Bridging Statistical Scattering and Aberration Theory: Ray Deflection Function -- I: Theoretical Framework

Netzer Moriya

REVIEW 5 major objections 4 minor 1 cited by

Bridging Statistical Scattering and Aberration Theory: Ray Deflection Function -- I: Theoretical Framework

T0 review · 5 major / 4 minor · reviewed 2026-08-16 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper argues that a rough mirror can be modeled as a deterministic ray deflection function whose statistics match Harvey-Shack scattering.

desk verdict The central RDF equation contradicts the paper's own derivation by a factor of two, and the Zernike coefficient independence claim is false; the framework's claimed equivalence to Harvey-Shack does not hold as written. read the letter →

arxiv 2505.01019 v1 pith:BRM7L742 submitted 2025-05-02 physics.optics math.OC

Netzer Moriya This is my paper · ORCID

classification physics.opticsmath.OC

keywords raydeflectionfunctionsurfaceroughnessscatteringpowerspectraldensityZernikepolynomialsHarvey-Shackmodelaberrationtheorytracingfocalvolume

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to establish that the optical effect of surface roughness can be replaced, for ray tracing, by a single deterministic field, the ray deflection function, that bends each reflected ray by the local gradient of a phase screen. The phase screen is built from Zernike polynomials, and the variance assigned to each polynomial coefficient is computed from the spectral overlap between the surface power spectral density (PSD) and that polynomial's Fourier transform. If this construction works, roughness no longer has to be modeled as a separate scattering channel; it becomes a set of ordinary aberration coefficients that can be added to the mirror's figure error. The paper argues that this statistical equivalence extends to the three-dimensional focal volume, reproducing the structure predicted by the standard Harvey-Shack scattering model for band-limited and Gaussian PSDs. The payoff would be a faster, more interpretable way to propagate manufacturing tolerances through optical system simulations.

What carries the argument

The load-bearing object is the spectral overlap integral $\omega_j=\int PSD(f_x,f_y)\,|F_j(f_r,\phi_f)|^2\,df_x\,df_y$, where $F_j$ is the Fourier transform of the $j$-th Zernike polynomial. This integral converts a continuous PSD into per-mode variances; the random coefficients $C_j=\sqrt{\omega_j}\,\xi_j$ then turn those variances into a statistically representative phase function. The companion identity is the Fourier transform of radially symmetric Zernike polynomials, $F_{n0}(f_r)\propto J_{n+1}(2\pi f_r R)/(f_r R)^{n+1}$, which lets the weights be computed analytically and shows that each Zernike order resonates at a specific spatial frequency. Together these pieces carry the argument: PSD in, aberration coefficients out, ray deflection by gradient, and Harvey-Shack BRDF recovered as the histogram of deflected rays.

What would settle it

Generate many random height maps from a known band-limited or Gaussian PSD on a finite circular aperture, project each onto the Zernike basis, and compute the sample covariance of the coefficients. If off-diagonal covariances are not small for modes sharing azimuthal order, or if the sample variance of a coefficient differs from $\omega_j$, then independent synthesis $C_j=\sqrt{\omega_j}\,\xi_j$ will not reproduce the PSD ensemble and the focal-volume ray density predicted by the ray deflection function will deviate measurably from the Harvey-Shack result.

Watch

Extended reading notes

Core claim

The central claim is that a rough mirror with height deviation $h(x,y)$ can be represented by a phase function $\Phi(x,y)=(4\pi/\lambda)h(x,y)$, and that the resulting ray deflection $D(\mathbf{r}_0)=-\frac{\lambda}{4\pi}\nabla\Phi(x,y)$, applied to an ensemble of rays, produces the same scattering statistics as the physical roughness. To give this construction statistical content, the paper expands $\Phi$ in Zernike polynomials and draws the coefficients from $C_j=\sqrt{\omega_j}\,\xi_j$, where $\xi_j$ are independent standard normals and $\omega_j$ is the integral of the surface PSD against the squared Fourier transform of the $j$-th Zernike mode. Because each $\omega_j$ is the exact variance contributed by that mode, the truncated sum captures a specified fraction of the roughness variance, and the ensemble statistics of the synthesized phase approach the target PSD. The paper then shows that the probability distribution of the deflected ray directions is proportional to the BRDF of the Harvey-Shack model, and derives closed-form spectral weights for uniform and Gaussian PSDs, including an exponentially decaying tail that provides a natural truncation order.

