aXe

Spectral Extraction and Visualization Software

The aXe software package is designed for the extraction, calibration, visualization, and simulation of spectra from slitless spectroscopic instruments.

The overall aXe package consists of three components:

aXe – automated extraction and calibration of spectra from slitless spectroscopic images
aXe2web – creates webpages for easy visualization and inspection of aXe output products
aXeSIM – simulation of slitless spectral images

aXe was originally developed by the Space Telescope - European Coordinating Facility (ST-ECF), a unit of the European Space Agency (ESA) hosted by the European Southern Observatory (ESO) headquarters in Garching, Germany. As of January 1, 2011, the Space Telescope Science Institute (STScI) has taken responsibility for support and distribution of the software.

To report problems, or seek help on aXe issues not addressed in these webpages, please contact the Help Desk.

aXe

The aXe reduction tasks enable automated, efficient extraction of thousands of spectra from slitless spectral images. The software drives the spectral extraction based on two sets of information:

a catalogue of objects in the field, and
configuration files which calibrate the positioning of the spectra and the wavelength dispersion solution, and convert from extracted detector units to physical flux units.

aXe is not instrument specific, and can accommodate data from different instrument/grism combinations by specifying the required parameters in the configuration file.

Software Access

aXe is a set of C-based routines, wrapped up into PyRAF/IRAF software which is comprised of several tasks and is distributed as a subpackage of the Space Telescope Data Analysis System (STSDAS) under analysis.slitless.axe.

Instructions on how to to install aXe can be found in the Installation tab below.

Documentation

GitHub – A Jupyter notebook is available to show how to extract WFC3 slitless data using aXe.
WFC3/IR Grism Cookbook – Short step-by-step guide to reducing G141 (IR) grism data using aXe. Includes link to download Grism data from the WFC3 ERS Program, which is used as an example in the guide.
aXe User Manual – Provides detailed documentation on all tasks and parameter settings in aXe, including more sophisticated reduction options not discussed in the WFC3/IR Grism Cookbook.
aXe PASP Paper – Overview of aXe methodology (Kummel et al. 2009).
aXedrizzle Overview – Description of procedure for the co-addition of the 2D spectra. Not to be confused with Astrodrizzle which is for the co-addition of direct imaging data.

What's New?

2018-05-14
A new Jupyter notebook is available on GitHub to show how to extract WFC3 slitless data using aXe. The notebook and sample data can be obtained from
2013-05-30
The WFC3 IR grism cookbook has been updated to accurately reflect availability of AstroDrizzle support in IRAFX.
2013-03-11
The WFC3 IR grism cookbook has been updated to reflect changes made to the aXe software to accommodate AstroDrizzle image products. An updated manual will be available shortly.

The instructions to install aXe and aXeSIM are the same as both packages are installed at the same time.

aXe has several software dependencies that need to be satisfied and the simplest way to install aXe is as part of a fresh Anaconda 3 installation using Python 2.7. It is therefore recommended that it be installed as a new Anaconda 3 environment, using the following steps:

Install anaconda 3 and follow the installation instructions.
Add the STScI channel to anaconda:
conda config --add channels http://ssb.stsci.edu/astroconda
Download the list of specific packages that should be installed and save the file as req262.txt
Create the new aXe anaconda environment:
conda create -n axe-wfc3 --file req262.txt
Once the axe-wfc3 is activated (in a new terminal, after issuing the command “source activate axe-wfc3), add pydrizzle and multidrizzle to axe-wfc3:
pip install multidrizzle
Before using aXe, activate this environment using:
source activate axe-wfc3
If needed, create a directory called iraf in your user directory, and from there issue the command:
mkiraf
to create the configuration files that are required by aXe and its Pyraf/iraf code

Note: We have verified that Anaconda3 version 5.2.0 works and if you encounter problems, please install that version specifically from https://repo.continuum.io/archive/

aXe2web

The program aXe2web provides a quick-look facility for slitless spectral data reduced with aXe. To facilitate examination of hundreds to thousands of spectra, axe2web produces a series of webpages that display the main results of the reduced data. With any web-browser the webpages can then be viewed to identify interesting objects or to reveal flaws in the data reduction. Examples of the webpage visualizations are linked at the bottom of this page.

Software Access

aXe2web is completely implemented in python. The installation uses the python standard module disutil. The latest version of aXe2web can be downloaded here: aXe2web-1.3.tar.gz. Important notes about the installation are given in the README file bundled there.

