Summer shutdown 2016

Currently, we are in summer shutdown phase, when the telescope is completely shut down for almost two months. During that period, the telescope mirrors, spectrograph & detectors are taken out, cleaned and re assembled. This is the time when some major upgrades are also made. This is typically a low work load period for me as we do not have daily observations. But, I have got a bunch of them now. This year we are starting a new program. The program is to observe a new type of targets called Emission Line Galaxies (ELGs). As we are attempting it new, there are lots of challenges associated with it. We have had several ELG test plates last year. In fact, a major share of my SDSS work was related to these ELG test plates. Almost everything is new with ELG observations as compared to a standard observations. As we are mainly chasing O III and O II emission lines in the redshift range 0.8 to 1.1, we have make sure we have good throughput in the red side of the spectrum. Hence, ELG targets will be designed at 7500 A. That means we will drill the hole so that 7500 \AA light will be at the centre of the hole. The telescope will also be guiding at 7500 \AA.

1. We have to drill more plates in the LST range 20-24 and 1-10 hours. Since we are starting new ELG program, we will have 125 plates in the SGC and 175 plates in the NGC. These plates will cover the gap in these LST ranges. But, that is not enough. We have to drill a batch of LRG/QSO plates to make sure to patch the gap. We have already drilled the LRG/QSO plates. ELG plates in the SGC are remaining. (Lots of issues with the plate-design)
2. Waiting for Demitri to finalise the plate design of the 45 ELG SGC plates. Although we started well in advance, we had to iterate a lot between tiling and plate design as we are doing it for the first time. It is already late for the drill shop and the need by date is in 2 weeks (25 August). I am sure I will get a lot thrashings from the drill shop.
3. Fix the SoS pipeline so that ELG plates gives correct feed back to the observers. DecALs survey only has u,g,z band observations. But, SDSS pipeline computes the SNR2 (signal to noise squared) in the g and i band wavelengths. It computes the SNR2 by fitting a line to the magnitudes vs snr2 plot and estimating the SNR2 at some fiducial magnitude(21). This currently will not work for ELG targets from DeCALs as it doesn't have i band magnitudes. The fix is to add a kluge to the pipeline by which if it is an ELG plate, it estimates the fits from z band magnitude
4. I have added a Kluge. But, it is not working as it is supposed to. Even if we estimate the SNR2 at z band wavelengths, we do not have reasonable values to decide the plate completeness. Need to explore more options

I cannot take rest peacefully untill I fix the SoS. Tough times ahead.