Load-bearing premise

The construction assumes that the Zernike coefficients of a stationary Gaussian rough surface on a circular aperture are statistically independent, so drawing each coefficient from a normal distribution with variance $\omega_j$ reproduces the full ensemble of surface statistics; orthogonality alone does not guarantee this, and finite-aperture edge effects can introduce correlations.

Editorial extensions

If this is right

Roughness effects can be added linearly to design aberrations by summing Zernike coefficients, so a single ray-trace model carries both figure error and finish error.
For Gaussian PSDs the Zernike expansion truncates naturally at order $n \approx 2\pi R/l_c$, reducing a typical meter-scale mirror to a few tens of coefficients instead of a dense phase screen.
The angular histogram of traced rays converges to the Harvey-Shack BRDF, so standard ray-tracing simulations can replace dedicated scattering simulations for small-slope surfaces.
The spectral weights classify roughness by physical effect: low orders change the PSF core, mid orders create small-angle scatter halos, and high orders drive wide-angle stray light, giving tolerance specifications a direct performance meaning.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

If the independence assumption is violated by finite-aperture edge effects, the remedy is to replace independent draws with correlated coefficients via the covariance of the Zernike projections; the paper sketches this Cholesky route for anisotropic PSDs but does not incorporate it into the main construction.
A testable extension would compare the variance of the focal-volume ray density, not just its mean, against direct phase-screen simulations; the paper reports mean agreement, but the framework's prediction for realization-to-realization fluctuations follows from the Gaussian coefficient model and could be checked in the companion numerics.
Because the method converts roughness into aberration coefficients, it should plug directly into gradient-based optical design and tolerance optimization, where PSD specifications would enter as prior covariance information on the coefficients.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Bottom line: this preprint isn't ready for review. The RDF formalism is essentially a repackaging of the standard small-angle phase-gradient relation, and the key new step—converting a PSD into Zernike variances—rests on load-bearing errors.

First, Eq. (3) defines D = -λ/4π ∇Φ, giving D = -∇h, while the derivation in the same section yields D = -λ/2π ∇Φ = -2∇h, which is the physically correct reflection deflection. Appendix A tries to justify the factor of 1/2 by saying reflection doubles the tilt, which is backwards. So the model underestimates scatter angles by half.

Second, Sec. 2.3.4 asserts that projecting a stationary Gaussian field onto Zernike polynomials gives independent coefficients because the basis is orthogonal. That's false on a finite aperture: the coefficient covariance is a double integral of the covariance function weighted by Zernike products, and it is not diagonal for any non-white PSD. The iid-normal synthesis of Eq. (29) therefore does not reproduce the target PSD's joint statistics. This isn't a minor gap; it breaks the claimed equivalence to Harvey-Shack.

The Fourier transform of the radial Zernikes also has normalization problems. Eq. (B.19)/(24) inserts an i^n factor and a 1/(2π) that don't follow from the derivation, and Eq. (28)'s sum rule requires sum_j |F_j|^2 = 1, which isn't true for these transforms. Eq. (37) has an exponent error. These are the kind of mistakes a careful referee would catch immediately.

What's worth keeping? The general strategy—spectral overlap to get modal variances, truncation based on correlation length, and a ray-based focal-volume model—is sensible and could be useful to optical engineers if done correctly. The validity criteria in Sec. 2.2.1 are fine. But none of that is new enough to outweigh the errors.

The paper also claims Monte Carlo and MoM validation in Sec. 5.3 and Sec. 4 without showing results; numerical work is deferred to the companion. That's acceptable for a theory paper only if the analytical results are solid, and they aren't.

Verdict: desk reject. If the author fixes the factor of two, replaces the independence claim with a proper covariance treatment (e.g., Karhunen-Loève or Cholesky with the true coefficient covariance), corrects the normalization, and includes the validation, the underlying idea could become a decent applied-optics paper. As submitted, the central framework doesn't hold together.

