How do Hypervelocity Stars End up Breaking The Speed Limit?

An artist's conception of a hypervelocity star that has escaped the Milky Way. Credit: NASA

The Sun is racing through the Galaxy at a speed that is 30 times greater than a space shuttle in orbit (clocking in at 220 km/s with respect to the galactic center). Most stars within the Milky Way travel at a relatively similar speed. But certain stars are definitely breaking the stellar speed limit. About one in a billion stars travel at a speed roughly 3 times greater than our Sun – so fast that they can easily escape the galaxy entirely!

We have discovered dozens of these so-called hypervelocity stars. But how exactly do these stars reach such high speeds? Astronomers from the University of Leicester may have found the answer.

The first clue comes in observing hypervelocity stars, where we can note their speed and direction. From these two measurements, we can trace these stars backward in order to find their origin. Results show that most hypervelocity stars begin moving quickly in the Galactic Center.

We now have a rough idea of where these stars gain their speed, but not how they reach such high velocities. Astronomers think two processes are likely to kick stars to such great speeds. The first process involves an interaction with the supermassive black hole (Sgr A*) at the center of our Galaxy. When a binary star system wanders too close to Sgr A*, one star is likely to be captured, while the other star is likely to be flung away from the black hole at an alarming rate.

The second process involves a supernova explosion in a binary system. Dr. Kastytis Zubovas, lead author on the paper summarized here, told Universe Today, “Supernova explosions in binary systems disrupt those systems and allow the remaining star to fly away, sometimes with enough velocity to escape the Galaxy.”

There is, however, one caveat. Binary stars in the center of our Galaxy will both be orbiting each other and orbiting Sgr A*. They will have two velocities associated with them. “If the velocity of the star around the binary’s center of mass happens to line up closely with the velocity of the center of mass around the supermassive black hole, the combined velocity may be large enough to escape the Galaxy altogether,” explained Zubovas.

In this case, we can’t sit around and wait to observe a supernova explosion breaking up a binary system. We would have to be very lucky to catch that! Instead, astronomers rely on computer modeling to recreate the physics of such an event. They set up multiple calculations in order to determine the statistical probability that the event will occur, and check if the results match observations.

Astronomers from the University of Leicester did just this. Their model includes multiple input parameters, such as the number of binaries, their initial locations, and their orbital parameters. It then calculates when a star might undergo a supernova explosion, and depending on the position of the two stars at that time, the final velocity of the remaining star.

The probability that a supernova disrupts a binary system is greater than 93%. But does the secondary star then escape from the galactic center? Yes, 4 – 25% of the time. Zubovas described, “Even though this is a very rare occurrence, we may expect several tens of such stars to be created over 100 million years.” The final results suggest that this model ejects stars with rates high enough to match the observed number of hypervelocity stars.

Not only do the number of hypervelocity stars match observations but also their distribution throughout space. “Hypervelocity stars produced by our supernova disruption method are not evenly distributed on the sky,” said Dr. Graham Wynn, a co-author on the paper. “They follow a pattern which retains an imprint of the stellar disk they formed in. Observed hypervelocity stars are seen to follow a pattern much like this.”

In the end, the model was very successful at describing the observed properties of hypervelocity stars. Future research will include a more detailed model that will allow astronomers to understand the ultimate fate of hypervelocity stars, the effect that supernova explosions have on their surroundings, and the galactic center itself.

It’s likely that both scenarios – binary systems interacting with the supermassive black hole and one undergoing a supernova explosion – form hypervelocity stars.  Studying both will continue to answer questions about how these speedy stars form.

The results will be published in the Astrophysical Journal (preprint available here)

 

Our Galactic Neighbors Shine In New Ultraviolet Pictures

An ultraviolet view of the Large Magellanic Cloud from Swift's Ultraviolet/Optical Telescope. Almost 1 million ultraviolet sources are visible in the image, which took 5.4 days of cumulative exposure to do. The wavelengths of UV shown in this picture are mostly blocked on Earth's surface. Credit: NASA/Swift/S. Immler (Goddard) and M. Siegel (Penn State)

Earth’s galactic next-door neighbors shine brighter than ever in new pictures taken by an orbiting telescope, focusing on ultraviolet light that is tricky to image from the surface.

