DustER: Dust Extinction with Photometric Red Galaxies
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Astrophysics. The day in astrophysics: observations, models and what the data forces.
Vera: We're looking at the paper: "DustER: Dust Extinction with Photometric Red Galaxies".
Jocelyn: Here's the gist of the paper. DustER presents a new method for reconstructing dust reddening maps from photometric galaxy survey data using red galaxies as "standard crayons" The recovered dust maps are statistically unbiased relative to the input dust field, with a mean bias of 0.5 mmag and an rms scatter of 6.3 mmag DustER uses red galaxies as "standard crayons," relying on the mean color of red-sequence galaxies being independent of their position in the sky Given the galaxy’s redshift, the difference between a galaxy’s observed color and its expected color measures the reddening by dust along the galaxy’s line of sight The method calibrates the color–redshift relation using redMaPPer to predict intrinsic galaxy colors as a function of redshift Red galaxies are selected by fitting every galaxy with the redMaPPer red-sequence model to determine a photometric redshift estimate zred Bias Mitigation and Noise Reduction The initial reconstruction is both biased and noisy because the red-sequence calibration and photometric redshifts are themselves affected by reddening DustER mitigates these effects through Wiener filtering and a statistical debiasing procedure that leverages the empirically determined one-point distribution of Galactic dust The raw DustER map is obtained by modeling the observed color of red-sequence galaxies in sky pixels as having Gaussian random noise This raw map is then processed through a Wiener filter to suppress noise, resulting in the filtered map Finally, a debiasing step models the bias as a power-law plus a constant Validation and Performance DustER was validated using simulated data based on the Cardinal catalog, where dust attenuation is applied using the SFD dust map in our fiducial simulation The recovered DustER map is statistically unbiased relative to the input dust field, with a mean bias of 0.5 mmag and an rms scatter of 6.3 mmag The recovered map is nearly perfectly correlated with the input map on large scales, with the correlation decreasing on small scales where the reconstruction is noise-dominated LSS Contamination Analysis The analysis detected statistically significant cross-correlations between the reconstructed DustER map and galaxy density fields in our simulations The non-zero DustER–red galaxy cross-correlation indicates that the DustER map itself is contaminated by large-scale structure Testing an LSS subtraction scheme showed that it does not remove the contamination, indicating that using all three redshift bins over-corrects the map The authors conclude that the proposed linear cleaning algorithm fails because it does not remove the contamination In future work, they plan to develop a forwardmodeling, field-level inference framework in which the dust and galaxy density fields are modeled jointly within a hierarchical Bayesian approach They anticipate this approach will yield posterior samples of the dust map and its uncertainties, with the maps being free of large-scale-structure correlations The authors plan to apply DustER to real photometric survey data, including observations from the Dark Energy Survey and the Dark Energy Camera All Data Everywhere (DECADE) program The resulting reconstructions will provide an independent reddening-based perspective on Galactic extinction and will allow detailed comparisons with emission-based maps such as SFD and P16-GNILC Pruning redundant reasoning tokens DustER uses red galaxies as “standard crayons” It relies on the mean color of red-sequence galaxies being independent of their position in the sky Given the galaxy’s redshift, the difference between a galaxy’s observed color and its expected color measures the reddening by dust along the galaxy’s line of sight How the training signal is built The calibration of our red-sequence model is performed using raw (i.e.
Vera: not dereddened) galaxy magnitudes This is done to ensure the DustER dust maps are not informed by the reddening values from existing dust maps The selection of red galaxies is performed using raw (i.e.
Jocelyn: not dereddened) photometry Where the gains come from The final dust map is obtained as the posterior expectation value of the extinction field, with pixel-level uncertainties derived from the posterior variance The authors found that the choice of prior used in DustER has little impact on the recovered dust map The correlation coefficient is nearly identical regardless of which dust map is used as the input to the simulations The resulting maps are nearly perfectly correlated with the input on large scales, with the correlation decreasing on small scales where the reconstruction is noise-dominated Pruning redundant reasoning tokens The method uses red galaxies as “standard crayons” It relies on the mean color of red-sequence galaxies being independent of their position in the sky Given the galaxy’s redshift, the difference between a galaxy’s observed color and its expected color measures the reddening by dust along the galaxy’s line of sight Where the gains come from The final dust map is obtained as the posterior expectation value of the extinction field, with pixel-level uncertainties derived from the posterior variance The authors found that the choice of prior used in DustER has little impact on the recovered dust map The correlation coefficient is nearly identical regardless of which dust map is used as the input to the simulations How the training signal is built The calibration of our red-sequence model is performed using raw (i.e.
