🔭 astrophysics

Surface Interactions in Photon Monte Carlo Simulations

This paper presents a comprehensive Monte Carlo simulation framework that models photon surface interactions—including wavelength-dependent reflection, micro-roughness scattering via Harvey-Shack theory, and Mie scattering from contaminants—to accurately predict complex observational effects like PSF wings and dust rings, demonstrating strong agreement with WIYN ODI telescope data.

Original authors: J. R. Peterson, D. Valls-Gabaud, A. Dutta, C. Kim, G. Sembroski

Published 2026-02-03

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: J. R. Peterson, D. Valls-Gabaud, A. Dutta, C. Kim, G. Sembroski

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 you are trying to take a perfect photograph of a star. You have a giant, polished mirror (a telescope) and a digital camera. In a perfect world, every ray of light from that star would bounce off the mirror and hit exactly the right pixel on your camera, creating a sharp, bright dot.

But in the real world, things aren't perfect. The mirror isn't perfectly smooth, it might have a slightly dirty coating, and there might be tiny specks of dust floating in front of the camera lens. These imperfections scatter the light, making the star look fuzzy, creating weird rings, or making the image dimmer than it should be.

This paper is essentially a digital "physics engine" that simulates exactly how light behaves when it hits these imperfect surfaces. The authors built a computer program (part of a larger tool called "PhoSim") that acts like a virtual laboratory. Instead of building a new telescope for every test, they shoot millions of "virtual photons" at a computer model of a mirror to see exactly what happens.

Here is how they break down the three main ways light gets messed up, using simple analogies:

1. The "Magic Coat" (Surface Coatings)

Think of a telescope mirror like a window. If you just look through plain glass, some light bounces off the front, and some goes through. But astronomers often put special "magic coats" (thin layers of chemicals like silver, silicon nitride, or magnesium fluoride) on their mirrors and lenses.

What the paper does: They figured out how to calculate exactly how these coats change the light. It's like knowing that a specific coat makes the mirror act like a super-reflective silver mirror for blue light but lets red light pass through.
The Catch: Sometimes these coats aren't perfectly even. Imagine a coat of paint that is slightly thicker in the middle and thinner on the edges. The paper shows how to simulate this unevenness, which causes the brightness of the image to vary across the picture (like a vignette effect).

2. The "Sandpaper Effect" (Micro-roughness)

Even the most polished mirror in the world isn't perfectly smooth if you look at it under a microscope. It has tiny hills and valleys, like very fine sandpaper.

The Analogy: If you shine a laser pointer at a perfect mirror, you get a sharp dot. If you shine it at a piece of sandpaper, the light scatters everywhere, creating a fuzzy glow.
What the paper does: They created a mathematical model (called Harvey-Shack theory) to predict exactly how much light scatters and in which direction based on how "rough" the surface is.
The Result: This scattering creates "wings" around bright stars in images. The authors tested their model against real photos from the WIYN telescope and found that their simulation perfectly matched the fuzzy "wings" seen around real stars. This proves they can tell the difference between the fuzziness caused by the atmosphere (turbulence) and the fuzziness caused by the mirror's own roughness.

3. The "Dust Bunny" Problem (Contamination)

Telescopes aren't in a vacuum; they can get dusty or have condensation (tiny water droplets) form on them.

The Analogy: Imagine a clean window with a few specks of dust. If you look through it, you might see a ring of light around a streetlamp. This is because the dust blocks some light and bends the rest.
What the paper does: They simulate millions of tiny dust particles and water droplets. Instead of tracking every single one (which would take too long for a computer), they use a clever trick:
- Big dust specks: They simulate these individually because they create distinct shadows or rings.
- Tiny dust specks: They treat them as a general "haze" that just dims the light slightly.
The Result: This allows them to predict "dust rings"—those annoying circles that sometimes appear around bright stars in telescope images—and calculate exactly how much light is lost because of the dirt.

Why Does This Matter?

The authors combined all three of these effects (coatings, roughness, and dust) into one powerful simulation.

For Astronomers: It helps them understand why their data looks the way it does. If a star looks dimmer or fuzzier than expected, they can use this tool to figure out if it's because the mirror is dirty, the coating is uneven, or the mirror is just naturally rough. They can even "clean" the data mathematically to get the true shape of the star.
For the Future: The paper mentions that this isn't just for space telescopes. Because this is just about how light hits objects, the same math could be used in computer graphics to make movies look more realistic, or to understand how we see everyday objects in the real world.

In short, this paper gives scientists a "virtual microscope" to see exactly how the tiny imperfections of a telescope's surface change the pictures of the universe we take.

Technical Summary: Surface Interactions in Photon Monte Carlo Simulations

Problem Statement
Modern astrophysics requires high-precision simulations of astronomical observations to interpret photometry, astrometry, and object shapes accurately. While previous iterations of the Photon Simulator (PhoSim) successfully modeled atmospheric effects, optical deformations, and sensor physics, a comprehensive treatment of photon interactions with optical surfaces was lacking. Specifically, existing models often relied on parameterized approximations rather than first-principles physics for surface reflectivity, transmission, scattering due to microroughness, and the effects of surface contamination (dust and condensation). These omissions lead to unmodeled photometric errors, complex point-spread-function (PSF) wings, and inaccurate throughput predictions, particularly in dense fields or for faint sources where the PSF profile extends below the background level.

