Wednesday, April 24, 2013

Happy Birthday, Hubble Space Telescope!

23 years ago today, NASA launched the space shuttle Discovery which carried the Hubble Space Telescope into low-Earth orbit. Since then, despite the infamous initial problems with Hubble's primary mirror, Hubble has been a workhorse of the astronomical community and is probably the most well-known telescope in human history. It's hard to encounter someone who hasn't seen one of the classic images from Hubble at some point in their lives, whether they know it or not. Even with its many scientific successes, I think that Hubble's greatest success has been its unmatched ability to bring the universe to the public through the beautiful, iconic images it has taken over the years. Without any doubt in my mind, Hubble lives up to being one of NASA's Great Observatories and may well be the greatest observatory we've ever had (for now).

In 2001 NASA conducted an online poll to determine what object Hubble should image to celebrate its eleventh year of operation, with the overwhelming winner being the Horsehead Nebula, in the Orion Molecular Cloud Complex. The result was the following image, taken in visible light (with some very near infrared).
Image Credit: Space Telescope Science Institute. http://www.spacetelescope.org/images/heic0105a/
The Horsehead Nebula is a cloud of cool, mostly molecular, gas and dust in which star formation is currently occurring. The gas and dust of the Horsehead Nebula are dark because, unlike stars or the brilliant planetary nebulae that form when stars like the Sun die (see: Ring Nebula), are cold. The average temperature for the cool molecular gas and dust found in star formation regions is between 10 and 50 Kelvin or so. That doesn't mean that all of the gas is this temperature. A large portion of the gas in the Horsehead Nebula is ionized because of young, massive nearby stars whose intense radiation are essentially blowing the nebula apart. One of these stars can just be seen through the clouds of the nebula toward the upper left side of the nebula.

The almost uniformly bright background that can be seen in the image against which the nebula is silhouetted is actually a bright background emission nebula known as IC-434. The Wikipedia page on this object isn't particularly enlightening, except to point out that it was first observed by William Herschel. Unlike the Horsehead Nebula, IC-434 is an emission nebula, which means that it is giving off light because it is being heated by some source. This source could be a particularly hot star or a white dwarf (as in a planetary nebula).

On Friday, April 19, the Space Telescope Science Institute released another anniversary image of the Horsehead Nebula that was taken in purely near-infrared light. Because it was taken at completely different wavelengths of light than the image taken in 2001, you'll see that the image will look very different.
Image Credit: Space Telescope Science Institute. http://www.spacetelescope.org/images/heic1307a/
But why does it look so different? Astronomers use infrared light to see through clouds of gas and dust because its longer wavelength means it doesn't get scattered as easily as visible light. So where you previously saw a dark cloud of gas and dust obscuring everything inside and behind it, now you see way more stars than you did before purely because their light can make it through the gas. (The wider field of vision may also aid the number of stars you can see compared to the previous image).

You've probably also noticed that the background cluster IC-434 no longer provides a luminous background  for the Horsehead Nebula. Instead, you can see through the optically bright gas to see the stars (and even some galaxies!) that were otherwise not visible in or behind IC-434 for the same reason as is stated above.

All of this talk of pretty astronomical pictures is not to diminish Hubble's importance as a scientific instrument, of course! Hubble has been involved in some of the most important scientific work of the past decade. Hubble is responsible for the first precision measurement of the Hubble Constant, the deepest visible light image of the universe (shown below), obtaining spectra of exoplanet atmospheres, detecting exoplanets through both direct and indirect techniques, and some of the best images of our neighbors in the solar system (like the picture of Mars also shown below). Hubble discovered the existence of dark energy, the ubiquity of central supermassive black holes in galaxies, protoplanetary disks, optical counterparts of gamma-ray bursts,  and the many moons of Pluto. And that's probably just a very small taste of what Hubble has accomplished during its time.

The Hubble Ultra Deep Field. To really get an idea of how awesome this picture is, you absolutely must view it in its full resolution, completely zoomed in. Image Credit: Space Telescope Science Institute
Hubble image of Mars. Image Credit: Space Telescope Science Institute
The Hubble Space Telescope is due to be replaced in the not-too-distant future by the James Webb Space Telescope, which is still on schedule to launch in 2018. JWST will have increased capabilities in the infrared part of the spectrum, along with a much larger collecting area than Hubble. Of course, this is very exciting for astronomers everywhere, because the advances we expect to make with JWST are on par with those the astronomical community was able to achieve with Hubble.

