NCAS Poster

A Search for Short Period Pulsation in All Hot Subdwarfs 

with NASA's GALEX Mission 

Thomas Boudreaux 1 , Brad N. Barlow 1 , Scott W. Fleming 2 , and Alan Vasquez Soto 1 

1: Department of Physics, High Point University, One University Parkway, High Point, NC 27268 USA 

2: Space Telescope Science Institute, 3700 San Martin Dr., Baltimore, MD 21218, USA 

Introduction 

Data Extraction 

Figure 3 

New Pulsators 

NASA's Galaxy Evolution Explorer (GALEX) provided 

near- and far-UV observations for approximately 77 

percent of the sky over a ten-year period. gPhoton 

allows time-series aperture photometry to be 

extracted easily from the raw GALEX data set. Here we 

use gPhoton to generate light curves for all hot 

subdwarf B (sdB) stars that were observed by GALEX, 

with the intention of identifying short--period, p-mode 

pulsations. 

We find that the spacecraft's short visit durations, 

uneven observing cadence, and dither pattern make 

difficult the detection of hot subdwarf pulsations. 

Nonetheless, we detect UV variations in four 

previously known pulsating targets and report their UV 

pulsation amplitudes and frequencies. Additionally, we 

find that several other sdB targets not previously 

known to vary show promising signals in their 

periodograms. Using optical follow--up photometry 

with the Skynet Robotic Telescope Network, we 

confirm p-mode pulsations in one of these targets, 

LAMOST J082517.99+113106.3, and report it as the 

most recent addition to the sdBV r class of variable 

stars. 

Analysis Methods 

Due to the large number of light curves generated by 

gPhoton, we cannot look at each individual light curve 

by eye for brightness variations. Instead we compute 

the Lomb-Scarlge periodogram (LSP) for each 

individual light curve in order to look for periodic 

signals. Candidate pulsators can be identified by 

comparing the highest peak in each LSP to its 

corresponding mean noise level. Summarizing our 

entire data set, we plot in Figure 2a the maximum 

peak amplitude in each LSP against the mean noise 

level. Additionally, we color each point according to 

the frequency associated with the highest LSP peak. 

Figure 2a reveals a predominance of visits with strong 

signals around 8300μ Hz (~120 s; green points). 

The Python package gPhoton was used for data extraction 

and target filtering. gPhoton leverages the design of 

GALEX, namely the use of a micro-channel plate detector, 

which records individual photon events with a high degree 

of time accuracy, to extract time-series photometry for any 

location observed by the spacecraft over its ten year 

lifespan. In order to generate light curves, gPhoton 

integrates raw photon events over a user-definable time 

and calibrates them using a recreation of the original 

GALEX pipeline (Million et al. 2016) outputting text files 

containing flux and time data. 

Targets that fall outside of our acceptable magnitude range 

(NUV=13-19) are not queried with gPhoton. We also 

discard any targets which have less than 600s of NUV or 

FUV exposure. After the initial 5613 sdBs provided by 

(Geier et al. 2016) were run through these filters, we find 

1881 targets with a sufficient number of GALEX 

observations for attempted photometry extraction. Output 

for each target include raw counts, calibrated flux, effective 

exposure time of each “integration bin”, mean observation 

time (in units of ”GALEX time”) for the integration bins, and 

the position of the target on the detector, as well as 

associated errors. Example gPhoton output for one of our 

targets is shown in Figure 1. 

Figure 1 

Example 

gPhoton 

output for one 

visit of one 

target 

including flux- 

-calibrated 

light curve 

(top), target 

distance from 

detector 

center 

(middle), and 

effective 

exposure time 

(bottom). 

Investigating the light curves exhibiting this signal by eye reveals a correlation between flux and position on the 

detector ("detrad"). The detrad variation and its frequency are consistent with the so--called "petal" dither pattern of 

the GALEX spacecraft. Consequently, we decided to pre-whiten all light curves of this instrumental artifact. The 

newly pre-whitened target light curves are used to generate Figure 2b. 

We investigated the large number of targets remaining with maximum peak frequencies below 1000μ Hz (dark blue 

points) -- a regime where the LSP is dominated by 1/f noise -- and find that many of these visits have a long-term 

variation introduced by a second, lower frequency spacecraft dither pattern. We elected to remove this frequency 

range from the calculation of the highest LSP peak. Figure 2c summarizes our full dataset after removing or 

ignoring instrumental effects and 1/f noise. In essence, Figure 2c provides an ordered list of targets to investigate 

for stellar pulsations, starting with the highest S/N objects. 

