DB - ASAP

Overview

The ASAP package installed in CASAPY has much of the required functionality for single-dish analysis. I find the tool-based approach cumbersome and hope that tasks can be written for some of the standard calibration and analysis as in the interferometry part of CASAPY. For example, trying to select specific scans, IFs, pols took several steps. Fitting baselines and Gaussians was all available and seemed to work but required many more steps than should be necessary.

I realize the documentation was not finished yet but it is important to have details about how the data are being processed. It was hard, for example, to know what calnod was doing. The message said it was processing "dototalpower" and "dosigref" and I am not sure what this means. The ability to use the tab key to get the different commands and some help by adding a "?" is really nice.

Although the canned datasets are essential I think it is critical to be able to load different datasets for testing. In my case two of the three datasets I understood since I was part of this GBT project. But right now other data in SDIFTS or MS formats cannot be loaded into single-dish CASAPY.

Other datasets can be loaded in but extra effort is required. This is due to the GBT conventions for Doppler tracking, the gbtmsfiller application's logic for separating the spectral windows and ASAPs desire to calibrate each IF separately. To load these datasets, we did a spectral window mapping of spectral indices to simplify/align with the observing intent.

Some Details:

1. The scan numbers appear to be one behind. For example, scan number 21-28 in GBTIDL (and in the SDFITS file) are 20-27 in CASAPY.

by design... It possible to add 1 somewhere in reader but there will be confusion since ASAP seems to treat scan numbers as regular indices.

2. Why is it necessary to run recalc_azel? I would think the (az,el) indicated positions would already exist.

It can be included in the filler from MS

3. The polarization type (poltype) is linear but the data were taken in circular. Also, how do you know which polarization channel is XX/YY or LL/RR, for example?

A filler (ASAP) problem. The filler determines poltype by telescope name. GBT is not in the list so the filler sets the default = linear for this case. The telescope name for GBT probably needs to be modify for this to accommodate both linear and circular.

I think there is nothing specifically to label the polarization so just by order?

4. I was able to use the selection module to select different subsets of the data for processing or plotting. But I could not figure out how to get the data into an array or list.

It is a bit hidden but there is scantable._getspectrum(row)

5. It looks like the rms calculation in stats is wrong. The data look reasonable (i.e., the peak-to-peak fluctuations seem close to expected).

6. The Tsys and noise looked reasonable for the calps calibration. But the Tsys values for calnod were different between GBTIDL and ASAP. For example, in the IRC+10216 data for scan 240 (241 in GBTIDL) I get near Tsys=170K for ASAP and Tsys=120K for GBTIDL. Yet the peak-to-peak fluctuations in the plotted data look to be the same for each data reduction package. So it seems the Tsys values are just different. This will of course affect the weighting when data are averaged.

Maybe partly due to the data is not current. Tcal values were transfered from SYSCAL TCAL column and the corresponding column in scantable. I found that some Tcal values used in Tsys caluculation are different from what appeared to used in GBTIDL, particularly for IRC data.

7. When plotting with the x-axis in units of velocity the scale appears to be incorrect.

I think it is correct but mislabeled-will investigate.

Some Error Messages:

1. When importing asap I always get these messages. Maybe not important but I thought I would mention it.

CASA <15>: import asap as sd First time ASAP use. Setting up ~/.asap /usr/lib/casapy/20.0.1345svn-007/lib/python2.5/asap/__init__.py:315: RuntimeWarning: Python C API version mismatch for module _asap: This Python has API version 1013, module _asap has version 1012. from asap._asap import Log as _asaplog /usr/lib/casapy/20.0.1345svn-007/lib/python2.5/asap/asapplotter.py:3: DeprecationWarning: The sre module is deprecated, please import re. import sre

Just a warning message about python versions. It is already fixed in the next version (they hadn't gone to Python 2.5 for this release)

2. I was never able to get the HI spectral line cube script to work which seemed to be a problem with the CV installation. Here is the message.

CASA <46>: res.save('FLS3a_calfs', 'MS2')
Traceback (most recent call last)

/export/home/duli/dbalser/data/casa/tests/ in ()

/export/home/duli/dbalser/data/casa/tests/asap/scantable.py in save(self, name, format, overwrite)
  1. from asap._asap import stwriter as stw
  2. writer = stw(format2)
--> 120 writer.write(self, name)
  1. print_log()
  2. return

: Table /export/home/duli/dbalser/data/casa/tests/FLS3a_calfs/POINTING does not exist

3. I get the following message when loading the IRC+10216 dataset. I'm not sure what is going on here and this may explain the problems with the Tsys above.

CASA <3>: s=sd.scantable('IRC+10216/IRC+10216_rawACSmod',False) Importing IRC+10216/IRC+10216_rawACSmod... Cannot find any matching Tcal at/near the data timestamp. Set Tcal=0.0 Cannot find any matching Tcal at/near the data timestamp. Set Tcal=0.0 . . .

4. I get the following message when trying to save data in SDFITS or MS2 format. The ASAP and ASCII formats seem to work.

ASA <56>: scal.save('testout_MS2', 'MS2')
Traceback (most recent call last)

/export/home/duli/dbalser/data/casa/tests/ in ()

/export/home/duli/dbalser/data/casa/tests/asap/scantable.py in save(self, name, format, overwrite)
  1. from asap._asap import stwriter as stw
  2. writer = stw(format2)
--> 120 writer.write(self, name)
  1. print_log()
  2. return

: ArrayColumn::put: Table array conformance error

This is a know issue. In the current version, saving to MS fails often after averaging of the data or calibration. There are a few sources of the problem (ASAP::STFiller, casa::PKSMS2writer, and casa::PKSMS2reader). Fixes are underway.

CASA <57>: scal.save('testout_SDFITS', 'SDFITS') Segmentation fault

Maybe, related the MS save problem but I need to look into more carefully.

-- JosephMcMullin - 28 Feb 2007
Topic revision: r1 - 2007-02-28, JosephMcMullin
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding NRAO Public Wiki? Send feedback