Archives

All posts for the month July, 2014

The K2 mission will observe sequential ecliptic campaigns with a duration of ∼83 days, where 75 days are dedicated to science.

The K2 mission will observe sequential ecliptic campaigns with a duration of ∼83 days, where 75 days are dedicated to science.

I’m a bit late to the game on this one, but I wanted to read Howell et al.’s (2014) paper describing the planned K2 mission, what is essentially the reincarnation of the Kepler mission.

Launched in 2009, the Kepler spacecraft was happily staring at about 150,000 target stars over about 4 years, looking for the shadows of planets as they passed between the Earth and their host star (called planetary transits). Groups using Kepler data have found thousands of planets outside our solar system, revolutionizing exoplanet studies. Unfortunately, two of the reaction wheels used to keep the spacecraft pointing stably at its target field failed by May 2013, ending the nominal science mission.

However, by carefully angling the telescope (see figure at left) and keeping it pointed along its orbital plane, NASA engineers realized they could use the photon pressure from Sun as a sort of third reaction wheel, allowing the astronomical revolution to continue — thus was born the K2 mission, which NASA selected for funding May this year.

The paper from Howell et al. (2014) describes the scientific and engineering capabilities of the K2 mission, which closely match those of the Kepler mission. So a lot of what Kepler could do, K2 can, too. In fact, since mission engineers have to turn the spacecraft every 80 days to satisfy the pointing requirements, K2 will look at lots of different fields on the sky, in contrast to Kepler, which only stared at the same field.

This variable pointing will enable a wider variety of scientific investigations, as discussed by Howell et al., including looking for more transiting planets, but also studies of other galaxies, supernovae, stellar clusters, and more. In fact, the astronomical community submitted more than 100 different ideas of things to do with K2. So the scientists and engineers at NASA have really done a spectacular job salvaging what otherwise would have been a disappointing loss.

Artist’s conception of a habitable exoplanet in orbit around its red dwarf star. Credit: NASA Ames/SETI Institute/JPL-Caltech. From http://themeridianijournal.com/2014/04/big-discovery-first-earth-sized-exoplanet-habitable-zone-another-star/#more-5509.

Artist’s conception of a habitable exoplanet in orbit around its red dwarf star. Credit: NASA Ames/SETI Institute/JPL-Caltech. From http://themeridianijournal.com/2014/04/big-discovery-first-earth-sized-exoplanet-habitable-zone-another-star/#more-5509.

Interesting paper from Prof. Michael Jura at UCLA and colleagues, in which they look for the chemical signatures of plate tectonics in white dwarfs that are accreting planetary and asteroidal material. The paper presents a really neat idea, combining several big concepts.

First, there is strong evidence that several white dwarfs (the ghostly remnants of Sun-like stars) are accreting rocky materials. The atmospheres of white dwarfs are very simple, hot hydrogen and helium cooling to space over billions of years. Any other, heavier elements quickly settle out of the atmospheres (on timescales of millions of years), and so if you find such, heavier elements in the atmospheres (via spectroscopy), those elements were probably recently dumped into the atmosphere — a process called pollution.

Such pollution has been observed for many white dwarfs, and the pollution typically consists of rocky elements, silicon, magnesium, etc. Consequently, the polluting materials probably come from rocky asteroids, falling into the white dwarfs. Jura and colleagues point out that some of the pollution may arise from the crusts of extrasolar rocky planets orbiting the white dwarfs, in the form of debris from large impacts with the planetary surfaces.

Second, on the Earth, the constant subduction and eruption of crustal materials from plate tectonics has the effect of sieving out certain elements and leaving them in the crust, producing a crustal composition distinct from that of other planets and asteroids.

Jura and colleagues went looking for such a chemical signature in the rocky pollution of white dwarfs but unfortunately don’t find it. However, this initial study may provide a novel to search for the signs of plate tectonics in other planetary systems. That’s important because plate tectonics is thought to be a key requirement for making a planet suitable for life, but observing it astronomically is almost impossible (there’s no strong evidence any solar system planets other than Earth experienced/s plate tectonics). Jura and colleagues may have provided us a new way to peer into the geophysical histories of extrasolar planets.

 

 

As part of my new push into data science, I read Shi & Malik’s paper on the perceptual grouping problem in which they develop a new algorithm for dividing an image up into coherent regions.

A portion of Figure 8 from Shi & Malik (2000).

A portion of Figure 8 from Shi & Malik (2000).

What does it mean to divide an image into coherent regions? Say, for example, you had a satellite image of a large storm system, such as the figure at left from Shi and Malik’s paper. Panel (a) shows the original image, while (b) and (c) show two portions of the image grouped together by Shi and Malik’s algorithm. The result makes intuitive sense (at least, to me): (b) is a large part of the storm, while (c) is the ground underneath.

Basically, Shi and Malik’s algorithm treats the entire image as a weighted graph, with portions of the image treated as nodes. Weights on the edges connecting the nodes are larger for portions that are closer to each other and for portions that are similar (where similarity may depend on the pixel brightnesses, textures, colors, etc.).

The algorithm decides which portions belong together as one segment using the weights for all the nodes connected in that segment. Shi and Malik developed a clever way to turn this process into an eigenvalue/vector problem, thereby dramatically facilitating the calculation. Their technique amounts to treating the pixels as individual masses connected by springs, with spring constants given by the edge weights, and then finding the normal modes of oscillation for the system: pixels that are strongly coupled are grouped together as one segment.

Conveniently, the algorithm is implemented in the scikit-learn python module. Using their example code, I was able to reproduce the segmentation of the Lena image easily (shown below), so I thought to try it on some VIMS observations of Titan. (Here’s the original Titan image. I took a small portion of it.)

Unfortunately, the result was not promising: the algorithm did NOT break up the image into the segments I expected. Instead, the segments seem pretty random. It also took about 40 minutes to work, even though the Titan image is smaller than the Lena image (100 x 100 pixels vs. 128 x 128).

(Left) Result from the spectral clustering example code on the scikit-learn page. (Right) My own attempt at segmenting a VIMS image of Titan.

(Left) Result from the spectral clustering example code on the scikit-learn page. (Right) My own attempt at segmenting a VIMS image of Titan.

Next things to try: it seems the algorithm needs me to tell it how many regions to use — I used the number given in the original example, 11. Maybe I should try a smaller number. There are also a few options as to how the graph weight are calculated in the original Shi and Malik algorithm (not sure if the scikit-learn module has that capability).