Vera: not dereddened) galaxy magnitudes This is done to ensure the DustER dust maps are not informed by the reddening values from existing dust maps The selection of red galaxies is performed using raw (i.e.
Title and authors: Vera: Now let's talk about how they frame this method in the paper. They call it DustER: Dust Extinction with Photometric Red Galaxies.
Jocelyn: It sets up the idea that we use red galaxies as these standard crayons, which means we rely on their mean color staying consistent across the sky.
Subrahmanyan: The implication there is that if you can reliably predict what a red galaxy's color should be based only on its redshift, you can infer the dust along its line of sight.
Vera: They explain that this difference between observed and expected color directly measures the reddening by dust along that specific path.
Jocelyn: The authors mention they calibrate this color-redshift relation using something called redMaPPer to get those intrinsic galaxy colors as a function of redshift.
Subrahmanyan: That calibration step is crucial because if the input colors are off, the resulting dust map will be wrong, so they use that to predict the true intrinsic color.
The paper's summary: Vera: So what’s actually happening in this DustER method, as described in this paper? It starts by using red galaxies as these standard crayons we just talked about.
Jocelyn: They then use the redshift of each galaxy to figure out how much dust is on its path by comparing its color to what redMaPPer predicts for that redshift.
Subrahmanyan: The method essentially measures the color excess of the sources themselves, rather than trying to guess reddening from some other tracer, which is a big advantage.
Vera: They then move into a process where they model the observed color in each sky pixel as having some kind of Gaussian random noise.
Jocelyn: They take that raw map and run it through what they call Wiener filtering to suppress that noise, which helps clean up the initial reconstruction.
The paper's improvements: Vera: The paper then details how they handle the inevitable problems with this approach, specifically looking at bias and noise reduction.
Jocelyn: They use a statistical debiasing procedure that relies on the empirically determined one-point distribution of Galactic dust to fix any systematic errors.
Subrahmanyan: They model this bias as a power-law plus a constant term, which is how they try to correct for the systematic errors in the reconstruction process.
Vera: They also show that they validate this by comparing their recovered maps against simulated data from the Cardinal catalog, where dust attenuation was applied using an existing SFD dust map.
Conclusion: Jocelyn: So to wrap up, what’s the main point they are making about DustER: Dust Extinction with Photometric Red Galaxies?
Vera: They conclude that their method recovers dust maps that are statistically unbiased relative to the input dust field, with a mean bias of zero point five mmag and an rms scatter of six point three mmag.
Subrahmanyan: That level of scatter is what we need to look at when we try to compare these new maps against existing ones like SFD or P16-GNILC.
Jocelyn: And they also found that on large scales, the recovered map stays nearly perfectly correlated with the input map, but that correlation drops off on smaller scales because it gets noise-dominated.
Vera: It’s a solid tool for getting an independent reddening perspective on Galactic extinction, even if it has those noise limits at small scales.
Subrahmanyan: And they flag one thing: this linear cleaning algorithm doesn't work when you look at large-scale structure contamination because testing an LSS subtraction scheme showed that it doesn't actually remove the contamination and using all three redshift bins over-correcting the map.
Jocelyn: So, in short, DustER is a new way to use galaxy colors to reconstruct dust maps, but they have some clear hurdles with large-scale structure and small scale noise.
E. Espinoza-Bustamante, E. S. Rykoff, E. Rozo, C. To, S. S. Boruahg
Department of Physics, University of Arizona · SLAC National Accelerator Laboratory · Kavli Institute for Particle Astrophysics and Cosmology Department of Astronomy and Astrophysics Department The University of Chicago Kavli Institute for Cosmological Physics NSF-Simons AI Institute for the Sky
astro-ph.IM, astro-ph.CO
Submitted: 2026-11-10
Updated: 2026-11-10
Comments: 27 pages, 9 figures, 6 tables, prepared for submission to JCAP
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 79/100
The gist: DustER presents a new method for reconstructing dust reddening maps from photometric galaxy survey data using red galaxies as "standard crayons" The recovered dust maps are statistically unbiased…
Key concepts
- Red Galaxies as Standard Crayons
- The method uses the mean color of red-sequence galaxies as a stable reference point across the sky. This assumption allows researchers to estimate how much dust lies along a galaxy's line of sight by comparing its actual color to what it should be based on its known redshift.