Referee Report

5 major / 4 minor

Summary. This manuscript (Part I of a two-part series) proposes a 'ray deflection function' (RDF) framework that recasts surface roughness as a phase gradient acting on reflected rays. The phase is expanded in Zernike modes whose coefficients are generated from spectral-overlap integrals between the surface power spectral density (PSD) and the Fourier transforms of Zernike polynomials. The paper claims that this construction preserves the PSD statistics, reproduces the Harvey-Shack BRDF in ray-tracing simulations, and allows roughness to be added linearly to deterministic aberration coefficients. Numerical validation is promised in the companion paper [1]; the present text contains analytical derivations, asymptotic estimates for band-limited and Gaussian PSDs, and validation claims (NCC>0.95, volumetric deviations <5%).

Significance. The intended bridge between statistical scattering and aberration theory is a worthwhile goal, and the manuscript has several positive features: the idea of spectral-overlap weights is natural, the validity-range criteria in Section 2.2.1 are explicit, and the dimensional analysis in Section 2.2.5 and the appendices attempt a self-contained derivation. If the central equivalence were established, the framework could offer a useful ray-tracing alternative to BRDF sampling and a common language for roughness and aberrations. However, the paper's central quantitative claims are not established: normalization errors in the Zernike Fourier transforms break the variance budget, the independence assumption for Zernike coefficients is false on a finite aperture, and the Harvey-Shack validation is enforced by construction rather than tested independently. These problems affect the load-bearing equivalence claim, not just presentation.

major comments (5)

[§2.2, Eqs. (1), (3), (5)] The derivation of the RDF contains a factor-of-two inconsistency. Eq. (5) gives Φ = (4π/λ)h, so Eq. (3) yields D = -(λ/4π)∇Φ = -∇h. In the same subsection, however, the text derives δθ = -(λ/2π)∇Φ = -2∇h and identifies D with δθ, while Eq. (1) with k = 2π/λ would give D = (λ/2π)∇Φ. The manuscript cannot simultaneously have these three expressions. Since the angular spread in Eq. (51) is quadratic in D, the predicted scattering angle differs by a factor of 2 (and the variance by a factor of 4) depending on which expression is used, directly affecting the claimed quantitative match with the Harvey-Shack BRDF in Section 4.
[Eqs. (24)/(B.19), (25), (28)] The normalization of the Zernike Fourier transform is inconsistent with the orthonormality convention used in Eq. (26). With the transform convention of Eq. (B.11), Parseval's theorem gives ∫|F_j(f)|^2 d^2f = ∫_disk Z_j^2 dA = 1, so for a complete orthonormal set ∑_j |F_j(f)|^2 equals the aperture area (πR^2) rather than unity. Consequently Eq. (28), ∑_j ω_j = ∫ PSD d^2f = σ^2, does not follow from the definition in Eq. (25); the left-hand side carries an extra factor of the aperture area. The redundant factor (−1)^{n/2}i^n in Eq. (B.19) for even n and the apparent discrepancy between Eqs. (B.18) and (B.19) should also be reconciled, because the spectral weights in Eq. (25) depend critically on the exact normalization.
[Eq. (37)] Substituting Eq. (24) into Eq. (36) does not produce Eq. (37). Using the stated F_{n0}, the pre-factor is P0/(2π R^{2n+2}) and the integrand is J^2_{n+1}(2πfR) f^{-2n-1} df, not P0/(2πR^2) J^2_{n+1}(2πfR) f^{-2n} df as written. The discrepancy in the powers of R and f changes the narrow-band estimate in Eq. (38) and the Gaussian asymptotic in Eq. (47), so the claimed dominance of modes near n ≈ 2πf_cR - 1/2 is not supported by the written derivation.
[§2.3.4, Eq. (31)] The assertion that Zernike orthogonality implies statistical independence of the projection coefficients of a stationary isotropic Gaussian field is not correct for a finite circular aperture. The coefficient covariance is Cov(C_i,C_j) = ∫∫_{D×D} C_h(r−r′) Z_i(r) Z_j(r′) dr dr′, where C_h(r−r′) is the surface autocovariance; this double integral is not diagonal for any non-white PSD because the aperture breaks translational invariance. Fourier modes diagonalize stationary spectra on the infinite plane; Zernike modes do not. Thus the i.i.d. prescription in Eq. (29) reproduces only the marginal variances ω_j and not the joint statistics required for a field with the target PSD, and the preservation properties 2 and 3 in Section 2.3.4 are unsupported. The remark that only anisotropic PSDs introduce coefficient correlations is also incorrect; isotropic band-limited PSDs already correlate modes, particularly those sharing azimuthal order.
[§4, Eqs. (59)–(75)] The claimed validation of the RDF against Harvey-Shack is largely circular. The phase function is deliberately synthesized from the same PSD used to evaluate the Harvey-Shack BRDF (e.g., Eqs. (64)–(72)), so agreement between the ray-density distribution and the BRDF is enforced by construction rather than tested independently. Moreover, the quantitative results invoked in Sections 4.2 and 4.3 (NCC>0.95, volumetric deviations <5%, convergence with N_rays≈10^5) are attributed to the companion paper [1], and no simulation data are shown here. An independent benchmark—for example, a direct Fourier-synthesis phase screen or a Beckmann-Kirchhoff/MoM computation—is needed before the claimed statistical equivalence can be accepted.

