Commit c5d1901f authored by Colin Clark's avatar Colin Clark
Browse files

Updated text at top of page

parent d34f0ab2
......@@ -10,15 +10,15 @@
<h2>Einstein@Home Gamma-ray Pulsar Discoveries in Fermi-LAT Data</h2>
<h3>Info</h3>
<p>Einstein@Home searches data from NASA's Fermi Gamma-ray Space Telescope for signals from
<p>Einstein@Home searches data from the Large Area Telescope (LAT) onboard NASA's <i>Fermi Gamma-ray Space Telescope</i> for signals from
gamma-ray pulsars. Pulsars are very compact stars with extreme physical properties compared to normal matter.
They are rapidly spinning neutron stars that emit pulses, observable from radio to
gamma-ray wavelengths.</p>
<p>Searching for new pulsars is an enormous computational challenge, because their spin
frequencies, sky position and other parameters are unknown in advance. Hence this is called a "blind" search, where one explicitly searches over a dense grid in parameter space. However, the
number of discrete grid points to cover such multi-dimensional parameter spaces is
tremendous and renders "brute forces approaches" computationally unfeasible.</p>
frequencies, sky position and other parameters are unknown in advance and must be covered a dense grid of search locations. The
number of grid points required to cover such multi-dimensional parameter spaces is
tremendous and renders "brute forces approaches" computationally unfeasible.</p>
<p>We have developed novel and much more efficient data-analysis methods for the volunteer
supercomputer Einstein@Home, which ranks among the fastest 25 computer systems worldwide.
......@@ -28,13 +28,27 @@ inaccessible on computational grounds.</p>
<p>These gamma-ray pulsar discoveries provide important contributions to advance
our (yet very poor) understanding these stellar objects, their population, and their role
in our Universe.</p>
<br>
<p>As always, we are extremely grateful to all of our volunteers, especially those whose computers
contributed to these new discoveries. To say "we couldn't have done it without you!" would be a terrible
understatement!</p>
<!-- <a target="blank" href="img/skymap.jpg"><img align="left" vspace="20" hspace="20" style="width: 450px;" alt="skymap" src="img/skymap_small.jpg"></a>
!-->
<hr>
<h3>The Einstein@Home surveys for new gamma-ray pulsars</h3>
<hr>
<h4> FGRP1 </h4>
<p>Einstein@Home's searches for gamma-ray pulsars began in 2011, using just over 3 years of <i>Fermi</i>-LAT data. Prior to this point, searches had been run "offline" on academic computing clusters. However, the computational cost of these searches increases rapidly with time, and so these searches quickly became infeasible. The much larger computational power offered by Einstein@Home allowed the search effort continue using longer data sets that provide increased sensitivity, and greater potential for discoveries. Four new pulsars were found in the first Einstein@Home <i>Fermi</i> Gamma-ray Pulsar survey (FGRP1).</p>
<h3>The Einstein@Home searches for new gamma-ray pulsars</h3>
<a href="#FGRP1">Jump to FGRP1 discoveries</a><br><br>
Publications:<br>
<a href="https://doi.org/10.1088/2041-8205/779/1/L11">Einstein@Home discovery of four young gamma-ray pulsars in <i>Fermi</i>-LAT data</a>
<hr>
<h4> FGRP4 </h4>
<p>In 2014, we began the 4<sup>th</sup> Einstein@Home survey for gamma-ray pulsars, or
"FGRP4". This survey incorporated many new advances that we learned during our previous
......@@ -42,19 +56,43 @@ searches and investigation of blind search methods. In addition, we were able to
data from the <i>Fermi</i>-LAT team, and search in longer data sets than ever before. In
combination, these improvements led to FGRP4 being our most sensitive survey to date.</p>