Methodology
The authors present a comprehensive Monte Carlo implementation within the PhoSim framework to simulate photon interactions with arbitrary surfaces. The methodology is divided into three primary physical components:

Electromagnetic Boundary Conditions for Coatings:
The simulation calculates wavelength and angle-dependent reflection, transmission, and absorption by solving electromagnetic boundary conditions.
- Bare and Semi-Infinite Interfaces: Uses Fresnel equations for uncoated interfaces and treats sufficiently thick coatings as semi-infinite media where transmission is interpreted as absorption.
- Monolayer and Multilayer Coatings: Implements recursive relations (based on Orfanidis 2016) to solve for reflectivity and transmission across multiple layers. For cases where full layer prescriptions are proprietary or unavailable, the authors introduce approximations to interpolate angular and non-uniformity dependencies using known transmission/reflection tables ( $R(\lambda, \theta)$ and $T(\lambda, \theta)$ ).
- Non-uniformity: Spatial variations in coating thickness are modeled using Zernike polynomials (for circular optics) and Chebyshev polynomials (for rectangular sensors), modifying the nominal reflectivity based on local thickness variations.
- Photo-conversion: In sensor interfaces, absorption events trigger the propagation of photo-electrons using a mean free path derived from the imaginary part of the refractive index.
Microroughness Scattering (Harvey-Shack Theory):
To model scattering from surface imperfections smaller than the photon wavelength, the authors implement a Monte Carlo approach based on Harvey-Shack scattering theory applied to an ABC (A, B, C) power spectrum model.
- Spectral Model: The surface power spectral density (PSD) is defined by parameters $A$ (normalization/roughness $\sigma$ ), $B$ (cutoff inverse frequency), and $C$ (slope).
- Monte Carlo Implementation: To avoid computationally prohibitive analytical evaluations of the Bidirectional Reflectance Distribution Function (BRDF) for millions of photons, the authors utilize a separation of variables approach. They pre-calculate cumulative distribution functions for the scattering angle parameter $\alpha$ and use rejection sampling for the azimuthal and polar angles. This allows for efficient generation of scattered photons that reproduce the correct angular distribution, including the PSF wings.
Contamination (Dust and Condensation):
Surface contaminants are modeled using Mie-Tyndall scattering theory.
- Particle Distribution: Contaminants are represented as spherical objects with sizes following a log-normal distribution.
- Efficient Simulation: To handle the vast number of potential contaminants, the surface is divided into a Cartesian grid. The simulation employs a three-regime approach:
  - Large particles: Simulated individually using full Mie-Tyndall calculations.
  - Moderate particles: Treated as an accumulated area loss within grid cells, capturing Poisson fluctuations without individual ray tracing.
  - Small particles: Modeled as a uniform transmission reduction across the entire surface.
- This hybrid approach allows the simulation of billions of contaminants with minimal computational overhead.

Key Results

Validation with WIYN ODI Data: The authors compared their microroughness model against observations of bright stars from the WIYN telescope's One Degree Imager (ODI). By fitting the PSF wings (1"–3" from the core) with a model combining atmospheric turbulence and mirror microroughness, they successfully disentangled the static surface scattering from time-variable atmospheric effects.
- The derived microroughness parameters for the WIYN mirror were consistent across $g$ , $r$ , and $i$ bands: $\sigma \approx 15 \pm 3$ nm, $B \approx 46 \pm 8$ mm, and $C \approx 3.1 \pm 0.3$ .
- The analysis confirmed that the PSF wings are dominated by a combination of turbulence residuals and mirror microroughness, with no significant correlation found between the derived roughness parameters and seeing conditions.
Simulation of Complex Scenarios: The authors demonstrated the combined effect of uncoated surfaces, microroughness, and contamination on a generic 1m telescope. The simulation successfully reproduced:
- Significant PSF wings due to high microroughness ( $\sigma = 100$ nm).
- Wavelength-dependent throughput losses due to the absence of anti-reflection coatings.
- Distinct "dust rings" caused by contamination on a filter surface located 1 mm from the sensor, illustrating the geometric projection of contaminants.

Significance and Claims
The paper claims that this implementation provides a "first principles" approach to surface interactions, moving beyond empirical models to physically grounded simulations.

Accuracy: The method accurately predicts complex, wavelength- and angle-dependent photometric errors, PSF wings, and dust ring patterns.
Efficiency: Despite the complexity of the physics (multilayer interference, fractal roughness, Mie scattering), the Monte Carlo implementation remains computationally efficient enough for large-scale astronomical simulations.
Reproducibility: The code is publicly available within PhoSim v6.1, allowing for reproducible research.
Broad Applicability: While focused on astronomical instrumentation, the authors note that the formalism is applicable to non-astronomical fields, such as computer graphics and the simulation of everyday object perception, as the physics governing light interaction with surfaces is universal.

The authors conclude that future work will involve validating these models against a wider variety of telescopes to refine particle size distributions and coating non-uniformity maps, further reducing systematic errors in photometric measurements.

1. The "Magic Coat" (Surface Coatings)

2. The "Sandpaper Effect" (Micro-roughness)

3. The "Dust Bunny" Problem (Contamination)

Why Does This Matter?

More like this