GB . PTCS . OffsetPointingAnalysis

Offset Pointing Analysis

October 11, 2007

Here we report investigations of how the pointing errors change with separation, both in time and angle on the sky. For any pointing run we may take all possible pairings of the peak data and calculate the difference in LPCs for the pair and the angular and temporal separation between the two peak measurements.

Methods

The pointing calibration runs were processed with "prepoint" with "-lpc" option to produce ascii datafiles that summarize the LPCs for each peak scan. IDL Software was written to further process these data files and produce from them two additional files:

1) a file with one line per peak sequence, giving the LPCs averaged over the forward and backward scans and over the two polarizations. These files have "_avgs" in their names.
2) a file with LPC differences between all pairs of lines in the first file. These file names contain "_pairs"

These files are stored in the directory: /home/groups/ptcs/data/offsetdata/ and the software is in ~fghigo/GBT/ptcs/src/astplot.pro

The format of an "avgs" file is as follows:

col 1 : the mean MJD of the peak sequence
col 2 : the source name
col 3 : the first scan number in the peak sequence
col 4 : the mean azimuth(degrees)
col 5 : the mean elevation (degrees)
col 6 : the mean azimuth (i.e., cross-El) LPC in arcsec.
col 7 : the mean elevation LPC in arcsec.

And the format of a "pairs" file :

col 1 : mean MJD of the two peak sequences
col 2 : time offset in minutes between the two peak sequences
col 3 : azimuth difference (degrees)
col 4 : elevation difference (degrees)
col 5 : great circle distance between the two measurements (degrees)
col 6 : cross-El difference in the LPCs of the two measurements (arcsec)
col 7 : elevation difference of the LPCs of the two measurements (arcsec)

Summary of processing

Data files have been produced for the following all-sky pointing runs. This tabulates the projects, means and standard deviations of the LPCs and differences of LPCs. (All the related files are in /home/groups/ptcs/data/offsetdata/ )

Date	Project	d/n	XEl_LPC	El_LPC	diff_XEl_LPC	diff_El_LPC
April 10, '07	TPTCSPNT_070410	night	3.21(3.08)	5.52(3.83)	-0.05(4.38)	-1.12(5.31)
April 25, '07	TPTCSPNT_070425	night	7.36(3.59)	9.29(5.57)	-0.86(5.02)	2.83(7.41)
Sept 26, '07	TPTCSPNT_070926	night	0.17(1.60)	5.25(3.14)	-0.07(2.24)	-0.41(4.43)
Sept 26, '07	TPTCSPNT_070926	day	0.83(4.76)	0.08(6.08)	2.94(6.07)	0.82(8.57)
Sept 29, '07	TPTCSPNT_070929	night	0.65(4.39)	3.32(4.00)	-2.39(5.74)	-1.12(5.56)
Sept 29, '07	TPTCSPNT_070929	day	2.16(1.91)	-0.94(6.97)	-0.34(2.68)	1.45(9.80)

Plots of the LPC differences versus time separation and angular separation are attached below.

Todd has processed these files with a C program that can select rows on the basis of data ranges in multiple columns and compute the rms of the resulting sample. Here are some quick results (subject to revision). For two sources observed within 60 minutes of each other, and separated by <10 deg in the axis in question, the rms of the LPC values are:

Dataset	Model	X-elev	Elev	Total	Timerange	X-elev #pairs	Elev #pairs
20070410+20070425	4c	3.7"	4.5"	5.8"	night (10PM-8AM)	92	49
20070926	5c	1.5"	4.0"	4.2"	night (10PM-8AM)	65	123
20070926	5c	4.8"	6.2"	7.8"	daytime (8AM-10PM)	71	178
20070929	5c	2.2"	3.3"	4.0"	night	114	203
20070929	5c	1.9"	3.4"	3.9"	daytime	104	239
20070926+20070929	5c	2.0"	3.6"	4.1"	night	178	325
20070926+20070929	5c	3.4"	5.0"	6.0"	daytime	174	417

within 30-60 minutes

Dataset	Model	X-elev	Elev	Total	Timerange	X-elev #pairs	Elev #pairs
20070926+20070929	5c	3.2"	5.8"		day	30	212
20070926+20070929	5c	1.9"	4.0"		night	27	186

within 15 deg in azim, 10 deg in elev, and 20-60 minutes

Dataset	X-elev	Elev	Total	Timerange	X-elev #pairs	Elev #pairs
200704	4.0"	5.0"		night	5	45
200709	3.1"	5.5"		day	99	356
200709	2.7"	3.9"		night	75	223

within 30 deg in azim, 15 deg in elev, and 30-60 minutes

20070425	5.2"	4.7"	night	22	53
200709	3.1"	5.4"	day	111	111
200709	3.6"	4.5"	night	142	50

As a reminder, the commands used to generate these numbers have the following format: rms -s 1 -f TPTCSPNT_070926_pairs_nite.txt -x 1 0 60 -x 2 0 10 -c 5

We would have to go back further in time to get daytime data for model 4c. In this processing scheme, the new model does appear to perform better than the old model did (just before the track refurb). However the values I get for the old model are not as good as what you had quoted below. Can you tell me more about the values you quoted: 2.8" and 3.2"? For example, were they computed for smaller intervals of time or for sources closer together than 10deg? Or perhaps they were measured before the track had degraded so far. one column that we should add to the output of your extraction tool is the windspeed (if it is easily available).

-- FrankGhigo - 11 Oct 2007

Attachment:	Action:	Size:	Date:	Who:	Comment:
TPTCSPNT_070410_pairs.gif	action	54906	11 Oct 2007 - 20:51	FrankGhigo	April 10, Night, LPC differences vs separations.
TPTCSPNT_070425_pairs.gif	action	46088	11 Oct 2007 - 20:51	FrankGhigo	April 25, night, LPC differences vs separation
TPTCSPNT_070926_pairs_nite.gif	action	67953	11 Oct 2007 - 20:52	FrankGhigo	Sept 26, night, LPC differences vs separations
TPTCSPNT_070926_pairs_day.gif	action	67896	11 Oct 2007 - 20:53	FrankGhigo	Sept 26, day, LPC differences vs separations
TPTCSPNT_070929_pairs_nite.gif	action	60665	11 Oct 2007 - 20:53	FrankGhigo	Sept 29, night, LPC differences vs separations
TPTCSPNT_070929_pairs_day.gif	action	53898	11 Oct 2007 - 20:54	FrankGhigo	Sept 29, day, LPC differences vs separations

Topic OffsetPointingAnalysis . { Edit | Attach | Ref-By | Printable | Diffs | r1.4 | > | r1.3 | > | r1.2 | More }

Revision r1.4 - 19 Nov 2007 - 20:42 GMT - ToddHunter Parents: WebHome > InformationForProjectTeam	Content copyright © 1999-2007 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.


NRAO Home > Green Bank \| Wiki Topic: GB > PTCS > InformationForProjectTeam > OffsetPointingAnalysis

Changes \| Index \| Search \| Statistics \| Go