<p>With the publication in <a href="http://advances.sciencemag.org/content/4/2/eaao7228">Science Advances</a> of the discoveries of two millisecond pulsars (J1035-6720 and J1744-7619), we have now published all pulsar discoveries made during FGRP4. These amount to 23 new pulsars, almost a third of all gamma-ray pulsars to be discovered in blind searches.</p>
<p>With the publication in <a href="http://advances.sciencemag.org/content/4/2/eaao7228">Science Advances</a> of the discoveries of two millisecond pulsars (J1035-6720 and J1744-7619), we have now published all pulsar discoveries made during FGRP4. These amount to 23 new pulsars, almost a third of all gamma-ray pulsars that had been discovered in blind searches at the time.</p>
<p>Of course, this does not mean that our gamma-ray pulsation searches on Einstein@Home are coming to an end! The Fermi mission continues to find new gamma-ray sources for us to target, so we are very optimistic that we will be able to build upon the success of FGRP4 in the near future. Currently FGRP5 is searching for pulsars in gamma-ray sources detected near the Galactic Centre, while FGRPB1 is searching for pulsations from binary millisecond pulsar candidates. </p>
<a href="#FGRP4">Jump to FGRP4 discoveries</a><br><br>
Publications:<br>
<a href="https://doi.org/10.1088/2041-8205/809/1/L2">PSR J1906+0722: An elusive gamma-ray pulsar</a><br>
<a href="https://doi.org/10.3847/2041-8205/832/1/L15">The braking index of a radio-quiet gamma-ray pulsar</a><br>
<a href="https://doi.org/10.3847/1538-4357/834/2/106">The Einstein@Home Gamma-ray Pulsar Survey. I. Search Methods, Sensitivity and Discovery of New Young Gamma-ray Pulsars</a><br>
<a href="https://doi.org/10.3847/1538-4357/aaa411">The Einstein@Home Gamma-Ray Pulsar Survey II. Source Selection, Spectral Analysis and Multi-wavelength Follow-up</a><br>
<a href="https://doi.org/10.1126/sciadv.aao7228">Einstein@Home discovers a radio-quiet gamma-ray millisecond pulsar</a><br>
<hr>
<h4> FGRP5 </h4>
<p>As always, we are extremely grateful to all of our volunteers, especially those whose computers
contributed to these new discoveries. To say "we couldn't have done it without you!" would be a terrible
understatement!</p>
<p>After the success of FGRP4, we began the fifth iteration of our survey, this time searching for pulsars in <a href="https://arxiv.org/abs/1705.00009">unidentified pulsar-like gamma-ray sources detected in the central region of the Milky Way</a>, using almost 9 years of data. Later on, FGRP5 was expanded to search for pulsars in sources from the latest <a href="https://fermi.gsfc.nasa.gov/ssc/data/access/lat/8yr_catalog/"><i>Fermi</i>-LAT Fourth Source Catalog (4FGL)</a> using 11.5 years of data.</p>
<p> To date we have made 14 new pulsar discoveries in FGRP5: 3 from the initial batch of inner-Galaxy sources, and 11 from 4FGL sources. FGRP5 is still ongoing, so we hope for more discoveries in the near future, and are preparing a publication with full details on the newly discovered pulsars. </p>
<a href="#FGRP5">Jump to FGRP5 discoveries</a>
<hr>
<h4> FGRPB1 </h4>
<p>In addition to the isolated pulsar surveys (the FGRPX series), Einstein@Home also searches for <i>binary</i> gamma-ray pulsars. These searches are orders-of-magnitude more difficult than isolated pulsar surveys; the pulsar's orbital motion modulates its pulsed signal, and three extra parameters describing the size, period and phase of the orbit are required to account for this. Optical observations of candidate "black widow" and "redback" counterparts can constrain these parameters, reducing the search space to a more manageable (but still enormous!) volume. </p>
<p>Two binary millisecond pulsars have been found in FGRPB1: an extremely compact "black widow" pulsar, PSR J1653-0158, whose 75-minute orbit is the shortest of any known pulsar binary; and PSR J2039-5617, a "redback" pulsar with interesting long-term variable behavior. </p>
<a href="#FGRPB1">Jump to FGRPB1 discoveries</a><br><br>
Publications:<br>
