, which consumes 100 W * 14.4 seconds / 3600 seconds/hour = 0.4 Watt-hours. It is significantly larger, and it would probably have a harder time handling rough seas (not to mention saltwater intrusion), but that's a lot less power.

Whichever modem you're using, you'd want to turn it on infrequently.

Edit: The Saildrone product brief is here:

https://assets.website-files.com/5beaf972d32c0c1ce1fa1863/61...

It describes a 23' or 7m boat. The 33'/10m larger version has 300W continuous sensor power/2kW peak available from the solar panels, which appear to be of a comparable size to those on the Saildrone.

Oh, on second thought this boat is a beast. It has a 75 HP diesel engine in it, along with the solar panels. It can surely crank out 100W continuously forever. Sadly, Starlink is not for mobile use.

Starlink will be available for marine uses: https://www.pcmag.com/news/spacex-preps-ruggedized-starlink-...

It will be a life-changing event for maritime robotics, assuming they don't get too greedy.

This particular saildrone seems to be a larger model: 72' long.

I always thought /s was 'sarcastic'. And, oh so very often, very not needed.

Just harvest the hurrican’s windpower /s

Can't we use these giant fans to blow the hurricane away?

/s because I am serious

Simply blow against the wind with equal force and hurricane will become neutralized.

/super cereal

I was just thinking to myself that maybe it's my age or something but the fact that I am sat a thousand miles away watching a video, on my phone, from inside a hurricane that was recorded, edited, and published all in less than a day, is one of those "I'm living in the future, aren't I?" Moments.

One thing I usually highlight about the Moon Landing (1969) is that the delivered not just people on the moon, but it was broadcast live!

We can still appreciate that as a mind-blowing achievement! And it might put a damper on the enthusiasm for a delayed video of a wet drone.. :)

Up until recently there were more people on moon than single handed non stop circumnavigations

I find this hard to believe. There have must've been 1-2 dozen solo circumnavigators in the 70's already.. how many people have been on the moon? quick search Apparently only 12 people have walked on the moon (24 total have "been to the moon")and I'd venture a guess that we hit that amount of solo circumnavigators in the 60's already thanks to the OSTAR.

Watching dashcam video footage of the Chelyabinsk meteorite a few hours after it had first been reported, on my smartphone, sitting in the garage after a grocery run, was it for me.

Watching it live was more impressive than videos, and the BOOM was really shaking.

(I actually live in Chelyabinsk)

I prefer living in my future at a distance ;-)

That must have been absolutely amazing.

Did you see the initial airburst itself? What were your thoughts / how would you describe your reaction as the event unfolded?

I saw part of the burst and to me it was immediately clear what it was. Also, the timing of the boom arrival helped determine the distance with good precision.

Thoughts, 'WOW', 'COOL', 'Did somebody film that??', and, of course, the rest of the workday was not very productive. It was nice to see that so much footage made.

One thing footage doesn't show is, however, the heat: the radiation was intense and open parts of the skin did feel hot, like , REALLLY BURNING HOT. Had it lasted longer, there would be burns on everybody.

I've seen estimates of the Chelyabinsk energy yield at about 500 kT TNT equivalent. How that was distributed as light, shock wave, and thermal energy (latter coupled with light) has been something I'd wondered at, and your comment on the heat is interesting.

I'd think that a larger impactor or one that survived further into Earth's atmosphere (and closer to the surface) might have changed that experience markedly. You're informing my own advice-to-self as to how to respond should I see a very large airburst at some point. "Stay away from glass" was already part of that, as well as "expect the shockwave after about 90 seconds". I think I'll add "avoid direct thermal exposure if it looks to be large" to the list.