The Large Magellanic Cloud (LMC) and Small Magellanic Cloud (SMC) — the two largest major galaxies near our own, the Milky Way — were imaged in 5.4 days and 1.8 days of cumulative exposure time, respectively. These produced two gorgeous, high-resolution photos in a spot of the light spectrum normally invisible to humans.

“Prior to these images, there were relatively few UV observations of these galaxies, and none at high resolution across such wide areas, so this project fills in a major missing piece of the scientific puzzle,” stated Michael Siegel, lead scientist for Swift’s Ultraviolet/Optical Telescope at the Swift Mission Operations Center at Pennsylvania State University.

Science isn’t interested in these pictures — taken in wavelengths ranging from 1,600 to 3,300 angstroms, mostly blocked in Earth’s atmosphere — because of their pretty face, however. Ultraviolet light pictures let the hottest stars and star-forming areas shine out, while in visible light those hotspots are suppressed.

“With these mosaics, we can study how stars are born and evolve across each galaxy in a single view, something that’s very difficult to accomplish for our own galaxy because of our location inside it,” stated Stefan Immler, an associate research scientist at NASA Goddard Space Flight Center and the lead of the SWIFT guest investigator program.

The Small Magellanic Cloud as seen by Swift's Ultraviolet/Optical Telescope. This composite of 656 separate pictures has a cumulative exposure time of 1.8 days. Credit: NASA/Swift/S. Immler (Goddard) and M. Siegel (Penn State)
The Small Magellanic Cloud as seen by Swift’s Ultraviolet/Optical Telescope. This composite of 656 separate pictures has a cumulative exposure time of 1.8 days. Credit: NASA/Swift/S. Immler (Goddard) and M. Siegel (Penn State)

Although the galaxies are relatively small, they easily shine in our night sky because they’re so close to Earth — 163,000 light-years for the LMC, and 200,000 light years for the SMC.

The LMC is only about 1/10 of the Milky Way’s size, with 1% of the Milky Way’s mass. The punier SMC is half of LMC’s size with only two-thirds of that galaxy’s mass.

Immler revealed the large images — 160 megapixels for the LMC, and 57 megapixels for the SMC — at the American Astronomical Society meeting in Indianapolis on Monday (June 3.)

Source: NASA

NASA Experiment Seeks Signatures of Formation of First Stars and Galaxies

NASA’s CIBER experiment seeks clues to the formation of the first stars and galaxies. It will study the total sky brightness, to probe the component from first stars and galaxies using spectral signatures, and searches for the distinctive spatial pattern seen in this image, produced by large-scale structures from dark matter. This shows a numerical simulation of the density of matter when the universe was one billion years old. Galaxies formation follows the gravitational wells produced by dark matter, where hydrogen gas coalesces, and the first stars ignite. Credit: Volker Springel/Virgo Consortium.

When did the first stars and galaxies form in the universe and how brightly did they burn?

Scientists are looking for tell-tale signs of galaxy formation with an experimental payload called CIBER.

NASA will briefly turn night into day near midnight along the mid-Atlantic coastline on June 4 – seeking answers to illuminate researchers theories about the beginnings of our Universe with the launch of the Cosmic Infrared Background ExpeRiment (CIBER) from NASA’s launch range at the Wallops Flight Facility along Virginia’s eastern shoreline. See viewing map below.

CIBER will blast off atop a powerful four stage Black Brant XII suborbital rocket at 11 PM EDT Tuesday night, June 4. The launch window extends until 11:59 PM EDT.

Currently the weather forecast is excellent.

The public is invited to observe the launch from an excellent viewing site at the NASA Visitor Center at Wallops which will open at 9:30 PM on launch day.

The night launch will be visible to spectators along a long swath of the US East coast from New Jersey to North Carolina; if the skies are clear as CIBER ascends to space to an altitude of over 350 miles and arcs over on a southeasterly trajectory.

Backup launch days are available from June 5 through 10.

Launch visibility map for the CIBER payload launch from NASA Wallops, Va, on June 4, 2013 at 11 PM EDT. Credit: NASA
Launch visibility map for the CIBER payload launch from NASA Wallops, Va, on June 4, 2013 at 11 PM EDT. Credit: NASA

“The objectives of the experiment are of fundamental importance for astrophysics: to probe the process of first galaxy formation. The measurement is extremely difficult technically,” said Jamie Bock, CIBER principal investigator from the California Institute of Technology

Over the past several decades more than 20,000 sounding rockets have blasted off from an array of launch pads at Wallops, which is NASA’s lead center for suborbital science.