The future of the Hubble Space Telescope itself is a bit more sad. Hubble was placed in low Earth orbit so it could easily be serviced and upgraded with shuttle missions (which has been a major part of its longevity). However, because NASA's shuttle program is now defunct, Hubble will not be retrieved at the end of its life, and will most likely be de-orbited to safely burn up in Earth's atmosphere. After all it's accomplished for humanity, I think Hubble deserves better. I, for one, would love to see a one-off mission designed to retrieve Hubble and bring it safely back to Earth. Hubble is part of our history that deserves its place in a museum for future generations. It was the first truly and universally great space telescope, which paved the way for greater still missions like JWST.

But let's save the sadness for another time. For now, as long as it's still chugging away, it deserves to be celebrated for what it has accomplished over the years, and what it will continue to accomplish well into the foreseeable future.

Happy birthday, Hubble Space Telescope!
Image Credit: NASA (except for the birthday hat; that was all me.)

Thursday, April 18, 2013

Kepler Press Conference: 4/18/2013

Today at 2:00 PM, NASA is holding a press conference to announce the latest results from the Kepler Mission. I will be watching the press conference live and updating this post as we go. When the press conference is completed, I will provide a summary of the information.

2:00. Panel members are Paul Hertz, NASA's astrophysics director, Roger Hunter, Kepler project manmager, William Borucki, principal science investegator for Kepler, Thomas Barclay, Kepler scientist, and Lisa Kaltenegger, research group leader at the Max Planck Institute for Astronomy.

2:05. And we're live. Opening remarks from Dr. Simon Worden "This is really cool."

2:11. Roger Hunter: Introduction and motivation of the mission. Outlining basic mission goals and mission findings so far. Announces two planetary systems. One with two habitable zone planets that are larger than Earth. One with one planet solidly in the habitable zone and one on the edge of the habitable zone. Kepler 62e and Kepler 62f are 1.6 and 1.4 times the size of the Earth respectively. Images from the press conference shown below.

(Yeah, clearly I lifted these images directly from the video presentation, as you can tell from the "Press Esc to exit full screen mode" dialog that I didn't wait to clear when screencapping.)

Image courtesy NASA.
Proper version of this image. Credit NASA and Kepler Mission.
William Borucki: "We have not measured the masses of these planets."

Other system announced: Kepler 69. Planet Kepler 69c is on inner edge of habitable zone as shown below, and is 1.7 times the radius of the Earth.
Image credit: NASA and Kepler Mission.
Kepler 62 is smaller than our Sun, and Kepler 69 is roughly the size of our Sun, but a tad smaller.

Image Courtesy NASA
From Lisa Kaltenegger: Kepler 62e needs clouds to be properly habitable. Kepler 62f requires a thicker carbon dioxide atmosphere because it is farther out.

Of course, determining the mass of the planet is one of the critical points remaining. However, these planets are definitely too small to be detected with the radial velocity method, which is how we typically determine the mass of an extrasolar planet. This looks more like a situation in which transit timing variations will be helpful.

Transit timing variations are the small changes that occur in the exact orbital period of a planet based on other planets in the system pulling on one another through their own gravity. While these changes tend to be very small, Kepler is good enough to measure these, and it has apparently become the leading way in which the masses of Kepler planets have been confirmed (according to Dr. Eric Ford).

3:00 update: William Borucki addresses transit timing variations: the planets are not close enough to have significant gravitational influence on one another. Guess TTVs won't work after all. Damn.

Wednesday, April 17, 2013

Neutrinos (Part 2)

If you have not read the first part of this post, I highly recommend you read it first. I'll even provide you a convenient link, because I'm nice like that. Neutrinos (part 1)

We shall pick up the neutrino story from the perspective of understanding the physics of the Sun. In 1957, Margaret and Geoffrey Burbidge, William Fowler, and Fred Hoyle published the groundbreaking paper "Synthesis of the Elements in Stars." Neil DeGrasse Tyson routinely cites this paper as one of the most important works in all of astrophysics. "Synthesis of the Elements in Stars" (often abbreviated B2FH, after the authors' last initials) proposed the existence of the nuclear fusion reactions necessary to create all of the elements we see in the universe apart from those initially created in the Big Bang (which I discuss briefly in my History of the Universe post).