minor comments (4)

[§2.2.3–2.2.4, Eqs. (14), (17)] The discrete Fourier synthesis in Eq. (14) omits the spectral sampling factors Δf_x Δf_y that appear in Eq. (17), so the variance of the synthesized phase is not evidently σ^2 as claimed; the dimensional and normalization conventions should be stated consistently for both cases.
[§2.4.1 and §4.3] There are several incorrect cross-references: Eq. (36) is said to use 'equation 28' when the intended formula is the Fourier transform in Eq. (24)/(B.19), and Section 4.3 says coefficients are generated 'using equation (31)' when the intended construction is Eq. (29). These should be corrected.
[§2.4.3, Eqs. (45)–(47)] The delta-function approximation for J^2_{n+1} in Eqs. (45)–(46) uses an undefined constant C and an unstated prefactor; the resulting exponent in Eq. (47) therefore should be checked carefully, since the cutoff estimate n ≈ 2πR/l_c depends on it.
[Throughout] The symbol k is introduced in Eq. (1) as 'proportional to the optical wavenumber' but is never fixed; the subsequent equations use λ explicitly. Defining k = 2π/λ at the outset and consistently using one form would remove avoidable ambiguity.

Circularity Check

3 steps flagged · score 7.0 of 10

Core scattering 'prediction' is enforced by construction: the phase is synthesized from the same PSD used to compute the Harvey-Shack benchmark, so the NZF≈NHS agreement reduces to the input BRDF∝PSD relation; no independent benchmark tests the RDF claim.

self definitional [Section 2.2, after Eq. (3)]
"By synthesizing Φ(x, y) with the same PSD as the true surface, equation (3) applied to an ensemble of rays yields statistically equivalent scattering to the physical roughness."

The RDF is defined as the gradient of a phase function, and the phase is then constructed so that its PSD equals the input surface PSD. Harvey-Shack theory (Section 2.1) already states BRDF ∝ PSD, so the claimed 'statistically equivalent scattering' is the input relation restated in ray coordinates. The derivation chain does not produce a new scattering prediction; it converts the PSD into a gradient and reads off the same PSD.
fitted input called prediction [Section 4.2, validation of uniform-PSD case]
"Statistical modeling with 105 sampled rays (see in [1] indicates that the resulting ray density NZF(x, y, z) matches the predicted NHS(x, y, z) with statistical deviations less than 5% when measured in volumetric bins of size (0.1λ f)3."

The validation target NHS(x,y,z) is computed from the Harvey-Shack BRDF, which Section 2.1 states is proportional to the same surface PSD used in Eq. (25) to compute the spectral weights ω_j and in Eq. (29) to generate the coefficients C_j. The compared quantities therefore share their defining input: the 'prediction' NZF is generated from the PSD, and the reference NHS is derived from the same PSD. Agreement to within sampling/truncation error is enforced by construction, not by an independent benchmark.

1 more flagged steps

self citation load bearing [Section 1, Introduction; repeated in Section 4.2 as 'see in [1]']
"A companion paper [1] provides comprehensive numerical validation of this framework through detailed simulations and comparisons with established scattering models."

The actual Monte Carlo evidence for the central equivalence is deferred to a submitted companion paper by the same author. Insofar as the present paper's validation rests on [1], it is a self-citation; and because the companion is said to compare against the same Harvey-Shack model that is derived from the construction's own PSD input, the self-citation does not provide independent, externally anchored support.

full rationale

The central equivalence is not tested against an independent benchmark. Section 2.2 builds the phase from the very PSD that Harvey-Shack theory (Section 2.1) states is proportional to the BRDF, so the subsequent 'agreement' between ray-traced NZF and Harvey-Shack NHS in Section 4 is the input relation re-derived in ray coordinates. The finite-Zernike truncation and Monte Carlo sampling are the only nontrivial numerical content; those are deferred to the author's own companion paper [1]. No external measured BRDF or independent scattering code is used as a benchmark. Separately, Section 2.3.4's claim that Zernike orthogonality implies statistical independence of the projection coefficients is not a consequence of orthogonality for finite-aperture stationary fields; this is a correctness risk rather than a circularity, but it breaks the stated mechanism by which the synthesized phase is supposed to reproduce the input PSD. Because the paper's central validation reduces by construction to its own inputs, the circularity score is high, though the eikonal phase-gradient relation itself is standard and non-circular.

Assumptions & free parameters 0 free parameters · 6 assumptions · 0 invented entities

No new physical entities (particles, forces, conserved quantities) are introduced. The ray deflection function is a renaming of the standard phase-gradient ray-tilt relation. The assumed axioms are the domain restrictions of scalar paraxial scattering, the Gaussian stationary surface model, the Zernike basis properties, and several normalization identities that are not correctly established in the paper.

assumptions (6)

domain assumption Small-angle, scalar, eikonal approximation: a transverse phase gradient produces ray deflection delta_theta = -lambda/(2 pi) grad Phi.
Used to derive Eq. (3); valid only for sigma/lc < 0.05 as quantified in Section 2.2.1, and it omits polarization and vector diffraction.
domain assumption Surface roughness is a stationary, zero-mean, isotropic Gaussian random process with a prescribed PSD.
Adopted in Section 3.1 and Appendix C to justify the random phase screen model; many real surfaces are non-Gaussian or anisotropic.
standard math Zernike polynomials form an orthonormal complete basis over the unit disk, and the Fourier transforms F_j given in Eq. (24) and Appendix B are correct.
The Bessel-function transform is a standard result, but the version printed contains an erroneous i^n factor and inconsistent normalization, so the paper's use of it is not reliable.
ad hoc to paper Projection coefficients of a stationary isotropic Gaussian field onto an orthonormal Zernike basis are statistically independent.
Invoked in Eqs. (29)-(31) to set C_j = sqrt(omega_j) xi_j. Orthogonality alone does not imply independence on a finite aperture; edge effects correlate modes, so this assumption is false in general.
ad hoc to paper The spectral weight sum equals the total variance: sum_j omega_j = integral PSD df (Eq. 28).
Requires sum_j |F_j|^2 to be a constant normalized to unity, but the basis normalization actually gives the aperture area; the identity is not established and appears dimensionally inconsistent.
domain assumption Harvey-Shack proportionality BRDF is proportional to the PSD of surface height.
Standard scattering model used throughout and in the validation target; valid only in the smooth-surface, small-slope regime.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Bridging Statistical Scattering and Aberration Theory: Ray Deflection Function -- I: Theoretical Framework." pith.science (2026). https://pith.science/paper/BRM7L742

@misc{pith2026250501019,
  author       = {Pith},
  title        = {Pith review of: Bridging Statistical Scattering and Aberration Theory: Ray Deflection Function -- I: Theoretical Framework},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BRM7L742}},
  note         = {Machine review of arXiv:2505.01019}
}

read the original abstract

This paper introduces a new conceptual framework that recasts surface roughness effects as a "ray deflection function" (RDF) which can be statistically represented through a modified Zernike-Fourier hybrid approach that directly connects the PSD with statistical aberration coefficients through spectral overlap integration. By establishing a direct mathematical relationship between the power spectral density (PSD) of surface imperfections and the statistical distribution of aberration coefficients, we develop a formalism that bridges known probabilistic scattering theory with deterministic aberration analysis. This transformation allows surface roughness to be seamlessly integrated with other optical aberrations by expressing its effects through equivalent modifications to the ideal mirror shape. This framework provides computational advantages for ray-tracing simulations while maintaining statistical fidelity to established scattering models, particularly for predicting the three-dimensional structure of imperfect focal bodies in optical systems.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

