Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.atnf.csiro.au/research/pulsar/ppta-obs/dashboard/docs/ppta_obsstrategy_rms.pdf
Дата изменения: Fri Dec 4 05:23:41 2015
Дата индексирования: Tue Apr 12 03:48:40 2016
Кодировка:
Поисковые слова: п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п

PPTA observing strategies
Ryan M. Shannon/21 Jan 2015

· Current state of play
Schedule lists pulsars in order of increasing right ascension Access to S/N diagnosDcs Observers typically observe either in increasing RA or in "closest-
non-
observed" order) While some are cognizant of S/N of observaDons many are not.

Why care?
· Strategy is essenDal to any observing program. · What is the PPTA observing strategy? · What mo5vates this strategy? · Observe 20 pulsars every ~two weeks in three bands to achieve 100 ns Dming precision. Detect GWB in 5 years
(Based on Jenet et al. 2005)

· How can we do beVer?

Parkes basics #1
· Sources locaDon: Right Ascension (RA) and declinaDon (DEC) · Local Sidereal Time (LST): Tells you what is visible when RA=LST sources are highest on sky
Sources with RA < LST are in the west (se]ng) Sources with RA > LST are in the east (rising)
· Generally we observe GalacDc plane from rise to set

· Parkes (AlDtude-
Azimuth telescope)

AlDtude limits -
> 30 degrees from horizon ~ 1 degree from zenith (tells you when source is above horizon) Azimuth limits -
> wraps (more about this on next side)

Parkes basics #2
· Telescope sensiDvity is defined by its gain (or system equivalent flux density), system temperature, received bandwidth, observing Dme
Parkes: 0.9 K/Jy Green Bank/MeerKAT 2 K/Jy Arecibo/FAST: 12 K/Jy (AO 20 K/Jy for 430 MHz line feed)

· Why Parkes/PPTA does well

LocaDon/laDtude (0437/1909), relaDvely clean 20cm-
band CalibraDon (PolarizaDon/instrument stability) InstrumentaDon (UWL/10cm system/mulDple backends) Cadence (lots of observing Dme sensiDvity ~ sqrt(t)) Clever people (y'all)

Parkes basics #3: Receivers
· Parkes receivers are mounted at prime focus (in the focus cabin) with a translator stage that physically moves between the receivers · To switch receivers currently requires the observer to park the telescope (drive toward zenith)
Drive Dme long RestarDng tcs, running LO scripts Even for experts, this process takes 15 minutes

· Strategy: minimize receiver changes by observing for ~ 24 hours with one receiver set-
up

·

Parkes is a large telescope

Strategy 1: Slew Dme

Slew rate 15 degrees per minute in azimuth "North and South" at transit require long slews to go from "northern sources" (dec > -
30 degrees) to "southern sources" (dec < -
30 degrees)

Long Dme to slew here

·

Wraps:

Will define slew Dmes Telescope can rotate 540 degrees. Wrap limits located in West (as sources are se]ng) If in Northern wrap, can have long slew to Southern source and vice versa
· · · J1600-
3053 and J1603-
7202 J1024-
0719 and J1045-
4509 J1744-
1134, J1713+0747

Wrap limit South wrap

·

Consequence: it is omen advantageous to not observe sources in increasing RA (as previously displayed on dashboard)

4 long slews over an observing session -
> one fewer pulsar observed

·

Strategy: Observe sources close to minimize slew 7me
Note dashboard doesn't correctly calculate slew 7me.

Wind
· ObservaDons cannot occur during high wind · During marginal wind condiDons, the telescope is more safe when the dish is pointed into the wind and close to zenith · If the winds are ~ 30-
40 km/h less likely to be wind-
stowed · Q: If LST is 14:45 and moderate winds are coming from north east, what would be a good pulsar to observe?

