Wind Map

by taylorbuley • 14 years ago

909 79 14 years ago

blueski• 14 years ago

Beautifully done. Just discovered you can zoom in by double clicking.

Edootjuh• 14 years ago

Very cool. It's interesting to see how elevation affects the wind here (http://www.theodora.com/maps/new9/usa_elevation_map.gif), but I'd wish there was a worldwide version. (I'm aware that you need a data source)

Should you be interested, here's data from the Netherlands: http://www.knmi.nl/samenw/hydra/cgi-bin/register.cgi

lutorm• 14 years ago

Beautiful visualization! It would be interesting to have an underlay showing the topography so one can correlate features in the wind pattern with mountain ranges, etc.

hoprocker2• 14 years ago

+1. Some of the linear disruptions in wind flow suggest the topography, it'd definitely be interesting to correlate changes in elevation w/ the wind patterns.

nostromo• 14 years ago

Great idea. You can clearly see how the wind patterns are distorted around the Sierra Nevadas and Rockies. While in the East wind patterns are much more smooth.

You could also add color to indicate the temperature of the air.

crisnoble3• 14 years ago

If you could get at the GIS data sets you would be golden. Anyone know of an easy way to integrate GIS data into web applications?

lbotos• 14 years ago

Would something like tilemill help? http://www.tilemill.com

gibybo• 14 years ago

I am used to seeing wind markings on aviation maps (e.g. http://www.aviationweather.gov/adds/winds/) but it's pretty cool how much clearer this map is. It makes me wish I could see a much larger portion of the world so I could get a better understanding of the currents that are entering/exiting the US.

th0ma53• 14 years ago

been wanting to do this myself for like forever. it looks like they are using high level summary data from the NWS ... they also have this amazing thing called MADIS which focuses on micro climate research (and more) ... so in theory you could get a high resolution version of this for san francisco or nyc, and that'd be really great and possibly rather useful too. I think the Weather Underground is a subscriber of MADIS, and their "rapid update" feature has some of this, and also I should mention some of this is collected by ham radio people in their spare time too.

crazygringo3• 14 years ago

That's amazing. It's so inspiring to see someone can have such a simple idea, execute, and create something not only incredibly intuitive and useful, but beautiful too.

It makes you wonder what other things are out there waiting to be done, and what you can do.

iamwil3• 14 years ago

I'm not so sure it's exactly simple to implement. A quick glance at the code, it seems like they had to implement vector field support from scratch.

malandrew• 14 years ago

I haven't looked at the code and how they implemented vector field support, but couldn't a library like Sylvester have been used?

http://sylvester.jcoglan.com/

Paper.js looks like it could work as well:

http://paperjs.org/about/

jjcm• 14 years ago

Really a stunning presentation of this data. A huge "whoa." moment when it loaded.

mahyarm• 14 years ago

They should add Canada, Mexico and the ocean if possible to see where these in and out flows are coming from.

th0ma53• 14 years ago

or tie it in with the ocean viz from nasa from the other day! unfortunately most of the data i could find for canada was only in the south, and the mexico data was similarly in only the more densely populated areas.

nextstep2• 14 years ago

This reminds me of the proof that there is always a point on the earth where there is no wind. This is a property of spherical surfaces: http://en.wikipedia.org/wiki/Hairy_ball_theorem

crazygringo3• 14 years ago

And some people think mathematicians have no sense of humor!

cbr• 14 years ago

There is only a windless spot on earth if you take wind to mean air moving horizontal to the earth's surface. Otherwise imagine that we have wind that points north along the ground, with wind pointing up at the north pole and down at the south pole (and higher up wind comes back south).

nextstep2• 14 years ago

That's a good point. I think the surface you are describing (that the wind would sweep out) would be a torus. The 2-torus is "combable."

joejohnson• 14 years ago

This visualization is incredible but it made my browser freeze for almost a minute :(

jordanb• 14 years ago

This is beautiful and a brilliant, intuitive "100,000 feet up" picture of wind movements.

But it's not horribly useful if you are interested in wind conditions at a particular place. Trying to figure out if a spot is experiencing 15mph wind vs 10mph is hard, for instance.

Compare it to these maps: http://passageweather.com/ which are not pretty nor are they intuitive but they are useful for someone trying to use the map to get detailed, localized wind condition information.

