Non-HST Filter Systems

In addition to the HST instruments, filters, and gratings, the graph table also contains entries for various standard bandpass from photometric systems that are not specific to HST. Actively supported systems (i.e., their data files are updated on CRDS as needed) are as tabulated below. Non-HST filters are specified using the name of the filter system, followed by the desired band name. For example:

>>> import stsynphot as stsyn
>>> bp ='cousins,i')  
>>> bp ='stromgren,u')  

If the name of the filter system is omitted for any of the common UBVRIJHK filters, the defaults are Johnson UBV, Cousins RI, and Bessell JHK. For example, the following are equivalent:

>>> bp ='v')  
>>> bp ='johnson,v')  

System Name

Band Name


r i


nuv fuv


u v b r i j k


u v b r i


u g r i z


u v b y

Comparison with Observed Non-HST Photometry

There are two issues that are sometimes overlooked when comparing synthetic photometry from stsynphot with observed photometry using a non-HST system.

Firstly, one should be careful whether the throughput data have been defined for a photon-counting or an energy-integrating detector. stsynphot always assumes that a throughput are of the former. In particular, some authors in the past have defined throughput curves for photomultipliers as if these detectors were energy integrators, which they are not. Such curves have to be converted into photon-counting form before they can be correctly used by stsynphot (Maiz Apellaniz 2006). Using the wrong definition can lead to errors of a few percent for broad-band filters.

Secondly, many systems (e.g., Johnson UBV) use Vega as a reference spectrum, but have been calibrated using secondary standards, leading to the existence of finite zero points. In some systems (e.g. Stromgren), those zero points are not even close to 0.0 for some filters. The table below defines the zero point corrections for ground-based filter systems from measurements of zero points collected from the respective literature; These values should be added to the VEGAMAG magnitude in stsynphot before they are compared with the observed data:



Zero point (mag)


Johnson and Landolt



Bohlin & Gilliland (2004)



Maiz Apellaniz (2006)



Maiz Apellaniz (2006)

Cousins and Landolt



Holberg & Bergeron (2006)



Holberg & Bergeron (2006)




Holberg & Bergeron (2006)



Maiz Apellaniz (2006)



Maiz Apellaniz (2006)



Maiz Apellaniz (2006)

The existence of these issues has led CRDS to divide the non-HST photometric systems into supported (as mentioned above) and not supported.

Systems for which there are analyses in the literature that deal with the issues mentioned above are as follow. CRDS Team is reasonably confident that the possible systematic errors in the stsynphot results for these systems are small:


The 2MASS JHK\(_s\) throughputs are taken from Cohen et al. (2003). These are normalized relative spectral response curves and include the throughputs of all of the appropriate optics from the 2MASS optical system, as well as the atmosphere above the two 2MASS telescopes.

Zero point reference fluxes for 2MASS reproduced from the IPAC 2MASS website are included here for reference (pay special attention to the units):


λ (µm)

Bandwidth (µm)

Fnu ref (Jy)

Flambda ref (W/cm^2/µm)


1.235 ± 0.006

0.162 ± 0.001

1594 ± 27.8

3.129E-13 ± 5.464E-15


1.662 ± 0.009

0.251 ± 0.002

1024 ± 20.0

1.133E-13 ± 2.212E-15


2.159 ± 0.011

0.262 ± 0.002

666.7 ± 12.6

4.283E-14 ± 8.053E-16

To use the 2MASS throughputs:

>>> import stsynphot as stsyn
>>> bp ='2mass,j')  
>>> bp ='2mass,h')  
>>> bp ='2mass,ks')  

(Note: 2MASS throughput curves were added to the TMG file in January 2020. Users must use a TMG/TMC file and associated throughput tables delivered after this date to use the 2MASS OBSMODEs.)


The Cousins RI throughputs are taken from Bessell (1983). They have been transformed into photon-counting form. For example:

>>> import stsynphot as stsyn
>>> bp ='cousins,i')  


The GALEX FUV and NUV throughputs were provided by Tom Barlow on behalf of the GALEX project, as described in Morrissey et al. (2007). They were measured on the ground in units of effective area, and were divided by the full area of the GALEX primary mirror (\(1963.495 \; \text{cm}^{2}\)) to convert them to the dimensionless transmission values required by stsynphot. Therefore, these curves represent the true total throughput, including obscuration by the secondary mirror, reflectivity of the mirrors, sensitivity of the detector, and so forth. For example:

>>> import stsynphot as stsyn
>>> bp ='galex,fuv')  


The throughput data for the Johnson UBV bands were obtained from Maiz Apellaniz (2006), while the RIJK bands from Johnson (1965). For example:

>>> import stsynphot as stsyn
>>> bp ='johnson,v')  


The Landolt (1983) UBVRI system is made up of the Johnson UBV and the Cousins RI bandpass. For example:

>>> import stsynphot as stsyn
>>> bp ='landolt,v')  


The Sloan Digital Sky Survey (SDSS) ugriz filter throughputs were provided by Sebastian Jester on behalf of the SDSS team, as described in Gunn et al. (2001). The filter curves are shown in the SDSS filter response plot. For example:

>>> import stsynphot as stsyn
>>> bp ='sdss,g')  

The throughput data give the system photon response to point sources of the 2.5-m SDSS survey telescope, including extinction through an airmass of 1.3 at Apache Point Observatory (to which all SDSS photometry is referenced). Originally, the ugriz system was intended to be identical to the \(u^{\prime} g^{\prime} r^{\prime} i^{\prime} z^{\prime}\) system described in Fukugita et al. (1996) and defined by the standard star system in Smith et al. (2002). However, in the course of processing the SDSS data, an unpleasant discovery was made that the filters in the 2.5-m telescope have significantly different effective wavelengths from the filters in the USNO telescope, which was used to observe the \(u^{\prime} g^{\prime} r^{\prime} i^{\prime} z^{\prime}\) standards; The difference originates from the USNO filters being exposed to ambient air, while the survey-telescope filters live in the vacuum of the survey camera. Therefore, it became necessary to distinguish between the primed and unprimed SDSS filter sets.

The response curves in r and i are slightly different for large extended sources (larger than about 80 pixels in size) because the extended IR scattering wings in these bands, which do not affect the photometry of point sources, begin to be included. The modified curves are shown in an updated SDSS system response plot.

The SDSS photometry is intended to be on the AB system (Oke & Gunn 1983), by which a 0-magnitude object should have the same counts as a source of \(F_{\nu} = 3631 \; \text{Jy}\) (except that it used the so-called “asinh” magnitudes defined by Lupton et al. 1999 instead of conventional Pogson magnitudes). However, this is known not to be exactly true, such that the photometric zero points are slightly off the AB standard. The SDSS team continues to work to pin down these shifts. Their estimate, based on comparison to the STIS standards of Bohlin et al. (2001) and confirmed by SDSS photometry and spectroscopy of fainter hot white dwarfs, is that the u band zero point is in error by 0.04 mag, \(u_{\text{AB}} = u_{\text{SDSS}} - 0.04 \; \text{mag}\), and that g, r, and i are close to AB; These statements are certainly not precise to better than 0.01 mag. The z band zero point is not as certain (as of January 2005), but there is mild evidence that it may be shifted by about 0.02 mag in the sense that \(z_{\text{AB}} = z_{\text{SDSS}} + 0.02 \; \text{mag}\).

See Holberg & Bergeron (2006) for a calibration of SDSS magnitudes using Vega as a reference spectrum. Further information about SDSS photometric calibration and the “asinh” magnitude system can be found at SDSS Photometric Flux Calibration webpage.


The Stromgren uvby throughputs are taken from Maiz Apellaniz (2006). For example:

>>> import stsynphot as stsyn
>>> bp ='stromgren,y')  


Phase C estimates of the Roman integrated system throughputs have been taken from the Roman Reference Information page at GSFC. For example:

>>> import stsynphot as stsyn
>>> bp ='roman,wfi,f062')  

Only the Wide Field Instrument (WFI) is currently supported with the following modes:




f062, f087, f106, f129, w146, f158, f184, f213


grism, grism0, prism

The grism component gives the throughput for the first order grism spectrum. This is the mode that most users should select. The zeroth order grism throughput (grism0) is provided only as a reference.

At this time, the estimated throughputs do not differentiate between the different sensor chip assemblies (SCAs). SCA-dependent throughputs will be delivered at a later time.

Deprecated Systems

As of March 2006, some non-HST bandpass systems were deprecated, as tabulated below. They remain accessible by stsynphot, but mostly for backward compatibility. There will be no updates from CRDS, so use these at your own risk.

System Name

Band Name


1550 1550n 1800 2200 2500 3300


f336w f439w f547m f555w f569w f606w f622w f675w f702w f725lp f785lp f791w f814w f850lp f1042m


j h k


88 97 100-102 104-106 109-119 121 122 125 127-130 132 136 140 141 145 149 152 154-157 159-161 163-166 168-170 172-179 181-183 185 186 189 192-194 196-199 201-207 209-234 236-242 244 247 248 253 254 260 264 265 537 538


j h k


j h k


v b l u w


The Astronomical Netherlands Satellite (ANS) system is a set of UV filters used by the satellite, as described in van Duinen et al. (1975). For example:

>>> import stsynphot as stsyn
>>> bp ='ans,1550')  


The Baum filter set is a set of 15 broadband and intermediate-band filters that are copies the ones onboard WF/PC-1 that were used as part of a ground-based calibration campaign for the instrument. In order to match the response of the in-flight bandpass as closely as possible, the throughputs for the Baum filters have been multiplied by the spectral response curve of the ground-based CCD (measured in the laboratory) and twice by the spectral reflectance of aluminum (Harris et al. 1991). For example:

>>> import stsynphot as stsyn
>>> bp ='baum,f336w')  


The Bessell JHK filter curves are taken from Bessell & Brett (1988), Table IV. These curves include the mean atmospheric transmission equivalent to 1.2 air masses of a standard KPNO atmosphere. For example:

>>> import stsynphot as stsyn
>>> bp ='bessell,k')  


The 530 ESO band throughput tables were received from Jan Koornneef in 1990. For example:

>>> import stsynphot as stsyn
>>> bp ='eso,198')  


The Kitt Peak National Observatory (KPNO) JHK filter curves are taken from the tracings of the Simultaneous Quad Infrared Image Device (SQIID) filter set, which were provided by Richard Joyce from the observatory. For example:

>>> import stsynphot as stsyn
>>> bp ='kpno,k')  


The Steward Observatory JHK filter curves are from data provided by Marcia Rieke from the observatory. For example:

>>> import stsynphot as stsyn
>>> bp ='steward,k')  


The throughput data for the Walraven VBLUW bands are from Lub & Pel (1977), Table 6. For example:

>>> import stsynphot as stsyn
>>> bp ='walraven,v')