JWST Operations Pipeline Build 11.3 Release Notes

A list of all relevant changes to build 11.3 of the JWST Operations Pipeline is provided in this article. 

On this page

Build 11.3 was released as a build candidate in March 2025, and installed as an operations build in May 2025. The information provided here focuses mainly on updates to the JWST Science Calibration Pipeline and associated reference files, but includes information relevant to changes in other subsystems that might affect what users see in their data.


Major updates

Words in bold are GUI menus/
panels or data software packages; 
bold italics are buttons in GUI
tools or package parameters.

Major pipeline updates in build 11.3

  • Added new clean_showers option to the straylight step to remove residual cosmic ray showers for MIRI MRS data. See Shower and Snowball Artifacts.
  • Significantly improved spectral extraction for NIRSpec BOTS data by ensuring the spectral extraction follows the curved spectral trace.
  • Updated the clean_flicker_noise step to handle flat field variations for NIRISS data.
  • Updated the extract_1d step for MIRI MRS; it is now no longer necessary to set the optional ifu_rfcorr parameter to apply residual fringe correction.  The 1D spectral tables now always contain columns for the extracted spectrum both with (RF_FLUX) and without (FLUX) 1D residual fringe correction applied.  Users should examine both and determine which is more appropriate for their science use case.
  • Significantly improved runtime and memory usage of the jump step in calwebb_detector1. Speed gains will be particularly noticeable for time-series observations.
  • Added a new algorithm  to the emicorr step for correcting striping artifacts in MIRI data due to electromagnetic interference (EMI). The new algorithm improves runtime and performance for data with fewer than 10 groups, and is most useful for LRS slitless time series observations. This new algorithm is optional in build 11.3 (see emicorr step arguments) and is under study to be made default in upcoming build 12.0.
  • Fixed a series of issues that had caused inaccurate S_REGION footprints for MIRI LRS fixed-slit observations.
  • Log messages created during pipeline processing are now stored in an extension of the output data files.
  • Updated all pipeline software for compatibility with numpy 2.0.

Major reference file updates in build 11.3

Build 11.3 is tied to the CRDS reference file context jwst_1363.pmap. Significant reference file deliveries included in this context include:

  • Updated 1D spectral extraction reference files for NIRCam grism TSO data to improve the extraction region (jwst_1327.pmap).
  • Updated flat fields for NIRCam longwave imaging and grism data (jwst_1330.pmap).
  • Updated multiple reference files for NIRSpec, MIRI, and NIRCam to account for continued evolution in the hot pixel population.
  • New NIRSpec parameter reference files for the dark_current step that improve the accuracy of the pipeline-estimated uncertainties for faint targets (jwst_1331.pmap).
  • Updated MIRI imaging flat fields to avoid masking out the LRS slit region in TA confirmation images (jwst_1338.pmap).
  • Updated the MIRI imaging calwebb_image2 parameter reference file to reduce the amount of residual artifacts from the background subtraction step (jwst_1344.pmap).
  • Multiple deliveries in support of upcoming NIRCam DHS observing mode.

Known issues in build 11.3

There were four known issues with previous build 11.2: (1) missing data in the MIRI LRS slit region for TA confirmation images in calibrated data products, (2) incorrect WCS information in the S_REGION for MIRI LRS slit data, (3) file saving issues for extremely long spectral TSO observations with hundreds of thousands of integrations, and (4) incorrect associations for some NIRSpec IFU data with leak calibration exposures. Issues (1) and (2) have been fixed in build 11.3.

Known issues in build 11.3 include:

  • Processing of recent extremely long spectral time-series observations have encountered a FITS file-save issue - the "x1dints" products have one extension per integration, and for some recent observations this can entail hundreds of thousands of extensions. STScI is exploring changes to the data structure of these products to improve runtime and data accessibility.
  • Some NIRSpec IFU leakcal observations are not being included in associations properly by the default pipeline. A workaround is to use the new NIRSpec IFU pipeline notebook to process data.

Data affected by this build


Provided below is a list of science data products in MAST generated by the JWST Science Calibration Pipeline that may be most significantly affected by the updates in build 11.3. The changes listed in the release notes below may affect other modes as well, but many of these will be transparent to the MAST user.

The data and pipeline stages most affected include:

  • Stage 1 processing for NIRCam, NIRISS, and MIRI (bad pixel mask updates).
  • Stage 2 processing for NIRSpec BOTS, MIRI LRS, and new options to improve offline rereduction of MIRI LRS and MRS data.
  • Stage 3 processing for NIRSpec BOTS, MIRI MRS, and NIRCam grism TSO.

Release notes

Build 11.3 of the JWST Operations Pipeline was released as a candidate build on April 10, 2025 and installed as an operations build (11.3) for the Barbara A. Mikulski Archive for Space Telescopes (MAST) on May 20, 2025. The build includes changes to the JWST Science Calibration Pipeline released in version 1.18.0 of the jwst calibration pipeline package via the PyPI repository. This release also includes changes to the JWST Science Data Processing subsystem (SDP) version JWSTDP-2025.X.0 which is the DMS sub-system that creates raw (uncal) products used as input to the calibration pipeline.

