BRC Optical Ground-Based Follow-Up
Contents
Big Picture
Optical data can be incredibly helpful in weeding out garbage from our YSO selection. Remember that Spitzer is phenomenally efficient, and with just a few seconds can reach objects at the edge of the Universe. Recall the discussion in the CG4 paper, and just at a gut level how much nicer the SEDs look if we have optical data too. Recall the stuff on the Resolution page (specifically example 4) and how much high spatial resolution matters. High spatial resolution optical imaging may allow us to determine that some of these objects are actually resolved galaxies. Once we add optical photometry to the SEDs, the shape of the SEDs may allow us to determine that some of these objects are actually AGN and not YSOs.
The Las Cumbres (LCOGT) Faulkes Telescope on Haleakala (specifially JD Armstrong) has kindly offered to help us get optical ground-based follow-up observations. We need to plan coherently (professionally!) for these observations to ensure we get the best quality data in the least amount of time possible. ("Ooops we screwed up" will not generally get you additional telescope time.)
Planning Observations
In order to plan for ground-based follow-up of our targets, we have many things to consider. These are the same issues, in more or less the same order, that I would consider before writing a proposal to any telescope for anything; this isn't fabricated just for you!
- Are the objects visible (ever) from the observatory(ies) to which we have access?
- Ex 1: the CG4 object from last year is a Southern Hemisphere object, so unless you have access to a Southern Hemisphere telescope, that's not possible.
- Ex 2: the BRC objects are Northern Hemisphere (both of them). When are they visible from the LCOGT telescope in Hawaii? See airmasses calculation below to help in answering that question.
- What filters are available at the telescope? And, which are best suited to the science we want to accomplish?
- JD says that they have currently: Sloan u,g,r,i, Halpha, and PannStarrs z,y.
- Any constraints on the Wien side of the SED will help us, so anything optical is scientifically helpful.
- I have no experience with z and y.
- While young stars are often bright in Halpha (and you know that from Ogura's papers), pretty much everything is fainter in narrow band filters (such as Halpha) than broad-band filters (such as the Sloan filters). So, anything will take much longer to detect to sufficient signal-to-noise in Halpha than in the other filters. It will take less time for us to get measurements in the broadband filters.
- I have some experience with the Sloan filters, so I know (for example) how to convert them into the right units for addition to the SEDs. Recall the shape of a stellar SED. Assuming all of our targets are really stars, will they be brighter at u or i?
- JD says that he is willing to do two filters for us. Which two would you do? I vote for r and i. Do you understand why?
- What is the spatial extent of sources on the sky that we care about?
- We don't have a final list of sources yet. But recall that a single IRAC field of view is 5 arcminutes on a side, and the region that we care most about (with four IRAC bands) is basically a single IRAC field of view. There are serendipitously obtained patches of just 2 bands offset from the main field; each of those patches is also basically a single IRAC field of view, so basically also 5 arcmin on a side. The centers of the serendipitously obtained fields are both a little more than 5 arcminutes offset from the center of the main field.
- What is the spatial extent of a single pointing of the ground-based telescope?
- In this case, 9 arcmin. So with a single pointing, we can obtain essentially the entire region with 4-band IRAC coverage in one of our fields, even if the fields are rotated nearly 45 degrees with respect to each other. So, great! No need to tile a large region with the LCOGT telescope.
- What is the surface density of the sources in our field of view, and what is the spatial resolution of the optical telescope?
- Recall the Resolution discussion to understand why this matters.
- The spatial resolution of IRAC is ~1.2-1.5 arcsec. There may be sources spaced closer together than that, but we wouldn't be able to tell based on the IRAC data.
- The spatial resolution of the LCOGT telescope is 1.1 arcsec, and is seeing-limited. So, if we have a bad weather night, then the resolution will be worse, but seems to commonly be comparable to IRAC. So we're good here too.
- What are the total integration times we need?
