29 ms·
The eye of hurricane Matthew passes directly over a weather buoy
- dandelany 10y agoCall me a cynic, but my first thought on looking at this was "we can't do any better than a datum per hour?!" :) Must be a bandwidth constraint.
- Zikes 10y agoCall me an optimist, but the data may have been binned by hour.
- dandelany 10y agoI don't see why binning would be advantageous here. If it's noisy, apply a moving average. But binning it is just throwing (time) data away. Don't get me wrong, I think it's great we have access to this data at all. Just curious why it's not higher-resolution.
- sathackr 10y agoIt could be power restraints -- the buoy is almost certainly solar powered and may not have enough reserve/design capacity to transmit at high resolutions over multiple days without sunlight. Or it may have just been designed that way -- perhaps the cost of the necessary equipment to transmit/log at higher frequencies didn't justify the benefit. Perhaps once per hour is sufficient for the purposes the buoy was deployed for.
- dandelany 10y agoYeah - power does seem a more likely constraint when you put it that way.
- sathackr 10y agoPhoto of the buoy http://www.ndbc.noaa.gov/images/stations/42089.jpg http://www.ndbc.noaa.gov/images/stations/42089.jpg Looks like at most 50 watt solar panels, probably closer to 25 watts. It probably has 4 of them but only 2 can make any appreciable amount of power at a time. Not much power at all considering the instrumentation and lighting load(I believe most are lit at night for collision avoidance) and that it has to power the buoy and recharge the batteries during the day. I considered why they wouldn't put larger panels/batteries -- cost, which, compared to what I would think is the overall cost, would be minimal. Wind loading is another story -- It looks fairly streamlined. Larger panels would significantly increase the wind load, and they already have a fair bit of trouble keeping them attached in storms(though more likely due to the waves).
- cr1895 10y agoI have experience working with directional waverider buoys, and ultimately what is useful to me is a statistical representation of sea conditions for a given period (typically 3 hours). I could obtain the raw timeseries of accelerations from the buoy but this is really much less interesting than the rolling spectral statistics that are processed by and output by the buoy.
- dandelany 10y agoThen use a 3-hour moving average of the timeseries... This does not imply one data point per 3 hours, the rolling average timeseries interval can be (& generally is) far more granular than the window size of the average.
- cr1895 10y agoI think you're not understanding me, or I'm not understanding you...unless for some specific reason you want to know a wave elevation timeseries (which given the question "how do I describe the current sea conditons" you probably don't; also, noting that significant wave height is not the same as wave elevation), the statistical parameters which are derived from the timeseries data are more useful. These of course can be given on a rolling basis.
- dandelany 10y agoYes, I think I misunderstood you originally - and instead of trying to explain it, I'll just say thanks for the thoughtful follow-up posts :)
- cr1895 10y agoTo comment on the other measurements (temp, pressure, wind spd/dir): these are different from Hs in that Hs is by definition a statistical parameter; it is 4*stdev of the wave elevations or more practically the mean of the highest third of measured wave heights. You could have higher granularity of temperature, pressure, etc. which arguably more precisely tells you about those conditions at a given time, but a more granular time series of Hs does not necessarily define the sea state any better.
- cr1895 10y agoAt least with regard to significant wave height (rather, describing the sea conditions): It's more useful to describe a "sea state" which is made up of infinitely many frequencies, directions, etc. than to have a timeseries of wave elevation at a given point. A snapshot of the overall sea conditions for a given time, if you will. Sea states can be well described by a couple statistical parameters, namely significant wave height (Hs), peak wave period (Tp), and peak wave direction. There are other refining parameters such as the directional spread but Hs, Tp, and direction are quite informative.
- lorenzhs 10y agoHere's a wind speed plot with a data point every ten minutes: http://www.ndbc.noaa.gov/show_plot.php?station=42058&meas=spd&uom=M&time_diff=-5&time_label=EST http://www.ndbc.noaa.gov/show_plot.php?station=42058&meas=sp... I prefer the hourly binned one. It's much cleaner. Just because the chosen visualisation doesn't have the accuracy you like doesn't mean that the data isn't there.
- dandelany 10y agoThe hourly one is cleaner and more readable only because they've insisted on using a symbol for each datum which obscures the line itself... I'd like this one 6 times more than the original, otherwise. Personally, I don't understand why anyone would want to look at timeseries data with anything less than the maximum possible information density & precision (accuracy is not the problem). Maybe if you're looking for cleanliness/aesthetic reasons, but that's not really the point of these charts. Obviously if they are noisy something needs to be done. But binning only really makes sense to me in histograms, not timeseries.
- webkike 10y agoI love this graph
- gilgoomesh 10y agoThat's a beautiful graph but NOAA's insistence on non-metric units frustrates me endlessly. Knots for speed are bad enough but "inches" for pressure is multiple levels of wrong.
- ared38 10y agoWhat's more wrong about inches than knots?
- Terribledactyl 10y agoIf taken literally the dimensionality is odd. Pressure is a force applied to an area (force/2-d), and that's just a length 1-d (as written)
- gizmo686 10y agoEven if not taken literally, the particular unit of "inches" is largly meaningless in this case. All we know (assuming the obvious mechanism of measurement) is that the air pressure in <choice of force units> is linear to the reported value. The conversion factor is dependent on the instrument itself (surface area, internal resistance, etc) although there may be a standard that is known specialists.
- tlb 10y agoInches of mercury -- the pressure exerted by a column of liquid mercury that high.
- labster 10y agoIt is a 1-D measurement. It's the pressure exerted by the atmosphere that would raise a column of mercury that 1-D length up a vacuum at 1g of gravity.
- caf 10y agoIt doesn't matter that you're measuring it in an odd way - pressure still has units of force per unit area. 1 "inch Hg" is 3386 Nm⁻². The fact that you can express length as "number of football fields" doesn't make length dimensionless.
- avs733 10y agoI would be immensely curious to see the raw data given how the point spacing changes (assuming those aren't actual data measurements). The drop and recovery around the eye are so staggeringly smooth.
- Taek 10y agoThey are most likely smooth because there are so few data points. The red line has the peak, one data point halfway down, one at the bottom, one halfway up, and then the next peak. 'halfway' could be anywhere between 20% and 80% and it'd still look like a super smooth line. Air pressure is less bad, but each data point is still moving a huge amount. If there was a lot of variance we'd have difficulties setting it at this resolution.
- sathackr 10y agohttp://www.ndbc.noaa.gov/station_realtime.php?station=42058 http://www.ndbc.noaa.gov/station_realtime.php?station=42058 http://www.ndbc.noaa.gov/data/hourly2/ http://www.ndbc.noaa.gov/data/hourly2/ http://www.ndbc.noaa.gov/data/hourly2/hour_01.cwind http://www.ndbc.noaa.gov/data/hourly2/hour_01.cwind I love the ndbc site, have been using it for 10+ years now
- avs733 10y agoThank you!
- sp527 10y agoIt's biuriful tears up And now back to our regularly scheduled React component.
- isomorphic 10y agoOne of the things I take away from this graph isn't just the peak wind speed, but the amount of time you'd be under hurricane-force winds. A couple minutes of 90mph winds is one thing... hours of 90mph winds is entirely another thing.
- bhrgunatha 10y agoOf course it depends on how fast the system is moving and how much area it covers. I've been kept up all night with the noise caused by the high winds, particularly when it is gusting. The sound of the wind and all the banging and clattering of scrap being tossed around, doors and windows rattling and slamming, car alarms and so on. The sound of the wind when it reaches its peak is equally awe inspiring and terrifying. I've only experienced one eye and it is quite surreal. Screaming winds start to drop and it becomes eerily quiet and calm for (in this case) a couple of hours. Then the chaos and fury start to ramp up again.
- dredmorbius 10y agoIt's not so much the wind as what is in the wind that really bowls you over.
- M_Grey 10y agoThat depends, at some point if the wind is fast enough and sustained for long enough, key parts of a building start to resemble wings.
- ihsw 10y agoThis is it -- much like how getting caught in a tsunami is not very interesting itself, it is the debris (which includes vehicles and other large objects) that will kill you and otherwise cause devastation.
- HeyLaughingBoy 10y agoIf I remember Ron White's line correctly, it's, "it's not that the wind is blowing; it's what the wind is blowing that matters."
- dluan 10y agoIf anyone wants to see where it is, the hurricane's moved further up north away from the buoy's location. https://earth.nullschool.net/#current/wind/surface/level/orthographic=-83.54,18.91,1666/loc=-76.935,14.522 https://earth.nullschool.net/#current/wind/surface/level/ort...
- harlanlewis 10y agoThat's a great visualization. Reminded me of one posted to hacker news a few weeks ago that let's you cycle through visualizations of cloud cover, atmospheric pressure, precipitation etc in addition to wind: https://www.windyty.com/?20.551,-74.487,5 https://www.windyty.com/?20.551,-74.487,5 I really love seeing all these conditions moving in concert at a global scale. As someone who's never experienced a hurricane, it makes the effects feel much less abstract.
- qvorak 10y agoAny idea what he used to make that? Super impressive
- shpx 10y agowindytv is a fork of earth [0], which is open source [1] and made with D3.js [2], I'm guessing windytv is also using d3. [0] https://news.ycombinator.com/item?id=12415777 https://news.ycombinator.com/item?id=12415777 [1] https://github.com/cambecc/earth https://github.com/cambecc/earth [2] https://earth.nullschool.net/about.html https://earth.nullschool.net/about.html
- anc84 10y agoClick the "Earth" link, there are much more options hidden.
- losteverything 10y agoWhat is this site about?
