🔭 astrophysics

Chromatic Effects Across the Roman Focal Plane: Implications for Supernova Photometry and Measurements of Cosmological Parameters

This paper demonstrates that while coherent filter wavelength calibration uncertainties are negligible for the Nancy Grace Roman Space Telescope, spatially varying focal plane array wavelength shifts introduce significant redshift-dependent biases in Type Ia supernova photometry that must be characterized to within 20% to prevent systematics from overwhelming statistical uncertainties in cosmological parameter measurements.

Original authors: Rujuta Purohit, Dillon Brout, Daniel Scolnic, Richard Kessler, Jillian Paulin, Stefano Casertano, Nora F. Sherman, Rebekah Hounsell, Eric R. Switzer, Benjamin M. Rose, Masao Sako, Lauren Aldoroty, Ken

Published 2026-07-13

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Rujuta Purohit, Dillon Brout, Daniel Scolnic, Richard Kessler, Jillian Paulin, Stefano Casertano, Nora F. Sherman, Rebekah Hounsell, Eric R. Switzer, Benjamin M. Rose, Masao Sako, Lauren Aldoroty, Kene Anumba, Maria Acevedo, David Rubin, Russell E. Ryan Jr., The Roman Supernova Cosmology Project Infrastructure Team

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the Nancy Grace Roman Space Telescope as a giant, ultra-precise camera floating in space, designed to take pictures of thousands of exploding stars called Type Ia supernovae. These stars are the universe's "standard candles"—they all shine with roughly the same true brightness. By measuring how dim they look from Earth, astronomers can calculate how far away they are and how fast the universe is expanding.

But here's the catch: to measure that brightness perfectly, the camera's "eyes" (its filters) need to be absolutely perfect. If the filters are even slightly off, the measurements get distorted, and the whole map of the universe could end up wrong.

This paper is like a rigorous stress test for Roman's camera. The authors ran massive computer simulations to see what happens if the camera's filters have tiny, invisible flaws. They found two main types of "glitches" and tested how badly they would mess up our understanding of dark energy.

The Two Glitches

First, there's the "FPA-dependent shift." Think of the camera's sensor as a giant mosaic made of 18 different tiles (detectors). The authors simulated a scenario where each tile sees colors slightly differently—like if one tile was slightly more sensitive to blue light and another to red light. In their simulations, these shifts ranged from +6 to -80 Ångströms (a unit of light wavelength).

Second, there's the "coherent shift." This is like if the entire camera was calibrated slightly wrong from the factory, shifting every single filter by the exact same tiny amount (0.06%). This represents the best-case scenario where we've done our homework, but a tiny, uniform error remains.

The Big Discovery: Location Matters

The paper's main finding is a bit of a warning: Where the star lands on the camera matters a lot.

When the authors simulated the "FPA-dependent shift" (where different tiles see colors differently), they found it introduced a sneaky bias that changed depending on how far away the star was. It's as if the camera told a lie that got bigger the further out you looked.

If they left this uncorrected, the simulations showed it would throw off the measurement of dark energy's equation of state ( $w_0$ ) by about -0.06 and the evolution of dark energy ( $w_a$ ) by 0.236.
To put that in perspective, the statistical "noise" or uncertainty the mission expects is only 0.025 for $w_0$ and 0.114 for $w_a$ .
In other words, the error from this glitch is larger than the random noise. If they don't fix it, the survey becomes "systematics-limited," meaning the camera's flaws, not the lack of data, would be the bottleneck stopping them from discovering new physics.

To keep the survey accurate, the authors suggest that they must characterize these tile-to-tile differences to within 20% of their current size. If they can do that, the error drops below the noise floor.

The Good News: Uniform Errors are Boring

On the bright side, the "coherent shift" (the uniform factory error) turned out to be harmless. Even with a 0.06% shift across the whole camera, the bias in the dark energy evolution ( $w_a$ ) was a tiny -0.0004. This is so small it's basically zero compared to the statistical noise.

This means the pre-launch calibration precision they've already achieved is good enough to handle uniform errors. The real villain isn't a general miscalibration; it's the fact that the 18 different tiles on the sensor aren't perfectly identical.

A New Way to Check the Camera

The paper also suggests a clever way to double-check the camera once it's in space: look at stars.

Stars are steady, predictable light sources. The authors simulated how stars of different colors would look through the flawed filters.
They found that redder stars would show a bigger shift in brightness than bluer ones, creating a clear pattern.
By measuring this pattern on the actual camera in orbit, astronomers could map out exactly how each of the 18 tiles is behaving and correct the data in real-time. This would act as a "self-healing" mechanism for the camera's calibration.

The Bottom Line

This paper doesn't say the mission is doomed; it says the mission needs a specific, careful fix. The authors simulated these effects and found that while a uniform error is negligible, the unevenness across the camera's 18 detectors is a major systematic threat.

If they ignore the differences between the tiles, the survey will hit a wall where the errors are bigger than the data's natural randomness. But if they use the star-based calibration method to correct for these tile-specific quirks, Roman should be able to measure the universe's expansion with the incredible precision it was built for. The paper concludes that these chromatic effects are a "required component" of the analysis—they can't be ignored, but they can be managed.

Technical Summary: Chromatic Effects Across the Roman Focal Plane

Problem Statement
Calibration uncertainties represent the leading systematic error in cosmological analyses utilizing Type Ia supernovae (SNe Ia). For the Nancy Grace Roman Space Telescope (Roman), specifically its High-Latitude Time Domain Survey (HLTDS), the primary source of chromatic effects is the spectrophotometric response of the 18 detector chips (SCAs) comprising the Wide Field Instrument (WFI). While atmospheric contributions are absent in space-based observations, spatial variations in filter transmission across the focal plane and uncertainties in absolute filter wavelength calibration introduce wavelength-dependent biases. These biases manifest as redshift-dependent errors in inferred distance moduli, potentially compromising constraints on dark energy parameters ( $w_0$ and $w_a$ ). Previous ground-based surveys have demonstrated that uncorrected chromatic effects can introduce biases at the millimagnitude (mmag) level, comparable to dominant systematic uncertainties. This paper quantifies the impact of these effects on Roman's cosmological goals, which aim for sub-percent precision in dark energy characterization.

Methodology
The authors employed the SuperNova ANAlysis (SNANA) software and the Pippin pipeline to simulate SNe Ia observations consistent with the "Sundial" cadence and observing strategy for the HLTDS. The simulation covered a redshift range of $0.1 < z < 3$ , generating light curves using the SALT3 model with population parameters drawn from DES-SN5YR and Popovic et al. (2023).

To isolate chromatic effects, three independent simulations were run using identical random seeds to ensure the same SN Ia population was analyzed under different filter conditions:

Nominal: No wavelength shifts.
FPA-Dependent Shifts: Realistic spatial variations in filter effective wavelengths across the 18 SCAs, derived from thermal vacuum (TVAC) testing data (Switzer et al. 2025). These shifts were applied by assigning SCA-specific filter transmission curves to each simulated observation.
Coherent Shifts: A uniform systematic offset of 0.06% applied to the central wavelengths of all filters across the entire focal plane. This scenario was tested in two modes: a fixed shift for all filters (COH-FIXED) and random uncorrelated shifts drawn from a Gaussian distribution for each filter (COH-RANDOM), representing the residual calibration uncertainty after FPA-dependent corrections.

Light curves were fitted using SALT3-NIR, and distance moduli were calculated using the Tripp estimator. Cosmological parameters ( $\Omega_M, w_0, w_a$ ) were inferred using the BEAMS with Bias Corrections (BBC) method and the wfit code, comparing results from shifted simulations against the nominal case.

Key Contributions

Quantification of FPA-Dependent Bias: The study provides the first detailed quantification of how spatial variations in filter transmission across Roman's focal plane affect SN Ia photometry. It demonstrates that FPA-dependent shifts range from +6 to -80 Å, introducing significant redshift-dependent biases.
Differentiation of Systematic Sources: The paper distinguishes between spatially varying effects (correctable per object) and coherent global offsets (residual calibration uncertainty), showing that their impacts on cosmological parameters differ drastically.
Stellar Validation Pathway: The authors propose and demonstrate that stellar photometry across the focal plane can serve as an independent probe to validate and refine chromatic corrections, leveraging the linear relationship between stellar color and chromatic magnitude offsets.

Results

Single-Epoch Photometry: FPA-dependent shifts induce magnitude differences ranging from +0.4 to -17.8 mmag depending on the filter and redshift. Notably, the F184 filter exhibits a positive shift due to its large wavelength offset (-39 Å) sampling the steep red end of the SN spectrum. Coherent shifts result in smaller, uniform corrections of roughly +3.3 to -5.3 mmag.
Redshift Dependence: The dominant systematic is a linear redshift-dependent bias in distance moduli. For FPA-dependent shifts, the bias is $\Delta\mu(z) = (1.56 \pm 1.00)z$ mmag. For coherent shifts, the bias is significantly smaller at $\Delta\mu(z) = (0.60 \pm 0.33)z$ mmag.
Nuisance Parameters: Chromatic effects primarily impact the peak magnitude term ( $m_B$ ) in the Tripp equation, with a redshift slope of $\Delta m_B(z) = (1.18 \pm 0.27)z$ mmag for FPA shifts. The stretch ( $\alpha x_1$ ) and color ( $\beta c$ ) terms show negligible redshift evolution.
Cosmological Constraints:
- FPA-Dependent Shifts: If left uncorrected, these shifts introduce biases of $\Delta w_0 \approx -0.06$ and $\Delta w_a \approx 0.236$ . These values exceed the forecast statistical uncertainties ( $\sigma_{stat, w_0} = 0.025$ , $\sigma_{stat, w_a} = 0.114$ ), rendering the survey systematics-limited.
- Coherent Shifts: A 0.06% coherent offset produces negligible bias ( $\Delta w_0 \approx 0.0001$ , $\Delta w_a \approx -0.0004$ ), remaining well below statistical noise.
Correction Requirements: To suppress the FPA-induced systematic to a level subdominant to statistical errors ( $\Delta |w_a| \lesssim 0.05$ ), the FPA-dependent wavelength shifts must be characterized and corrected to within approximately 20% of their current amplitude.

Significance and Claims
The paper asserts that chromatic effects are a required component of SN Ia cosmology with Roman. The results establish that while pre-launch calibration precision (0.06%) is sufficient to keep coherent global offsets subdominant, the spatial variations across the focal plane are significant enough to bias cosmological parameters if not properly modeled. The authors conclude that detector-specific filter curves must be employed to recover unbiased cosmological constraints. Furthermore, they highlight that stellar observations can provide an on-orbit validation mechanism to refine these corrections, potentially tightening cosmological constraints by 10–15% for $w_a$ . The study underscores that as statistical uncertainties decrease with larger samples, even biases currently below $0.5\sigma$ will become critical, necessitating rigorous characterization of chromatic effects.

More like this