To make a rather long story short, the Sun exists, and continues to exist, because it is fusing hydrogen into helium in its very hot, very dense core. The net reaction (ignoring the intermediate steps) is graphically shown below. This process releases energy and keeps the Sun hot, which is incidentally what prevents the Sun from collapsing on itself from its own weight. There are a lot of nuclear reactions that contribute to the production of helium in the Sun (occupies multiple pages of An Introduction to Modern Astrophysics), the most important point is that a number of these reactions result in beta-plus decays where protons are forced to change into neutrons to keep newly formed nuclei stable. This is the opposite of the reaction that was shown in Part 1, so rather than getting electrons and anti-electron neutrinos, we get positrons and electron neutrinos. (Recall an electron neutrino is just the type of neutrino that is specifically associated with electrons.)
Image slightly modified to include neutrinos. Image credit: Addison Wesley Publishing
If you crunch the numbers, you find that the Sun should be producing something like 1034 neutrinos every second, and that's just the ones that are a high enough energy for us to detect. It's pretty natural that astrophysicists would want to confirm the existence of these neutrinos, because that would tell us whether or not we were right about the physics going on deep inside the Sun. Remember, neutrinos don't like interacting with anything, so they can fly right out of the center of the Sun no problem. Unfortunately, the same generally holds true for our detectors.

Fortunately, we know what reactions need to take place for us to detect a neutrino. Therefore, we can do experiments to measure the probability that a neutrino interacts with your detector before you actually put it to use.

The first dedicated solar neutrino detector was the Raymond Davis and John Bacall's Homestake detector, which came online in 1970. Like most neutrino detectors, the Homestake detector was placed in an old mine in North Dakota. This shielded the detector from cosmic rays entering Earth's atmosphere that can produce neutrinos. The detector itself was quite literally a giant tank of dry cleaning fluid, a molecule made up of 2 carbon and 4 chlorine atoms. The chlorine atoms are the part that matters. When an energetic neutrino interacts with chlorine, the chlorine atom can be changed into a radioactive isotope of argon. Every few weeks, Davis would extract the argon and count how much was created, which would tell him about how many neutrinos had interacted with his detector.

While Davis was in charge of the experimental measurements, Bacall calculated the number of argon atoms that the detector was expected to see. When the data were compared to the calculations, it turns out that Davis' measurements came up with roughly one third of the neutrinos that Bacall predicted.

At first, everyone thought that Bacall's calculations were just wrong, or that he hadn't accounted for something properly. Eventually, more neutrino detectors were built: Kamiokande in 1983 which became Super-Kamiokande in 1996, GALLEX in 1991, SNO in 1999. While I won't discuss the finer points of these detectors, each of them works in a slightly different way, but saw the same result; the Sun was producing only one third of the electron neutrinos it was expected to. The distinction of electron neutrino is important because that is both the type of neutrino the Sun creates and the type that our detectors are primarily sensitive to. In textbooks, this is routinely referred to as the "solar neutrino problem".

Not all neutrino experiments use solar neutrinos though. As I mentioned above, cosmic rays entering Earth's atmosphere can produce neutrinos that can either be annoying background noise or a useful signal! For obvious reasons, we call these "atmospheric neutrinos" and we've been observing these since 1965. Neutrinos are also created in nuclear reactors from beta decays that occur in the reactors. Because these "reactor neutrinos" tend to be pretty high energy, they can be used for neutrino physics experiments. We can also use particle accelerators to produce "neutrino beams". Basically, this involves creating a bunch of particles travelling in the same direction that will eventually break apart and create neutrinos (generally muon neutrinos, actually) that are all travelling in the same direction as one another. Then you stick a detector in their way (or better yet, aim the beam at a detector that already exists) and you have your very own neutrino experiment!