If you've not already seen the Sandia Labs modelling based on the 1908 Tunguska event, the shockwave dynamics suggest to me why and how the multiple shockwave arrivals at a given point on the ground occur:

https://newsreleases.sandia.gov/releases/2007/asteroid.html

Particularly this simulation: http://www.sandia.gov/videos2007/2007-6514Pfire.hv1.1.mpg

Nukemap has settings for airburst height (under "advanced options"): https://nuclearsecrecy.com/nukemap/

For a 10mt explosion at 20km height it shows a third degree burn radius of 27km. Chelyabinsk was ~0.5kt at 29km. Larger objects are expected to penetrate further into the atmosphere before exploding: https://en.wikipedia.org/wiki/Meteor_air_burst

I'm not sure how much time you'd have to evaluate size or distance, videos of Chelyabinsk show it pretty bright just a second or two after becoming visible. Length of infrared exposure determines severity of burn, so reacting early is helpful.

"Resist the urge to stare at it through the window"

The one reason Halifax has such a large amount of expertise on eye surgery was people not resisting that urge.

That was 191x. And still translates to experience/quality/clustering of eye surgery there?

I'm unaware of something like that in locations which suffered from large explosions around a similar timeframe.

There's apparently some literature on the topic:

https://www.researchgate.net/publication/5633081_The_Halifax...

https://scholar.google.com/scholar?q=eye%20surgery%20halifax...

From 1917?

"Do not look into laser with remaining eye."

"Do not gaze upon meteorite armageddon through window with remaining face."

If you are serious about that, you'll need to retrain one of your instincts — there is a nearly reflexive response to orient towards a new or bright flash of light. In normal situations, this is highly adaptive. However, in raare events like meteors, nuclear or other really large explosions, it results in literally frying the retinas before the viewer has time to sort out what is happening, nevermind judging whether it "looks to be large".

You'll need to retrain your instincts to instantly close your eyes and flinch away in response to bright light, then judge the "looks to be large (or not)" through your closed eyelids. This should work fine as I remember reading that some observers of the Trinity nuclear test blast saw the bones of their hands through their closed and covered eyes...

Alternatively, make a habit of wearing welding glasses with 100% UV protective glass that will auto-darken to Shade 14++

&yes, those Sandia Labs simulations are really amazing!

There's some trainability.

If you're close enough to a fireball that you're instantly incinerated, you might as well just enjoy the show. There's nothing you can do.

If you're within the zone of survivability, then there's cause to take action, and responses over seconds, minutes, and hours can make a difference. A 20--30 mile airburst gives 2--3 minutes before he

The prompt heat flash lasts several seconds. Ducking and sheltering quickly behind any shading barrier will provide protection. Infrared is no penatrating radiation. And bollides as blackbody emitters release mostly IR and visible light. Short-term flash-blindness, likely, permanent blindness ... probably not?

Blast effects lag blast by seconds to minutes. A 20--30 mi altitude bollide burst (32--48km) gives 2--3 minutes before the blast will hit.

Fragments might be another risk. Again, they'll lag considerably and arrive with fairly low terminal velocity for any likely impactor.

TL;DR: Killing effects cover several modalities and don't arrive instantaneously or simultaneously.

Oh, yes, there absolutely is trainablity, you just need to do the training as it's against 'natural' orientation instincts.

Totally agree on the survivability. While most ppl just immediately think 'it's a nuke/meteorite, you're just fried', even a quick look will show that the lower effects zone is at least 10x the area of the 'you're fried' zone.

So yes, just instantly closing your eyes & looking away, stepping away from the window, behind a tree or lamppost, etc. can do a lot in the first 5 seconds, then using the next 10sec that get behind something to be on the leeward side of the shockwave, and you'll be way ahead of the situation. Probably best measure is to avoid important cities.

In my case, it was seeing the photos of Comet Shoemaker–Levy 9's [1] impact with Jupiter on the web very soon after the impact date (latency was but a few days if memory serves, date was July 1994).

[1] https://en.wikipedia.org/wiki/Comet_Shoemaker%E2%80%93Levy_9

There were some earlier moments for me as well.

Being impacted by the Morris Worm, and having a (text-based) copy of the Pons-Fleischman paper, both circa 1988, via the uni Unix server, was pretty cool.

But Chelyabinsk was a massively-shared instance, where a random news event in a place that was absolutely not a media centre, was still accessible in very short order with multiple coverages.

Sci-Hub / LibGen give a similar feeling, though in a different sense. Wells's World Brain and Bush's Memex, delivered. Even if the Establishment is being dragged kicking and screaming.

