Merge pull request #8 from radicamc/dev

Dev

README.md

@@ -8,14 +8,14 @@ ___
 
 **exoTEDRF** (formerly supreme-SPOON) is an end-to-end pipeline for JWST exoplanet time series observations.  
 
-Currently supports: **NIRISS/SOSS**, **NIRSpec/BOTS** (coming soon), **MIRI/LRS** (in development).
+Currently supports: **NIRISS/SOSS** and **NIRSpec/BOTS**. **MIRI/LRS** is in development.
 
 ## Documentation
 An installation guide, code documentation, and tutorials are included in the [exoTEDRF documentation](https://exotedrf.readthedocs.io/en/latest/index.html).
 
 ## Citations
-If you make use of this code in your work, please cite [Radica et al. (2023)](https://ui.adsabs.harvard.edu/abs/2023MNRAS.524..835R/abstract) 
-and [Feinstein et al. (2023)](https://ui.adsabs.harvard.edu/abs/2023Natur.614..670F/abstract).  
+If you make use of this code in your work, please cite [Radica (2024)](), as well as [Radica et al. (2023)](https://ui.adsabs.harvard.edu/abs/2023MNRAS.524..835R/abstract) 
+and [Feinstein et al. (2023)](https://ui.adsabs.harvard.edu/abs/2023Natur.614..670F/abstract) for its first uses.  
 
-A list of other works you should consider citing depending on which functionalities you are using can be found [here]().
+A list of other works you should consider citing depending on which functionalities you are using can be found [here](https://exotedrf.readthedocs.io/en/latest/content/citations.html).
 

changelog.md

@@ -1,6 +1,20 @@
 # Changelog
 All notable changes to this project will be documented in this file.
 
+### [2.0.0] -- 2024-07-05
+#### Added
+- Support for NIRSpec/BOTS observations. 
+- Functions to refine SOSS timestamps for limb asymmetry studies.
+- Light curve fitting with exoUPRF.
+- Updates to documentation.
+- NIRSpec/G395H reduction tutorial. 
+- Misc. bug fixes.
+
+#### Removed
+- Light curve plotting routines
+- juliet dependencies. 
+- Light curve fitting tutorial.
+
 ### [1.4.0] -- 2024-03-08
 #### Added
 - Package name changed to exoTEDRF.

docs/content/citations.rst

@@ -1,18 +1,19 @@
 Citations
 ==========
 
-If you make use of this code in your work, please cite `Radica et al. (2023) <https://ui.adsabs.harvard.edu/abs/2023MNRAS.524..835R/abstract>`_ and `Feinstein et al. (2023) <https://ui.adsabs.harvard.edu/abs/2023Natur.614..670F/abstract>`_.
+If you make use of this code in your work, please cite `Radica (2024) <xxx>`_, as well as `Radica et al. (2023) <https://ui.adsabs.harvard.edu/abs/2023MNRAS.524..835R/abstract>`_ and `Feinstein et al. (2023) <https://ui.adsabs.harvard.edu/abs/2023Natur.614..670F/abstract>`_ for its first uses.
+
+For NIRSpec capabilities, please also cite `PLACEHOLDER <xxx>`_.
 
 Additional Citations
 --------------------
 If you use the ATOCA extraction algorithm for NIRISS/SOSS, please also cite `Radica et al. (2022) <https://ui.adsabs.harvard.edu/abs/2022PASP..134j4502R/abstract>`_
 and `Darveau-Bernier et al. (2022) <https://ui.adsabs.harvard.edu/abs/2022PASP..134i4502D/abstract>`_.
 
 If you make use of the light curve fitting routines, also include the following citations for
-`juliet <https://ui.adsabs.harvard.edu/abs/2019MNRAS.490.2262E/abstract>`_,
-`batman <https://ui.adsabs.harvard.edu/abs/2015PASP..127.1161K/abstract>`_,
-`dynesty <https://ui.adsabs.harvard.edu/abs/2020MNRAS.493.3132S/abstract>`_, and
-`Kipping et al. (2013) <https://ui.adsabs.harvard.edu/abs/2013MNRAS.435.2152K/abstract>`_ for the limb-darkening sampling.
+`exoUPRF <https://zenodo.org/records/12628067>`_,
+`batman <https://ui.adsabs.harvard.edu/abs/2015PASP..127.1161K/abstract>`_, and
+`dynesty <https://ui.adsabs.harvard.edu/abs/2020MNRAS.493.3132S/abstract>`_.
 