"Well that's cool and all, but what does this have to do with the solar neutrino problem?" you may be asking. This goes back to work done (and promptly ignored) by Italian physicist Bruno Pontecorvo in 1957. Pontecorvo suggested that neutrinos could spontaneously change flavors as they traveled through space. But for this to be the case, neutrinos had to have mass, which, according to the Standard Model of particle physics, they did not. Because the Standard Model had been so bloody successful, there was no reason to believe the theory was wrong.

Until 1998. In 1998, Super-Kamiokande began to see hints of neutrinos changing their flavor in its atmospheric and solar neutrino data. When SNO came online a year later, it also saw evidence that neutrinos were changing flavors. This marks the first, and so far only, time that the Standard Model was shown to be wrong, which required an extension of the theory to allow for neutrinos with mass whose flavors could change more or less spontaneously. Now we even have long baseline (hundreds of miles) neutrino beam experiments such as MINOS and T2K that are designed to detect these flavor changes (and they do). This phenomenon is now referred to as "neutrino oscillations".

This may seem like a strange term for "changing flavors", but it actually makes sense, in an odd sort of way. In particle physics jargon, neutrinos can change their flavor because neutrino mass eigenstates don't line up with neutrino flavor eigenstates. This means that when a neutrino of a certain flavor, let's say an electron neutrino, is created, it is a pure electron neutrino in flavor, but its mass is some combination of electron, muon, and tau neutrino masses. As the neutrino moves through space, as shown in the diagram below for a simple two-flavor case, the mass combination changes a bit (for the life of me, I couldn't tell you why). The change in the mass combination will change the flavor of the neutrino, so rather than being a pure electron neutrino, it has some probability of being a muon or tau neutrino as well.

The oscillation comes in because (in a simple two-flavor picture), as the neutrino travels through space, it will flip back and forth between the two flavor states. Not shown in the diagram, the rate of oscillation depends largely on the energy of the neutrino.
Figure adapted from Scientific American, March 19, 2013 "Neutrino Experiments Light the Way to New Physics"
Oscillations can also occur as neutrinos travel through not-empty space. In fact, they happen much quicker. This is called the Mikheyev-Smirnov-Wolfenstein (yeah, I just wanted to write out those names) or MSW effect. It turns out that the MSW effect is mostly responsible for the oscillation of solar neutrinos rather than their trip through empty space to Earth. This explains the solar neutrino problem because we expect that the neutrinos will become roughly evenly divided amongst all three flavors during their trip from the center of the Sun to our detectors on Earth.

Phew. That was probably the hardest three paragraphs I've ever tried to write. These last paragraphs alone were why this post took so damn long to write. On the bright side, you now all know why neutrinos are so weird and part of why I love them so much. I hope this post was understandable to most of you, and there will be a Part 3 in the works. But before that, I expect you'll be getting some exoplanet news some time tomorrow. Stay tuned!

Sunday, April 7, 2013

Neutrinos (Part 1)

Because I've been pretty busy lately, I haven't really had time to post too much, which also means I missed out on the opportunity to write an April Fools post. As a way of making up for that, I'm going to start my series of articles on neutrinos, which are easily my favorite type of particle in the universe. These guys are so weird that they may as well be nature's April Fools joke on all of us.

Before I dive into this, let me point out that we physicists love our conservation laws. Energy (or mass-energy if you're talking about something relativistic), momentum (normal and angular), charge, lepton number, baryon number (and many more) had better be conserved in nature (I'll explain what leptons and baryons are later on). Violating some kind of conservation law is the easiest way for a physicist to determine that a certain theory or model is wrong.

To start our discussion, let's jump back in time to 1930. Quantum mechanics and relativity were both coming of age, and with them, our understanding of the physics that governs how atoms work. One of the more interesting new results was the understanding of radioactivity as changes that took place within the nucleus of an atom.

There are three different types of radiation that physicists know about: alpha, beta, and gamma (yeah, we're really creative), and each consists of a different sub-atomic particle. An alpha particle is the combination of two protons and two neutrons (like the nucleus of helium-4). Gamma radiation is just very high energy photons, or light particles. Beta radiation comes in two different types: negatively charged and positively charged. Turns out that these are just electrons and their anti-matter counterpart positrons.