Documentation

Detailed documentation is bundled in the above tar file, and can be found in the subdirectory aXe2html/doc/aXe2web_help.html. A higher-level overview follows below.

For each detected object aXe2web produces a table row giving the reference number, magnitude in the magnitude system of the direct object, the (x,y) image coordinate of the direct object, its RA and Dec, a cut-out image showing the direct object, the spectral stamp image showing the 2D spectrum, a 1D extracted spectrum in counts and the same in calibrated flux units. To facilitate the navigation within a data set, overview and index pages accompany the detailed object pages. The index page includes a table with the ordered reference number of all objects. Direct links from both the overview page and the index page take you to the corresponding locations of the objects in the object pages.

The details of the html summary produced by aXe2web are controlled by a set of about 20 keywords or settings, which influence the presentation of the objects in various ways. There exist, for example, keywords/settings to:

order the objects in a catalog column (usually by magnitude)
specify the number of objects that are displayed
limit the wavelength range of the spectra
produce spectra with or without error bars

A list of all settings can be found in the detailed documentation above.

aXeSIM

A dedicated simulation software package aXeSIM has been developed as part of the support for the WFC3 slitless spectroscopic modes. However, aXeSIM can also simulate spectral images for other slitless spectroscopic instruments, and has been used, for example, in the planning of observations with JWST NIRISS and Euclid NISP.

aXeSIM is available in two forms:

As part of the regular aXe Anaconda installation
To verify an aXeSIM installation and to help users taking their first steps with aXeSIM, we have composed test packages for some HST instruments with slitles spectroscopic observing modes.
For the simulation of WFC3 UVIS and IR grism spectra, aXeSIM is also available via the aXeSIMweb application.

In addition to the aXeSIM task that is available in Anaconda, aXeSIM can also be run as a web application for the WFC3 grism modes. The web application offers only a subset of all the options available in aXeSIM in order to simplify use for new, less experienced or occasional users of the software. The user can select a WFC3 slitless mode, choose or adjust additional parameters and start the simulation. Once ready, the simulated images and data can then be downloaded from the web.

Disclaimer: aXeSIMweb was developed independently of the STScI WFC3 spectroscopic Exposure Time Calculators (ETC). While every effort has been made to ensure that aXeSIMweb provides similar estimates of detected counts as the ETC, differences will inevitably result. The STScI ETC is to be taken as the official tool and should be used to determine exposure times for the purposes of critical time justification in HST Phase I and II proposals.

What's New?

2009-07-15
Release of aXeSIM version 1.4
2008-06-26
The sensitivity files of WFC3 grism modes, which are used in aXeSIM, were updated to reflect the status of Thermal Vac 3 (TV3) ground calibrations. While the UV channel sensitivity remained largely unchanged since TV2, there are relative sensitivity gains ranging from 25 to 70% for the IR grisms due to the installation of the new IR detector (IR4; FPA165).
2008-02-14
Release of aXeSIM version 1.1. The changes with respect to version 1.0 are an improvement in the computational speed and a small change in the definition of the simulated beams, which corrects a problem for the simulations of large objects with WFC3/G280.

Documentation

The aXeSIM Users’ Manual describes how aXeSIM works and how to perform simulations with it.

Date	Details
2018-05-14	A Jupyter-based cookbook using aXe and WFC3 slitless data is available at GitHub. The cookbook shows how to run aXe and aXedrizzle and comes with test data.
2018-05-11	aXe is currently distributed as part of Anaconda and the STScI astroconda channel.
2015-05-29	aXe is currently distributed as part of Ureka.
2013-05-30	STSDAS 3.16 has been released, but aXe users should note that the version of aXe bundled there ONLY supports processing with Multidrizzled direct images. Support of processing with AstroDrizzled products is currently provided only in versions of aXe in IRAFX with release dates after 2013-03-28. aXe software release notes can be found below. The WFC3 IR grism cookbook has been updated to accurately reflect availability of AstroDrizzle support in IRAFX.
2013-04-01	Fix for iolprep bug reported on 2013-03-13 in aXe v2.x available in IRAFX (the STSDAS testing environment).
2013-03-13	Bug Report 2013-03-13 (IRAFX 2013-03-07; aXe v2.x iolprep)
2013-03-11	The WFC3 IR grism cookbook has been updated to reflect changes made to the aXe software to accommodate AstroDrizzle image products. An updated manual will be available shortly.
2013-03-07	aXe was originally designed to support Multidrizzle style astrometric information. Current (aXe v2.3) and earlier versions of the software rely on output files from the Multidrizzle process to transform object pixel coordinates between the user supplied "drizzled" direct image and corresponding spectral images. We are currently working on updating aXe to accommodate the use of new Astrodrizzle products, and maintain back-compatability with Multidrizzle.

