Evolution of the scattering screen of PSR B0834+06 Dana Simard, - PowerPoint PPT Presentation

Evolution of the scattering screen of PSR B0834+06 Dana Simard, Caltech J.-P. Macquart, Ue-Li Pen, Franz Kirsten, Robert Main, Marten van Kerkwijk, Walter Brisken Scintillometry 2019 MPIfR @ Bonn, November 2019

PSR B0834+06 has long been used for scintillation arc studies FAINT SCATTERING AROUND PULSARS: PROBING THE INTERSTELLAR MEDIUM ON SOLAR SYSTEM SIZE SCALES Stinebring et al. 2001 ApJ 549 L97 Dana Simard Scintillometry 2019 @ MPIfR 6 November 2019

PSR B0834+06 has long been used for scintillation arc studies • Brisken et al. 2010 measured the first VLBI distance to a pulsar scattering screen using PSR B0834+06 Brisken et al. 2010 ApJ 708 232 Dana Simard Scintillometry 2019 @ MPIfR 6 November 2019

PSR B0834+06 shows evidence for compact structures Hill et al. 2005 ApJ 619 L171 Brisken et al. 2010 ApJ 708 232 ∆ θ ∝ λ γ Arclet Apex Group γ τ ∼ 1 ms f D < 0 0 . 062 ± 0 . 006 0 . 1 ms � τ � 0 . 4 ms 0 . 01 ± 0 . 01 τ > 0 . 4 ms f D > 0 0 . 019 ± 0 . 004 • Evidence for compact lensing structures: • No detectable dependence of location of sub-images on pulsar position & very small dependence on observing frequency Dana Simard Scintillometry 2019 @ MPIfR 6 November 2019

Models of pulsar scintillation arcs rely on geometry Gwinn 2019 (MNRAS 486 2809); pulsar Gwinn & Sosenko 2019 Pen & Levin 2014 (MNRAS 489 3692) (MNRAS 442 3338); Simard & Pen 2018 Noodle model for (MNRAS 478 983) scintillation arcs Inclined corrugated i sheets lens observer Also confined turbulence in filaments (e.g. ESEs associated with hot stars Walker et al. 2017) or sheets (e.g. SNe remnants) Dana Simard Scintillometry 2019 @ MPIfR 6 November 2019

Models of pulsar scintillation arcs rely on geometry pulsar i lens observer Simard & Pen 2018 MNRAS 478 983 Dana Simard Scintillometry 2019 @ MPIfR 6 November 2019

The corrugated sheet model predicts evolution of secondary spectrum 3 simple predictions: (Underdense) • Asymmetric lens profile: See a di ff erent set of images from a di ff erent set of crests after a corrugated region passes by the pulsar • Frequency evolution: At higher frequencies , a steeper gradient is needed -> if underdense (overdense), separation between the pulsar and the image decreases (increases) • Temporal evolution: As the crest moves away from the pulsar, a steeper gradient is needed -> if underdense (overdense), separation between the pulsar and image decreases (increases) Simard & Pen 2018 MNRAS 478 983 Dana Simard Scintillometry 2019 @ MPIfR 6 November 2019

The corrugated sheet model predicts evolution of secondary spectrum 3 simple predictions: (Overdense) • Asymmetric lens profile: See a di ff erent set of images from a di ff erent set of crests after a corrugated region passes by the pulsar • Frequency evolution: At higher frequencies , a steeper gradient is needed -> if underdense (overdense), separation between the pulsar and the image decreases (increases) • Temporal evolution: As the crest moves away from the pulsar, a steeper gradient is needed -> if underdense (overdense), separation between the pulsar and image decreases (increases) Simard & Pen 2018 MNRAS 478 983 Dana Simard Scintillometry 2019 @ MPIfR 6 November 2019