- JD says "magnitude 19 is easily within reach."
- What are the brightnesses we expect from our targets? Ah, this is the meat of it! More below.
- NEW -- Will any targets (of interest or accidentally in our field of view) saturate the images at these exposure times? (Do we need shorter exposure times to avoid saturation)?
- How many dithers will be needed?
- Typically, one does at least 3 dithers in order to clean out cosmic rays, but if the exposures are very short, just one can be sufficient. It depends on how much our total integration time works out to be.
- What kinds of calibrators do we need?
- We need to find (stable) objects of known brightnesses in r and i bands (preferably comparable brightnesses to our targets), at a comparable airmass to our targets, and plan to observe them (request time to observe them) before and after our science targets.
- JD says that there are some Faulkes archives of standards from which we can pick; I hope he can help with this.
- Who will actually do the observing? (and from where?)
- No one is going to Hawaii!
- JD says that both queue and internet-based remote observing is possible. TBD.
- In what format will we get the data back? On what timescale?
- The images will arrive back to us in FITS format, flatfielded and bias subtracted but nothing else -- no coordinates, not calibrated, nothing else. We will need to take care of all of that in some fashion.
- Timescale? Presumably immediately, or within a few days of the observations.
- Who will process the data (and how)?
- You guys will have to lead this!
- My thought was to sidestep the coordinates issue, use the IRAC images in essence as finding charts (though 2MASS might be better choices) to find the subset of sources in which we are interested (surviving best YSO candidates and calibrators in those fields), and have you guys do APT photometry on those targets. Once we have measurements in total counts (or counts/sec), we can solve for extinction and color corrections and calibrate the measured photometry (as opposed to calibrating the images themselves).
- Russ has offered (11 Aug) to take on the bulk of the data reduction! Yay!
- What will we do with the data when we have it?
- See the CG4 paper for examples!
Brightness Estimations For Our Targets
We know some information about our targets, but we need to make some educated guesses and assumptions in order to get some estimates of how bright our targets are likely to be at the optical bandpasses.
We have JHK and IRAC-1 for most if not all of our 'interesting' targets. Of the data we have, these bandpasses are closest to the optical, and least likely to be affected by any excess due to a circumstellar disk. We need to use this information to get an educated guess as to the distribution of brightnesses we will expect for our targets in the optical.
We have no good and simple way of estimating any extinction (reddening, or Av). So, we will assume no reddening. This is not necessarily a good assumption, but it's all we can do at this time.
Most of our stars will probably end up being late K stars or early M stars. The mass function is such that there are more M stars than K stars, but our brightness limits are such that we probably only reach to early M stars, which cuts off the mass function effect.
Conceptually, the process we need to follow is:
- Take the observed distribution of brightnesses for our 'interesting' targets.
- Assume that they are all really stars, and have minimal or no Av.
- Get an estimate of where the photosphere is based on the NIR.
- Extrapolate a guess at the expected brightness of the photosphere in the optical.
I can think of two different ways we can do this. One is conceptually harder and easier to implement in Excel ("option 1" below); the other is conceptually easier but harder to implement in Excel ("option 2" below).
Option 1
As you know, young stars can have a LOT going on. They have circumstellar disks, reddening, accretion... it is tough to get a really good, solid empirical estimate of what truly plain disk-free young stars actually look like, what colors they really have. There are a few different estimates in the literature, and they are unevenly good, and sometimes even internally inconsistent. My own favorite list kinda falls apart at about M2 (and later). I have copied it below. These are not quite just main sequence colors because these stars have not yet completely settled onto the main sequence, but you could also start from a distribution of main sequence colors if you needed to.
You can use the brightness at just one band (say, J), hope that that band is minimally affected by any of the other things that are going on, assume that all of the stars that we have are about K7, read off the V-J color below for that type, and shift the distribution of J accordingly. How does this change if you assume they are all a different type? Or use a different band to extrapolate to V?
SpTy code Mv Ic U-V V-Ic U-B V-R V-J V-H V-K R-I B0 V 1.0 -4.00 -3.74 -1.38 -0.30 -0.31 -0.12 -0.70 -0.31 -0.83 -0.18 B1 V 1.1 -3.08 -2.86 -1.23 -0.26 -0.31 -0.11 -0.61 -0.31 -0.74 -0.15 B2 V 1.2 -2.40 -2.21 -1.08 -0.23 -0.31 -0.10 -0.55 -0.31 -0.66 -0.13 B3 V 1.3 -1.60 -1.44 -0.94 -0.19 -0.31 -0.09 -0.45 -0.31 -0.56 -0.10 B4 V 1.4 -1.39 -1.26 -0.84 -0.18 -0.31 -0.08 -0.30 -0.31 -0.49 -0.09 B5 V 1.5 -1.20 -1.10 -0.72 -0.16 -0.31 -0.07 -0.35 -0.31 -0.42 -0.09 B6 V 1.6 -0.80 -0.71 -0.61 -0.14 -0.31 -0.07 -0.30 -0.31 -0.36 -0.07 B7 V 1.7 -0.51 -0.43 -0.50 -0.12 -0.31 -0.06 -0.25 -0.31 -0.29 -0.06 B8 V 1.8 -0.20 -0.05 -0.40 -0.09 -0.31 -0.04 -0.17 -0.31 -0.24 -0.05 B9 V 1.9 0.27 0.35 -0.17 -0.04 -0.11 -0.03 -0.09 -0.16 -0.13 -0.01 A0 V 2.0 0.60 0.60 0.00 0.00 0.00 -0.02 -0.01 0.00 0.00 0.02 A1 V 2.1 1.00 0.98 0.05 0.02 0.02 0.01 0.06 0.06 0.07 0.03 A2 V 2.2 1.30 1.25 0.10 0.07 0.04 0.03 0.11 0.13 0.14 0.04 A3 V 2.3 1.50 1.42 0.15 0.11 0.06 0.03 0.18 0.21 0.22 0.05 A4 V 2.4 1.70 1.59 0.20 0.14 0.08 0.04 0.25 0.28 0.30 0.07 A5 V 2.5 1.90 1.76 0.24 0.17 0.10 0.09 0.27 0.36 0.38 0.08 A6 V 2.6 2.05 1.84 0.26 0.20 0.10 0.10 0.31 0.41 0.44 0.10 A7 V 2.7 2.20 1.93 0.28 0.22 0.09 0.11 0.35 0.47 0.50 0.11 A8 V 2.8 2.40 2.08 0.30 0.27 0.07 0.13 0.43 0.54 0.57 0.14 A9 V 2.9 2.55 2.18 0.32 0.33 0.05 0.16 0.50 0.61 0.64 0.16 F0 V 3.0 2.70 2.28 0.34 0.37 0.03 0.19 0.58 0.67 0.70 0.18 F1 V 3.1 2.92 2.48 0.36 0.40 0.03 0.22 0.63 0.73 0.76 0.19 F2 V 3.2 3.15 2.69 0.37 0.43 0.02 0.23 0.68 0.79 0.82 0.20 F3 V 3.3 3.27 2.79 0.38 0.45 0.01 0.24 0.72 0.87 0.91 0.21 F4 V 3.4 3.38 2.87 0.39 0.47 0.00 0.25 0.75 0.97 1.01 0.22 F5 V 3.5 3.50 2.96 0.40 0.50 -0.02 0.26 0.79 1.06 1.10 0.24 F6 V 3.6 3.67 3.09 0.44 0.53 -0.02 0.28 0.85 1.17 1.21 0.25 F7 V 3.7 3.83 3.21 0.49 0.56 -0.01 0.30 0.90 1.27 1.32 0.26 F8 V 3.8 4.00 3.34 0.54 0.59 0.02 0.31 0.96 1.30 1.35 0.28 F9 V 3.9 4.20 3.51 0.59 0.61 0.04 0.32 0.99 1.33 1.38 0.29 G0 V 4.0 4.40 3.69 0.64 0.63 0.04 0.33 1.03 1.36 1.41 0.30 G1 V 4.1 4.55 3.83 0.64 0.65 0.04 0.34 1.08 1.39 1.44 0.31 G2 V 4.2 4.70 3.97 0.64 0.67 0.04 0.35 1.09 1.41 1.46 0.32 G3 V 4.3 4.83 4.09 0.71 0.68 0.08 0.35 1.11 1.44 1.49 0.33 G4 V 4.4 4.97 4.22 0.79 0.69 0.15 0.36 1.15 1.47 1.53 0.34 G5 V 4.5 5.10 4.34 0.86 0.71 0.20 0.36 1.16 1.52 1.58 0.35 G6 V 4.6 5.20 4.43 0.90 0.73 0.22 0.37 1.18 1.58 1.64 0.36 G7 V 4.7 5.38 4.59 0.95 0.75 0.24 0.38 1.22 1.66 1.72 0.37 G8 V 4.8 5.50 4.69 1.00 0.77 0.27 0.39 1.26 1.69 1.76 0.38 G9 V 4.9 5.70 4.87 1.13 0.83 0.35 0.44 1.30 1.73 1.80 0.39 K0 V 5.0 5.90 5.02 1.27 0.88 0.45 0.47 1.41 1.88 1.96 0.41 K1 V 5.1 6.10 5.17 1.44 0.93 0.59 0.50 1.53 2.00 2.09 0.43 K2 V 5.2 6.40 5.47 1.52 0.95 0.63 0.53 1.60 2.13 2.22 0.45 K3 V 5.3 6.60 5.54 1.80 1.06 0.83 0.57 1.79 2.33 2.42 0.49 K4 V 5.4 7.00 5.85 2.01 1.15 0.94 0.62 1.91 2.53 2.63 0.53 K5 V 5.5 7.40 6.08 2.22 1.32 1.06 0.70 2.13 2.74 2.85 0.62 K6 V 5.6 7.75 6.37 2.43 1.38 1.16 0.73 2.24 2.88 3.00 0.66 K7 V 5.7 8.10 6.65 2.54 1.45 1.18 0.76 2.52 3.03 3.16 0.69 K8 V 5.8 8.55 6.85 2.64 1.70 1.20 0.86 2.66 3.18 3.32 0.84 M0 V 6.0 8.86 7.06 2.65 1.80 1.22 0.89 2.80 3.48 3.65 0.91 M0.5 V 6.05 9.14 7.26 2.66 1.88 1.24 0.92 2.91 3.58 3.76 0.96 M1 V 6.1 9.41 7.45 2.67 1.96 1.26 0.94 3.02 3.67 3.87 1.02 M1.5 V 6.15 9.79 7.74 2.68 2.05 1.27 0.97 3.14 3.79 3.99 1.08 M2 V 6.2 10.16 8.00 2.70 2.16 1.27 1.00 3.28 3.91 4.11 1.16 M2.5 V 6.25 10.70 8.39 2.73 2.31 1.27 1.05 3.53 4.16 4.38 1.26 M3 V 6.3 11.18 8.71 2.75 2.47 1.28 1.10 3.84 4.40 4.65 1.37 M3.5 V 6.35 12.00 9.34 2.77 2.66 1.28 1.16 4.08 4.69 4.95 1.50 M4 V 6.4 12.71 9.85 2.89 2.86 1.29 1.23 4.39 4.98 5.26 1.63 M4.5 V 6.45 13.20 10.18 2.98 3.02 1.27 1.36 4.78 5.39 5.69 1.66 M5 V 6.5 14.69 11.30 3.07 3.39 1.28 1.48 5.22 5.80 6.12 1.91 M5.5 V 6.55 15.75 11.99 3.10 3.76 1.29 1.70 5.78 6.37 6.71 2.06 M6 V 6.6 16.49 12.48 3.35 4.01 1.30 1.80 6.26 6.93 7.30 2.17 M6.5 V 6.65 17.14 12.82 3.48 4.32 1.32 2.03 6.63 6.93 7.30 2.30 M7 V 6.7 17.76 13.22 3.60 4.54 1.33 2.15 7.03 6.93 7.30 2.41 M8 V 6.8 18.26 13.60 3.85 4.66 1.33 2.15 7.61 6.93 7.30 2.51 M8.5 V 6.85 18.13 13.56 3.98 4.67 1.33 2.15 7.63 6.93 7.30 2.52 M9 V 6.9 19.29 14.93 4.10 4.68 1.33 2.15 7.65 6.93 7.30 2.53 SpTy code Mv Ic U-V V-Ic U-B V-R V-J V-H V-K R-I
You should look at the distribution of V values you derive. How faint do we have to go in order to catch 50% of our targets of interest? 80%? 90%? All of them? How sensitive is this value to your choice of type or color to leverage? Hint: I would assess the fraction of objects brighter than a certain level by doing a histogram, which you certainly can do in Excel. But another formula of use would be "countif" -- this is an Excel function that you can use to tell it "count up the number of objects brighter than x" and compare it to the total number of objects.
This is pretty easy to implement in Excel. It's just one column more of calculation, and if you code it up right, you can make absolute cell references, such that you can change one single cell value to play with the effects of the assumption of the type, and another single cell value to play with the color offset (e.g., which band you choose to leverage from). BUT you are depending that whatever known color from which you are leveraging is uncontaminated (from Av, disk, accretion, variability, or just plain errors).
V is Johnson V, not Sloan r or i. V is bluer than either Sloan r or i, so if our estimates of V are any good, getting to a V of x will almost certainly get to an r or i of x. BUT Johnson V is a Vega-based magnitude system, and most people working with Sloan filters are working in AB mags (more on this below). This makes a difference too, and we just need to keep track of this; it's kind of like Celsius vs. Fahrenheit -- converting between the two is not as simple as moving the decimal point, but both systems measure things just fine, and you just have to be clear which system you're working in.
We're assuming no reddening. That's almost certainly going to affect what fraction of stars we actually do detect.
Option 2
Look again at the SEDs that were included as part of the CG4 paper. Note that for the stars for which we did not have a spectral type, we just assumed a type of M0. I have a model grid that includes M0 stars. It's really hard to make really good models, especially for the late type stars, because of the complicated chemistry and quantum mechanics (molecules, line emission) involved. The models I have are a hybrid of the best available over a wide range of masses -- they are called Kurucz-Lejeune models (because Dr. Lejeune took Dr. Kurucz's models and updated them slightly). (Note that above, it was hard to get a guess at a plain photosphere from empirical observations of real young late-type stars; here it's hard to get a guess at a plain photosphere from the perspective of calculating models.)
What we did in the CG4 paper was normalize the model to the energy at K (if available), or other NIR bands if K was not available. We can do the same thing for our BRC SEDs. One can then use the model to read off a guess at what the expected energies are at the central wavelengths of the two Sloan bands (r and i). Then you have to back out the expected magnitude by "undoing" the calculation you did to get it into an SED in the first place : log lambdaFlambda (erg/s/cm^2) -> lambdaFlambda (erg/s/cm^2) -> Flambda (erg/s/cm^3) -> Flambda (Jy) -> magnitude (mag).
Here is the original full grid of models. Note that you can get them for "a large range of stellar parameters : Teff=50,000K to K, logg=5.50 to -1.02 and [M/H]=-5.0 to +1.0." (the latter is metallicity, in case that wasn't clear-- it's also written [Fe/H]).
Here is a 'standard' M0 model in log lambdaFlambda (erg/sec/cm^2). File:Justm0model.txt Note that there are a lot of points here because it is a model and is designed to provide a well-sampled spectral energy distribution. Watch your units! You can then use this to get a guess at r and i.
Just like you did above, you should look at the distribution of r and i values you derive. How faint do we have to go in order to catch 50% of our targets of interest? 80%? 90%? All of them? You don't really have the machinery to do this, but in theory you can pick your own model to experiment with, and you can also experiment with different values of Av.
This is harder to implement in Excel. You'll have to take a spreadsheet in which you calculated SEDs, figure out how to normalize the model to K, make a plot, check that normalizing at K is the best thing to do (or is it better to normalize at a different band?). Then interpolate the expected energy at the r and i bands, and back out a magnitude at the two bands.
Here, you are no longer depending that whatever single band from which you are leveraging is uncontaminated (from Av, disk, accretion, variability, or just plain errors) because you will make a plot to check on the model fit and see if normalizing at a different band is better.
Here, you are explicitly estimating Sloan r and i, not just V. BUT you do have to be careful to keep track of whether you are working with Vega magnitudes or AB mags. Most Sloan folks work in AB mags instead. Central wavelengths and zero points is our local repository of central wavelengths and zero points. For AB mags, instead of a Vega zero point, you always use a flat reference spectrum, so the zero point is 3631 Jy for all bands.
We're still assuming no reddening. That's still almost certainly going to affect what fraction of stars we actually do detect. We can redden the model and check, but I am worried that this will be a lot of trouble for you, for not a lot of gain. In some cases, the combination of even just J, H, and K will clearly indicate some degree of reddening (e.g., the unreddened model will be a poor fit when you make the plot to check the fit).
More concrete observation plans!
When can we observe? Hourly almanac (airmasses) for our targets
calculated for Hawaii.
*** Hourly airmass for brc27 *** Epoch 2000.00: RA 7 04 00.0, dec -11 22 55 Epoch 2011.87: RA 7 04 33.4, dec -11 24 01 At midnight: UT date 2011 Nov 16, Moon 0.72 illum, 31 degr from obj Local UT LMST HA secz par.angl. SunAlt MoonAlt 18 00 4 00 21 18 -9 47 (down) -71.1 -4.8 ... 19 00 5 00 22 18 -8 47 (down) -73.4 ... ... 20 00 6 00 23 18 -7 47 (down) -73.5 ... ... 21 00 7 00 0 18 -6 46 (down) -72.2 ... ... 22 00 8 00 1 18 -5 46 (v.low) -69.9 ... -1.6 23 00 9 00 2 18 -4 46 4.439 -66.3 ... 11.5 0 00 10 00 3 19 -3 46 2.261 -61.0 ... 24.8 1 00 11 00 4 19 -2 46 1.602 -52.8 ... 38.4 2 00 12 00 5 19 -1 46 1.318 -40.0 ... 52.1 3 00 13 00 6 19 -0 45 1.195 -19.8 ... 65.9 4 00 14 00 7 19 0 15 1.172 6.7 ... 79.6 5 00 15 00 8 19 1 15 1.240 30.7 ... 84.8 6 00 16 00 9 20 2 15 1.430 47.0 -7.7 71.6 At midnight: UT date 2011 Dec 16, Moon 0.67 illum, 51 degr from obj Local UT LMST HA secz par.angl. SunAlt MoonAlt 18 00 4 00 23 16 -7 49 (down) -73.5 -4.0 ... 19 00 5 00 0 16 -6 48 (down) -72.3 -17.1 ... 20 00 6 00 1 16 -5 48 (v.low) -70.0 ... ... 21 00 7 00 2 16 -4 48 4.599 -66.5 ... ... 22 00 8 00 3 17 -3 48 2.297 -61.2 ... ... 23 00 9 00 4 17 -2 48 1.616 -53.2 ... 2.1 0 00 10 00 5 17 -1 48 1.324 -40.5 ... 15.7 1 00 11 00 6 17 -0 47 1.197 -20.6 ... 29.4 2 00 12 00 7 17 0 13 1.171 5.7 ... 42.9 3 00 13 00 8 17 1 13 1.236 30.0 ... 56.0 4 00 14 00 9 18 2 13 1.420 46.6 ... 67.7 5 00 15 00 10 18 3 13 1.830 57.0 ... 74.2 6 00 16 00 11 18 4 13 2.884 63.7 -11.5 69.5
*** Hourly airmass for brc34 *** Epoch 2000.00: RA 21 33 30.0, dec +58 04 32 Epoch 2011.62: RA 21 33 50.9, dec +58 07 39 At midnight: UT date 2011 Aug 16, Moon 0.93 illum, 61 degr from obj Local UT LMST HA secz par.angl. SunAlt MoonAlt 19 00 5 00 16 15 -5 19 2.653 -87.8 -3.0 ... 20 00 6 00 17 15 -4 19 1.997 -100.9 -16.4 -1.4 21 00 7 00 18 15 -3 18 1.639 -115.3 ... 12.4 22 00 8 00 19 16 -2 18 1.434 -131.9 ... 26.2 23 00 9 00 20 16 -1 18 1.322 -151.3 ... 39.8 0 00 10 00 21 16 -0 18 1.277 -173.2 ... 52.9 1 00 11 00 22 16 0 42 1.288 164.1 ... 64.6 2 00 12 00 23 16 1 42 1.359 143.1 ... 71.8 3 00 13 00 0 16 2 43 1.504 124.9 ... 69.1 4 00 14 00 1 17 3 43 1.760 109.2 ... 59.1 5 00 15 00 2 17 4 43 2.212 95.4 -14.8 46.5 6 00 16 00 3 17 5 43 3.080 82.7 -1.4 33.2 At midnight: UT date 2011 Sep 16, Moon 0.86 illum, 66 degr from obj Local UT LMST HA secz par.angl. SunAlt MoonAlt 19 00 5 00 18 17 -3 17 1.631 -115.8 -9.2 ... 20 00 6 00 19 17 -2 16 1.430 -132.5 ... ... 21 00 7 00 20 18 -1 16 1.320 -151.9 ... 6.6 22 00 8 00 21 18 -0 16 1.276 -173.9 ... 20.0 23 00 9 00 22 18 0 44 1.289 163.4 ... 33.6 0 00 10 00 23 18 1 44 1.362 142.5 ... 47.4 1 00 11 00 0 18 2 44 1.510 124.3 ... 61.3 2 00 12 00 1 18 3 45 1.771 108.8 ... 75.2 3 00 13 00 2 19 4 45 2.232 95.0 ... 87.5 4 00 14 00 3 19 5 45 3.119 82.3 ... 76.4 5 00 15 00 4 19 6 45 5.239 70.0 -17.1 62.6 6 00 16 00 5 19 7 45 14.754 57.7 -3.1 48.8 At midnight: UT date 2011 Oct 16, Moon 0.84 illum, 79 degr from obj Local UT LMST HA secz par.angl. SunAlt MoonAlt 18 00 4 00 19 15 -2 18 1.435 -131.9 -1.2 ... 19 00 5 00 20 16 -1 18 1.323 -151.2 -15.2 ... 20 00 6 00 21 16 -0 18 1.277 -173.2 ... ... 21 00 7 00 22 16 0 42 1.288 164.2 ... 4.2 22 00 8 00 23 16 1 42 1.358 143.1 ... 17.1 23 00 9 00 0 16 2 42 1.503 124.9 ... 30.3 0 00 10 00 1 16 3 43 1.759 109.3 ... 43.6 1 00 11 00 2 17 4 43 2.211 95.5 ... 57.2 2 00 12 00 3 17 5 43 3.076 82.7 ... 70.8 3 00 13 00 4 17 6 43 5.122 70.4 ... 84.0 4 00 14 00 5 17 7 43 13.931 58.1 ... 81.2 5 00 15 00 6 17 8 43 (down) 45.4 ... 67.7 6 00 16 00 7 17 9 44 (down) 32.1 -4.8 54.1
Target brightnesses
DSS images for planning purposes: 
and File:Brc27 dss.fits
and 
and File:Brc34 dss.fits
These are 10 arcmin on a side, and centered on the same centers as our IRAC observations. North is up.
