###################################################################
### Simulated NIRCam imaging and reduction for CEERS Pointing 5 ###
###################################################################

*** We recommend starting with part1/ceers_nircam_reduction.ipynb.
    This Jupyter notebook provides info and examples of running raw CEERS 
    NIRCam images through all stages of the JWST Calibration Pipeline as 
    well as custom processing steps we've developed for the simulated data.
    Note total disk space after running the pipeline: ~15 GB

*** If you wish to reduce the full CEERS 5 pointing (90 raw images in total),
    see part2/README.txt (and below for more info). 
    Note total disk space after running the pipeline: ~140-150 GB


############
## Summary
Instrument: NIRCam
Mode: Imaging
Source of Simulations: MIRAGE v2.1.0
Calibration Pipeline Used: jwst v1.2.3
Target: CEERS Pointing 5 (in parallel with prime NIRSpec MSA)

Filters: SWC F115W, F150W, F200W; LWC F277W, F356W, F444W
Readout: MEDIUM8
Observation specification: 9 groups, 1 integration
Dithers: 3 dithers, steps specified by primary instrument
APT (version 2021.2): CEERS ERS 1345

Contact: Micaela Bagley (mbagley@utexas.edu)


#######################
## Simulated data set
CEERS 5 is a single NIRCam pointing, observed in imaging mode in parallel 
to prime NIRSpec MSA observations. The pointing is observed with 6 filters:
short wavelength channel (SWC) filters F115W, F150W, and F200W, and long 
wavelength channel (LWC) filters F277W, F356W, and F444W. There is a three 
point dither in each filter, defined by the 3 slitlet nods of the NIRSpec
primary observations.

We have simulated NIRCam images using MIRAGE version 2.1.0, with input 
sources taken from a mock catalog created using the Santa Cruz semi-analytic 
model for galaxy evolution. The images were simulated using pointing and 
XML files based on the CEERS APT (with modifications required for MIRAGE to 
simulate the custom primary-parallel dither patterns planned for CEERS
observations).

In total there are 90 raw simulated images:
 - 24 each for the SWC filters (8 SWC detectors * 3 dithers)
 - 6 each for the LWC filters (2 LWC detectors * 3 dithers)


#############
### Mosaics
We provide fully reduced mosaics in all 6 filters in the mosaics directory.

The mosaics are called ceers5_[filt]_i2d.fits, where filt is one of:
SWC filters: f115w, f150w, f200w
LWC filters: f277w, f356w, f444w

The mosaics are gzipped and unzip to 5 GB (SWC) and 1.2 GB (LWC).

These images were processed through jwst version 1.2.3, with custom steps 
developed to handle the specifics of the simulated data. The mosaics are on 
pixel scales of 0.015"/pix (SWC) and 0.03"/pix (LWC). 

***Note that the images are NOT pixel-aligned. Pixel-aligned images across 
   all filters will be part of our next data release. 

We also include preliminary photometric catalogs, produced using the 
default options for the SourceCatalog step of the Stage 3 pipeline. 
The SourceCatalog parameters are not yet optimized, and so the photometric
catalogs are not final. We include them here as an example of the 
pipeline outputs. See the mosaics directory for a comparison of the input 
and recovered photometry.


###################
## Data Reduction
We provide instructions on how to reduce the simulated data in two parts:
 - part1: All input files necessary to reduce exposures for one detector
          of the F115W imaging and one detector of the F277W imaging, 
          including a jupyter notebook that demonstrates running the 
          jwst pipeline steps and our custom processing steps. 
 - part2: Batch scripts to reduce all 90 raw images through all pipeline
          and custom steps. 

Note: The reference file mapping used to create the simulated images has 
since been deprecated. The main difference between the old mapping and the 
one we now recommend (`jwst_0674.pmap`) is in the bad pixel maps. Therefore, 
a small number of pixels in your reduced images may be incorrectly flagged 
as bad or incorrectly assumed to be good.

#########################
## part1.tar.gz, ~650 MB
Untars to ~1.2 GB, takes up ~8 GB including all calibration pipeline 
outputs, i.e., after running all steps in the notebook (plus an 
additional ~7 GB of downloaded JWST reference files)

This directory contains everything needed to produce 2 partial mosaics, 
each combining 3 exposures of a single NIRCam detector:
 - F115W: 3 dithered exposures of the A1 detector
 - F277W: 3 dithered exposures of the A5 detector
including all parameter files, association files, and custom reference files.

See the Jupyter notebook ceers_nircam_reduction.ipynb for information, 
including installation and set-up, examples of running the pipeline using 
the three calling methods, and detailed descriptions of each processing step. 
While we have processed the mosaics with JWST Calibration Pipeline version 
1.2.3, we recommend using version 1.3.3 with the Jupyter notebook.

#########################
## part2.tar.gz, ~8.3 GB
Untars to ~13 GB, takes up **~95 GB** after performing all processing steps

This directory contains everything needed to produce 6 full mosaics,
each combining 3 exposures and all NIRCam detectors:
 - F115W, F150W, F200W: Each filter has 8 detectors and 3 dithered exposures
 - F277W, F356W, F444W: Each filter has 2 detectors and 3 dithered exposures

We provide batch scripts for each pipeline stage/processing step that 
call all necessary parameter, association, and reference files as 
described in part1/ceers_nircam_reduction.ipynb. 

See part2/README.txt for instructions. 
 - rundetector1: Runs the 90 raw *uncal.fits images through calwebb_detector1
 - remstriping.py: Removes 1/f noise striping patterns from countrate images
 - runimage2: Runs the 90 countrate *rate.fits images through calwebb_image2
 - applyflat.py: Removes a large-scale feature present on the A5 detector
 - runskymatch: Runs sky subtraction (SkyMatchStep) on the 90 *cal.fits images
 - runimage3: Runs the 6 filter associations through calwebb_image3