General changes

  • The log messages generated during processing are now stored in output datamodels under the model.cal_logs  attribute.
  • The pipeline now installs with numpy >=2.0 by default.
  • The pipeline Python version compatibility has been updated; Python 3.10 support has been dropped, and Python 3.13 support has been added.
  • The fits_generator submodule was unused and has been removed.

Pipeline modules

  • In calwebb_spec2, added an additional "_fs"  suffix to the interpolated flat output from the flat_field step for NIRSpec fixed slits processed as part of a combined MOS/FS observation, to avoid overwriting the MOS flat output.

Product associations

  • For imprint images in spec2 associations, the strict constraint on dither position matching has been removed. Instead, extra imprints intended for other targets or other dither positions are pruned if necessary during association finalization.
  • Spec2 associations for NIRCam pure parallel WFSS associations now select the long wavelength imaging product, if present, rather than the first image listed.

Datamodels

  • The error array in 4D ramp models was unused and has been removed.
  • The datamodel.read()  and write() methods have been deprecated; use __init__  and save() instead.
  • NIRSpec MSA slit and spectral datamodels now include keywords detailing the slit quadrant and slit area scales.

Step Changelist

Changes to specific steps in the pipelines are listed below. 

assign_wcs step

  • Calculate slit scaling factors for open area to full slit pitch for NIRSpec MOS data, and store the values as slit_xscale and slit_yscale in slit and spectral datamodels.
  • For MIRI LRS data, fix incorrect S_REGION values.
  • Fixed crash in transform from MSA to detector coordinates for NIRSpec imaging WCS with OPAQUE filter setting.

background step

  • Store background scaling factor for WFSS data.

barshadow step

  • Update barshadow correction to scale correctly to NIRSpec MOS data in long slit mode. Also correct the scale factors for error propagation when the correction is inverted.
  • Extrapolate barshadow corrections to cover wavelength gaps in reference files.

clean_flicker_noise step

  • Add optional flat handling to avoid fitting and removing flat structure.

emicorr step

  • Refactor step to improve runtime and reduce memory usage.
  • Implement a new EMIcorr algorithm that optimizes the phase and amplitude of a reference waveform of known frequency. The best phase and amplitude are the ones that, when applied to the waveform and subtracted from the data, produce ramps for the pixels that are as linear as possible. Also, fix a minor bug in the existing algorithm that prevented saving intermediate reference files.

extract_1d step

  • Add options for PSF-based optimal extraction for point sources in MIRI LRS fixed slit exposures.
  • Expanded the use_source_posn option to calculate a source trace from WCS and expected source positions for unresampled NIRSpec and MIRI LRS fixed slit data. Added the step parameter position_offset to allow an additional aperture offset in pixels.
  • For MIRI MRS data, add residual fringe corrected columns to extracted spectrum table. Also apply aperture correction to the residual fringe corrected flux and surface brightness columns.

extract_2d step

  • Add ability to select multiple slit names or source ids in extract_2d.

flatfield step

  • Remove assignment of DO_NOT_USE to all NON_SCIENCE flagged pixels in the flatfield step for MIRI, NIRCam, and NIRISS.

guider_cds step

  • Remove hard-wired reference file information from step and apply style updates for increased readability.

imprint step

  • Handle imprint matching separately for science and dedicated background targets.

jump step

  • Refactor step for improved runtime, reduced memory usage and improved readability.

outlier_detection step

  • Made in-memory processing the default for outlier detection imaging observations.
  • Fix bad input type to median computation for coron data.

pixel_replace step

  • Change from the default BFGS algorithm to Nelder-Mead when calling scipy.minimize within the fit_profile approach to pixel replacement in order to fix numpy 2.0 compatibility issues. Additionally, add safety catch to ensure that pixel replacement profile fitting doesn't attempt to scale based on noise.
  • Base output filenames on input filenames when the step is called outside the pipeline, in order to create sensible names when the input is a list of models.

ramp_fitting step

  • Add code to enable users to select firstgroup and lastgroup parameters when performing ramp fitting. It works by setting the DO_NOT_USE bit in the "GROUPDQ" extension for groups outside the selected range. Added a unit test and updated the docs.

refpix step

  • Add basic reference pixel corrections for multistripe data.
  • Add "sirskernel" to the reference types for the step, so that override_sirskernel is available as a step parameter.
  • Fix an indexing bug in the SIRS kernel convolution code.

resample step

  • Fix failure to correctly load single datamodels from file.

residual_fringe step

  • Refactor for maintainability, fix intermediate filenames when input datamodel is read from memory, and stop producing an unused intermediate output file.

skymatch step

  • Add option to pass in user-defined sky levels.

straylight step

  • Add an additional optional step to the MIRI MRS straylight correction code to measure and remove residual cosmic ray showers using information from the inter-slice pixels.

wavecorr step

  • Fixed recording cal_step status for wavecorr for NIRSpec BOTS mode.

JWST Science Data Processing subsystem changes

  • Changes were made to how the EFFEXPTM, EFFINTTM, and DURATION keywords are set, and redundant keyword TEXPTIME was removed (see Understanding Exposure Times).
  • Obsolete keyword R_DRZPAR  was removed now that the resample step no longer uses "drizpar" files.
  • Additional checks have been added to stop "New Science" notifications being sent for reprocessed data products.
  • Updates were made to the APERNAME keyword to add values for new NIRCam multi-stripe subarray modes.



Notable updates


Originally published