aXe is PyRAF/IRAF software which is comprised of several tasks and is distributed as a subpackage of the Space Telescope Data Analysis System (STSDAS) under analysis.slitless.axe. New releases of STSDAS become available approximately twice a year. More frequent updates and bug fixes are available in IRAFX, the development version of STSDAS, which is released weekly. Download and installation information for the STSDAS package can be found at STSDAS Home, and for IRAFX at http://stsdas.stsci.edu/irafx.

Updates Being Tested in IRAFX

Version	Date	Release Summary	Detailed Release Notes
aXe-2.X	2013-03-28	STScI: iolprep bug fix implemented	aXe was originally designed to support Multidrizzle style astrometric information. Current (aXe v2.3) and earlier versions of the software rely on output files from the Multidrizzle process to transform object pixel coordinates between the user supplied "drizzled" direct image and corresponding spectral images. We are currently working on updating aXe to accommodate the use of new Astrodrizzle products, and maintain back-compatability with Multidrizzle. Updates to iolprep and fcubeprep are being tested. Bug Report 2013-03-13 (IRAFX 2013-03-07): The output catalogs produced by iolprep inadvertently drops the first entry for the x,y image coordinates with the result that the remaining x,y coordinate entries are then offset by one. Bug fix for iolprep reported on 2013-03-13
aXe-2.X	2013-03-07	STScI: Updates for Astrodrizzle in iolprep and fcubeprep

Version

Date

Release Summary

Detailed Release Notes

aXe-2.X

2013-03-28

STScI: iolprep bug fix implemented

aXe was originally designed to support Multidrizzle style astrometric information. Current (aXe v2.3) and earlier versions of the software rely on output files from the Multidrizzle process to transform object pixel coordinates between the user supplied "drizzled" direct image and corresponding spectral images. We are currently working on updating aXe to accommodate the use of new Astrodrizzle products, and maintain back-compatability with Multidrizzle. Updates to iolprep and fcubeprep are being tested.

Bug Report 2013-03-13 (IRAFX 2013-03-07): The output catalogs produced by iolprep inadvertently drops the first entry for the x,y image coordinates with the result that the remaining x,y coordinate entries are then offset by one.

Bug fix for iolprep reported on 2013-03-13

aXe-2.X

2013-03-07

STScI: Updates for Astrodrizzle in iolprep and fcubeprep

Official Releases through STSDAS

Version	Date	Release Summary	Detailed Release Notes
aXe-2.3	2011-08-23	ST-ECF: replacement of IRAF calls with equivalent Python, PyFITS, and numpy procedures; bug fixes	Part of STSDAS 3.14, which was released in August 2011. Fixed a bug in "drzprep" that was causing 2D stamp images to have the wrong size (too short in the x/dispersion direction) when setting DRZRESOLA in the .config file to value smaller than the physical pixel size, in order to oversample the drizzled image. Fixed a bug in "drzprep" that was causing the calculated errors for spectra extracted from sub-sampled 2D images with axedrizzle to be too high by about a factor of 6. Modified the behavior of low-level routines called by the "axeprep" task so that it no longer quits with an error if it encounters an object in the input object list (IOL) with mag=99 (which is put there by SExtractor when it can't compute a magnitude for an object). It will now only issue a warning message and continue with processing. Entries for these objects will be ignored anyway by later aXe processing step due to the fact that their magnitudes exceed the MMAG_EXTRACT limit in the .config file. Fixed a bug in the "axedrizzle" task to prevent a divide-by-zero error for object spectra that have all of their pixels rejected due to excessive contamination. Upgraded the syntax of Python "raise" error statements in several modules to conform to new Python 2.7 syntax rules. Modified the drizzle method used by the "axedrizzle" task to check for input/output file names that are >80 characters long, in order to prevent the IRAF drizzle task from crashing on long file names. Upgraded several methods within the "axeprep" task to replace calls to IRAF tasks imcopy, imrename, imarith, imexpression, and imstatistics with equivalent Python, PyFITS, and numpy procedures. axeprep is now completely free of IRAF tasks. Removed all calls to IRAF tasks from the drizzleobjects module and replaced them with equivalent Python and PyFITS file utilities and numpy array arithmetic. Removed all calls to IRAF tasks imcopy, imrename, and imexpression from the fcubeobjs module and replaced them with equivalent Python, PyFITS, and numpy operations. fcubeprep still contains a call to the IRAF/STSDAS dither package task "blot".
aXe-2.1	2010-07-15	ST-ECF: optimized and tested for WFC3 grism data	Part of STSDAS 3.13, which was released in January 2011. Also released in July 2010 as the external PyRAF/IRAF package taxe21. Extensively used and tested in the reduction of Wide Field Camera 3 (WFC3) grism data. Extended aXedrizzle to be able to detect deviant pixel values in the same way MultiDrizzle does for direct imaging. This new functionality was developed for and tested on WFC3 data. Using it for other instruments, such as ACS, cannot be guaranteed to produce meaningful results; tuning of the input parameters is strongly advised. Starting with aXe version 2.1 and aXeSIM version 1.4, the two packages have been united on the code level. As a consequence, the package taxe21 automatically contains taxesim14 and vice versa.