<a href="https://doi.org/10.3847/2041-8213/abbc02">Discovery of a Gamma-ray Black Widow Pulsar by GPU-accelerated Einstein@Home</a><br>
<a href="https://doi.org/10.1093/mnras/staa3484">Einstein@Home discovery of the gamma-ray millisecond pulsar PSR J2039-5617 confirms its predicted redback nature </a>
<hr>
<h3> Timeline of gamma-ray pulsar discoveries in searches of <i>Fermi</i>-LAT data.</h3>
<a href="img/psrs_vs_time.png"><img style="width: 550px;" alt="Timeline" src="img/psrs_vs_time.png"></a>
<br>
<p>The above plot illustrates the number of gamma-ray pulsars discovered in blind searches using NASA's Fermi Gamma-ray Space Telescope as a function of time (when the discoveries were published). Since the Fermi satellite was launched in 2008, it has been continuously scanning the entire sky and thus is providing an ever increasing data set. In principle, having more data available allows us to do more sensitive pulsar searches. However, at the same time, the computational cost also increases rapidly with the longer data time spans. Thus, as the graphic shows, over the last few years the only new such discoveries were made with Einstein@Home, thanks to the massive collective computing power provided by the Einstein@Home volunteers.</p>
<p>The above plot illustrates the number of gamma-ray pulsars discovered in blind searches using NASA's <it>Fermi Gamma-ray Space Telescope</it> as a function of time (when the discoveries were published). Since the Fermi satellite was launched in 2008, it has been continuously scanning the entire sky and thus is providing an ever increasing data set. In principle, having more data available allows us to do more sensitive pulsar searches. However, at the same time, the computational cost also increases rapidly with the longer data time spans. Thus, as the graphic shows, over the last few years nearly all discoveries were made with Einstein@Home, thanks to the massive collective computing power provided by the Einstein@Home volunteers.</p>
<br>
<hr>
<h3>The discoveries made by Einstein@Home volunteers in detail</h3>
......@@ -69,7 +107,7 @@ how much the pulsar is slowing down over time. The energy required to emit elect
radiation is drawn from the pulsar rotation. The characteristic age is a rough estimate of
the pulsar's age, computed from the spin frequency and its derivative. Finally, the
spin-down power is a measure of the total energy emitted by the pulsar. For comparison, our
Sun outputs roughly 4 x 10<sup>33</sup> erg per second. All pulsars below have a much higher
Sun outputs roughly 4 x 10<sup>33</sup> erg per second. Most pulsars below have a much higher
spin-down power.</p>
<p>The graphics on the right show the pulse profile of each pulsar in green, and the phase-folded
......@@ -109,7 +147,7 @@ gamma-ray emission as a function of pulsar rotation phase and resolve the pulse
<hr>
<tr>
<td style="vertical-align: top; text-align: center;">
<b>FGRPB1 Discoveries</b></td>
<b><a id="FGRPB1">FGRPB1 Discoveries</a></b></td>
</tr>
<!-- psr start //-->
......@@ -177,7 +215,7 @@ gamma-ray emission as a function of pulsar rotation phase and resolve the pulse
<tr>
<td style="vertical-align: top; text-align: center;">
<b>FGRP5 Discoveries</b></td>
<b><a id="FGRP5">FGRP5 Discoveries</a></b></td>
</tr>
<!-- psr start //-->
<tr>
......@@ -796,14 +834,9 @@ gamma-ray emission as a function of pulsar rotation phase and resolve the pulse
</tr>
<!-- psr end //-->
<tr>
<td style="vertical-align: top; text-align: center;">
<b>FGRP4 Discoveries</b></td>
<b><a id="FGRP4">FGRP4 Discoveries</a></b></td>
</tr>
<!-- psr start //-->
<tr>
......@@ -1649,7 +1682,7 @@ gamma-ray emission as a function of pulsar rotation phase and resolve the pulse
<!-- psr end //-->
<tr>
<td style="vertical-align: top; text-align: center;">
<b>FGRP1 Discoveries</b></td>
<b><a id="FGRP1">FGRP1 Discoveries</a></b></td>
</tr>
<!-- psr start -->
<tr>
......