Alpha and beta radiation from an atom will change one type of atom into another. Alpha radiation makes a nucleus smaller by two protons and two neutrons, and is commonly observed in heavy, unstable nuclei like uranium. Beta radiation can either change a proton into a neutron (by ejecting a positron) or a neutron into a proton (by ejecting an electron). This also makes nuclei more stable by bringing it closer to a combination of protons and neutrons that makes the nucleus more happy. Gamma radiation just brings nuclei into a lower energy state, which makes it more stable (a lot like taking your energetic kid to the park until he or she tires him/herself out).

While studying beta decay, Austrian physicist Wolfgang Pauli realized that the reaction disobeyed particular conservation laws (energy, momentum, angular momentum). Of course, this is bad, so Pauli predicted the existence of another particle that is involved in beta decays. He knew that it had to be very light (massless), neutrally charged, and weakly interacting (or it would have been detected already). In 1933, Italian physicist Enrico Fermi coined the name "neutrino" for this hypothetical particle, which means "little neutral one". Pauli is later said to have lamented "I have a done a terrible thing. I have postulated a particle that cannot be detected."

Fortunately for us, Pauli was wrong in one major respect: neutrinos can be detected. It's just not easy.

To explain how this was first done, I need to spend some time talking about how beta decays actually work. So to help out, I'll show you a Feynman diagram of a beta decay reaction. A Feynman diagram is a great way to visualize what occurs in particle physics reactions, but they also represent some incredibly complicated mathematics regarding interaction probabilities and such (check out the Wikipedia page if you want to read something that is utterly incomprehensible to non-particle physicists). But for our purposes, Feynman diagrams are almost brilliantly simple graphical representations of these type of reactions.
Feynman diagram of beta decay reaction that changes a neutron into a proton by kicking out an electron showing the quark structure of both the neutron (bottom) and proton (top). Time progresses forward along the y axis.
The figure above shows what we call a beta-minus decay, which just means that the ejected particle (an electron) has a negative charge. The particle in the middle of the reaction is a W-minus boson, which is one of the particles that is responsible for one of our four fundamental forces. Specifically, the W-minus boson is one of three particles that is involved in the "weak force", which you pretty much only see in beta decay type reactions (with some exceptions). In this reaction, the W-minus is not actually an observable particle because it is just a force carrier. What we can observe are the electron and the neutrino. In this case, the neutrino is specifically an anti-neutrino (as indicated by the vertical bar above the particle's symbol).

Why does it have to be an anti-neutrino? This comes down to some of those conservation laws I mentioned earlier. Electrons and neutrinos both belong to a family of particle known as leptons. Because lepton number is one of those things that has to be conserved, if we have 0 leptons at the beginning of the reaction, we had better get 0 net leptons out. When we count leptons (or any type of conserved particle, for that matter), leptons like electrons and neutrinos have a lepton number of +1, and the anti-leptons (positrons, anti-neutrinos) have a lepton number of -1. Therefore, with an electron and an anti-neutrino, we still have a lepton number of 0 afterwards.

We also need to conserve baryon number in particle reactions. Baryons are particles that are made up of three quarks, like neutrons and protons. It's easier to see that baryon number is also conserved in this reaction, because you get one baryon (neutron) in and one baryon (proton) out. We also conserve charge. Neutrons (as indicated by their name) and neutrinos are neutrally charged, whereas electrons and protons have opposite charges (negative and positive, respectively). Because this reaction is allowed to occur by the laws of nature, this makes particle physicists happy.

Interestingly, you can also run this reaction backwards. If an energetic anti-neutrino interacts with a proton, it can cause that proton to change into neutron and kick out a positron in the process. (Reality check: does this reaction conserve all of the quantities I mentioned above?) The positron emitted will quickly encounter an electron (because our universe is chock full of electrons) and annihilate, producing two gamma ray-energy photons.

This was successfully detected by Cowan et al. in 1956, which eventually got them the Nobel Prize in Physics in 1995. Amusingly enough, the discovery of a the muon neutrino won a Nobel Prize before the discovery of neutrinos themselves did (1988 compared to 1995).