Strategy 2: RFI
· · Monitor RFI using web monitors (CASPSR and PSRMON) and SPD (real-
Dme) Satellite: (20cm)
Use FROG to monitor locaDon of Beidou satellites Strategy: Change sources Strategy: Wait it out (DR2 pipeline will edit it out)

·

Dish CafИ: (Azimuth and Diurnal dependence on RFI)
Observe at night Observe with telescope towards East (dish blocks visitors centre)

·

Midweek RFI

DevastaDng 20cm RFI Coordinate with ATNF Sciops to figure out when it is occurring (RMS has been asking Phil Edwards before recent sessions) IdenDfy on Parkes RFI weather monitor Switch to 10cm/50cm band

ScinDllaDon
· All of our pulsars experience refracDvely and diffracDvely scinDllate · Some can easily see factor of 20 variaDon in flux density with long exponenDal tails · If pulsar is bright, stays bright for some Dme (hours) · If pulsar is faint, remains faint for a similar length · Results in large variaDons in TOA uncertainty · Non-
detecDons (5-
sigma) detecDons are worthless · ScinDllaDon can change the gain of Parkes to Green Bank (or even larger) · We want to observe as much as possible on those exponenDal tails!

ScinDllaDon #2
· TOA error ~ (Pulse EffecDve Width)/(S/N raDo) · SensiDvity to GWs ~ (1/TOA error)^2
(True for clock signal as well/planet ephemeris)

· We want high S/N observa7ons of pulsars with narrow profiles (narrow features)

Strategy #3 ScinDllaDon
· Where does the current observaDon stand · Monitor flux using the CASPSR flux monitor · S/N state of play is plots provided on dashboard · Black-
belt -
-
Get TOAs and compare S/N of observaDon with historical observaDons

Pulsar prioriDsaDon
· · · · Grey area Handful of pulsars central to our array (Priority 1) Northern pulsars are well observed (Priority 3)
How does data quality compare to NANOGrav/EPTA?

Some pulsars show evidence for strong Dming noise (Priority 4)
J1024-
0719, J0613-
0200, J1643-
1224, J1824-
2452A, J1939+2134 Probably aren't of much use to any of the goals

· ·

Legacy Dming baseline for SKA era/Moderately useful southern pulsars (Priority 2) M. Keith/X. Siemens detecDon significance accelerates with addiDon of more pulsars
Assume that pulsars are relaDvely equal Work of acDve research that we should be looking at (Kobi?) in the realiDes of PPTA/IPTA observaDons.

Current scheme
· 1: Highest priority (dominate limit, searches for correlated signals) · 2: Important southern pulsars · 3: Northern pulsars/legacy southern pulsars · 4: Poor pulsars of liVle use to Dming array · This is a logarithmic scale
RMS wouldn't observe priority 4 pulsar if there was a priority 1 pulsar visible

Histogram

Time series for PSR J2145-
0750 Histogram for PSR J1017-
7156

ScinDllaDon observing strategy
· If a high priority (1/2) pulsar is in a high scinDllaDon state (top quarDle), it should be re-
observed · If any pulsar is a low state amer 5 minutes it observing should be abandoned and a nearby (unobserved/non-
weak pulsar should be observed)
Nearby in slew Dme

· Regardless of state of second pulsar observe it for full hour
Mindful of wasDng Dme slewing between sources

Concluding thoughts
· Our high grade and generous Dme allocaDon allow us to be clever with how we observe. "Observing every pulsar for the sake of checking them all off" every session is a waste of this allocaDon · ObservaDons of pulsars when they are bright are important. If a pulsar is bright (as seen on a web monitor/SPD) it should be re-
observed.
Benefits both our primary scienDfic objecDves, secondary scienDfic objecDves

· If a pulsar is faint it, should be ditched

We don't understand the staDsDcs fully yet (Renee started working on this), but as a general rule of thumb, I wouldn't re-
observe it in the same band unDl it rises again.

· All of us should be thinking "why are we doing this". A lot of these thoughts were formed amer trips to Parkes, talking to Mike Keith, MaVhew Bailes