John calculated brightnesses using 'option 1' above. He writes: "If I am doing it correctly I have V magnitudes for M0 ranging from around 15 to 19. 60% are brighter than magnitude 18 and 85% brighter than 19 which is what was quoted as our limit. R and I should be somewhat brighter."
Note to JD (1 Sep 11)
hi JD -
I think that we have made some progress on our end for possible LCOGT observations of our two targets.
Our two targets are :
- BRC27 at 07:03:58.71, -11:23:10.2 (most easily done second half of the night Nov/Dec)
- BRC34 at 21:33:32, +58:04:30 (up now, transiting middle of the night)
Our fields of view of interest are more or less 5' diameter, centered on the coordinates above, so a single pointing (modulo any dithers) from Faulkes ought to cover the area of interest.
We'd like Sloan r and i filters. You'd mentioned that "magnitude 19 is easily within reach" and that would be perfect for our targets of interest -- roughly 85% of our targets should be brighter than that. Our brightest object of interest we estimate to be ~15th mag at V. What is the exposure time we need for that? I looked for, but could not find, this information on the LCOGT pages.
We will need some help finding calibrators. I looked for this kind of information on the LCOGT pages as well, but could not find it.
Is this a feasible time request?
If you would like to talk to all of us at once, we have a weekly telecon at 3:45 pacific time on Wednesdays.
Our working area for this aspect of our project is here: http://coolwiki.ipac.caltech.edu/index.php/BRC_Optical_Ground-Based_Follow-Up Near the bottom of that page, there are some airmass calculations and DSS finding charts in jpg and FITS.
Calibration fields
(From Luisa, 14 Sep): Here is what my friend Peregrine found for us in regards to nearby Sloan calibration fields. There are some pretty nearby Sloan stripes that work, and our expected brightnesses are well-matched to the regime where Sloan gets good photometry.
He (Peregrine) also provided fits binary tables of the photometry -- we'll need this to be able to calibrate our data!
Note that the Sloan data are in AB mags, not Vega mags. Most Sloan folks work in AB mags instead. For AB mags, instead of a Vega zero point, you always use a flat reference spectrum, so the zero point is 3631 Jy for all bands. Marginally more information here.
I've forwarded this to JD, who will let us know what happens next.
Note from Peregrine:
Hi Luisa,
Here are the SDSS photometry for point sources in the two calibration fields. I have applied standard QA checks (rejected sources flagged as BRIGHT, EDGE, SATURATED, MOVED, or TOO_FEW_GOOD_DETECTIONS) and used a SNR > 20 cut in r, i, and z.
Field (ra,dec) ra dec Nsrcs BRCCAL27 (114,-11) 113.9 to 114.1 -11.1 to -10.9 412 BRCCAL34 (310,+58) 309.8 to 310.2 +57.9 to +58.1 314
The fields that are probably of the most interest to you are:
- ra
- dec
- psfMag_r
- psfMag_i
Best regards, Peregrine
UPDATE 14 Sep - File:Brccal peregrin.xlsx - file from Russ with the Sloan data that Peregrine sent for nearby Sloan calibration fields. There are two worksheets, for fields 27 and 34.
Actual observations and data reduction!
BRC 34 was observed on Oct 22, by JD Armstrong and some of his students.
See Russ' page for the data reduction.
