Page 1 PACS NHSC Data Processing Workshop – Pasadena 26 th - 30 th Aug 2013 Overview of SPIRE Photometer Pipeline Kevin Xu NHSC/IPAC on behalf of the SPIRE

page 1PACS

NHSC Data Processing Workshop – Pasadena26th- 30th Aug 2013

Overview of SPIRE Photometer Pipeline

Kevin Xu

NHSC/IPAC

on behalf of the SPIRE ICC

PACS page 2


The Goal

• Show how SPIRE Photometer pipeline works (functionalities of major modules).

Reference: “SPIRE Data Reduction Guide” in HIPE (under “Help”) or in:http://herschel.esac.esa.int/hcss-doc-11.0/load/spire_drg/html/spire_drg.html

PACS page 3


5’

Pipeline & Data Products

Level 2.5Data Product

Level 3Data Product

Maps

• Maps of combined parallel observations with cross-scans

• Maps of overlap observations belonging to the same proposal

New in HIPE 11(only in archive data from HSA)

Scan Map Pipe- line (for small- map, large-map, & SPIRE/PACSparallel modes).

(in HIPE)

PACS page 4

NHSC Data Processing Workshop – Pasadena26th- 30th Aug 2013 Scan Map Pipeline Flow Chart

A “reversed sequence” relative to the chain of data acquisition

Level 0.5Products

Photometer Detector Timeline

Remove electrical crosstalk

Electriccrosstalkmatrix

Concurrent deglitcher

Waveletdeglitcher

ThermistorJump detector

Correct forelectrical filter

Convert toflux density

Temperaturedrift correction

Temperaturederivatives

LPFparameters

Conversion table

Correlationparameters

First leveldeglitching

PACS page 5


Level 0.5Products





Waveletdeglitcher






LPFparameters

Conversion table




significant for very bright sources

PACS page 6


Level 0.5Products





Waveletdeglitcher






LPFparameters

Conversion table



flag thermistors affected by “signal jumps”


PACS page 7


Level 0.5Products





Waveletdeglitcher






LPFparameters

Conversion table




detect & remove glitches shared by multiple detectors

PACS page 8


Level 0.5Products





Waveletdeglitcher






LPFparameters

Conversion table




Single detector deglitcher Alternative: σ-κ deglitcher

PACS page 9


Level 0.5Products





Waveletdeglitcher

Thermistorjump detector





LPFparameters

Conversion table




FFT based (run after deglitchto avoid ringing)

PACS page 10


Level 0.5Products





Waveletdeglitcher






LPFparameters

Conversion table




Include:(1) Non-linearity correction(2) Volt to Jy/beam conversion

PACS page 11


Level 0.5Products





Waveletdeglitcher






LPFparameters

Conversion table



Remove 1/f noisedue to T-drift


PACS page 12


Level 0.5Products





Waveletdeglitcher






LPFparameters

Conversion table



Remove 1/f noisedue to T-drift


flag thermistors affected by “signal jumps”

Exclude flagged thermistors

PACS page 13


Level 0.5Products





Waveletdeglitcher






LPFparameters

Conversion table



• FFT based


PACS page 14


Level 0.5Products





Waveletdeglitcher






LPFparameters

Conversion table




• telescope pointing & orientation • SPIRE detectors positions

PACS page 15

NHSC Data Processing Workshop – Pasadena26th- 30th Aug 2013 Scan Pipeline Flow Chart

Level 0.5Products





Waveletdeglitcher






LPFparameters

Conversion table




Destriper + NaiveMapper

PACS page 16

NHSC Data Processing Workshop – Pasadena26th- 30th Aug 2013 Destriper: Offsets Removal

Destriper: remove the relative offsets of timelines of individual bolometers by minimizing the dispersions in overlap sky pixels (using the Naïve-Mapper iteratively).

PACS page 17

NHSC Data Processing Workshop – Pasadena26th- 30th Aug 2013 SPIRE Naïve Mapper

signal samplings

Sky pixelTwo options:(1) No weighting (pipeline default): Flux of a sky pixel is the simple average of all signal samplings (by all bolometers) in the pixel:

f3

f1 f2

f4

f5

(2) Inverse variance (of instrument noise) weighted:Flux of a sky pixel is the inverse variance weighted

mean of all signal samplings in the pixel, the variance is calculated using the white noise of the bolometer with which a given sampling is taken:

,

PACS page 18

NHSC Data Processing Workshop – Pasadena26th- 30th Aug 2013 Details on Level 2 Products Generation

point sourcemaps (PLW,PMW, PSW)

extendedemission maps

sso maps(only for solar systemobjects)

only for sso:change toextendedemissioncalibraiton

PACS page 19


Status of SPIRE Photometer Pipeline in HIPE 11.0

• General assessment Overall, It works very well. In most cases, data from HSA are already of science quality!• The official calibration accuracy is ±6% (4% from model, 2% RMS).• An example (on the right): The image from HSA looks good.

SPIRE 3-color map of NGC 5315 (a planetary nebula)

(Public data taken from HSA)

PACS page 20


Summary• SPIRE Photometer Scan Map Pipeline handles data in the following observational modes: Small Map, Large Lap, SPIRE/PACS Parallel Mode.• Corrections for instrumental effects (between Level 0.5 and Level 1 products) follow a “reversed sequence” relative to the chain of data acquisition.• The map-making (between Level 1 and Level 2 products) is carried out using a Destriper-NaiveMapper combination. • The current pipeline (HIPE 11.0.1) does a good job (“science ready”) in general.• In subsequent talks, we will discuss how to use the “user pipelines” (HIPE scripts) and other tools in HIPE to resolve data problems caused by known issues.

Documents

Page 1 PACS NHSC Data Processing Workshop – Pasadena 26 th - 30 th Aug 2013 Overview of SPIRE Photometer Pipeline Kevin Xu NHSC/IPAC on behalf of the SPIRE