Version

Date

Release Summary

Detailed Release Notes

aXe-2.3

2011-08-23

ST-ECF: replacement of IRAF calls with equivalent Python, PyFITS, and numpy procedures; bug fixes

Part of STSDAS 3.14, which was released in August 2011.

Fixed a bug in "drzprep" that was causing 2D stamp images to have the wrong size (too short in the x/dispersion direction) when setting DRZRESOLA in the .config file to value smaller than the physical pixel size, in order to oversample the drizzled image.

Fixed a bug in "drzprep" that was causing the calculated errors for spectra extracted from sub-sampled 2D images with axedrizzle to be too high by about a factor of 6.

Modified the behavior of low-level routines called by the "axeprep" task so that it no longer quits with an error if it encounters an object in the input object list (IOL) with mag=99 (which is put there by SExtractor when it can't compute a magnitude for an object). It will now only issue a warning message and continue with processing. Entries for these objects will be ignored anyway by later aXe processing step due to the fact that their magnitudes exceed the MMAG_EXTRACT limit in the .config file.

Fixed a bug in the "axedrizzle" task to prevent a divide-by-zero error for object spectra that have all of their pixels rejected due to excessive contamination.

Upgraded the syntax of Python "raise" error statements in several modules to conform to new Python 2.7 syntax rules.

Modified the drizzle method used by the "axedrizzle" task to check for input/output file names that are >80 characters long, in order to prevent the IRAF drizzle task from crashing on long file names.

Upgraded several methods within the "axeprep" task to replace calls to IRAF tasks imcopy, imrename, imarith, imexpression, and imstatistics with equivalent Python, PyFITS, and numpy procedures. axeprep is now completely free of IRAF tasks.

Removed all calls to IRAF tasks from the drizzleobjects module and replaced them with equivalent Python and PyFITS file utilities and numpy array arithmetic.

Removed all calls to IRAF tasks imcopy, imrename, and imexpression from the fcubeobjs module and replaced them with equivalent Python, PyFITS, and numpy operations. fcubeprep still contains a call to the IRAF/STSDAS dither package task "blot".

aXe-2.1

2010-07-15

ST-ECF: optimized and tested for WFC3 grism data

Part of STSDAS 3.13, which was released in January 2011.

Also released in July 2010 as the external PyRAF/IRAF package taxe21.

Extensively used and tested in the reduction of Wide Field Camera 3 (WFC3) grism data.

Extended aXedrizzle to be able to detect deviant pixel values in the same way MultiDrizzle does for direct imaging. This new functionality was developed for and tested on WFC3 data. Using it for other instruments, such as ACS, cannot be guaranteed to produce meaningful results; tuning of the input parameters is strongly advised.

Starting with aXe version 2.1 and aXeSIM version 1.4, the two packages have been united on the code level. As a consequence, the package taxe21 automatically contains taxesim14 and vice versa.

Spectral Extraction and Visualization Software

Extraction and Calibration

aXe

Software Access

Documentation

What's New?

Installation

Visualization

aXe2web

Software Access

Documentation

Simulation

aXeSIM

What's New?

Documentation

News

Release Notes

Updates Being Tested in IRAFX

Official Releases through STSDAS