img/psrs_vs_time.png

19.3 KB | W: | H:

img/psrs_vs_time.png

103 KB | W: | H:

img/psrs_vs_time.png
img/psrs_vs_time.png
img/psrs_vs_time.png
img/psrs_vs_time.png
  • 2-up
  • Swipe
  • Onion skin
J1203-6242 2020-05-26 FGRP5
J1742-3321 2017-12-17 FGRP5
J1358-6025 2020-07-21 FGRP5
J1447-5757 2021-03-20 FGRP5
J1231-6511 2021-04-28 FGRP5
J0744-2525 2020-05-15 FGRP5
J2034+3632 2020-06-11 FGRP5
J1231-5113 2020-06-12 FGRP5
J1139-6247 2020-07-19 FGRP5
J1335-5656 2020-08-13 FGRP5
J1736-3422 2020-08-25 FGRP5
J0802-5613 2020-12-18 FGRP5
J1748-2815 2018-07-18 FGRP5
J1649-3012 2019-03-21 FGRP5
J0002+6216 2014-12-31 FGRP4
J0359+5414 2014-12-15 FGRP4
J0631+0646 2015-01-20 FGRP4
J1035-6720 2014-12-09 FGRP4
J1057-5851 2014-10-28 FGRP4
J1105-6037 2014-10-03 FGRP4
J1208-6238 2014-10-03 FGRP4
J1350-6225 2015-06-19 FGRP4
J1422-6138 2011-12-14 FGRP1
J1522-5735 2011-09-30 FGRP1
J1528-5838 2015-03-25 FGRP4
J1623-5005 2014-10-04 FGRP4
J1624-4041 2014-12-20 FGRP4
J1641-5317 2015-12-30 FGRP4
J1650-4601 2014-10-17 FGRP4
J1744-7619 2014-12-07 FGRP4
J1817-1742 2015-12-24 FGRP4
J1827-1446 2014-12-19 FGRP4
J1844-0346 2014-11-18 FGRP4
J1906+0722 2014-09-11 FGRP4
J1932+1916 2011-09-19 FGRP1
J2017+3625 2014-09-09 FGRP4
J0554+3107 2012-01-06 FGRP1
J2039-5617 2019-01-22 FGRPB1
J1653-0158 2018-05-09 FGRPB1
J0007+7303 2008-12-01 USC
J0357+3205 2009-08-14 USC
J0632+0633 2009-08-14 USC
J1418-6058 2009-08-14 USC
J1459-6053 2009-08-14 USC
J1732-3131 2009-08-14 USC
J1741-2054 2009-08-14 USC
J1809-2332 2009-08-14 USC
J1813-1246 2009-08-14 USC
J1826-1256 2009-08-14 USC
J1836+5925 2009-08-14 USC
J1907+0602 2009-08-14 USC
J1958+2846 2009-08-14 USC
J2021+4026 2009-08-14 USC
J2032+4127 2009-08-14 USC
J2238+5903 2009-08-14 USC
J1023-5746 2010-11-19 USC
J1044-5737 2010-11-19 USC
J1413-6205 2010-11-19 USC
J1429-5911 2010-11-19 USC
J1846+0919 2010-11-19 USC
J1954+2836 2010-11-19 USC
J1957+5033 2010-11-19 USC
J2055+2539 2010-11-19 USC
J0734-1559 2011-01-01 USC
J1135-6055 2011-05-09 USC
J0106+4855 2011-09-02 AEI
J0622+3749 2011-09-02 AEI
J1620-4927 2011-09-02 AEI
J1746-3239 2011-09-02 AEI
J1803-2149 2011-09-02 AEI
J2028+3332 2011-09-02 AEI
J2030+4415 2011-09-02 AEI
J2111+4606 2011-09-02 AEI
J2139+4716 2011-09-02 AEI
J1838-0537 2012-06-01 AEI
J1311-3430 2012-11-06 AEI
J1111-6039 2018-10-17 USC
J1714-3830 2018-10-17 USC
J1615-5137 2019-07-22 AEI
J1901-0125 2016-08-01 AEI
J1827-0849 2016-08-01 AEI
\ No newline at end of file
import matplotlib.pyplot as plt
from matplotlib.ticker import ScalarFormatter
from datetime import datetime
import numpy as np
plt.rc('text',usetex=True)
plt.rc('font',family='serif')
import cycler
n = 6
color = plt.cm.Blues(np.linspace(1/n, 1,n))
plt.rc('axes',prop_cycle =cycler.cycler('color', color))
# Timeline of pulsar discoveries: [month since launch, total psrs]
......@@ -23,61 +29,59 @@ plt.rc('font',family='serif')
# FGRP4_idx = 8
# -------------------------------------------------------------------------------------
# By publication date
psr_timeline = np.array([[0,0],
[4,1], # CTA 1
[12,15], # 16 PSRs - CTA1, Abdo et al., 2009
[27,8], # 8 PSRs, Saz Parkinson et al., 2010
[33,2], # 2 PSRs, Saz Parkinson et al., 2010
[39,9], # 9 Pulsars, Pletsch et al. 2012
[47,1], # J1838