- 10y ago
- sathackr 10y agoWonder why we don't see winds at the reported speeds(130mph+) in the data? Maybe it's a limit of the measuring device?
- cr1895 10y agoIt could be a number of things: - the max gust didn't occur at the buoy; - the reported wind speed is a 10-min/1-hr mean or some other variation that wouldn't show the peak gust; I really doubt that it's a limitation of the sensor.
- VLM 10y agoCheck out http://www.nhc.noaa.gov/aboutwindprofile.shtml http://www.nhc.noaa.gov/aboutwindprofile.shtml In summary the peak is highly variable for every storm but is usually somewhere around 500-1000 feet above ground level. Its not irrelevant that its so high off the ground ... a tree thats uprooted, picked up to 500 feet, accelerated to 150 MPH, and dropped on you might land on you in merely 100 MPH winds, but it'll still be flying at 150 MPH (maybe faster?) Note there are plenty of broadcast transmitter towers where the top will get winds 50 MPH faster than the surface. 500 feet is only maybe 50 stories for a tall building. So there is some impact directly on the largest constructed things. "How come the 150 MPH rated water tower survived but the 160 MPH radio antenna next to it collapsed?" well the water tower was in 130 MPH winds and the antenna was in 180 MPH winds at the same time so ...
- sathackr 10y agoThank you -- I climb towers and skydive occasionally so I'm aware that winds aloft are often greater than surface winds. But I was under the impression that the forecast wind speeds were surface winds, and assumed that surface meant ground level. I shall research.
- matwood 10y agoBeing in the eye of a hurricane is an odd feeling. Growing up I went through Hugo, and when the eye finally passed over, it went from chaos to an eerie dead calm. We went outside and to see the destruction, but knew there was not much we could do yet. A tree had clipped the corner of the house so my parents did what they could do to protect the exposed inside against the rain of back half of the storm. Beyond that, we just prepared hold tight for another few hours.
- chiph 10y agoI lived near Charlotte during Hugo. The thing I remember most is going out afterwards and smelling all the pine sap (smells like the Pine Sol cleanser) from all the downed trees. IIRC, the storm depressed the market for lumber for a year or more afterwards, as it was flooded with wood cut from the trees that were uprooted and then harvested.
- matwood 10y agoI was in Charleston, SC. We literally had firewood for years afterwards. The storm itself was bad, but no power for almost 4 weeks was really rough.
- jpalomaki 10y agoI was surprised some time ago when I found out that the modern wind speed measuring devices work without moving parts. I had expected them to all have these spinning ping pong ball halves. To do it without moving parts, they are for using for example ultrasound. These devices have pairs on transducers, placed 10-20cm apart and the device then measures how long it takes for the sound to pass through that distance. https://en.wikipedia.org/wiki/Anemometer#Ultrasonic_anemometers https://en.wikipedia.org/wiki/Anemometer#Ultrasonic_anemomet...
- swehner 10y agoDoes a speaker count as a moving part?
- HeyLaughingBoy 10y agoWell, if you want to be pedantic about it, hot-wire anemometers have absolutely no moving parts :-) https://en.wikipedia.org/wiki/Anemometer#Hot-wire_anemometers https://en.wikipedia.org/wiki/Anemometer#Hot-wire_anemometer...
- chadclan 10y agoSome other neat data from the same buoy. Wave height http://www.ndbc.noaa.gov/show_plot.php?station=42058&meas=wvht http://www.ndbc.noaa.gov/show_plot.php?station=42058&meas=wv... Wind direction http://www.ndbc.noaa.gov/show_plot.php?station=42058&meas=wdir http://www.ndbc.noaa.gov/show_plot.php?station=42058&meas=wd... Air temperature http://www.ndbc.noaa.gov/show_plot.php?station=42058&meas=atmp http://www.ndbc.noaa.gov/show_plot.php?station=42058&meas=at... Water surface temperature http://www.ndbc.noaa.gov/show_plot.php?station=42058&meas=wtmp http://www.ndbc.noaa.gov/show_plot.php?station=42058&meas=wt... Dewpoint http://www.ndbc.noaa.gov/show_plot.php?station=42058&meas=dewp http://www.ndbc.noaa.gov/show_plot.php?station=42058&meas=de... I just tried the data format codes from the link someone posted earlier http://www.ndbc.noaa.gov/measdes.shtml http://www.ndbc.noaa.gov/measdes.shtml Cool stuff!
- josh_carterPDX 10y agoBeing a former QuarterMaster in the Navy, all of this is so fascinating. 20 years ago this data was not easily accessible to the public so you had to wait for some sort of published journal to come out to see this information. I love that we all have access to it so easily.
- prplhaz4 10y agoAdditionally, it is located in the Central Caribbean - 210 NM SSE of Kingston, Jamaica http://www.ndbc.noaa.gov/station_page.php?station=42058 http://www.ndbc.noaa.gov/station_page.php?station=42058