 If you use Gaussian Processes please cite `celerite <https://ui.adsabs.harvard.edu/abs/2017AJ....154..220F/abstract>`_,
 and if you use ExoTiC-LD for limb darkening priors cite `Grant & Wakeford (2022) <https://doi.org/10.5281/zenodo.7437681>`_.

docs/content/installation.rst

@@ -24,15 +24,6 @@ The default pip installation only includes Stages 1 to 3. Stage 4 can be include
     pip install exotedrf[stage4]
 
 
-.. attention::
-    The `radvel <https://github.com/California-Planet-Search/radvel>`_ package may fail to build during the installation of Stage 4. If so, simply run:
-
-    .. code-block:: bash
-
-        pip install cython
-
-    and then proceed with the **exoTEDRF** installation as before.
-
 Alternatively, the package can be grabbed from GitHub (inlcludes all pipeline stages as well as tutorial notebooks, etc.) via:
 
 .. code-block:: bash

docs/content/usage.rst

@@ -1,9 +1,9 @@
 exoTEDRF Usage Guide
 ====================
 
-Currently supported instruments/modes: **NIRISS/SOSS**, **NIRSpec/BOTS** (coming soon), **MIRI/LRS** (in development).
+Currently supported instruments/modes: **NIRISS/SOSS** and **NIRSpec/BOTS**. **MIRI/LRS** is in development.
 
-The pipeline is divided into four stages, which closely mirror the STScI pipeline:
+The pipeline is divided into four stages, which closely mirror the STScI JWST calibration pipeline:
 
  * Stage 1: Detector-Level Processing
  * Stage 2: Spectroscopic Processing
@@ -20,8 +20,11 @@ Below are several tutorials that will walk you through the basics of JWST data a
    :maxdepth: 2
 
    notebooks/tutorial_niriss-soss
-   notebooks/tutorial_light-curve-fitting
+   notebooks/tutorial_nirspec-g395h
 
+exoTEDRF also has the capabilities for basic transit and eclipse light curve fitting in Stage4 through the exoUPRF library.
+If you are looking for some guidance on basic light curve fitting for JWST data, you can check out the exoUPRF tutorial notebooks
+`here <https://exouprf.readthedocs.io/en/latest/content/usage.html#tutorial-notebooks>`_.
 
 Scripting
 ---------

docs/index.rst

@@ -5,7 +5,7 @@ exoTEDRF: the EXOplanet Transit and Eclipse Data Reduction Framework!
 
 **exoTEDRF** (formerly supreme-SPOON) is an end-to-end pipeline for JWST exoplanet time series observations.
 
-Currently supported instruments/modes are: **NIRISS/SOSS**, **NIRSpec/BOTS** (coming soon), **MIRI/LRS** (in development).
+Currently supported instruments/modes are: **NIRISS/SOSS** and **NIRSpec/BOTS**. **MIRI/LRS** is in development.
 
 .. toctree::
    :maxdepth: 2

docs/requirements.txt

@@ -18,7 +18,5 @@ pastasoss
 pyyaml
 dynesty==1.2.2
 exotic_ld
+exouprf
 h5py
-
-#TODO: docs for Stage 4 will not build as they require juliet, which in turn requires radvel. radvel is not installing
-#TODO: on readthedocs for some reason. It's installing fine (either as part of juliet or on its own) in a fresh conda env.