The Black Brant XII sounding rocket is over 70 feet tall.

The launch pad sits adjacent to the newly constructed Pad 0A of the Virginia Spaceflight Authority from which the Orbital Sciences Antares rocket blasted off on its maiden flight on April 21, 2013.

“The first massive stars to form in the universe produced copious ultraviolet light that ionized gas from neutral hydrogen. CIBER observes in the near infrared, as the expansion of the universe stretched the original short ultraviolet wavelengths to long near-infrared wavelengths today.”

“CIBER investigates two telltale signatures of first star formation — the total brightness of the sky after subtracting all foregrounds, and a distinctive pattern of spatial variations,” according to Bock.

Preparing the CIBER instrument for flight. The optics and detectors are cooled by liquid nitrogen to -19C (77 K, -312F) during the flight to eliminate infrared emission from the instrument and to achieve the best detector sensitivity. Photo: NASA/Berit Bland
Preparing the CIBER instrument for flight. The optics and detectors are cooled by liquid nitrogen to -19C (77 K, -312F) during the flight to eliminate infrared emission from the instrument and to achieve the best detector sensitivity. Photo: NASA/Berit Bland

This will be the fourth launch of CIBER since 2009 but the first from Wallops. The three prior launches were all from the White Sands Missile Range, N.M. and in each case the payload was recovered and refurbished for reflight.

However the June 4 launch will also be the last hurrah for CIBER.

The scientists are using a more powerful Black Brant rocket to loft the payload far higher than ever before so that it can make measurements for more than twice as long as ever before.

The consequence of flying higher is that CIBER will splashdown in the Atlantic Ocean, about 400 miles off the Virgina shore and will not be recovered.

You can watch the launch live on NASA Ustream beginning at 10 p.m. on launch day at: http://www.ustream.com/channel/nasa-wallops

I will be onsite at Wallops for Universe Today.

And don’t forget to “Send Your Name to Mars” aboard NASA’s MAVEN orbiter- details here. Deadline: July 1, 2013

Ken Kremer

…………….
Learn more about Conjunctions, Mars, Curiosity, Opportunity, MAVEN, LADEE and NASA missions at Ken’s upcoming lecture presentations

June 4: “Send your Name to Mars on MAVEN” and “CIBER Astro Sat, LADEE Lunar & Antares Rocket Launches from Virginia”; Rodeway Inn, Chincoteague, VA, 8:30 PM

June 11: “Send your Name to Mars on MAVEN” and “LADEE Lunar & Antares Rocket Launches from Virginia”; NJ State Museum Planetarium and Amateur Astronomers Association of Princeton (AAAP), Trenton, NJ, 730 PM.

June 12: “Send your Name to Mars on MAVEN” and “LADEE Lunar & Antares Rocket Launches from Virginia”; Franklin Institute and Rittenhouse Astronomical Society, Philadelphia, PA, 8 PM.

NASA’s CIBER payload will launch from a suborbital launch pad located directly behind this Antares rocket erected at Pad 0A at the NASA Wallops Flight Facility along the Eastern shore of Virginia. Only a few hundred feet of beach sand and a miniscule sea wall separate the Wallops Island pads from the Atlantic Ocean waves and Mother Nature. Credit: Ken Kremer (kenkremer.com)
NASA’s CIBER payload will launch from a suborbital launch pad located directly behind this Antares rocket erected at Pad 0A at the NASA Wallops Flight Facility along the Eastern shore of Virginia. Only a few hundred feet of beach sand and a miniscule sea wall separate the Wallops Island pads from the Atlantic Ocean waves and Mother Nature.
Credit: Ken Kremer (kenkremer.com)

Crashed Asteroid Has a Tail that Keeps Getting Longer

The dust tail on Asteroid P/2010 A2 (LINEAR) has grown to over 1 million kilometers long. Image taken with the new One Degree Imager (ODI), a wide field optical camera at the WIYN telescope on Kitt Peak.

A strange comet-like object discovered in 2010 ended up being an asteroid that had been the victim of a head-on collision from another space rock. The object created a bit of buzz because of its mysterious X-shaped debris pattern and long, trailing streamers of dust. Named P/2010 A2 (LINEAR), the object is located in the asteroid belt between Mars and Jupiter, and has been the focus of much study, including images taken by the Hubble Space Telescope and many ground-based observatories. But over time, the asteroid’s long dust tail has grown to be so long that the entire object can’t fit into the field of view of most observatories.