It seems like there should be some kind of middle ground between this wind map and maps like passageweater.com, maps that are intuitive and pretty but also provide detailed, useful information.

hoprocker2• 14 years ago

Maybe using colors instead of shades of gray would be clearer?

iamwil3• 14 years ago

actually, if you click on the map, you can zoom in, and you can see whether a particular spot is experiencing fierce winds or not. I haven't figured out how to zoom out yet though.

link2009• 14 years ago

NINJA EDIT: There is an 'unzoom' button on the left!

I can't figure out how to zoom out either because I don't think they've implemented that functionality. You can also zoom in infinitely...which is strange.

WALoeIII• 14 years ago

http://windonthewater.com/ (use http://windonthewater.com/i.htm on your phone) http://windalert.com/ (iOS app for mobile)

NOAA Point Forecasts: ex: http://forecast.weather.gov/MapClick.php?lat=20.91&lon=-...

dfc• 14 years ago

Is a 5 mile an hour difference important?

recursive• 14 years ago

To me? Yes.

lacerus• 14 years ago

Yes, absolutely, for many sports that rely on wind, such as wind- or kite-surfing.

RockyMcNuts• 14 years ago

Also wouldn't mind seeing high pressure/low pressure systems, fronts, so you can see how they generate/impact wind.

eagsalazar• 14 years ago

they just need to add mouseover details. Rough wind speeds are already indicated by line density, using color would just waste that variable so you couldn't use it for things like pressure. That would be a super cool visualization and again details could be provided by mouseover data.

It would also be cool if you could zoom in. There is definitely a lot of local variation in wind speed/direction that you have to filter out when looking at the entire US but would still be very interesting to see.

jmilloy• 14 years ago

You just hover over the spot of interest and then it tells you to one decimal place...

th0ma53• 14 years ago

some previous visual inspiration possibly for project: (based on lidar data) http://www.datatelling.com/2011/07/29/wind-flow/

http://www.shiffman.net/itp/classes/nature/week06_s09/flowfi...

as well as the big nasa ocean viz from the past few days

spot• 14 years ago

predates both of those by quite a bit: http://nixweb.com/dust/

white_raven• 14 years ago

I want this as a screen saver!

DivisibleByZero• 14 years ago

I was thinking a desktop background version.

samstave2• 14 years ago

I can zoom really far in by double-clicking, but I cant zoom back out.

I have to refresh to zoom out.

This is BEAUTIFUL.

It would be cool to see the different speed ranges reflected by [color of my choice] as well!

unreal37• 14 years ago

There's an UNZOOM button on the left.

samstave2• 14 years ago

What do you want? Me to actually READ the websites!?

Uchikoma2• 14 years ago

Brilliant.

On a side node: Anyone else reminded of Deluxe Paint palette cycling animations by this?

ssdsa• 14 years ago

Yes, of course! The master of this technique is here: http://www.effectgames.com/demos/canvascycle/

Uchikoma2• 14 years ago

Thanks!

iamwil3• 14 years ago

It's also neat because you can tell where the mountains are. They're the sections where the wind was pretty slow, and then it speeds up and then slows down again. You can see it over the Rocky Mountains.

krosaen• 14 years ago

very pretty. related: http://windhistory.com

tpurves• 14 years ago

Where is the wind over the water? Who cares about what the is on land? As a sailor, I find this map frustrating and upsetting for showing me beautiful wind everywhere I can't sail a boat.

jeremyarussell• 14 years ago

I got lost for quite awhile looking at all the wind and how it moves around the area I'm pretty sure is my mountainous home. Kudos to the maker(s).

Slie• 14 years ago

I love this i think it is really cool. Where is Alaska and the rest of the united states?

afterburner• 14 years ago

Wind power, baby.

uvdiv• 14 years ago

Not actually a good representation for wind energy potential, because wind power is cubic in wind speed (v^2 for kinetic energy per unit volume, times v^1 for volume/(cross sectional area * unit time)). So energy potential is concentrated in the "long tail", infrequent high winds. (E.g. 30 mph is 8x more powerful than 15 mph, 27x more than 10 mph). The distribution looks much more "uneven" -- both in space (clustered) and time (fluctuating) -- than a linear-in-windspeed map shows.

Off topic: if anyone's looking for a slightly gigantic weather history dataset ("oodles of terabytes"), look at [1]. It has accumulated "a few thousand" weather stations, at 5 minute intervals, going back "a few" years (I forget the numbers).