- RedMaPPer Calibration
- This tool is used to predict the intrinsic colors of galaxies as a function of their redshift. This calibration step ensures that the measured difference between observed and expected color accurately reflects dust reddening, rather than just galaxy evolution effects.
- Wiener Filtering
- This technique is applied after an initial map reconstruction to suppress noise. It models the observed color variations as Gaussian random noise and filters this signal to produce a cleaner, de-noised map of the dust field.
- LSS Contamination
- The reconstructed dust maps are contaminated by large-scale structure (LSS) because the method correlates with galaxy density fields. This contamination is significant, indicating that current cleaning methods over-correct when using multiple redshift bins.
Summary
DustER presents a new method for reconstructing dust reddening maps from photometric galaxy survey data using red galaxies as As with IR-based maps, PAH-based
maps exhibit residual large-scale structure contamination from the diffuse PAH background,
whereas HI maps appear to be free from large-scale structure contamination [2]. However,
HI-based maps assume a fixed dust-to-gas ratio and miss dust associated with molecular gas,
which restricts them to regions of low column density (E(B − V ) ≲ 45 mmag) [4]. As with IR-based maps, PAH-based
maps exhibit residual large-scale structure contamination from the diffuse PAH background,
whereas HI maps appear to be free from large-scale structure contamination [2]. However,
HI-based maps assume a fixed dust-to-gas ratio and miss dust associated with molecular gas,
which restricts them to regions of low column density (E(B − V ) ≲ 45 mmag) [4].standard crayons
The recovered dust maps are statistically unbiased relative to the input dust field, with a mean bias of 0.5 mmag and an rms scatter of 6. p. 2Where the paper says it
Where the paper says it
Methodology for Dust Reconstruction
DustER uses red galaxies as standard crayons,
relying on the mean color of red-sequence galaxies being independent of their position in the sky Given the galaxy’s redshift, the difference between a galaxy’s observed color and its expected color measures the reddening by dust along the galaxy’s line of sight The method calibrates the color–redshift relation using redMaPPer to predict intrinsic galaxy colors as a function of redshift Red galaxies are selected by fitting every galaxy with the redMaPPer red-sequence model to determine a photometric redshift estimate zred
Bias Mitigation and Noise Reduction
The initial reconstruction is both biased and noisy because the red-sequence calibration and photometric redshifts are themselves affected by reddening DustER mitigates these effects through Wiener filtering and a statistical debiasing procedure that leverages the empirically determined one-point distribution of Galactic dust The raw DustER map is obtained by modeling the observed color of red-sequence galaxies in sky pixels as having Gaussian random noise This raw map is then processed through a Wiener filter to suppress noise, resulting in the filtered map Finally, a debiasing step models the bias as a power-law plus a constant
Validation and Performance
DustER was validated using simulated data based on the Cardinal catalog, where dust attenuation is applied using the SFD dust map in our fiducial simulation The recovered DustER map is statistically unbiased relative to the input dust field, with a mean bias of 0.5 mmag and an rms scatter of 6.3 mmag The recovered map is nearly perfectly correlated with the input map on large scales, with the correlation decreasing on small scales where the reconstruction is noise-dominated
LSS Contamination Analysis
The analysis detected statistically significant cross-correlations between the reconstructed DustER map and galaxy density fields in our simulations The non-zero DustER–red galaxy cross-correlation indicates that the DustER map itself is contaminated by large-scale structure Testing an LSS subtraction scheme showed that it does not remove the contamination, indicating that using all three redshift bins over-corrects the map
Conclusion and Future Work
The authors conclude that the proposed linear cleaning algorithm fails because it does not remove the contamination In future work, they plan to develop a forwardmodeling, field-level inference framework in which the dust and galaxy density fields are modeled jointly within a hierarchical Bayesian approach They anticipate this approach will yield posterior samples of the dust map and its uncertainties, with the maps being free of large-scale-structure correlations The authors plan to apply DustER to real photometric survey data, including observations from the Dark Energy Survey and the Dark Energy Camera All Data Everywhere (DECADE) program The resulting reconstructions will provide an independent reddening-based perspective on Galactic extinction and will allow detailed comparisons with emission-based maps such as SFD and P16-GNILC
Pruning redundant reasoning tokens
DustER uses red galaxies as “standard crayons” It relies on the mean color of red-sequence galaxies being independent of their position in the sky Given the galaxy’s redshift, the difference between a galaxy’s observed color and its expected color measures the reddening by dust along the galaxy’s line of sight
How the training signal is built
