spec_analysis.features¶

The features is responsible for measuring the properties of individual spectral features. Individual classes are used to represent separate calculations of feature properties. These classes may be used individually, however, it is recommended you use the ObservedFeature class which combines all of the calculations into a single object.

`spec_analysis.features.FeatureArea`	Represents the area calculation for a spectroscopic feature
`spec_analysis.features.FeaturePEW`	Represents the pEW calculation for a spectroscopic feature
`spec_analysis.features.FeatureVelocity`	Represents the velocity calculation for a spectroscopic feature
`spec_analysis.features.ObservedFeature`	Represents a spectral observation spanning a single absorption feature

Usage Examples¶

Included in the ObservedFeature class is the ability to calculate the area, pseudo equivalent width (PEW), and flux of a given feature. Feature classes are instantiated using the rest frame wavelength, flux, and binned flux (e.g., the flux through a median filter).

import numpy as np
from scipy.ndimage.filters import median_filter

from spec_analysis.features import find_peak_wavelength
from spec_analysis.simulate import gaussian

wave = np.arange(1000, 2000)
flux, flux_err = gaussian(wave, stddev=100)
bin_flux = median_filter(flux, size=10)

feature = ObservedFeature(wave, flux, bin_flux)

Feature properties can be accessed using attributes:

# From the area calculation
area = feature.area            # Feature area

# From the PEW calculation
continuum = feature.continuum  # Sudo continuum flux per wavelength
norm_flux = feature.norm_flux  # Normalized flux per wavelength
pew = feature.pew              # Pseudo equivalent width

# From the velocity calculation
velocity = feature.velocity    # velocity of the feature in km / s
amp = feature.gauss_amplitude  # Amplitude of fitted gaussian
avg = feature.gauss_avg        # Average of fitted gaussian
stddev = feature.gauss_stddev  # Standard deviation of fitted gaussian
offset = feature.gauss_offset  # y-offset of fitted gaussian

Note that the velocity is determined by fitting the normalized flux using an inverted Gaussian of the form

\[- A * e^{(-(x - \mu)^2 / (2 \sigma^2)} + c\]

This function is accessible programmatically as spec_analysis.features.gaussian, or the evaluated fit can be determined using

gauss_fit = feature.gaussian_fit()

API Documentation¶

spec_analysis.features.gaussian(x, depth, avg, std, offset)[source]¶

Evaluate a negative gaussian

f = -depth * e^(-((x - avg)^2) / (2 * std ** 2)) + offset

Parameters:	x (ndarray) – Values to evaluate the gaussian at depth (float) – Amplitude of the gaussian avg (float) – Average of the gaussian std (float) – Standard deviation of the gaussian offset (float) – Vertical offset
Returns:	The evaluated gaussian

class spec_analysis.features.FeatureArea(wave, flux, bin_flux)[source]¶

Represents the area calculation for a spectroscopic feature

__init__(wave, flux, bin_flux)[source]¶

Calculates the area of a spectroscopic feature

Parameters:	wave (ndarray) – The wavelength values of the feature flux (ndarray) – The flux values for the feature bin_flux (ndarray) – The binned flux values for the feature

area¶

The area of the feature

Area is determined between the sudo continuum of the binned flux and the flux values from the non-binned flux.

class spec_analysis.features.FeaturePEW(wave, flux, bin_flux)[source]¶

Represents the pEW calculation for a spectroscopic feature

__init__(wave, flux, bin_flux)[source]¶

Calculates the pEW of a spectroscopic feature

Parameters:	wave (ndarray) – The wavelength values of the feature flux (ndarray) – The flux values for the feature bin_flux (ndarray) – The binned flux values for the feature

continuum¶: Array of values for the fitted sudo continuum

norm_flux¶: The flux normalized by the sudo continuum

pew¶

Calculate the pseudo equivalent-width of the feature

Returns:	The pseudo equivalent-width of the feature

class spec_analysis.features.FeatureVelocity(wave, flux, bin_flux, rest_frame)[source]¶

Represents the velocity calculation for a spectroscopic feature

__init__(wave, flux, bin_flux, rest_frame)[source]¶

Calculates the pEW of a spectroscopic feature

Parameters:	wave (ndarray) – The wavelength values of the feature flux (ndarray) – The flux values for each feature

gauss_amplitude¶: The fitted gaussian amplitude

gauss_avg¶: The fitted gaussian average

gauss_offset¶: The fitted gaussian offset

gauss_stddev¶: The fitted gaussian standard deviation

gaussian_fit()[source]¶

Evaluate the gaussian fit of the normalized flux for feature wavelengths

Returns:	An array of normalized flux values

velocity¶

Calculate the velocity of a feature

Fit a feature with a negative gaussian and determine the feature’s velocity. Returned value is np.nan if the fit fails.

Returns:	The velocity of the feature in km / s

class spec_analysis.features.ObservedFeature(wave, flux, bin_flux, rest_frame)[source]¶

Represents a spectral observation spanning a single absorption feature

__init__(wave, flux, bin_flux, rest_frame)[source]¶

Represents a spectral observation spanning a single absorption feature

Parameters:	wave (ndarray) – The wavelength values of the feature flux (ndarray) – The flux values for each feature