[1] http://madis.noaa.gov/

peter_l_downs• 14 years ago

Hey, thanks for that weather dataset link. I had tried to find something similar a couple of months ago and completely failed — somehow, I missed this site entirely.

cshimmin• 14 years ago

Interesting point. Of course a wind-powered device cannot extract all of the power of the wind; the power goes more like v^2 times (delta)v, where the delta is basically how effectively the device can slow the wind. Of course, one might expect this term might be proportional to the original windspeed, so it still goes roughly as v^3. For a related analysis, see Betz' Law: https://en.wikipedia.org/wiki/Betz_law

scblock2• 14 years ago

It's easy enough to convert something like that to energy potential, but when determining whether a site has good potential for wind energy development we really do tend to think of the wind in terms of speed (in m/s) and not energy density.

Depending on the capture technology (even specific wind turbine models using a certain technology) and project arrangement the variation even at a single location is very high.

And this is a beautiful visualization of the flow.

doubtfully• 14 years ago

If you are in the business of determining whether a site has good potential for wind energy development, answer me this if you can: what is the maximum realistic potential percentage of the U.S. energy consumption that windpower could satisfy in 10, 20, 30, 40, and 50 years given realistic conditions (funding, available land, etc.)?

pragmatic• 14 years ago

A nice addition might be the current locations of wind farms (if that data is accessible someplace).

toomuchtodo• 14 years ago

Have already been working on an app for this (just for lulz); haven't had time to finish it yet. I pull from public DOE data regarding generation locations.

crisnoble3• 14 years ago

I wish they would make this open source. It is beautiful.

dgudkov• 14 years ago

Looks beautiful but barely usable. Winds have different speed and direction at different altitudes.

kfury• 14 years ago

This. The site should clearly state whether they're showing 1000ft prevailing winds, jetstream-level currents (around 30,000 feet) or more surface-level effects. All the people talking about power generation from this data should be aware that the only wind that matters there is below 1,000 ft (in many cases, far below 1,000 ft.)

crazygringo3• 14 years ago

"Surface wind data from the National Digital Forecast Database."

mbostock• 14 years ago

This appears to be an animated form of line integral convolution. Cabral and Leedom paper's from SIGGRAPH '93, Imaging Vector Fields Using Line Integral Convolution (PDF):

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.115...

datatelling• 14 years ago

This is great, Mike - thanks for sharing the doc.

xbryanx• 14 years ago

Odd that they didn't label any of the cities in the places that seem to have the greatest generation potential.

MyNewAccount• 14 years ago

beautiful, but not sure how useful it is?

dfj225• 14 years ago

This would be even better if it could display change over time. To support, for example, animate the change in wind direction for the past week.

ww520• 14 years ago

Wow that's cool. Where do you get the wind data?

brownbat2• 14 years ago

My favorite wind potential graph, NPR's "Visualizing the U.S. Electric Grid"

http://www.npr.org/templates/story/story.php?storyId=1109973...

Makes offshore look pretty good.

brownbat2• 14 years ago

So... I want this for all of the world, extrapolated over the oceans.

prbuckley• 14 years ago

This would be even cooler if you could display wind speeds at different elevations above the ground. I am guessing that at a certain elevation you have higher winds that kites might be able to capture energy from.

nix• 14 years ago

They are using a bad dataset. Some of the discontinuities you see are administrative boundaries between National Weather Service forecast offices.

http://nixweb.com/dust/ is my version, for the SF Bay Area only. It was 15 years ago so pardon my Java applet.

sidwyn• 14 years ago

Looks a lot like a chunk of hair. Very neat implementation!

voodoochilo• 14 years ago

great stuff, but my cputemp is in the clouds

kevinslin8• 14 years ago

its been said by others but this is an incredible visualization of a massive amount of data in a beautiful and intuitive medium. it'll be interesting to have this for other large datasets eg. real time web usage...

paraschopra• 14 years ago

I want color!

BHSPitMonkey• 14 years ago

Never before has a piece of software made me wish for a hurricane to ravage my country simply out of curiosity.

run4yourlives• 14 years ago

This is beautiful, but I find it lacking because of the focus on the US mainland exclusively. It would be nice to see North America as a whole along with a good part of the Pacific and Atlantic, because these areas have a massive effect on wind patterns.

For example, the wind over the Chicago is travelling "backwards" for no apparent reason. This is most likely due to this being the "backward half" of a cyclonic system centred over Canada.

The other thing that struck me is the effect the Southwest desert has on wind patterns over all... just, wow.

adlep• 14 years ago

This will be a great tool for paragliders. I will post this on paraglidingforums Thanks for the link.

Wind Map

Related Stories