Frequency evolution consistent with Brisken et al. 2010 measurement ∆ θ ( λ ) Calculate Brisken et al. 2010 find small positive and fit a power law of the form dependence of the angular separation ∆ θ ∝ λ γ on wavelength for one group of arclets. 𝛿 Underdense Arclet Apex Group γ 0.019 sheet τ ∼ 1 ms f D < 0 0 . 062 ± 0 . 006 0 . 1 ms � τ � 0 . 4 ms 0 . 01 ± 0 . 01 Overdense -0.022 τ > 0 . 4 ms f D > 0 0 . 019 ± 0 . 004 sheet Brisken et al. 2010 ApJ 708 232 *For a lensed image with magnification 0.01 and angular separation 10 mas at 314.5 MHz Simard & Pen 2018 MNRAS 478 983 Dana Simard Scintillometry 2019 @ MPIfR 6 November 2019

Multi-epoch observations of PSR B0834+06 PI J.-P . Macquart 10 Arecibo Observations, 4 VLBA + Ar + VLA + Ef observations 150 100 τ ( µ s) 50 0 150 100 τ ( µ s) 50 0 − 20 0 20 − 20 0 20 − 20 0 20 − 20 0 20 − 20 0 20 f D (mHz) f D (mHz) f D (mHz) f D (mHz) f D (mHz) Simard et al. in prep. Dana Simard Scintillometry 2019 @ MPIfR 6 November 2019

Evidence of asymmetric lens profiles in PSR B0834+06 Two arclets 150 approaching pulsar: 100 τ ( µ s) 50 0 − 20 0 20 − 20 0 20 − 20 0 20 f D (mHz) f D (mHz) f D (mHz) One arclet leaving 150 pulsar: 100 τ ( µ s) 50 0 − 20 0 20 Simard et al. in prep. f D (mHz) Dana Simard Scintillometry 2019 @ MPIfR 6 November 2019

day 0 Evolution of PSR B0834+06 day 9 60 day 21 day 30 day 42 day 51 Two arclets approaching pulsar: day 59 day 66 150 50 day 68 day 89 100 τ ( µ s) 50 40 0 − 20 0 20 − 20 0 20 − 20 0 20 f D (mHz) f D (mHz) f D (mHz) One arclet leaving pulsar: 30 150 100 τ ( µ s) 20 50 0 − 20 0 20 10 f D (mHz) After avg. motion of the screen removed 0 Dana Simard Scintillometry 2019 @ MPIfR FOE, August 2019 0 2 4 6 ∆ θ (mas)

To compare with models we need the distance to the screen Ve ff ,|| = 270 +/- 30 km s -1 Angle of 8 +/- 2 degrees W of N 120 100 80 delay ( µ s) 60 40 20 0 − 20 − 10 0 10 20 − 20 − 10 0 10 20 f D (mHz) f D (mHz) 100 phase (deg) 0 − 100 − 20 − 10 0 10 20 − 20 − 10 0 10 20 f D (mHz) f D (mHz) Simard et al. in prep. Dana Simard Scintillometry 2019 @ MPIfR 4 November 2019

To compare with models we need the distance to the screen Ve ff ,|| = 270 +/- 30 km s -1 Angle of 8 +/- 2 degrees W of N 120 100 Curvature of 0.537 +/- 0.010 s 3 80 delay ( µ s) × 10 14 60 1 . 50 average power per pixel 40 1 . 25 1 . 00 20 0 . 75 0 − 20 − 10 0 10 20 − 20 − 10 0 10 20 0 . 50 f D (mHz) f D (mHz) 0 . 25 0 . 4 0 . 6 0 . 8 100 η (s 3 ) phase (deg) 0 − 100 − 20 − 10 0 10 20 − 20 − 10 0 10 20 f D (mHz) f D (mHz) Simard et al. in prep. Dana Simard Scintillometry 2019 @ MPIfR 4 November 2019

