March 6 2018 telecon

online: Karl, Eric, Brendan, Maurizio

• Karl: been discussing GRASP with Maura, no breakthrough yet. In parallel has worked with Julian and Andrea Z to insert a FSL beam into the correct file format for toast. This worked well but need to double-check normalization.
• Karl has also been working on Earth/Moon avoidance calculation. He's created a model for these purposes, results coming soon.
• Eric: Pointing error post. Added a curve showing how the “r” uncertainty compares to the pointing error leakage term. The pointing noise is assumed to be white. Eric chose 2 microK - arcmin, but actual PICO noise level in polarization is 0.6 microK - arcmin. The purple curve will drop below the blue curve (Eric will redo it): level should be quadratic in noise rms. so we can say that instrument noise will be subdominant to the pointing error leakage into B-modes.
• Eric gives the caveat that this is very back-of-the-envelope given the white noise assumptions about the pointing which is not realistic. But for these purposes, we think it's safe to say that at the r ~ 1e-4 level of sensitivity, 1 arcsecond pointing reconstruction may start to become important but it is not a dominant effecft.
• Maurizio has a half-year set of sims for 2 detectors from Andrea, but there was a bug in toast (not in Maurizio or Andrea's code). Strange jumps seen between different data sets produced on different processes. Maurizio did an end-to-end test of his code just to check performance: performed very well 6 months of data takes 2 hours on edison with 48 cores. The final results look correct, but the jump problem needs to be solved before we can claim victory. Ted Kisner looking into it. Problem must be trivial because it hasn't been seen before. Full 2-year test will come next once bug is solved. Getting close!!