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

TL;DR

ZodiSURF provides an optical-wavelength zodiacal-light model by reworking the Kelsall framework with a wavelength-dependent albedo and a three-component Henyey-Greenstein phase function, calibrated against thousands of HST sky-SB measurements from the SKYSURF project. The model integrates an updated solar irradiance spectrum and a data-driven DGL estimator to isolate the IPD scattering signal, delivering a ~4.5% overall uncertainty and flat residuals across Sun angle and ecliptic latitude. A notable finding is a blue, isotropic diffuse-light residual of 0.013 ± 0.006 MJy sr^{-1}, for which a toy spherical IPD component at ~4–5 AU is proposed as a possible explanation. ZodiSURF, released publicly, significantly improves optical zodiacal-light predictions for current and future missions and provides a pathway to better constrain IPD composition and potential isotropic foreground contributions.

Abstract

We present an improved zodiacal light model, optimized for optical wavelengths, using archival Hubble Space Telescope (HST) imaging from the SKYSURF program. The Kelsall et. al. 1998 model used infrared imaging from the Diffuse Infrared Background Experiment (DIRBE) on board the Cosmic Background Explorer to create a 3D structure of the interplanetary dust cloud. However, this model cannot accurately represent zodiacal light emission outside of DIRBE's nominal wavelength bandpasses, the bluest of which is 1.25 micron. We present a revision to this model (called ZodiSURF) that incorporates analytical forms of both the scattering phase function and albedo as a function of wavelength, which are empirically determined across optical wavelengths (0.3-1.6 micron) from over 5,000 HST sky surface brightness (sky-SB) measurements. This refined model results in significantly improved predictions of zodiacal light emission at these wavelengths and for Sun angles greater than 80 deg. Fits to HST data show an uncertainty in the model of ~4.5%. Remarkably, the HST sky-SB measurements show an excess of residual diffuse light (HST Sky - ZodiSURF - Diffuse Galactic Light) of 0.013 +/- 0.006 MJy/sr. The blue color of our diffuse light signal makes it unlikely to be of extragalactic origin. Instead, we suggest that a very dim spherical dust cloud may need to be included in the zodiacal light model, which we present here as a toy model.

Figures (18)

• Figure 1: Comparison of the sky model from this work (teal stars) with HST observed sky-SB (black circles) shows good agreement at 0.3--1.6 . We show other commonly used zodiacal light models Reach_1997Kelsall_1998Wright_1998Aldering_2001San_2024Rigby_2023, which tend to overpredict at $\lambda \lesssim1.0$. The SKYSURF HST sky-SB measurements are from OBrien_2023, and represent pointings that are within 45° of the ecliptic poles and are dominated ($>$$90\%$) by zodiacal light. We also show sky-SB measurements from Giavalisco_2002 as open black squares, and direct measurements of zodiacal light from Kawara_2017 as open black circles. The SKYSURF error bars are the standard deviation of sky-SB measurements divided by $\sqrt{N}$, where $N$ is the number of measurements for that filter. The top axis lists the HST filter corresponding to each SKYSURF measurement. For these same SKYSURF measurements, we estimate the total sky-SB using the modeling in this work: ZodiSURF$+$DGL (large empty teal stars). We also propose for the addition of an isotropic component to zodiacal light (large filled teal stars). We use the IPAC IRSA Background Model to retrieve the Kelsall and Wright predictions for those same observations. We use the Cosmoglobe ZodiPy code San_2024, the JWST Background Tool, and the Gunagala model (implementation of the Aldering_2001 model) to calculate zodiacal light predictions. We fit a two-dimensional polynomial to the Kelsall, Wright, ZodiPy, JWST Background Tool, and Gunagala predictions, which are shown as dashed and solid grey lines.
• Figure 2: Geometry illustrating the scattering angle ($\theta$), phase angle ($\alpha$), and Sun angle ($\epsilon$) for a dust particle as seen from Earth. The distance from the Sun (S) to the Earth (E) is $R_E$ and from the Sun to the dust particle (D) is $R$. The distance from the Earth to the dust particle is $s$, which represents the model’s line of sight. The scattering angle is given by $\theta = \pi - \alpha$, where $\alpha = \arcsin[(R_E/R)\sin\epsilon]$.
• Figure 3: Example of the relative contribution of each $g$ parameter from Equation \ref{['eq:new_phase_func']}, compared with the Kelsall scattering phase function for $\lambda = 1.25$. $g_1$ represents the forward scattering component of the scattering phase function, $g_2$ represents the backward scattering component, and $g_3$ represents the gegenschein component. We use values of $g_1$ = 0.43, $w_1$ = 0.05, $g_2$ = -0.24, $w_2$ = 0.03, $g_3$ = -0.87, and $w_3$ = 0.0003. The x-axis is limited to 0.8 radians ($\sim$45°), just below the minimum Sun Angle observable by HST.
• Figure 4: Ratio of DGL intensity (nW m$^{-2}$ sr$^{-1}$) to 100 intensity (MJy sr$^{-1}$). The 100 intensity for both the IPAC IRSA Background Model and this study's comparisons come from the IRIS+SFD maps released in https://irsa.ipac.caltech.edu/data/Planck/release_2/external-data/. DGL estimates from this work are shown as large pink symbols, where HST's three main cameras are distinguished by different symbols: squares for WFC3/UVIS, circles for ACS/WFC, and triangles for WFC3/IR. The error bars are calculated as the standard deviation of the measurements divided by $\sqrt{N}$, where $N$ is the number of independent pointings. The grey symbols represent DGL estimates from various other studies Guhathakurta_1989Zagury_1999Witt_2008Brandt_Draine_2012Ienaka_2013Arai_2015Kawara_2017Onishi_2018Symons_2023, where it is typically calibrated with 100 maps. The SKYSURF project originally used the IPAC IRSA Background Model to estimate DGL, shown as large grey circles. The top axis lists the HST filter corresponding to each DGL measurement from this work.
• Figure 5: Albedo measurements from this work (solid blue line) compared to values from previous studies. Blue symbols with error bars show the best-fit albedos for each HST filter. The error bars represent the 68.27th-percentile distribution from the posterior. Different symbol shapes represent different HST cameras: squares represent WFC3/UVIS, circles represent ACS/WFC, and triangles represent WFC3/IR. For reference, the solid black circles show the original albedo model from Kelsall_1998. Black symbols show albedo measurements from other studies: Hanner_1974 (Helios A and Pioneer 10), Lumme_1985 (using polarimetric data), Renard_1995 (using nodes of lesser uncertainty method), Ishiguro_2013 (from the WIZARD instrument), and San_2024 (a reanalysis of COBE/DIRBE data). The first two report geometric albedos, which are converted to single-particle albedos for this comparison (conversion shown in Appendix \ref{['app:geometric_albedo']}). The purple curve shows the modeled albedo of interstellar dust from Zubko_2004. In the bottom panel, we show the best fit albedo for each individual HST filter (blue symbols) with the linear fit (solid blue line) subtracted. The y-axis of the bottom panel is scaled linearly from -0.1 to 0.1, outside of which it is scaled logarithmically.