JPL's mission page for SPHEREx says:

The mission will create a map of the entire sky in 96 different color bands, far exceeding the color resolution of previous all-sky maps.

Wikipedia's SPHEREx Spacecraft/telescope says:

The triple mirror telescope will have an aperture diameter of 20 centimeters with a 3.5° x 11° field of view and six 2k x 2k mercury cadmium telluride (HgCdTe) photodetector arrays.12, 8 Each 2K x 2K focal plane array is covered with a linear variable filter, providing narrow-band response with a band center that varies along one axis of the array. SPHEREx obtains spectra through multiple exposures, placing a given source at multiple positions in the field of view, where it is measured at multiple wavelengths by repointing the spacecraft.8



Question: How exactly does SPHEREx implement 96 spectral bands with a linear variable filter across its focal plane?

Does the telescope constantly rotate like GAIA and use fancy array readout techniques to produce the 96 discrete spectral bands in software?

Or does it move discretely such that nearby objects are actually recorded at slightly different central wavelengths since the filter is linear.

How does this work exactly?

It seems SPHEREx has adopted the KISS principle:

  • Low-Risk Implementation
  • No Moving Parts
  • Single Observing Mode
  • Large Technical & Scientific Margins
  • Follows successful CIT/JPL mgt. model of NuSTAR

1 Answer 1


I think it's more like your second suggestion. A linear variable filter is a filter where the spectral response changes continuously in one direction across the filter. For an instrument like SPHEREx, with the filter mounted in the focal plane, they need to scan in a direction parallel to the filter's variable direction to build up a spatial/spectral cube.

According to this graphic in their 2014 paper, they do ~100 second integrations separated by 8.8 arcsecond slews.

Interestingly though, they don't scan through their entire 7-degree field of view before slewing to a new position. After 4-8 8.8' slews (which covers about 1 degree of right ascension), they slew 60 degrees to a different part of the sky and start observing there. They return to the original FOV the next day, and complete the entire 7-degree FOV over a period of a week. Observing 1 degree of RA in this manner keeps the FOV of SPHEREx in sync with the 1 degree of orbital precession it experiences in its sun-synchronous orbit.

Spherx observing visualization


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .