A bandpass can be constructed by one of the following methods:

It has various photometric properties and these main components:

  • name (Description of the bandpass; Also accessible via __str__())
  • throughput
  • wave (a.k.a. waveset)
  • waveunits (see Wavelength Units)

To evaluate its transmission at a given wavelength, use its sample() or __call__() method, as given in the following example. Internally, evaluation uses numpy.interp().

>>> bp = S.ObsBandpass('acs,wfc1,f555w')
>>> bp.sample(5000)
>>> bp(5000)

Observation Mode

The observation mode (obsmode) parameter defines the passband, i.e., the wavelength-dependent sensitivity curve of the photometer or spectrophotometer. It is controlled by Reference Data. It can also be used to specify color index.

The obsmode is usually given as a string of keyword arguments to the ObsBandpass class. The list of keywords identify the light path through the telescope and the instrument, or through a non-HST filter system. For example, "wfc3,uvis1,f555w" creates a bandpass that is calculated by taking the product of the individual throughputs of the Wide Field Camera 3 (WFC3) UVIS Channel 1, the F555W filter, the detector sensitivity, and the HST Optical Telescope Assembly (OTA). There are special considerations for OTA and COSTAR. Another example, "johnson,v" creates a Johnson V bandpass that does not account for HST optics. A complete list of obsmode keywords can be found in Appendix B.

Quick Example

Create a bandpass for HST/ACS instrument with its WFC1 detector and F555W filter:

>>> bp_acs = S.ObsBandpass('acs,wfc1,f555w')

To see which throughput tables are being used, as set by Reference Data:

>>> bp_acs.showfiles()

Create a bandpass for Johnson V:

>>> bp_v = S.ObsBandpass('johnson,v')

Compare them in a plot:

>>> plt.plot(bp_acs.binset, bp_acs(bp_acs.binset), 'b',
...          bp_v.wave, bp_v.throughput, 'g--')
>>> plt.xlim(4000, 7000)
>>> plt.xlabel(bp_acs.waveunits)
>>> plt.ylabel('throughput')
>>> plt.legend([, 'Johnson V'], loc='best')
Bandpasses for ACS WFC1 F555W and Johnson V.

Pixel and Wavelength Ranges

The pixel_range() and wave_range() methods can be used to calculate the pixel and wavelength ranges, respectively, spanned by the observation mode given its binset, if available. For example:

>>> bp = S.ObsBandpass('wfc3,ir,f105w')

To calculate the number of pixels covered from 8600.5 to 12400.5 Angstroms:

>>> bp.pixel_range([8600.5, 12400.5])

To calculate starting and ending wavelengths in Angstroms covered by 3800 pixels centered at 10500 Angstroms:

>>> bp.wave_range(10500.0, 3800)
(8600.5, 12400.5)

Thermal Background

For IR detectors (e.g., NICMOS and WFC3), thermal background can be calculated using the thermback() method. The thermal component is defined by thermtable in Reference Data. For non-IR detectors, calling this method would raise NotImplementedError. For example:

>>> bp = S.ObsBandpass('wfc3,ir,f105w')
>>> bp.thermback()
>>> bp = S.ObsBandpass('acs,wfc1,f555w')
>>> bp.thermback()
NotImplementedError: No thermal support provided for acs,wfc1,f555w


A box-shaped bandpass is a rectangular window centered on a given wavelength with a given width, both in Angstroms. It is defined as:

\[\begin{split}\mathrm{throughput} = \left \{ \begin{array}{ll} 1 & : x_0 - w/2 \geq x \geq x_0 + w/2 \\ 0 & : \mathrm{else} \end{array} \right.\end{split}\]


  • \(x_{0}\) is the central wavelength
  • \(x\) is the wavelength array
  • \(w\) is the width of the box

The example below creates and plots a box-shaped bandpass centered at 6000 Angstroms with a width of 100 Angstroms:

>>> bp = S.Box(6000, 100)
>>> plt.plot(bp.wave, bp.throughput)
>>> plt.ylim(0, 1.1)
>>> plt.axvline(6000, ls='--', color='k')
>>> plt.xlabel(bp.waveunits)
>>> plt.ylabel('throughput')
>>> plt.title(
Box bandpass.


UniformTransmission generates a uniform (flat) bandpass that has a constant throughput at any wavelength value.

The example below creates and samples a bandpass with a uniform transmission value of 0.8:

>>> bp = S.UniformTransmission(0.8)
>>> bp.sample(5000)
>>> bp.sample(np.arange(1000, 10000))
array([ 0.8,  0.8,  0.8, ...,  0.8,  0.8,  0.8])

From File

A bandpass can also be defined using a FITS or ASCII table containing columns of wavelength and throughput. See File I/O for details on how to create such tables.

The example below loads a bandpass from FITS table:

>>> filename = os.path.join(
...     os.environ['PYSYN_CDBS'], 'comp', 'acs', 'acs_wfc_ccd2_019_syn.fits')
>>> bp = S.FileBandpass(filename)
>>> bp.throughput
array([  0.00000000e+00,   0.00000000e+00,   1.87380003e-27, ...,
         4.14354995e-09,   0.00000000e+00,   0.00000000e+00])
>>> bp.sample(5100)
>>> bp.sample(2100)

Tutorial 10: Spectrum from Custom Text File offers hints on how to load a bandpass from an ASCII table of any format.

From Arrays

To create a bandpass from arrays, use ArraySpectralElement (also callable as pysynphot.ArrayBandpass). Note in the example below that the bandpass is explicitly tapered at both ends to avoid extrapolation; Also, unlike source spectrum, its negative throughput value is not automatically set to zero:

>>> w = np.array([999, 1000, 2000, 3000, 3001])  # Angstroms
>>> t = np.array([0, 0.1, -0.2, 0.3, 0])
>>> bp = S.ArrayBandpass(w, t, name='MyBandpass')
>>> bp.throughput
array([ 0. ,  0.1, -0.2,  0.3,  0. ])
>>> bp.sample(2500)
>>> bp.sample(4000)

Overlap Checks

To check whether the wavelength range of other bandpass or spectra is defined everywhere within the main bandpass, you can use the check_overlap() method, which returns "full", "partial", or "none". The example below checks whether the main bandpass overlap with another bandpass of the same detector but with a different filter, and with a box-shaped one:

>>> bp = S.ObsBandpass('wfc3,ir,f105w')
>>> other_bp = S.ObsBandpass('wfc3,ir,f110w')
>>> bp.check_overlap(other_bp)
>>> box_bp = S.Box(10000, 10000)
>>> bp.check_overlap(box_bp)

To check if the lack of overlap is insignificant, you can use the check_sig() method. The example below shows that the partial overlap above is not a concern:

>>> bp.check_sig(box_bp)