-
Notifications
You must be signed in to change notification settings - Fork 0
/
notes.txt
92 lines (72 loc) · 4.54 KB
/
notes.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
Lynne's new NEO runs: cumulative completeness for H<22 objects:
Run 3p/30 |3p/15| SNR3 SNR0 singlePair singleDet Modification
astro_lsst_01_1014 76.1 72.6 79.1 94.4 88.8 92.0 12 deg Ecliptic + WFD, 60sec
astro_lsst_01_1015 79.2 74.8 81.7 96.4 90.3 93.0 15 deg Ecliptic + WFD, 60sec
astro_lsst_01_1016 80.0 77.5 83.1 95.6 88.9 92.1 NEO baseline
astro_lsst_01_1017 80.8 77.8 83.3 95.3 90.1 93.1 EB + 30s, gri
lucy_1001 76.5 73.7 79.9 95.3 88.8 92.1 15 deg Ecliptic (ri, 60 sec), NES gz
astro_lsst_01_1015:
- the best (but only about half of WFD visits due to 60 sec exposures)
- how about 20 deg band around the Ecliptic? Going from 12 to 15 has a
big impact.
- 3 pairs in 30 days at SNR=4 should be similar in completeness to 3
pairs in 15 days at SNR=3, but more doable
Run 3p/30 |3p/15| SNR3 SNR0 singlePair singleDet Modification
astro_lsst_01_1015 79.2 74.8 81.7 96.4 90.3 93.0 15 deg Ecliptic + WFD, 60sec
- at the bright end (H<20), C for single detection and single pair
saturates at ~99.4%; this shows that 12-year survey is long enough
to get even those objects that can ``hide behind the Sun''
- 3 pairs in 15 nights with SNR=0 results in C=96.4%; this shows that
the impact of finite cadence on completeness for *large* objects is
only 3.0%
- 3 pairs in 15 nights with SNR=5 gives C=75%, which compared to
a single detection with SNR=5 and C=93% represents a loss of 18%;
therefore, the impact of cadence on *small* objects is much larger
than for large objects; going from a single detection to 3 pairs in
15 nights is equivalent to losing 2 mag of depth!
- 3 pairs in 15 nights with SNR=5 crosses C=90% at H~20.5; data
deeper by 1.5 mag would bring H<22 completeness to 90% (or we'd
need to go to SNR=1.3); but to go deeper by 1.5 mag, the exposure
time needs to be 16 times longer!!
- going from 15 nights to 30 nights boosts C by 4.4%; this is
equivalent to about 0.3 mag deeper data and 1.2 mag of ``missing''
depth; in this case, the exposure time would have to be ``only'' 9
times longer to reach C=90% for H<22.
So, reaching C=90% is becoming more and more hopeless. My last
desperate traces of hope are:
- we could look at the median brightness of (say) H=22 objects vs.
ecliptic latitude (and perhaps solar elongation to account for
sweet spots) and adjust exposure time as a function of latitude
(we know that the faintest H=22 detections are close to the
Ecliptic)
- if the required band width is much less than 15 deg, we would
perhaps have enough observing time budget to really go with 5-10
minutes long exposures. Of course, we'd have to keep them down
to 1 min or so and shift-and-coadd multiple exposures for a grid
of possible motions, which would be a major impact on DM
- we could look at H~22 objects which we missed with 3 pairs in 15
nights, but which we did *not* miss with ``single detection'' and
ask what were their SNRs when they were in the FOV but didn't
have SNR>5. And where they were on the sky. This sounds like
something for Joachim to look at.
- NASA says ``we'd be happy with even C=80%" :)
- 75% of the PHAs: Ivezic & Jones, 2014 (AAS)
TODO from May 29 reading:
- ref to Chesley & JPL, find his simulations plan
- in Introduction, add info about counts of transients from my AAS
time-domain talk, also refer to Ridgway paper
-
10 km/s at 100 Lunar Distances is:
384,000 km * ang. speed (radian/s) = 10 km/s
ang. speed = 5.4 arcsec / sec = 5.4 * 3600 * 24 arcsec / day = 5.4*24 ~ 100 deg/day
5.4 arcsec/sec * 30 sec ~ 3 arcmin trail!
The impact of trailing: for 0.8 arcsec seeing and 30 sec exp (tau=1)
slow: 0.25 deg/day, fast: 1 deg/day
tau dmT dmTotSlow dmTotFast dmSNRSlow dmSNRFast
1.0 0.00 0.03 0.38 0.04 0.25
2.0 0.38 0.25 1.26 0.10 0.48
5.0 0.87 0.52 1.78 0.31 0.88
Optimal (trailed) vs. PSF detection for v=1 deg/day is worth 0.13 mag
for nominal exposures, and 0.23 mag for 60 sec exposures (0.15 mag
for v=0.25 deg/day). Adopting 0.2 mag for doubling exposure time,
this corresponds to about 0.02 in completeness.