Physical origin and functional form of exposure-long trends in JWST/NIRSpec time-series

Characterize the physical origin and derive an appropriate functional form for the exposure-long decreasing, nonlinear background trends observed in JWST/NIRSpec PRISM Bright Object Time Series exposures, in order to inform robust trend modeling strategies for high-precision transit analyses including exomoon searches.

Background

The authors find that each ~10 hour exposure exhibits a gradually decreasing, nonlinear trend that strongly impacts exomoon inference because the trend timescale is similar to the transit duration. They emphasize that accurate trend modeling is critical and that heuristic functions (polynomials, exponentials, or GPs) currently dominate practice.

They explicitly state that the cause and functional form of these exposure-long trends are not well-understood, leaving analysts to guess models without physical guidance, and recommend multiple trend models to assess sensitivity and encourage deeper study of the systematic’s origin and best practices.

References

To begin with, like the HST visit long trends, the cause and functional form of these trends is not well-understood and thus we are left with little alternative but to essentially guess heuristic models.

— A JWST Transit of a Jupiter Analog: II. A Search for Exomoons  (2511.15317 - Kipping et al., 19 Nov 2025) in Section 6.3 (Lessons for Future Searches)