🔭 astrophysics

SKYSURF-11: A New Zodiacal Light Model Optimized for Optical Wavelengths

This paper introduces ZodiSURF, an improved zodiacal light model optimized for optical wavelengths (0.3–1.6 micron) using over 5,000 Hubble Space Telescope measurements, which significantly reduces prediction uncertainties to ~4.5% and reveals a residual diffuse light excess suggesting the potential existence of a dim spherical dust cloud.

Original authors: Rosalia O'Brien, Richard G. Arendt, Rogier A. Windhorst, Tejovrash Acharya, Annalisa Calamida, Timothy Carleton, Delondrae Carter, Seth H. Cohen, Eli Dwek, Brenda L. Frye, Rolf A. Jansen, Scott J. Ken

Published 2026-04-29

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Rosalia O'Brien, Richard G. Arendt, Rogier A. Windhorst, Tejovrash Acharya, Annalisa Calamida, Timothy Carleton, Delondrae Carter, Seth H. Cohen, Eli Dwek, Brenda L. Frye, Rolf A. Jansen, Scott J. Kenyon, Anton M. Koekemoer, John MacKenty, Megan Miller, Rafael Ortiz, Peter C. B. Smith, Scott A. Tompkins

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

The Problem: A Glaring Fog in Space

Imagine you are trying to take a photo of a very faint, distant star using a telescope. But right in front of your lens, there is a thick, glowing fog made of dust. This fog is so bright that it washes out the faint star you are trying to see.

In our solar system, this "fog" is called Zodiacal Light. It isn't a cloud of gas, but a massive cloud of tiny dust particles (interplanetary dust) left over from comets and asteroids. When sunlight hits these particles, they scatter the light, creating a diffuse glow that fills the sky.

For decades, astronomers have used a "recipe" (a computer model) called the Kelsall model to calculate how bright this fog is so they can subtract it from their photos and see the faint universe behind it. However, this old recipe was written specifically for infrared light (heat radiation), like a recipe for baking a cake that only works if you use a specific type of oven.

The problem is that modern telescopes like the Hubble Space Telescope (HST) often look at optical light (the colors our eyes can see). When scientists tried to use the old "infrared-only" recipe to predict the brightness of the dust fog in visible light, it failed. It was like trying to use a cake recipe to bake bread; the results were wrong, often predicting the fog was much brighter than it actually was.

The Solution: A New Recipe (ZodiSURF)

The authors of this paper, led by Rosalia O'Brien, created a new, improved model called ZodiSURF.

Instead of guessing how the dust behaves in visible light, they went back to the source. They used over 5,000 measurements from the Hubble Space Telescope, which acted like a giant, high-precision light meter. They looked at the sky in many different colors (from ultraviolet to near-infrared) and at many different angles relative to the Sun.

They used this data to rewrite the "physics" part of the recipe. Specifically, they updated two key ingredients:

Albedo (Reflectivity): How shiny the dust particles are. They found that the dust gets slightly shinier as the light gets redder (longer wavelength).
Phase Function (Scattering Angle): How the dust bounces light. They discovered that the dust doesn't just scatter light randomly; it has a preference for bouncing light forward or backward depending on the color of the light.

By feeding these new, empirically measured rules into the model, ZodiSURF can now accurately predict the brightness of the dust fog in visible light, something the old models could not do.

The Results: A Clearer View

When the team tested their new model against the Hubble data, it worked beautifully.

Accuracy: The model predicts the brightness of the dust fog with an uncertainty of only about 4.5%.
The "Flat" Residual: When they subtracted their new model from the actual sky photos, the leftover "noise" was flat and even. This is a good sign. It means the model correctly accounted for how the dust changes brightness as you look at different angles from the Sun.

The Mystery: A Ghostly Excess

However, there is a small mystery left over. Even after subtracting the new, perfect model of the dust fog and the light from our own galaxy (Diffuse Galactic Light), there is still a tiny bit of extra light left in the photos.

The Excess: The leftover light is about 0.013 MJy sr⁻¹. It's very dim, but it's there.
The "Toy" Explanation: The authors suggest this might be caused by a very faint, spherical shell of dust far away from the Sun (perhaps 4 to 5 times the distance of Earth). Because this shell is so far away and spread out, it looks the same no matter where you look in the sky (isotropic).
The Caveat: They call this a "toy model." It's a simple mathematical sketch to show that such a shell could explain the extra light, but they haven't proven it exists yet. It's like seeing a shadow and guessing it's a cat, but you need more evidence to be sure.

Why This Matters

This new model is a vital tool for current and future space missions.

Cleaning the Lens: Missions like the James Webb Space Telescope (JWST), Euclid, and the upcoming Roman Space Telescope need to know exactly how bright the dust fog is to subtract it. If they use the old, wrong recipe, they might think they are seeing a faint galaxy when they are actually just seeing a mistake in the dust calculation.
Better Efficiency: With the new ZodiSURF model, these telescopes can plan their observations better, potentially cutting down the time they need to stare at the sky to get a clear picture.

Summary

In short, the authors took a massive amount of data from the Hubble Space Telescope to create a new, highly accurate map of the "dust fog" in our solar system for visible light. This new map (ZodiSURF) fixes the errors of the old maps, allowing astronomers to see the faint universe beyond our solar system much more clearly. They also found a tiny, unexplained "ghost" of light that might be a distant, spherical shell of dust, but that requires further investigation.

1. Problem Statement

Zodiacal light (ZL), caused by sunlight scattering off interplanetary dust (IPD), is the dominant foreground contaminant in space-based ultraviolet-to-infrared observations, accounting for over 90% of the total sky signal. Accurate subtraction of ZL is critical for studying the Diffuse Galactic Light (DGL) and the Extragalactic Background Light (EBL).