“Here, we are watching the death of an asteroid,” said Jayadev Rajagopal, a scientist at the WIYN (Wisconsin Indiana at Yale NOAO) Telescope, speaking today at the American Astronomical Society meeting in Indianapolis, Indiana. “We know of dozens of asteroids this has happened to in the past, but this is the only one showing us the event as it is happening.”

A graphic showing the orbit of Asteroid P/2010 A2. Credit: WIYN telescope.
A graphic showing the orbit of Asteroid P/2010 A2. Credit: WIYN telescope.

Using the new wide-field camera at the WIYN 3.5 meter telescope, Rajagopal and his team have found that the peculiar asteroid P/2010 A2’s tail is much longer than was previously supposed. The tail is about a million kilometers long, roughly three times the distance from the Earth to the Moon. The new One Degree Imager (ODI) can currently image an area of the sky about the size of the full moon: a future upgrade will increase the size of the field to about four times as large.

“Three and a half years after the initial disruption, and almost a full orbit around the Sun, the tail is still visible and growing,” said Rajagopal. “One of the reasons it is so long is that radiation pressure and gravity are stretching out the tail. It will progressively grow and sweep out into the ecliptic.”

He added that imaging the full extent of the tail will help pin down the total mass in the dust tail, as well as helping to determine the size of dust particles.

Hubble Views of Comet-like Asteroid P/2010 A2. Credit: NASA, ESA, and D. Jewitt (UCLA)
Hubble Views of Comet-like Asteroid P/2010 A2. Credit: NASA, ESA, and D. Jewitt (UCLA)

Asteroid collisions are thought to be a commonplace occurrence, and are responsible for kicking up dust in our Solar System and probably other planetary systems, too. Just how much dust is produced, and how frequent the collisions happen is still a hazy topic. But the observations of P/2010 A2 are helping astronomers to better model this phenomenon. By figuring out how much dust is produced by the process of ‘collisional grinding,’ astronomers can better model the dusty debris disks of other planetary systems, as well as our own.

“This object is giving us insight into the interplay between asteroids and debris disks,” Rajagopal said. “How much dust do objects like this contribute to our zodiacal dust disk to keep it replenished? This dust must be constantly replenished because it is constantly being destroyed by radiation. The very unusual tail of this active asteroid will help us pin down the mass of the tail, and in a broader context, help us understand how asteroids brought organics and other materials into the inner planets.”

Rajagopal also said this the tail of Asteroid P/2010 A2 is a meteor stream in the making. “It will eventually sweep into the Earth’s orbit and give us a meteorite stream, sending some meteorites our way, maybe a million years from now.”

More info and images: WIYN

How Many Stars are There in the Universe?

How Many Stars in the Universe?
How Many Stars in the Universe?

When we look at the night sky, filled with stars, it’s hard to resist counting. Just with the unaided eye, in dark skies, you can see a few thousand.

How many stars are there in the entire Universe? Before we get to that massive number, let’s consider what you can count with the tools available to you.

Perfect vision in dark skies allows us to see stars down to about magnitude 6. But to really make an accurate census of the total number of stars, you’d need to travel to both the Northern and Southern Hemispheres, since only part of the sky is visible from each portion of the Earth. Furthermore, you’d need to make your count over several months, since a portion of the sky is obscured by the Sun. If you had perfect eyesight and traveled to completely dark skies in both the Northern and Southern Hemispheres, and there was no Moon, you might be able to get to count up almost 9,000 stars.

With a good pair of binoculars, that number jumps to about 200,000, since you can observe stars down to magnitude 9. A small telescope, capable of resolving magnitude 13 stars will let you count up to 15 million stars. Large observatories could resolve billions of stars.

But how many stars are out there? How many stars are there in the Milky Way?

Milky Way. Image credit: NASA
Milky Way. Image credit: NASA

According to astronomers, our Milky Way is an average-sized barred spiral galaxy measuring up to 120,000 light-years across. Our Sun is located about 27,000 light-years from the galactic core in the Orion arm. Astronomers estimate that the Milky Way contains up to 400 billion stars of various sizes and brightness.