Surface-Encoded Partial Coherence Transformation: Modeling Source Coherence Effects in Wave Optics
physics.optics 2025-05 reject novelty 3.0 of 10

A reformulation of standard partial coherence propagation as a surface operator followed by propagation, whose equivalence to Van Cittert-Zernike is true by construction and whose claimed computational gains rely on a...

Reference graph

Works this paper leans on

25 extracted references · 21 canonical work pages · cited by 1 Pith paper

[1]

Bridging Statistical Scattering and Aberration Theory: Ray Deflection Function - II: Numerical Validation

N. Moriya, “Bridging Statistical Scattering and Aberration Theory: Ray Deflection Function - II: Numerical Validation”, submitted to Applied Optics (2025).. 40

work page 2025
[2]

Di ffracted radiance: New approach to the prediction of surface scatter,

J. E. Harvey and R. Shack, “Di ffracted radiance: New approach to the prediction of surface scatter,”, Applied Optics, vol. 19, no. 22, pp. 3723–3724, 1980

1980
[3]

Beckmann and A

P. Beckmann and A. Spizzichino, The Scattering of Electromagnetic Waves from Rough Surfaces. Artech House, 1987. (Original publication: Pergamon Press, 1963)

1987
[4]

E., Krywonos, A., & Vernold, C

Harvey, J. E., Krywonos, A., & Vernold, C. L. (2007). Modified Beckmann- Kirchhoff scattering model for rough surfaces with large incident and scattering angles. Optical Engineering, 46(7), 078002

2007
[5]

Surface characterization techniques for determining the root-mean-square rough- ness and power spectral densities of optical components,

A. Duparré, J. Ferre-Borrull, S. Gliech, G. Notni, J. Steinert, and J. M. Bennett, “Surface characterization techniques for determining the root-mean-square rough- ness and power spectral densities of optical components,” Applied Optics, vol. 41, no. 1, pp. 154–171, 2002

2002
[6]

Dai, G. M. (1996). Modal wave-front reconstruction with Zernike polynomials and Karhunen-Loève functions. Journal of the Optical Society of America A, 13(6), 1218-1225

work page 1996
[7]

Van Trees, H. L. (2001). Detection, Estimation, and Modulation Theory, Part I. Wi- ley, ISBN:9780471095170, pp. 247-252

work page 2001
[8]

Goodman, J. W. (2017). Introduction to Fourier Optics (4th ed.). W. H. Freeman, pp. 142-145

work page 2017

Show all 25 references

[9]

V oronovich, A. G. (2001). Wave Scattering from Rough Surfaces (2nd ed.). Springer, pp. 83-89

2001
[10]

Papoulis, A., & Pillai, S. U. (2002). Probability, Random Variables, and Stochastic Processes (4th ed.). McGraw-Hill, pp. 383-389

2002
[11]

Mahajan, V . N. (2013). Optical Imaging and Aberrations, Part III: Wavefront Anal- ysis. SPIE Press, pp. 123-157

2013
[12]

E ffects of the optical transfer function in surface profile measurements,

E. L. Church and P. Z. Takacs, “E ffects of the optical transfer function in surface profile measurements,”Proceedings of SPIE, V ol. 680, Surface Characterization and Testing, pp. 125-136, 1986

1986
[13]

Church, E. L. (1988). Fractal surface finish. Applied Optics, 27(8), 1518-1526

1988
[14]

Effects of surface scatter on the optical performance of x-ray synchrotron beam-line mirrors,

J. E. Harvey, K. L. Lewotsky, and A. Kotha, “Effects of surface scatter on the optical performance of x-ray synchrotron beam-line mirrors,”Applied Optics, vol.34, no.16, pp. 3024–3032, 1995

1995
[15]