exotedrf/extra_functions.py

@@ -183,3 +183,104 @@ def make_smoothed_2d_lightcurve(spec, baseline_ints, nint, dimx, filename,
     # Save file.
     suffix = 'lcestimate_2d_o{}.npy'.format(order)
     np.save(filename + suffix, ref_file)
+
+
+def refine_soss_timestamps(mid_int_times, centroids, subarray='SUBSTRIP256',
+                           outfile=None):
+    """SOSS read times are long compared to other instruments meaning that
+    there are several seconds of time difference between when red and blue
+    wavelengths are read. For limb asymmetry studies, this can be important.
+    This function refines the default mid integration times to yield a
+    seperate timestamp for each wavelength, which can then be outputs as 2D
+    maps, or appended to an existing spectrum file.
+    Note, when binning these time stamps it is important to also consider the
+    relative contributions of each wavelength to the bin via the brightness of
+    the stellar spectrum.
+
+    Parameters
+    ----------
+    mid_int_times : array-like(float)
+        Default mid-integration timestamps.
+    centroids : dict
+        Centroids dictionary.
+    subarray : str
+        SOSS subarray identifier, either "SUBSTRIP256" or "SUBSTRIP96".
+    outfile : str, None
+        Spectrum file to which to append output.
+
+    Returns
+    -------
+    int_times_o1 : ndarray(float)
+        Integration times for each wavelength in order 1.
+    int_times_o2 : ndarray(float), None
+        Integration times for each wavelength in order 2.
+    """
+
+    tpix = 1e-5  # Read time for individual pixel
+    tgap = 1.2e-4  # Gap time between columns
+    # Note using these times results in a frame time which is ~5ms shorter
+    # than the actual frame time of 5.494s. Not sure why, since these are
+    # apparently the correct pixel and gap times. The frame time formula
+    # (Equ. 3 in Albert+ 2023) gives the 5.494s (for SUBSTRIP256).
+
+    # Get readout pattern and subarray dimensions.
+    if subarray == 'SUBSTRIP256':
+        dimx, dimy = 2048, 256
+    elif subarray == 'SUBSTRIP96':
+        dimx, dimy = 2048, 96
+    else:
+        raise ValueError('Unknown subarray: {}'.format(subarray))
+
+    # Create map of readout times of individual pixels.
+    time = 0
+    time_map = np.zeros((dimy, dimx))
+    for x in range(dimx - 1, -1, -1):
+        if x != dimx - 1:
+            time += tgap
+        for y in range(dimy - 1, -1, -1):
+            time += tpix
+            time_map[y, x] = time
+    # Assume default mid-integration time is the time at the frame center.
+    time_map -= time_map[dimy // 2, dimx // 2]
+
+    # Get time shift for each wavelength.
+    t_o1 = []
+    for x in range(dimx):
+        ypos1 = round(centroids['ypos o1'][x], 0).astype(int)
+        t_o1.append(time_map[ypos1, x])
+    t_o1 = np.array(t_o1)
+
+    int_times_o1 = np.repeat(mid_int_times[:, np.newaxis], dimx, axis=1)
+    int_times_o1 = int_times_o1 - t_o1 / 3600 / 24
+
+    # Only do order 2 for SUBSTRIP256.
+    if subarray != 'SUBSTRIP96':
+        t_o2 = []
+        for x in range(dimx):
+            ypos2 = round(centroids['ypos o2'][x], 0).astype(int)
+            if 0 <= ypos2 < dimy:
+                t_o2.append(time_map[ypos2, x])
+            else:
+                t_o2.append(np.nan)
+        t_o2 = np.array(t_o2)
+
+        int_times_o2 = np.repeat(mid_int_times[:, np.newaxis], dimx, axis=1)
+        int_times_o2 = int_times_o2 - t_o2 / 3600 / 24
+    else:
+        int_times_o2 = None
+
+    # If a spectrum file is passed, tack on the new timestamps.
+    if outfile is not None:
+        file = fits.open(outfile)
+        hdr = fits.Header()
+        hdr['EXTNAME'] = "Time O1"
+        hdr['UNITS'] = "MJD"
+        file.append(fits.ImageHDU(int_times_o1, header=hdr))
+        if subarray != 'SUBSRIP96':
+            hdr = fits.Header()
+            hdr['EXTNAME'] = "Time O2"
+            hdr['UNITS'] = "MJD"
+            file.append(fits.ImageHDU(int_times_o2, header=hdr))
+        file.writeto(outfile, overwrite=True)
+
+    return int_times_o1, int_times_o2