To compare with models we need the distance to the screen Ve ff ,|| = 270 +/- 30 km s -1 Angle of 8 +/- 2 degrees W of N 120 100 Curvature of 0.537 +/- 0.010 s 3 80 delay ( µ s) × 10 14 60 1 . 50 average power per pixel 40 1 . 25 1 . 00 20 0 . 75 0 − 20 − 10 0 10 20 − 20 − 10 0 10 20 0 . 50 f D (mHz) f D (mHz) 0 . 25 0 . 4 0 . 6 0 . 8 100 η (s 3 ) phase (deg) η = λ 2 D e ff 0 V 2 2 c e ff , k − 100 Derive a distance to the screen of Dscr = 390 +/- 20 pc − 20 − 10 0 10 20 − 20 − 10 0 10 20 f D (mHz) f D (mHz) Simard et al. in prep. Dana Simard Scintillometry 2019 @ MPIfR 4 November 2019

To compare with models we must remove the motion of the screen 8 6 V psr * 0.5 min 4 0 . 27 V earth * 5 min 0 . 58 Day 0 2 d t (mHz d − 1 ) 0 . 26 v (Mm) η (s 3 ) 0 . 56 0 0 . 25 d f D Day 89 d t d f D − 2 0 . 24 0 . 54 η 0 . 23 − 4 0 . 52 0 20 40 60 80 − 6 day α s − 8 − 5 0 5 u (Mm) ✓ ◆ d f D 1 d η = 1 − ν f D d t d t 2 ην Simard et al. in prep. Dana Simard Scintillometry 2019 @ MPIfR 6 November 2019

Temporal evolution suggests underdense structures • Largest uncertainty is the curvature of the parabola in the secondary spectrum • Combining VLBI w/ incoherent distance measurement will improve constraints on distance to, velocity of the screen & the orientation of the scattering ( | ∆ θ | − | β | ) − ( | ∆ θ | − | β | ) 0 (mas) 0.008 cm − 3 0.0008 cm − 3 0 . 4 -0.008 cm − 3 -0.0008 cm − 3 0 . 2 0 . 0 − 0 . 2 Structure A Structure B − 0 . 4 0 2 4 6 8 0 2 4 6 8 | β | (mas) | β | (mas) * for the same model, Δ n e = -0.007 cm -3 was consistent with the frequency evolution of the scattering screen measured by Brisken et al. 2010 (see previous slide) Simard et al. in prep. Dana Simard Scintillometry 2019 @ MPIfR 6 November 2019

Precise curvature measurements are crucial! • We need precise curvatures to remove the motion of the screen and measure the motion of the images within the screen • We also need precise measurements to monitor changes in the screen over time - in thin screen model these mean change in orientation of screen • Discrepancy between measurements from di ff erent observations of PSR B0834+06: Evolution over time? Hard to explain by changes in screen parameters. Dana Simard Scintillometry 2019 @ MPIfR 6 November 2019

VLBI observations & special transformations can improve curvature 0 150 Hough Transform 100 τ ( µ s) × 10 14 50 1 . 50 average power per pixel 1 . 25 0 − 20 0 20 1 . 00 f D (mHz) 0 . 75 0 . 50 0 . 25 0 . 4 0 . 6 0 . 8 η (s 3 ) Dana Simard Scintillometry 2019 @ MPIfR 6 November 2019

VLBI observations & special transformations can improve curvature 0 Example on 2005 data from 150 Brisken et al. 2010 ApJ 708 232 Hough Transform 𝜾 j , l 𝜾 j , m 100 τ ( µ s) × 10 14 50 1 . 50 average power per pixel 1 . 25 0 − 20 0 20 1 . 00 f D (mHz) 𝜾 k , m 𝜾 k , l 0 . 75 0 . 50 0 . 25 0 . 4 0 . 6 0 . 8 η (s 3 ) 1-D screen models & VLBI + intensities position-position spectrum See talks by Olaf Wucknitz & Daniel Baker Power profiles used by Daniel Reardon Dana Simard Scintillometry 2019 @ MPIfR 6 November 2019