WIDE component discussion

Posts: 8
Joined: Tue Feb 25, 2014 3:16 pm

WIDE component discussion

Postby gtrichards » Mon Mar 10, 2014 11:10 am

WIDE VLASS component range of possibilities:

Min Area = 3603 sq. deg (DES+HSC)
Max Area = 13055 sq. deg (LSST+DES+HSC)
(see below for detailed notes)

Min Depth = 50uJy (3x FIRST)
Max Depth = 30uJy (5xFIRST)

Min Exp Time = 30s
Max Exp Time = 90s

Multi-epoch: Up for discussion. These times are such that we don't
*need* to do multiple passes to allow variability investigations, but
we could choose to do so.

Bandpass = S (default, but L up for discussion)

Polarization: Yes (default, just need to figure out how that affects
the overhead and thus the depth/area)

Array: B (default, but A or A+B or A+B or A+B+C or Custom up for discussion)


WIDE area calculations:

LSST will observe up to 20k of the 23500 sq deg between
dec=+10 and -75. Where 23500 is (41253/4*pi)*2*pi*(cos(80)-cos(165))

There are 16840 sq. degrees between dec=+10 and dec=-40
(where the VLA and LSST overlap). If we assume that
all of the -40 to -75 area will be observed by LSST, that
leaves 20,000 - (23500-18840) = 13,000 sq. deg. that we
could, in principle, observe with the VLA that will have
LSST coverage.

The DES (i~25th mag) fields north of dec=-40 area are:
Viking -30≤α≤60 -40≤δ≤-25 = 1135 sq. deg
Round 82 -3≤α≤45 -25≤δ≤3 = 1306 sq. deg
Stripe 82 -43≤α≤-3 -1≤δ≤1 = 80 sq. deg
for a total of 2521 sq. deg

These are all within the LSST area, so don't add area, but represent
"must do" regions if we target a region smaller than the LSST area.

The HSC (i~26th mag) fields are all north of dec=-40 with a small
field north of the LSST area. Specifically:

Fall Equatorial
-30<RA<40, -1<dec<7
(adds to DES Stripe82/Round82:
-30<RA<3, 1<dec<7 = 197.4 sq. deg
-3<RA<40, 3<dec<7 = 147.4 sq. deg
27.5<RA<40, -7<dec<-1
(adds no new area to DES)

Spring Equatorial
127.5<RA<225, -2<dec<5 = 682 sq. deg
(all in addition to DES)

Northern Sky
200<RA<250, 42.5<dec<44 = 55 sq. deg
(all in addition to DES)

So, HSC adds 1082 sq. deg to the DES area and provide a nice (if
perhaps relatively too small--consider expanding to cover all of
KIDS-N?) Sprint complement. In all DES+HSC would be 3603 sq. deg. We
consider this to be the *minimum* area for the WIDE component of

If we also add the HSC Northern Sky component to LSST, that takes the
maximum area to 13055 sq deg.

To the extent that the WIDE fields don't overlap with the DEEP fields
and if the two components are done in different bandpasses, then we
would add those DEEP fields to the WIDE component as well, so these
areas could increase slightly.

Please discuss the merits and optimal configuration for a WIDE tier.
Either by responding to this message or posting to the google
spreadsheet at
https://docs.google.com/spreadsheet/ccc ... sp=sharing

Posts: 4
Joined: Thu Feb 20, 2014 2:04 am

Re: WIDE component discussion

Postby bgaensler » Mon Mar 10, 2014 6:26 pm

(Comment below is also pasted into the Google Docs spreadsheet.)

Hi Gordon,

I'm responding about the combined choice of observing frequency and array configuration for VLASS WIDE.

In my opinion, an S-band B-array survey for VLASS WIDE is not the optimal choice, because an overriding consideration needs to be that a VLASS WIDE survey either be completely distinct from EMU/POSSUM/WODAN, or be highly complementary to these other surveys; if it's neither highly distinct nor highly complementary, then in my opinion the large amount of effort required isn't warranted. This dictates one of three possibilities for array configuration and frequency choice:

(1) S-band C or D array, to match angular resolution of EMU/POSSUM/WODAN. This would be bad news for optical IDs, but absolutely spectacular for broadband spectropolarimetry, Faraday rotation and spectral curvature studies (admittedly my main interest!). A factor of 30 in lambda^2 coverage by combining ASKAP and VLASS would be far far beyond anything else contemplated or otherwise possible.

(2) L-band A array, to provide high-resolution counterpart to EMU/POSSUM/WODAN (cf. FIRST compared w NVSS); or

(3) S-band A array, to provide the highest possible angular resolution for morphologies and optical IDs, but with the disadvantage that the results will not be directly comparable or complementary in resolution or frequency to EMU/POSSUM/WODAN.

The default proposal of S-band B-array does not meet any of these criteria - the results will not be able to be merged with EMU/POSSUM/WODAN because of the mismatch in resolution, and the improved angular resolution vs EMU/POSSUM/WODAN is not an improvement by a large enough amount to make the effort of a whole new survey warranted. Bear in mind that the signal-to-noise ratios in the ASKAP/AperTIF surveys will be higher (because of brighter sources and deeper pointings), which will partly offset their poorer angular resolution when it comes to source positions and optical IDs. E.g. for a spectral index of -0.7 and an RMS sensitivity for EMU of 10 uJy/beam, EMU will have signal-to-noise ratios 8x better than VLASS WIDE at S-band at 50 uJy/beam sensitivity, which (in the simplest case) pretty much cancels out the difference in angular resolution (15 arcsec for EMU vs 2 arcec for VLASS wide in S-band B-array). So the positional IDs for S-band B array will often be no better than EMU/POSSUM/WODAN.

Thus we should either go for even higher resolution in VLASS WIDE (S-band in A array), or lower resolution to match EMU/POSSUM/WODAN (S-band in C array). If there are other reasons why S-band B array woudl be useful, then a possible compromise would be S-band B+C, or S-band A+B, which retains the B-array survey but also adds in the extra advantages of A array or C array. My personal preference between these combined scenarios would be S-band B+C arrays.


PS Note that the argument for VLASS DEEP is different, depending on whether DEEP is a single pointing or a mosaic of >~10 deg^2.

Return to “Extragalactic Working Group”

Who is online

Users browsing this forum: No registered users and 1 guest