Dear friends,
Our students have made amazing progress since my last newsletter. In just a few weeks, they designed an observing sequence for the Green Bank Telescope, wrote code to compute Fourier transforms, and started to look for signals indicative of a technological civilization in a large data set. Let me expand on each of these steps.
First, the students figured out the observability of sources by an observer located in Green Bank, West Virginia. As you may have noticed, the apparent trajectories that stars trace in the sky are strongly affected by the observer's latitude. If you were standing at the North Pole, Polaris (a.k.a. the North Star) would remain straight overhead (the zenith direction) at all times of the day. If you were located at 45 degrees north latitude, Polaris would be seen 45 degrees above the horizon in the northern direction at all times. Celestial navigation (e.g., with a sextant) is based on this principle. Because of the Earth's spin, stars appear to trace concentric circles around Polaris. Stars located near Polaris never rise or set; they are called circumpolar stars. Stars located further away dip below the horizon at certain times of the day. I asked the students to create a general diagram that shows the apparent motion of stars in the Green Bank sky so that we could figure out when specific stars and planets are observable.
|
|
|
|
Diagram showing the apparent motion of stars in the Green Bank sky. Azimuth is the direction on the horizon expressed in degrees, with cardinal directions given by 0 (north), 90 (east), 180 (south), 270 (west), 360 (north). Elevation is the angle above the horizon. Black lines show the apparent motion of stars. The lines are labeled in increments of 10 degrees in declination, which is one of the celestial coordinates. Thin blue lines show the time before/after the star reaches the highest elevation in the sky, labeled in increments of 1 hour. Stars rise in the east and set in the west, but circumpolar stars remain above the horizon at all times. Credit: UCLA Spring 2017 SETI student Sara Gallagher.
|
|
|
The students selected the planetary systems that they wanted to observe. The Kepler field remains an attractive choice because several Kepler planets are known to orbit in the habitable zone. The students selected ten Kepler planetary systems. They also chose to observe TRAPPIST-1 and LHS 1140, two recently discovered planetary systems with habitable planets. In order to minimize the time spent moving the telescope from one planetary system to the next, students solved the "traveling salesperson" problem (described in the April 19, 2016 newsletter) with a variety of techniques, including the brute force approach that consists of computing all possible combinations and selecting the most efficient one. The students also considered various observing frequencies and settled on a portion of the electromagnetic spectrum between 1.1 and 1.9 GHz. This part of the spectrum includes a protected band near 1420 MHz, where emission from hydrogen in the Milky Way can be detected.
|
|
|
|
An artist's conception of the TRAPPIST-1 planetary system. Three of the planets (e, f, g) are in the habitable zone. Credit: NASA.
|
|
|
|
|
|
Second, prior to our observations, the students wrote programs to compute Fourier transforms, which we use to display the amount of power received at various frequencies (a.k.a. a power spectrum). A wave of satisfaction rippled through the classroom as each student computed and displayed his or her first power spectrum. When the students were comfortable solving the problem in one dimension, I asked them to add a time dimension by computing many consecutive power spectra and arranging them in a time-frequency diagram. These diagrams are quite useful because they show not only the signal power at various frequencies, but also how the signal evolves as a function of time. Some of the students used their program during our observing run to verify that our calibration signal behaved exactly as expected, indicating that our telescope and data-taking configurations were correct.
Third, I challenged the students to use their program to identify the signal from Voyager 1 in a test data set. The Voyager 1 data set provides excellent practice because the signal is recognizable as technological (as opposed to natural), while also having the needle-in-a-haystack character that introduces the students to large data sets.
Our observations took place on Thursday, May 4 from 8 a.m. to 10 a.m. PDT. Although we controlled the telescope and data-taking systems remotely, we were fortunate to have the assistance of astronomer Ryan Lynch, who monitored the systems on-site at Green Bank. We had enough time to perform two 170-second scans of each one of the ten Kepler planets, TRAPPIST-1, and LHS 1140. We are now in the process of downloading several terabytes of data to our storage server at UCLA. The students will start analyzing the data next week. I can't wait to see what they will find.
Warm regards,
Jean-Luc Margot
|
|
|
|
|