My wife was hiking in the high mountains here in Oregon, above 2000 meters, and received a video call from her mom who was in Florence, Italy.

I am old enough to remember rotary phones and I am not that old.

Sitting in the Intercity Express, not on one of it's maiden (record) runs, but right when they went into public service,

or very shortly before that on a promotion, on the new high speed track between Cologne and Frankfurt.

Gently sloped up- and downhill, sometimes right next and parallel to the Autobahn A3.

At 331 to 333 kph. Not shaking at all, and relatively silent.

Looking out of the window, seeing Porsches, Mercedes-Benz, BMW

trying to overtake slower traffic permanently blinking left, flashing their lights.

Appearing almost stationary.

Thinking: Who needs Transrapid?

For me it was "so I'm living in the future, and look how relatively boring it is". Boring compared to the wild storm chaser fantasies of Bruce Sterling, and fortunately boring.

But well, there are still ten years to go until 2031, let's hope we still won't be enticed to think about an F-6 by then...

The stream had no audio and my brainchip implant couldn't feel me the salty smell of the sea and the blast of wind on my face.

about a year and a half ago i had the exact same thought, only i was watching a live stream on my phone of the iss crew doing an eva to repair a component. i was genuinely awestruck at the fact that i could see humans servicing a space station — live — whilst i was on the toilet.

Even better watch the stream on a phone or tablet while watching the ISS fly overhead. There is something about watching humans working on that bright star flying overhead live on a device in your hand. :)

All that...by a US federal agency. Respect.

Saildrone is a startup; its primary customer is NOAA, the US Federal agency in the question.

From my impressions, NOAA is a very useful agency that delivers on its mission pretty well. But I never interacted with them directly.

I've used NOAA data to investigate how weather effects production and energy consumption in manufacturing environments.

I found the data to be of good quality, free, and a simple interface to interact with.

One day I went to export data from their web portal and it never seemed to be ready. I shot an email off with no expectation of a response, but a little while later I got a nice response from their system administrator that they were doing an upgrade and some jobs got backed up in the queue. My limited experience with them has been all positive.

NOAA is very hit or miss with their data. I spent a lot of time this summer with the NOAA buoy data. So much of it is available and well documented. The historical and live CSV's are useful, but there is also the stuff where a column name doesn't link up to any of the docs and you have to dig through papers and reverse engineer the correct equation.

The biggest downside is the buoy's are rather old so you don't get a lot of data. Nowadays we could design a buoy that streamed back all of its raw data. But the buoys are designed with bandwidth constrained hardware so they do the analysis on the machine and return the summary results infrequently. It really limits what you're able to do with the data. Especially holding back from machine learning capability.

I was using very basic factors - temperature and humidity primarily. I never ran into that issue, but I could certainly see it being a challenge.

Out of curiosity, what have you been using buoy data for?

Their forecasts are used by basically all the news reports in the entire country. They might actually have the most direct effect on your life of any agency out there. When there is a warning or watch it's basically their call.

Something about NOAA brushing shoulders with what could have been existential disaster[1] to the detriment of the public, only to resume their mission of diving head first into natural ones supported by the first-to-market ethos of a modern startup has poetic justice vibes to it. A win for both Saildrone and the general public at large.

[1] https://www.bloomberg.com/news/features/2018-06-14/trump-s-p...

NOAA's most obvious citizen-facing product for me has been the National Hurricane Center.

Right here is a link to the web site: https://www.nhc.noaa.gov/

Their forecasting graphics (probability distributions of tropical cyclone tracks, wind speeds, rainfall, et cetera, all overlaid on maps) are direct and easy-to-read, and do a good job of conveying the uncertainty of the behavior of these storms in a way that's legible to a lay person.

When those tracks are thought to be inaccurate they can easily be edited with a sharpie.

They run most of the weather radar in the US.

I do and it's true!

NOAA has always had it together. I think working for them might be a lot of fun (unless you have to interface with politics, which I think only a few people there have to).

Hijacking: Can someone ELI5 me why two pockets of differing air temperature create such violent weather patterns? It feels like dark magic.