[47,1], # J1311-3430
[64,4], # 4 E@H Pulsars, 2013
[84,1], # J1906
[91, 13], # Next bunch of FGRP4 pulsars without J1208
[100,1], # J1208
[106, 2], # J1817, J1641
[116,2], # 2 E@H MSPs
[117,0]])
FGRP1_idx = 8
FGRP4_idx = 9
AEI_idx = 5
# -------------------------------------------------------------------------------------
start = 54682
mjd2008 = 54466
months = 2008 + (psr_timeline[:,0] * (365.25 / 12) + (start - mjd2008))/365.25
psrs = np.cumsum(psr_timeline[:,1])
startdate = datetime.fromisoformat("2008-08-04")
jan2008 = datetime.fromisoformat("2008-01-01")
with open('psr_timeline.dat','r') as dat:
lines = dat.readlines()
months = np.array([(datetime.fromisoformat(line.split()[1]) - startdate).days/365.25 for line in lines]) + (startdate - jan2008).days / 365.25 + 2008
survey = np.array([line.split()[2] for line in lines])
USC_mask = np.where(survey == 'USC')
AEI_mask = np.where(survey == 'AEI')
FGRP1_mask = np.where(survey == 'FGRP1')
FGRP4_mask = np.where(survey == 'FGRP4')
FGRP5_mask = np.where(survey == 'FGRP5')
FGRPB1_mask = np.where(survey == 'FGRPB1')
fig,ax = plt.subplots(1,1,figsize=(6,4))
plt.hist([months[USC_mask],
months[AEI_mask],
months[FGRP1_mask],
months[FGRP4_mask],
months[FGRP5_mask],
months[FGRPB1_mask]],
int(np.ceil(months.max()/12)),
cumulative=True,stacked=True,edgecolor='face')
plt.hist(months,int(np.ceil(months.max()/12)),cumulative=True,color='black',histtype='step')
# -------------------------------------------------------------------------------------
xfmt = ScalarFormatter(useOffset=False)
fig = plt.figure(figsize=(10,5.5))
ax = fig.add_subplot(111)
plt.step(months, psrs,where='post',color='red',linewidth=2.0)
plt.xticks(np.arange(2009,2019))
plt.xticks(np.arange(2009,2022))
plt.ylabel('Total number of pulsars discovered')
plt.title('Timeline of gamma-ray pulsar discoveries in blind searches with Fermi LAT')
#plt.title('Timeline of gamma-ray pulsar discoveries in blind searches with Fermi LAT')
ax.xaxis.set_major_formatter(xfmt)
plt.xlim(months[0],months[-1])
plt.xlim(2008.5,2021.3)
# Survey labels
plt.text(months[FGRP1_idx] + 0.8, 33,'E@H\nFGRP1',ha='center')
plt.text(months[AEI_idx] + 1.0, 29,'ATLAS Cluster\n AEI Hannover',ha='center')
plt.text(months[FGRP4_idx] + 1.40, 35,'E@H\nFGRP4',ha='center')
# Survey shading
plt.fill_between(months[FGRP4_idx:],psrs[FGRP4_idx:],color='grey',step='post')
plt.fill_between(months[AEI_idx:FGRP1_idx+1],psrs[AEI_idx:FGRP1_idx+1],color='lightgrey',step='post')
plt.fill_between(months[FGRP1_idx:FGRP1_idx+2],psrs[FGRP1_idx:FGRP1_idx+2],color='darkgrey',step='post')
# plt.text(months[FGRP1_idx] + 0.8, 33,'E@H\nFGRP1',ha='center')
# plt.text(months[AEI_idx] + 1.0, 29,'ATLAS Cluster\n AEI Hannover',ha='center')
# plt.text(months[FGRP4_idx] + 1.40, 35,'E@H\nFGRP4',ha='center')
# plt.text(months[FGRP5_idx] + 1.40, 40,'E@H\nFGRP5',ha='center')
ax.spines['right'].set_visible(False)
ax.spines['top'].set_visible(False)
ax.xaxis.set_ticks_position('bottom')
ax.yaxis.set_ticks_position('left')
plt.ylim(ymin=0)
plt.savefig('img/psrs_vs_time.png',bbox_inches='tight')
plt.xlabel('Year')
plt.savefig('img/psrs_vs_time.pdf',bbox_inches='tight')
plt.show()
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment