JP-3588: Use Pastasoss datamodel for NIRISS SOSS transform solution (#…

…8763)

CHANGES.rst

@@ -122,6 +122,12 @@ emicorr
 
 - Removed unnecessary copies, and created a single copy at step.py level. [#8676]
 
+extract_1d
+----------
+
+- Updated NIRISS SOSS extraction to utilize ``pastasoss``
+  rotation solution. [#8763]
+
 first_frame
 -----------
 

docs/jwst/extract_1d/arguments.rst

@@ -169,11 +169,6 @@ The ``extract_1d`` step has the following step-specific arguments.
   used when soss_wave_grid is not provided to make sure the computation time or the memory
   used stays reasonable. Default value is 20000.
 
-``--soss_transform``
-  This is a NIRISS-SOSS algorithm-specific parameter; this defines a rotation to
-  apply to the reference files to match the observation. It should be specified as
-  a list of three floats, with default values of None.
-
 ``--soss_tikfac``
   This is a NIRISS-SOSS algorithm-specific parameter; this is the regularization
   factor used in the SOSS extraction. If not specified, ATOCA will calculate a

jwst/extract_1d/extract_1d_step.py

@@ -174,14 +174,13 @@ class Extract1dStep(Step):
     soss_estimate = input_file(default = None)  # Estimate used to generate the wavelength grid
     soss_rtol = float(default=1.0e-4)  # Relative tolerance needed on a pixel model
     soss_max_grid_size = integer(default=20000)  # Maximum grid size, if wave_grid not specified
-    soss_transform = list(default=None, min=3, max=3)  # rotation applied to the ref files to match observation.
     soss_tikfac = float(default=None)  # regularization factor for NIRISS SOSS extraction
     soss_width = float(default=40.)  # aperture width used to extract the 1D spectrum from the de-contaminated trace.
     soss_bad_pix = option("model", "masking", default="masking")  # method used to handle bad pixels
     soss_modelname = output_file(default = None)  # Filename for optional model output of traces and pixel weights
     """
 
-    reference_file_types = ['extract1d', 'apcorr', 'wavemap', 'spectrace', 'specprofile', 'speckernel']
+    reference_file_types = ['extract1d', 'apcorr', 'pastasoss', 'specprofile', 'speckernel']
 
     def process(self, input):
         """Execute the step.
@@ -432,8 +431,7 @@ def process(self, input):
                     return input_model
 
                 # Load reference files.
-                spectrace_ref_name = self.get_reference_file(input_model, 'spectrace')
-                wavemap_ref_name = self.get_reference_file(input_model, 'wavemap')
+                pastasoss_ref_name = self.get_reference_file(input_model, 'pastasoss')
                 specprofile_ref_name = self.get_reference_file(input_model, 'specprofile')
                 speckernel_ref_name = self.get_reference_file(input_model, 'speckernel')
 
@@ -444,7 +442,6 @@ def process(self, input):
                 soss_kwargs['tikfac'] = self.soss_tikfac
                 soss_kwargs['width'] = self.soss_width
                 soss_kwargs['bad_pix'] = self.soss_bad_pix
-                soss_kwargs['transform'] = self.soss_transform
                 soss_kwargs['subtract_background'] = self.subtract_background
                 soss_kwargs['rtol'] = self.soss_rtol
                 soss_kwargs['max_grid_size'] = self.soss_max_grid_size
@@ -458,8 +455,7 @@ def process(self, input):
                 # Run the extraction.
                 result, ref_outputs, atoca_outputs = soss_extract.run_extract1d(
                     input_model,
-                    spectrace_ref_name,
-                    wavemap_ref_name,
+                    pastasoss_ref_name,
                     specprofile_ref_name,
                     speckernel_ref_name,
                     subarray,

jwst/extract_1d/soss_extract/atoca_utils.py

@@ -252,80 +252,6 @@ def get_wv_map_bounds(wave_map, dispersion_axis=1):
     return wave_top, wave_bottom
 
 
-def check_dispersion_direction(wave_map, dispersion_axis=1, dwv_sign=-1):
-    """Check that the dispersion axis is increasing in the good direction
-    given by `dwv_sign``
-    Parameters
-    ----------
-    wave_map : array[float]
-        2d-map of the pixel central wavelength
-    dispersion_axis : int, optional
-        Which axis is the dispersion axis (0 or 1)
-    dwv_sign : int, optional
-        Direction of increasing wavelengths (-1 or 1)
-
-    Returns
-    -------
-    bool_map : array[bool]
-        Boolean 2d map of the valid dispersion direction, same shape as `wave_map`
-    """
-
-    # Estimate the direction of increasing wavelength
-    wave_left, wave_right = get_wv_map_bounds(wave_map, dispersion_axis=dispersion_axis)
-    dwv = wave_right - wave_left
-
-    # Return bool map of pixels following the good direction
-    bool_map = (dwv_sign * dwv >= 0)
-    # The bad value could be from left or right so mask both
-    bool_map &= np.roll(bool_map, 1, axis=dispersion_axis)
-
-    return bool_map
-
-
-def mask_bad_dispersion_direction(wave_map, n_max=10, fill_value=0, dispersion_axis=1, dwv_sign=-1):
-    """Change value of the pixels in `wave_map` that do not follow the
-    general dispersion direction.
-
-    Parameters
-    ----------
-    wave_map : array[float]
-        2d-map of the pixel central wavelength
-    n_max : int
-        Maximum number of iterations
-    fill_value : float
-        Value use to replace pixels that do not follow the dispersion direction
-    dispersion_axis : int, optional
-        Which axis is the dispersion axis (0 or 1)
-    dwv_sign : int, optional
-        Direction of increasing wavelengths (-1 or 1)
-
-    Returns
-    -------
-    wave_map : array[float]
-        The corrected wave_map.
-    convergence flag : bool
-        Boolean set to True if all the pixels are now valid, False otherwise.
-    """
-    # Do not modify the input
-    wave_map = wave_map.copy()
-
-    # Make the correction iteratively
-    for i_try in range(n_max):
-        # Check which pixels are good
-        is_good_direction = check_dispersion_direction(wave_map, dispersion_axis, dwv_sign)
-        # Stop iteration if all good, or apply correction where needed.
-        if is_good_direction.all():
-            convergence_flag = True
-            break
-        else:
-            wave_map[~is_good_direction] = fill_value
-    else:
-        # Did not succeed! :(
-        convergence_flag = False
-
-    return wave_map, convergence_flag
-
-
 def oversample_grid(wave_grid, n_os=1):
     """Create an oversampled version of the input 1D wavelength grid.