Muon neutrinos are a type, or "flavor" (as we like to call different varieties in particle physics for some reason), of neutrino that is specifically associated with muons, which are very similar to electrons, but more massive. Muons are also leptons, but because they're more massive than electrons (about 200 times more massive), they're unstable in nature and tend to decay into electrons. Muons are typically created in high-energy particle collisions like one would create at the Large Hadron Collider or from energetic cosmic rays plowing into Earth's atmosphere. So up until now, we've strictly been discussing electron neutrinos, which is why the neutrino in the diagram above has a subscript e.

Turns out, as shown by Perl et al. in 1975 with data from the Stanford Linear Accelerator Center (SLAC), electrons have yet another lepton cousin, the tau (we don't call these guys tauons because that just sounds dumb). Tau particles are, themselves, about 15 times more massive than muons, and have their own neutrino, the (creatively named) tau neutrino. This discovery shared the 1995 Nobel Prize with the discovery of neutrinos in the first place.

So, now you know the basics of how neutrinos work and what kinds of neutrinos are out there. Next time, I'll be talking about the things that make neutrinos truly weird and unique particles, and how they tie in to astrophysics.

Thursday, April 4, 2013

Gravitational microlensing with...Kepler?

Here's a random cool result from the Kepler mission.

Kepler, which I have mentioned before as a dedicated mission for finding exoplanets, simultaneously measures the brightnesses of 160,000 stars to try to find exoplanets that transit, or pass in front of, their host stars. Because Kepler was designed to detect planets like Earth transiting stars like our Sun (which results in a very small signal), Kepler is made to be excruciatingly sensitive to very small changes in a star's brightness. As such, it has been able to see some very strange things, including planets orbiting binary star systems (Kepler-16, -34, -35, -38, -47) and even one planet orbiting a binary star system that is itself part of a quadruple star system!

Earlier today, in a press release, NASA announced that Kepler had detected the gravitational lensing of a red dwarf star by its binary white dwarf companion. White dwarfs are a type of stellar remnant, consisting of the leftover core of a star that was once of similar size to our Sun or a bit more massive. White dwarfs squeeze an amount of mass near that of the Sun into a sphere that is about the size of Earth, so it's no surprise that they would have strong gravitational fields.

Gravitational lensing is an interesting effect predicted by Albert Einstein's general theory of relativity. In general relativity, massive bodies can actually cause light to bend around them as a result of the mass of the object warping spacetime. The general analogy is a heavy ball placed on a rubber sheet, where the rubber sheet represents spacetime. Of course, this is only a two-dimensional analog, but it works for visualization purposes.

There are also different types of gravitational lensing that depend on the mass of the object warping spacetime and the amount of distortion you get in the images of background objects whose light is getting bent. Strong and weak lensing both occur on the scale of entire clusters of galaxies. Strong lensing produces multiple images of a background object and what have come to be called Einstein rings, where the light from the background object becomes smeared out into a circular pattern around the lensing cluster. Weak lensing is far more common, and causes a small distortion in the appearance of the background object.

Gravitational microlensing works on the scale of individual stars, and can even be used as a way of detecting exoplanets! Microlensing doesn't produce the same spectacular images as strong and weak lensing do. Rather, it re-directs some of the light emitted by the background object that would otherwise completely miss us towards Earth. This causes the background object to appear temporarily brighter when the alignment of the two objects is just right.

We've detected white dwarf/red dwarf binary systems before, so we know pretty well what to expect from the observationally. Kepler, however, is sensitive enough to observe that the light curve (the graph of the system's brightness over time) was a bit wonky from what we expected. Sadly, because the paper hasn't been published yet, I can't show you what this actually looks like, but I will post an update when the data become available. I can however, show you this nifty video representation of what is going on in this system. The main observable distortion occurs when the white dwarf is first moving in front of the star and the edge of the star becomes distorted. Here's a frame from that video that shows the beginning of the transit.
Image credit: NASA, JPL, Kepler Team
This completely interrupted the other post I was working on today, but oh well. Results like this are just some of those cool, totally unexpected things that happen when you're studying astronomy that you have to just sit back and admire for a bit. As we like to say, one man's signal is another man's noise.