Daily satellite imagery from NASA GIBS

[Patrick Cozzi]

Folks,

NASA has developed a few cool examples integrating their Global Imagery Browse Services (GIBS) and Cesium.  I particularly like this one where you can use the timeline to see daily imagery:

https://earthdata.nasa.gov/labs/gibs/examples/cesium/time.html

Check out all their examples on GitHub:

https://github.com/nasa-gibs/gibs-web-examples

Patrick

--

twitter.com/pjcozzi

[Hyper Sonic]

Awesome! Not only the current cloud cover map, but easily accessed historical cloud cover! I noticed that it also has the odd blank areas that http://openweathermap.org/help/sat.html has. Are those gaps out of the FOV of the satellite cameras that take these pictures by chance?

I suppose a JavaScript/WebGL engine has broader coverage(number of devices) than a Java based engine (such as Nasa's World Wind engine http://worldwind.arc.nasa.gov/java/ ) 

Such as Chrome blocking Java. https://support.google.com/chrome/answer/1247383?hl=en 

[Scott Reynolds]

You are correct about the gaps being caused by the size of the MODIS instrument's field of view.  This snapshot from Systems Tool Kit (STK) byAGI shows the extent after about 8 hours which means worldwide updates 3 times per day.

 

[Patrick Cozzi]

Hyper - some of the NASA GIBS folks are here at FOSS4G NA.  I can introduce you tomorrow if you want.

Patrick

[Hyper Sonic]

You can scroll back in time such as to 2-28-2015 and see the huge dust storm blow off the Sahara into the Atlantic

http://modis.gsfc.nasa.gov/gallery/individual.php?db_date=2015-03-14

This would be useful for tracking volcanic eruptions and hurricanes as well. Setting speed at 21000x or so it a good way see day to day changes quickly.

Are there multiple satellites taking these photos? If just one I'd imagine there would be quite a time difference when the first and last photo are taken. Actually if just one it would have to continuously alter it's orbit otherwise it would just cover the same ground over and over again, so it's probably multiple. It would be nice if there were no gaps with either more paths or a wider FOV.

[Patrick Cozzi]

I think it is just one satellite.  The imagery becomes available within three hours of the observation.

Patrick

[Hyper Sonic]

For some reason I had initially thought that the satellite spun with the Earth in the Earth Fixed reference frame. Then I realized that wouldn't make sense, especially for a polar orbit. Rather the satellite would be in the Celestial reference frame which basically shares the z axis of Earth Fixed reference frame, but doesn't spin with the Earth's rotation. So when the satellite travels from the north pole to the south pole in the Celestial frame, it appears to drift west in the Earth Fixed frame due to the Earth's spin. However, what about when the satellite travels from the south to north pole, the path would then slant in the other direction. Then I realized that when travelling from the south pole to the north pole it would be on the dark side of the Earth not taking photos, while north to south would be on the bright side of the Earth.

I count 15 paths total. I'm thinking that rather than 1 there are 2 satellites. When one is done with the north to south path on the bright side the other starts with the 'next over' path on the bright side. It takes 24 hours to get every part of the Earth on the brightest part of the bright side, and I'm guessing that photos are taken only in this section. 24/15=1.6 hours, so I assume the orbit time is twice this at 3.2 hours if there are in fact 2 satellites, about half the orbit speed of the ISS. If there are 2 pairs of satellites with 2 adjacent photo paths taken at the same time then they'd travel at about the same orbit speed as the ISS.

[Hyper Sonic]

Actually if there were 2 adjacent paths taken at the same time then half of the time photos wouldn't be taken, also they'd orbit at the same speed (as 1 pair) and wouldn't need partners, so scratch out the idea of 2 pairs. The longitude of the bottom of a path roughly matches up with the longitude of the top of the next path so when one is done it's partner immediately begins with the next adjacent path.

[Scott Reynolds]

The image depicts the Terra satellite's MODIS sensor footprints.  I've attached two movies showing about one half orbit of Terra and Aqua (Terra's sister spacecraft).  In 3D, the "yellow" lines are the orbit paths and the "blue" lines are the ground traces.

Scott

[Hyper Sonic]

Thanks for sharing the videos Scott. I'm a bit confused though, while Terra is going north to south on the bright side, Aqua is going north to south on the dark side. I thought Terra would have a partner along the same orbit just 180deg behind it.

Perhaps Aqua is obtaining photography south to north on the bright side, but its north-west photography path isn't used? Only Terra's south-west photography path seems to be used judging by the slant angle of the gaps. If these 2 criss-crossing paths were combined perhaps the gaps would be eliminated, but then the clouds will have moved and it wouldn't quite match. If adjacent criss-crossing occurred at the same time, the time difference wouldn't be as much near where they criss-cross.

Perhaps Aqua could have a orbital inclination that also created north-east (=south-west reversed) photo paths by moving easterly faster than the Earth spins easterly, but it seems that would result in a more horizontal path.

[Hyper Sonic]

Looking up the satellites I've found a wiki page here with Terra satcat # 25994

http://en.wikipedia.org/wiki/Terra_(satellite)

No wiki page for Aqua 27424, but I found this page

http://satellites.findthedata.com/l/208/Aqua

Both are listed as a sun synchronous orbit

http://en.wikipedia.org/wiki/Sun-synchronous_orbit

quote from the wiki page

A Sun-synchronous orbit (sometimes called a heliosynchronous orbit^[1]) is a geocentric orbit which combines altitude andinclination in such a way that an object on that orbit will appear to orbit in the same position, from the perspective of the Sun, during its orbit around the Earth. Or in other words orbit in such a way that it precesses once a year. The surface illumination angle will be nearly the same every time. This consistent lighting is a useful characteristic for satellites that image the Earth's surface in visible or infrared wavelengths (e.g. weather and spy satellites) and for other remote sensing satellites (e.g. those carrying ocean and atmospheric remote sensing instruments that require sunlight).

endquote

It would be interesting if NASA were to show the satellites move as the timeline progresses along with the weather imagery.

[Hyper Sonic]

some other links I ran across

http://en.wikipedia.org/wiki/Moderate-Resolution_Imaging_Spectroradiometer

http://modis.gsfc.nasa.gov/about/specifications.php

http://modis.gsfc.nasa.gov/about/design.php

"A ±55-degree scanning pattern at the EOS orbit of 705 km achieves a 2,330-km swath and provides global coverage every one to two days"

These instruments seem to be running well beyond their 6 year design range at around 15 years now. Both satellites have about a 98 minute orbital period, so 1440/98 = 14.7 orbits per day.

Can track them real time with Cesium here (unfortunately this doesn't indicate which is the bright side of the Earth, the side the sun is shining directly upon)

http://science.nasa.gov/iSat/?group=SMD&satellite=27424 Aqua

http://science.nasa.gov/iSat/?group=SMD&satellite=25994   Terra

[Hyper Sonic]

(sorry for all the post, just thought I'd share just a few more things I've come across)

Orbit circles are shown in the celestial frame on the Nasa site using Cesium. It would be nice if they could also draw orbit lines in Earth fixed frame where they'd be warped due to the Earth spinning.

AGI's Space Book site has more options and shows the bright side

http://apps.agi.com/SatelliteViewer/?Status=Operational

but it lacks orbit lines (celestial frame or Earth fixed frame)

Accelerating time with AGI's site you can see that:

-Aqua (27424) clearly moves NW in the bright side of the Earth frame.

-Terra (25994) clearly moves SW in the bright side of the Earth frame.

Though it would be nice if AGI's site would also allow the camera to be stationary in the Celestial frame and not just within the Earth fixed frame (or is that already an option? perhaps via the console at least)