The calibration of our red-sequence model is performed using raw (i.e. not dereddened) galaxy magnitudes This is done to ensure the DustER dust maps are not informed by the reddening values from existing dust maps The selection of red galaxies is performed using raw (i. p. 8 not dereddened)
galaxy magnitudes. This is done to ensure the DustER dust maps are not informed by
the reddening values from existing dust maps. not dereddened)
galaxy magnitudes. This is done to ensure the DustER dust maps are not informed by
the reddening values from existing dust maps.Where the paper says it
Where the paper says it
Where the gains come from
The final dust map is obtained as the posterior expectation value of the extinction field, with pixel-level uncertainties derived from the posterior variance The authors found that the choice of prior used in DustER has little impact on the recovered dust map The correlation coefficient is nearly identical regardless of which dust map is used as the input to the simulations The resulting maps are nearly perfectly correlated with the input on large scales, with the correlation decreasing on small scales where the reconstruction is noise-dominated
The method uses red galaxies as “standard crayons” It relies on the mean color of red-sequence galaxies being independent of their position in the sky Given the galaxy’s redshift, the difference between a galaxy’s observed color and its expected color measures the reddening by dust along the galaxy’s line of sight
The final dust map is obtained as the posterior expectation value of the extinction field, with pixel-level uncertainties derived from the posterior variance The authors found that the choice of prior used in DustER has little impact on the recovered dust map The correlation coefficient is nearly identical regardless of which dust map is used as the input to the simulations
The calibration of our red-sequence model is performed using raw (i.e. not dereddened) galaxy magnitudes This is done to ensure the DustER dust maps are not informed by the reddening values from existing dust maps The selection of red galaxies is performed using raw (i.e.
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Update DustER to incorporate a forwardmodeling, field-level inference framework in which dust and galaxy density fields are modeled jointly within a hierarchical Bayesian approach to explicitly parameterize and marginalize over their correlations. This will allow for posterior samples of the dust map and its uncertainties that are free of large-scale structure correlations.
-
Enhance the Wiener filtering step to better account for the non-Gaussian nature of residuals in low extinction regions by refining the modeling of the power spectrum C obs l as a sum of a power-law and a constant noise term Cn. This will lead to more accurate de-noising of raw DustER maps.
-
Improve map debiasing by refining the estimation of the one-point function Pdust using statistical properties from both SFD and P16-GNILC dust maps to better constrain the parameters a and u. This will result in a more robust recovery of the true dust field, as demonstrated by comparing results across different prior choices.
-
Develop methods to remove LSS contamination using LSS subtraction based on Chiang's work, specifically refining the linear combination coefficients Ai l through better handling of pseudo-C l cross-power spectra at effective multipoles. This will improve the effectiveness of cleaning maps, as the current method is found to be over-corrected when using all three redshift bins.
-
Implement a mechanism to generate correlated posteriors for pixel-level uncertainties instead of marginalizing them, which is currently limited to Gaussian assumptions. This will allow the system to fully characterize the spatial structure of statistical uncertainties in DustER maps.
Abstract
We present DustER, a new method for reconstructing dust reddening maps from photometric galaxy survey data. DustER uses red galaxies as "standard crayons". That is, it relies on the mean color of red-sequence galaxies being independent of their position in the sky. Given the galaxy's redshift, the difference between a galaxy's observed color and its expected color measures the reddening by dust along the galaxy's line of sight. When using photometric galaxy survey data, the method must rely on photometric redshifts, which, combined with selection effects, bias the recovered dust maps. DustER uses a novel approach to characterize and correct for these biases by leveraging the known statistical properties of the distribution of dust around the Milky Way. Testing on simulations, we find that the recovered dust maps are contaminated by large-scale structure. However, when we use the DustER map to deredden and select red-sequence galaxies, we find the resulting two-point function is not contaminated by dust.
Sources
- The Buzzard Flock: Dark Energy Survey Synthetic Sky Catalogs
- Assessing Galaxy Limiting Magnitudes in Large Optical Surveys
Related papers
- On-sky demonstration of a vector Zernike wavefront sensor in a cascaded adaptive optics system
- Euclid Data Release 1 (DR1): Calibration strategy and observations
- The General Antiparticle Spectrometer (GAPS) Antarctic Balloon Payload
- Detection and excision of radio-frequency interference in radio arrays
- Unified matched-filter and excess-power pipeline for gravitational-wave detector characterization
- Ushering in sub-arcsecond angular resolution observatories in the far-infrared: The case for a double-Fourier interferometer