Specifications: Figure and finish are not enough,

R. E. Parks, “Specifications: Figure and finish are not enough,”Proceedings of SPIE, vol.7071, Optical Manufacturing and Testing VIII, 70710B, 2008. 41

2008
[16]

A., & Maradudin, A

Simonsen, I., Letnes, P. A., & Maradudin, A. A. (2004). Numerical simulation of electromagnetic wave scattering from random surface fractal profiles. Physical Re- view A, 73(1), 013806

2004
[17]

Y ., & Silva, D

Wang, J. Y ., & Silva, D. E. (2009). Wave-front interpretation with Zernike polyno- mials. Applied Optics, 19(9), 1510-1518

2009
[18]

J. C. Stover, Optical Scattering: Measurement and Analysis , 3rd ed., SPIE Press, Bellingham, W A, 2012. DOI: 10.1117/3.899758

2012 doi
[19]

Sidick, E. (2009). Power spectral density specification and analysis of large optical surfaces. Proceedings of SPIE, V ol. 7390, p. 73900L

2009
[20]

Born, M., & Wolf, E. (1999). Principles of Optics (7th ed.). Cambridge University Press, pp. 459-461

1999
[21]

Mahajan, V . N. (1998). Optical Imaging and Aberrations, Part I. SPIE Press, pp. 21-25

1998
[22]

E., & Pfisterer, R

Harvey, J. E., & Pfisterer, R. N. (2014). Understanding Diffraction Grating Behavior. Proceedings of SPIE, V ol. 9271, p. 92710J

2014
[23]

Stover, J. C. (2012). Optical Scattering: Measurement and Analysis (3rd ed.). SPIE Press, pp. 46-50

2012
[24]

E., & Choi, N

Krywonos, A., Harvey, J. E., & Choi, N. (2011). Linear systems formulation of scattering theory for rough surfaces with arbitrary incident and scattering angles. Journal of the Optical Society of America A, 28(6), 1121-1138

2011
[25]

H., & Harvey, J

Schröder, S., Duparré, A., Coriand, L., Tünnermann, A., Penalver, D. H., & Harvey, J. E. (2011). Modeling of light scattering in di fferent regimes of surface roughness. Optics Express, 19(10), 9820-9835. 42

2011

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.

[1] [1]

Bridging Statistical Scattering and Aberration Theory: Ray Deflection Function - II: Numerical Validation

N. Moriya, “Bridging Statistical Scattering and Aberration Theory: Ray Deflection Function - II: Numerical Validation”, submitted to Applied Optics (2025).. 40

work page 2025

[2] [2]

Di ffracted radiance: New approach to the prediction of surface scatter,

J. E. Harvey and R. Shack, “Di ffracted radiance: New approach to the prediction of surface scatter,”, Applied Optics, vol. 19, no. 22, pp. 3723–3724, 1980

1980

[3] [3]

Beckmann and A

P. Beckmann and A. Spizzichino, The Scattering of Electromagnetic Waves from Rough Surfaces. Artech House, 1987. (Original publication: Pergamon Press, 1963)

1987

[4] [4]

E., Krywonos, A., & Vernold, C

Harvey, J. E., Krywonos, A., & Vernold, C. L. (2007). Modified Beckmann- Kirchhoff scattering model for rough surfaces with large incident and scattering angles. Optical Engineering, 46(7), 078002

2007

[5] [5]

Surface characterization techniques for determining the root-mean-square rough- ness and power spectral densities of optical components,

A. Duparré, J. Ferre-Borrull, S. Gliech, G. Notni, J. Steinert, and J. M. Bennett, “Surface characterization techniques for determining the root-mean-square rough- ness and power spectral densities of optical components,” Applied Optics, vol. 41, no. 1, pp. 154–171, 2002

2002

[6] [6]

Dai, G. M. (1996). Modal wave-front reconstruction with Zernike polynomials and Karhunen-Loève functions. Journal of the Optical Society of America A, 13(6), 1218-1225

work page 1996

[7] [7]

Van Trees, H. L. (2001). Detection, Estimation, and Modulation Theory, Part I. Wi- ley, ISBN:9780471095170, pp. 247-252

work page 2001

[8] [8]

Goodman, J. W. (2017). Introduction to Fourier Optics (4th ed.). W. H. Freeman, pp. 142-145

work page 2017