A few are supergiants, like Betelgeuse or Rigel. Many more are average-sized stars like our Sun. The vast majority of stars in the Milky Way are red dwarf stars; dim, low mass, with a fraction of the brightness of our Sun.

As we peer through our telescopes, we can see fuzzy patches in the sky which astronomers now know are other galaxies like our Milky Way. These massive structures can contain more or less stars than our own Milky Way.

Elliptical galaxy ESO 325-G004. ESO
Elliptical galaxy ESO 325-G004. ESO

There are spiral galaxies out there with more than a trillion stars, and giant elliptical galaxies with 100 trillion stars.
And there are tiny dwarf galaxies with a fraction of our number of stars.

So how many galaxies are there?

According to astronomers, there are probably more than 170 billion galaxies in the observable Universe, stretching out into a region of space 13.8 billion light-years away from us in all directions.

And so, if you multiply the number of stars in our galaxy by the number of galaxies in the Universe, you get approximately 1024 stars. That’s a 1 followed by twenty-four zeros.

That’s a septillion stars.

But there could be more than that.

It’s been calculated that the observable Universe is a bubble of space 47 billion years in all directions.

It defines the amount of the Universe that we can see, because that’s how long light has taken to reach us since the Big Bang.

This is a minimum value, the Universe could be much bigger – it’s just that we can’t ever detect those stars because they’re outside the observable Universe. It’s even possible that the Universe is infinite, stretching on forever, with an infinite amount of stars. So add a couple more zeros. Maybe an infinite number of zeroes.

That’s a lot of stars in the Universe.

Additional Resources:
How Many Stars Can you See?
Astronomy Cast: How Big is the Universe?
How Big is Our Observable Universe
Astronomy Cast: The Observable Universe
How Many Galaxies in the Universe?

Our Place in the Galactic Neighborhood Just Got an Upgrade

The sun's newly classified neighborhood -- the Local Arm, as shown in this picture -- is more prominent than previously supposed. Credit: Robert Hurt, IPAC; Bill Saxton, NRAO/AUI/NSF

Some cultures used to say the Earth was the center of the Universe. But in a series of “great demotions,” as astronomer Carl Sagan put it in his book Pale Blue Dot, we found out that we are quite far from the center of anything. The Sun holds the prominent center position in the center of the Solar System, but our star is just average-sized, located in a pedestrian starry suburb — a smaller galactic arm, far from the center of the Milky Way Galaxy.

But perhaps our suburb isn’t as quiet or lowly as we thought. A new model examining the Milky Way’s structure says our “Local Arm” of stars is more prominent than we believed.

“We’ve found there is not a lot of difference between our Local Arm and the other prominent arms of the Milky Way, which is in contrast what astronomers thought before,” said researcher Alberto Sanna, of the Max-Planck Institute for Radio Astronomy, speaking today at the American Astronomical Society’s annual meeting in Indianapolis, Indiana.

Sanna said that one of the main questions in astronomy is how the Milky Way would appear to an observer outside our galaxy.

If you imagine the Milky Way as a rippled cookie, our star is in a neighborhood in between two big ripples (the Sagittarius Arm and the Perseus Arm). Before, we thought the Local Arm (or Orion Arm) was just a small spur between the arms. New research using trigonometric parallax measurements, however, suggests the Local Arm could be a “significant branch” of one of those two arms.

In a few words, our stellar neighborhood is a bigger and brighter one than we thought it was.

Astrophoto: Colorado Milky Way by Michael Underwood
Colorado Milky Way. Credit: Michael Underwood

As part of the BeSSeL Survey (Bar and Spiral Structure Legacy Survey) using the Very Long Baseline Array (VLBA), astronomers are able to make more precise measurements of cosmic distances. The VLBA uses a network of 10 telescopes that work together to figure out how far away stars and other objects are.

It’s hard to figure out the distance from the Earth to other stars. Generally, astronomers use a technique called parallax, which measures how much a star moves when we look at it from the Earth.

VLBA telescope locations, courtesy of NRAO/AUI
VLBA telescope locations, courtesy of NRAO/AUI

When our planet is at opposite sites of its orbit — in spring and fall, for example — the apparent location of stellar objects changes slightly.

The more precisely we can measure this change, the better a sense we have of a star’s distance.

The VLBA undertook a search for spots in our galaxy where water and methanol molecules (also known as masers) enhance radio waves — similar to how lasers strengthen light waves. Masers are like stellar lighthouses for radio telescopes, the National Radio Astronomy Observatory stated.

