search

desihub / desisurvey

Code for desi survey planning and implementation
BSD 3-Clause "New" or "Revised" License
2 stars 7 forks source link

Fix transparency and seeing in exposure factor and ETC computations #128

Closed schlafly closed 3 years ago

schlafly commented 3 years ago

There was a small bug in the signal accumulation rate in the ETC. This PR fixes that, replacing the signal accumulation rate with the square root of the weather factor, rather than the weather factor. This PR also changes the transparency to be the square of the transparency in the weather factor.

In combination with a related PR on surveysim, this slightly improves survey margin. That's a bit of a surprise, since the transparency fix makes things strictly worse, but the bug in the survey simulations more than compensates.

Original margin: image New margin: image

The fix to the srate in the ETC looks like it should have a huge impact, but its effect was muted since it only effects how survey speed varies within an exposure; the bug didn't affect the initial expectations of how long an exposure should last from the scheduler---though those were affected by the transparency issue.

dkirkby commented 3 years ago

Please update doc/changes.rst then this looks ready to go.