Limitations of Existing Models: The standard models used today (Kelsall et al. 1998 and Wright 1998) were optimized using data from the COBE/DIRBE instrument, which operated primarily in the infrared (1.25–240 µm).
The Optical Gap: These models fail to accurately extrapolate to optical wavelengths ( $\lambda \lesssim 1.0$ µm). They rely on non-linear phase function parameters fitted only to IR bands, leading to significant over-predictions of sky brightness in the optical regime.
Consequences: Inaccurate ZL subtraction leads to systematic errors in EBL measurements, potentially mimicking an excess "diffuse light" signal that could be misinterpreted as new physics (e.g., decaying dark matter) or missing galaxy populations.

2. Methodology

The authors present ZodiSURF, a revised 3D zodiacal light model optimized for the optical and near-IR range (0.3–1.6 µm).

Data Source

SKYSURF Project: The model is calibrated using over 5,000 high-precision sky surface brightness (sky-SB) measurements from the Hubble Space Telescope (HST) archival data.
Quality Control: The dataset includes rigorous flagging to exclude images with high contamination from stars/galaxies, stray light, or Earth/Moon proximity. Only "independent pointings" (separated by >10 arcmin and 2 days) are used to avoid bias.
Geometry: The data spans Sun angles ( $\epsilon$ ) from $80^\circ$ to $180^\circ$ , providing a broad range of scattering angles crucial for constraining the phase function, particularly in the backscattering (gegenschein) regime.

Modeling Framework

The authors retained the geometric structure of the Kelsall model (smooth cloud, dust bands, circumsolar ring) but replaced the scattering physics with new analytical forms:

Solar Irradiance: Updated using the Hybrid Solar Reference Spectrum (HSRS) from TSIS-1.
Scattering Phase Function: Replaced the Kelsall non-linear formulation with a weighted sum of three Henyey-Greenstein (HG) functions:
- $i=1$ : Forward scattering.
- $i=2$ : Backward scattering.
- $i=3$ : Gegenschein component.
- Parameters ( $g_i, w_i$ ) are allowed to vary linearly with wavelength.
Albedo: Modeled as a wavelength-dependent parameter.
DGL Subtraction: A new DGL estimator was implemented, correlating optical DGL with Planck 350 µm and 550 µm dust maps (following Postman et al. 2024), rather than the less accurate 100 µm correlation.

Fitting Procedure

A three-step Markov Chain Monte Carlo (MCMC) fitting pipeline was used:

Joint Fit: Simultaneously fit albedo and phase function parameters for each HST filter independently.
Refined Fit: Fix the albedo to a linear wavelength trend derived in Step 1 and refit the phase function parameters.
Wavelength Dependence: Fit linear trends to the phase function parameters across all filters to create a continuous model valid for any wavelength between 0.3–1.6 µm.

3. Key Contributions

ZodiSURF Model: The first 3D zodiacal light model specifically optimized for optical wavelengths (0.3–1.6 µm), publicly available as Python and IDL packages.
Analytical Phase Function: Introduction of a wavelength-dependent, 3-component Henyey-Greenstein phase function that accurately captures forward, backward, and gegenschein scattering in the optical.
Improved DGL Estimation: Implementation of a DGL subtraction method based on sub-millimeter dust maps (350/550 µm), reducing uncertainties in the optical regime.
Toy Model for Isotropic Dust: A proposal for a faint, spherical, isotropic dust component to explain residual diffuse light.

4. Key Results

Model Accuracy: ZodiSURF reduces the uncertainty in zodiacal light predictions to $\sim 4.5\%$ at optical wavelengths.
Comparison with Data:
- Unlike previous models (Kelsall, Wright, ZodiPy) which overpredict sky brightness by significant margins at $\lambda < 1.0$ µm, ZodiSURF matches HST observations closely.
- The residuals (HST Sky – DGL – ZodiSURF) are flat as a function of Sun angle and ecliptic latitude, indicating the scattering physics and 3D geometry are correctly modeled.
Albedo and Phase Function:
- Albedo: Found to increase linearly with wavelength: $A(\lambda) \approx 0.113\lambda + 0.082$ .
- Scattering: Forward scattering becomes more prominent at longer wavelengths, while backward scattering decreases. The gegenschein component is relatively constant with wavelength.
- Composition Implications: The derived geometric albedo ( $p_V \sim 0.07$ ) and scattering properties suggest the IPD is dominated by low-albedo, cometary dust grains, likely larger than 2 µm.
Diffuse Light Residual:
- After subtracting ZodiSURF and DGL, a residual diffuse light signal of $0.013 \pm 0.006$ MJy sr $^{-1}$ remains.
- This signal is isotropic and cannot be explained by the Integrated Galaxy Light (IGL) or standard EBL models.
- Toy Model: The authors propose this residual could be a very dim, spherical dust shell (isotropic component) located at $\sim 4\text{--}5$ AU with a density profile $n \propto R^{-1.78}$ . Adding this component to the model significantly improves the fit to the residuals.

5. Significance

Cosmology: By providing a more accurate foreground model, ZodiSURF allows for more precise measurements of the EBL, resolving discrepancies between direct measurements and integrated galaxy light counts.
Exoplanet Science: Accurate modeling of ZL is essential for future missions (e.g., Habitable Worlds Observatory) aiming to detect exoplanets, as it informs the modeling of exozodiacal light around other stars.
Mission Planning: The model enables more efficient exposure time calculations for current and future observatories (JWST, Euclid, Roman, SPHEREx) operating in the optical/NIR, potentially halving required exposure times by reducing foreground subtraction errors.
Future Work: The paper sets the stage for SKYSURF-IR (using JWST data) to extend these constraints to longer wavelengths and utilize polarimetry to further constrain the isotropic dust component.