Trigonometric Parallax method determines distance to star or other object by measuring its slight shift in apparent position as seen from opposite ends of Earth's orbit. CREDIT: Bill Saxton, NRAO/AUI/NSF
Trigonometric Parallax method determines distance to star or other object by measuring its slight shift in apparent position as seen from opposite ends of Earth’s orbit. CREDIT: Bill Saxton, NRAO/AUI/NSF

Between 2008 and 2012, the VLBA tracked the distances to (and movements of) several masers to higher precision than previously, leading to the new findings.

Will the findings help ease our “inferiority complex” after all those great demotions?

“I would say yes, that’s a nice conclusion to say we are more important,” Sanna told Universe Today. “But more importantly, we are now mapping the Milky Way and discovering how the Milky Might appear to an outside observer. We now know the Local Arm arm is something that an observer from afar would definitely notice!”

The results will be published in the Astrophysical Journal, (preprint available here) and were presented today (June 3) at the AAS meeting.

Source: National Radio Astronomy Observatory

Stunning View of Earth and Orbital Sunset from the Space Station

An orbital sunset puts Earth in a unique light, as seen from the International Space Station. Credit: NASA, via astronaut Karen Nyberg.

This is just a gorgeous shot of our home planet from the International Space Station, shared by astronaut Karen Nyberg via Twitter. While many pictures of Earth from space show a bright view of our planet, this view of the world plunging into darkness provides a unique, not-often-seen view. If a picture can be this beautiful, imagine what must look like in person.

Nyberg is sharing her experiences via Twitter and also — I believe she is the first astronaut sharing on Pinterest. She describes herself as “Aspiring quilter, crafter, artist” (perfect for the Pinterest crowd) in addition to being an astronaut by day, and said she hopes to do some crafting in space if she has any spare time. Nyberg has a special board for “Hair in Space” (which includes both bald pates and gravity defying hair,) hoping to inspire the younger generation of women to get interested in space exploration. “When girls see pictures of ponytails, don’t you think it stirs something inside them that says, that could be ME up there!” Nyberg writes.

Weekend Aurora Surprise

The aurora borealis over Crater Lake National Park taken on June 1, 2013. This image also includes the flyover of the International Space Station (ISS). Credit and copyright: Brad Goldpaint.

An unexpected arrival of a surprisingly strong (6 KP) geomagnetic storm from the Sun provided an amazing weekend for astrophotographers. Stargazers from both hemispheres were treated with seeing the aurora. We already posted the images from Mike Hollingshead seeing the aurora and red sprite lightning in Iowa, but here are some more great views, including this gorgeous shot of the aurora over Crater Lake in Oregon, from astrophotographer Brad Goldpaint, with the added intrigue of the International Space Station flying over at 2:35 am, local time. He’s also provided an amazing video, too, below.

“I drove to Crater Lake National Park last night to photograph the Milky Way rising above the rim,” Goldpaint said via email to Universe Today. “I’ve waited months for the roads to open and spring storms to pass, so I could spend a solitude night with the stars. Near 11pm, I was staring upward towards a clear night sky when suddenly, without much warning, an unmistakable faint glow of the aurora borealis began erupting in front of me. I quickly packed up my gear, hiked down to my truck, and sped to a north facing location. With adrenaline pumping, I raced to the edge of the caldera, set up a time-lapse sequence, and watched northern lights dance until sunrise. The moon rose around 2am and blanketed the surrounding landscape with a faint glow, adding depth and texture to the shot.”

A green aurora even colors the waters in this image from May 31, 2013, taken near Leith, Ontario, Canada. Credit and copyright: Adam Wipp.
A green aurora even colors the waters in this image from May 31, 2013, taken near Leith, Ontario, Canada. Credit and copyright: Adam Wipp.
Aurora seen on May 31, 2013 near Leith, Ontario, Canada. Credit and copyright: Adam Wipp.
Aurora seen on May 31, 2013 near Leith, Ontario, Canada. Credit and copyright: Adam Wipp.

This video from Loic Le Guilly shows the aurora australis (southern lights) and the glow of the Milky Way in the skies over Tasmania he saw at Signal Station, near Hobart, Tasmania:

Aurora Australis, Signal Station, Hobart from Loic Le Guilly on Vimeo.

This great aurora video comes from Yuichi Takasaka, taken near Lumby, British Columbia:

Mid-Latitude Auroras in May 31, 2013 from Yuichi Takasaka on Vimeo.

Want to get your astrophoto featured on Universe Today? Join our Flickr group or send us your images by email (this means you’re giving us permission to post them). Please explain what’s in the picture, when you took it, the equipment you used, etc.

LRO Makes a Map of the Moon’s Water

The blue areas show locations on the Moon's south pole where water ice is likely to exist (NASA/GSFC)

The Moon might seem like a poor place to hunt for water, but in fact there’s a decent amount of the stuff dispersed throughout the lunar soil — and even more of it existing as ice deposits in the dark recesses of polar craters. While the LCROSS mission crashed a rocket stage into one of these craters in October 2009 and confirmed evidence of water in the resulting plume of debris, there haven’t been any definitive maps made of water deposits across a large area on the Moon — until now.

Over the course of several years, NASA’s Lunar Reconnaissance Orbiter scanned the Moon’s south pole using its Lunar Exploration Neutron Detector (LEND) to measure how much hydrogen is trapped within the lunar soil. Areas exhibiting suppressed neutron activity — shown above in blue — indicate where hydrogen atoms are concentrated most, strongly suggesting the presence of water molecules… aka H2O.

The incredibly-sensitive LEND instrument measures the flux of neutrons from the Moon, which are produced by the continuous cosmic ray bombardment of the lunar surface. Even a fraction of hydrogen as small as 100 ppm can make a measurable change in neutron distribution from the surface of worlds with negligible atmospheres, and the hydrogen content can be related to the presence of water.

No other neutron instrument with LEND’s imaging capability has ever been flown in space.

Watch the video below for more details as to how LRO and LEND obtained these results:

“While previous lunar missions have observed indications of hydrogen at the Moon’s south pole, the LEND measurements for the first time pinpoint where hydrogen, and thus water, is likely to exist.”

What’s so important about finding water on the Moon? Well besides helping answer the question of where water on Earth and within the inner Solar System originated, it could also be used by future lunar exploration missions to produce fuel for rockets, drinking water, and breathable air. Read more here.

Video credit: NASA Goddard Space Flight Center

Incredible Astrophoto: Rare Red Sprite Lightning Shows Up During Aurora

Red sprite lightning appears in the sky with an aurora, as well as a distant thunderstorm, near Denison, Iowa on May 31, 2013. Credit and copyright: Mike Hollingshead, extremeinstability.com

“Holy crap, this is the rarest scene I’ve ever captured and likely ever will,” said photographer Mike Hollingshead. “I was standing there just watching when bam, big red sprites ‘squirting’ up into the air in the aurora.”

Mike said was hoping to see the aurora the night of May 31, 2013, and felt lucky when he saw a faint yellow glow begin to rise in the skies. At the same time, a thunderstorm could be seen off on the horizon and almost before he could even ponder the possibility of seeing something unusual, sprites started appearing.

This is an extremely rare event to be captured on film; in fact an image appearing just a few days ago on Astronomy Picture of the Day (APOD) on May 22 showed red sprite lighting with an aurora, and the APOD team said the image was a “candidate for the first color image ever recorded of a sprite and aurora together.”

“Sprites were first imaged in 1989 accidentally and first color photograph in 1994,” wrote Mike on his Extreme Instability website. “Recent. But with auroras, evidently it is possible the very first time was a couple freaking weeks before this one of mine. It’s that crazy rare.”

Sprites are huge electrical discharges that occur high above thunderstorm clouds. They are rare, but at least one has been captured on film from the International Space Station. They are triggered by the discharges of positive lightning between an underlying thundercloud and the ground. They often occur in clusters within the altitude range 50–90 km above the Earth’s surface.

You can read all the details on Mike’s website. And Mike also got his wish for seeing great auroras that night:

Aurora captured on May 31, 2013, seen from near Denison, Iowa. Credit and copyright: Mike Hollingshead.
Aurora captured on May 31, 2013, seen from near Denison, Iowa. Credit and copyright: Mike Hollingshead.

Stunning! Thanks to Mike Hollingshead for sharing his amazing photos, and congratulations on capturing such a rare event!

Want to get your astrophoto featured on Universe Today? Join our Flickr group or send us your images by email (this means you’re giving us permission to post them). Please explain what’s in the picture, when you took it, the equipment you used, etc.