Merge pull request #9 from lsst-pst/tickets/DM-45247

DM-45247: Add amp offset description paragraph to Instrument Signature Removal section of Science Pipelines paper

authors.yaml

@@ -7,3 +7,4 @@
 - sullivanis
 - watersc
 - findeisenk
+- nourbakhshe

isr.tex

@@ -1,2 +1,17 @@
 \section{Instrument Signature Removal}
 \label{sec:isr}
+
+\subsection{Amplifier Offset Correction}
+\label{sec:isr:ampoffset}
+The amplifier offset correction (commonly referred to as amp-offset correction, or pattern continuity correction) runs as part of the instrument signature removal (ISR) process.
+This correction is designed to address systematic discontinuities in background sky levels across amplifier boundaries.
+We believe that these discontinuities arise from electronic biases between adjacent amplifiers, persisting even after application of dark and flat corrections.
+
+Drawing on the \texttt{PANSTARRS}' Pattern Continuity algorithm \citep{2020ApJS..251....4W}, our method aims to eliminate these offsets, thereby preventing problems such as background over-/under-subtraction at amplifier boundaries caused by discontinuities across the detector.
+
+The amp-offset algorithm initially computes a robust flux difference measure between two narrow strips on opposite sides of each amplifier-amplifier interface.
+Regions containing detected sources, or pixel data which have been masked for other reasons, are not considered.
+These amp-interface differences are stored in an amp-offset matrix; diagonal entries represent the number of neighboring amplifiers, and off-diagonal entries encode information about the associations between amplifiers.
+A complementary interface matrix encodes directional information for these associations.
+Using this information, a least-squares minimization is performed to determine the optimal pedestal value to be added or subtracted to each amp which would reduce the amp-offset between that amplifier and all of its neighboring amplifiers.
+This method is generalized to support 2D amplifier geometries within a detector, as with LSSTCam, incorporating length-based weighting into the matrices to account for amplifiers that are not square.

local.bib

@@ -331,3 +331,22 @@ @INPROCEEDINGS{2022ASPC..532..559B
        adsurl = {https://ui.adsabs.harvard.edu/abs/2022ASPC..532..559B},
       adsnote = {Provided by the SAO/NASA Astrophysics Data System}
 }
+
+@ARTICLE{2020ApJS..251....4W,
+       author = {{Waters}, C.~Z. and {Magnier}, E.~A. and {Price}, P.~A. and {Chambers}, K.~C. and {Burgett}, W.~S. and {Draper}, P.~W. and {Flewelling}, H.~A. and {Hodapp}, K.~W. and {Huber}, M.~E. and {Jedicke}, R. and {Kaiser}, N. and {Kudritzki}, R. -P. and {Lupton}, R.~H. and {Metcalfe}, N. and {Rest}, A. and {Sweeney}, W.~E. and {Tonry}, J.~L. and {Wainscoat}, R.~J. and {Wood-Vasey}, W.~M.},
+        title = "{Pan-STARRS Pixel Processing: Detrending, Warping, Stacking}",
+      journal = {\apjs},
+     keywords = {Astronomy data reduction, CCD observation, Sky surveys, 1861, 207, 1464, Astrophysics - Instrumentation and Methods for Astrophysics},
+         year = 2020,
+        month = nov,
+       volume = {251},
+       number = {1},
+          eid = {4},
+        pages = {4},
+          doi = {10.3847/1538-4365/abb82b},
+archivePrefix = {arXiv},
+       eprint = {1612.05245},
+ primaryClass = {astro-ph.IM},
+       adsurl = {https://ui.adsabs.harvard.edu/abs/2020ApJS..251....4W},
+      adsnote = {Provided by the SAO/NASA Astrophysics Data System}
+}