HS 0815+4243: 

Amp: 12±4 ppt | Freq: 7880±12μ Hz 

EC 14026-2647: 


Known Pulsators 

We cross checked all known sdBV r with our data 

set. We find that of the thirteen known sdBV r 

stars with sufficient GALEX observations for 

analysis, only four had signals present above the 

noise level. Their NUV light curves and 

corresponding LSPs are shown in Figure 3. We 

use non-linear, least squares fitting of sine waves 

to the data to determine pulsation amplitudes 

and frequencies, which are shown under each 

light curve. 

Figure 2 

HS 2201+2610: 


GALEX J08069+1527: 


Maximum Peak in the Lomb-Scargle Periodigram 

(LSP) vs. Mean Noise in the LSP. (a) - No prewhitening, 

all visits plotted, (b) - pre-whitening, all 

visits plotted, (c) - pre-whitening, visits with a 

maximum peak amplitude lower than 1000μ Hz not 

plotted. Stars mark locations of known pulsators. 

Dashed lines show sigma values for visits. 

Single--visit light 

curves and Lomb- 

Scargle 

periodograms for 

previously--known 

sdBV r stars 

detected in the 

GALEX NUV data 

set. Clockwise from 

top--left, targets 

shown include HS 

0815+4243, HS 

2201+2610, GALEX 

J08069+1527, and 

EC 14026-2647 

21±4 ppt pulsation mode at 6810±56μ 

Hz. This data is from the GALEX data-set 

Challenges 

Due to the low frequency resolution , spacecraft dither patterns, 

easily saturated sensor, small aperture, a host of other spacecraftrelated 

reasons, and the relatively small p-mode pulsation 

amplitudes which hot subdwarfs exhibit the noise properties of the 

data set from GALEX proved to be non-ideal for the study and 

classification of sdBV r s. Note that in Figure 5 we see a noise baseline 

(at bright targets) of ~5ppt, already inside the standard 

pulsation amplitude range 

for sdBV r s. Due to this we 

cannot easily extract most 

sdB pulsations. 

Figure 5 

Noise properties in V-mag 

for all visits in data-set 

Using Figure 2c as a guide we selected 

three targets with promising signals for 

follow-up observation using the PROMPT- 

Chile telescope array (Reichart et al. 

2005).We observe LAMOST 

J082517.99+113106.3, HS 0740+3734, and 

EC 12186-3014. Each of these targets 

show signals greater than 4σ in their LSPs. 

Of these three, LAMOST J082517.9 shows 

a signal in the follow up consistent with 

what we find in GALEX (below), and the 

other two do not. We are continuing to take 

follow up data on LAMOST J082517.9 in 

hopes of confirming it as a sdB pulsator. 

This data is from ground based follow up 

and is consistent with the GALEX data 

References 

- Baran, A. S., Gilker, J. T., Reed, M. D., et al. 2011, MNRAS, 413, 

2838 

- Geier, S., Østensen, R. H., Nemeth, P., et al. 2016, ArXiv e-prints, 

arXiv:1612.02995 

- Kilkenny, D., Koen, C., O’Donoghue, D., & Stobie, R. S. 1997, 

MNRAS, 285, 640 

- Million, C. C., Fleming, S. W., Shiao, B., et al. 2016, gPhoton: 

Time-tagged GALEX photon events analysis tools, Astrophysics 

Source Code Library, , , ascl:1603.004 

- Østensen, R., Solheim, J.-E., Heber, U., et al. 2001, A&A, 368, 175 

- Østensen, R. H., Oreiro, R., Solheim, J.-E., et al. 2010, A&A, 513, 

A6 

- Reichart, D., Nysewander, M., Moran, J., et al. 2005, Nuovo 

Cimento C Geophysics Space Physics C, 28, 767 

Acknowledgments 

This Research has made use of NASA’s Astro-Physics Data 

System (ADS). The Skynet Robotic Telescope Network was used to 

collect data as part of this research. This research was made 

possible by the Department of Physics at High Point University 

(HPU) and the Summer Astronomy Student Program (SASP) at the 

Space Telescope Science Institute (STScI). Thanks to the North 

Carolina Academy of Sciences (NCAS) for hosting this conference.

NCAS Poster

Create successful ePaper yourself

Delete template?

Save as template?