Pith. Sign in

REVIEW

GalaxyCount: a JAVA calculator of galaxy counts and variances in multiband wide-field surveys to 28 AB mag

PaperPDF from arXiv

data files

ReviewThe objections and verdict, as JSON Paper breakdownThe authors' claims, conditions and measurements, as JSON Signed receiptProof this review has not been altered
X Bluesky LinkedIn Reddit Hacker News
S.C. Ellis, J. Bland-Hawthorn (AAO)

Astrophysics · Submitted 27 Feb 2006 · 103 references

Pith's verdict: Useful public tool, but the abstract overstates the validation: the R-band success is partly a calibration check rather than an independent prediction.

T0 · AI review

How this review was made: An AI referee (deepseek-v4-flash) read the paper on August 28, 2026 and answered a fixed list of questions about it. T0 is the starting level every paper gets. The levels describe how deeply Pith has checked the paper, never the authors.

  1. T0 AI review An AI read the paper and answered a fixed list of questions.
  2. T1 Audit Automatic checks that the references and files the paper relies on exist and match.
  3. T2 Compute The paper's code or calculations re-run with the same results.
  4. T3 Formal statement The claims written out in a formal math language a computer can read.
  5. T4 Proof Those claims proved, and the proof checked by computer.

How Pith checks papers →

· 3 major objections · 6 minor · Ask about this paper · The compiled record · Read the full review →
Challenge this review Comment (0) Read on arXiv ↗
Cite ▾
arXiv IDastro-ph/0602573 (version 2)
Pith ID7CGWWHMV, Pith's permanent record of this paper and its review
CitationJ. Bland-Hawthorn (AAO), S.C. Ellis. "GalaxyCount: a JAVA calculator of galaxy counts and variances in multiband wide-field surveys to 28 AB mag." arXiv:astro-ph/0602573 (2006). https://pith.science/paper/7CGWWHMV
Cite this reviewBibTeX for Pith's review of this paper
Claim paper

Why Pith says this

Highlight any text to comment on it.

Referee report

How Pith checks papers → · Full audit of this paper →

The paper

The paper claims

A consistent framework predicts galaxy counts and their scatter to 28 AB mag.

The reading

GalaxyCount assembles three observational ingredients—galaxy number densities compiled from published surveys, a magnitude-dependent amplitude for the angular correlation function, and a sigmoid description of detection completeness—into a public calculator that predicts, for any waveband, area, and depth, how many galaxies an image should contain and how much that count will scatter. The scatter matters because the uncertainty is not purely Poisson: galaxies cluster, so the variance carries an extra cosmic-variance term that grows with area and depth. The authors compare the model against random counts-in-cells drawn from the Subaru Deep Field and show that, when the input counts are right, the predicted standard deviation matches the measured scatter; they also show that cosmic variance alone can explain the difference between the Hubble Deep Field North and South. Two worked uses follow: a reanalysis of the NGC 300 outer-disc star counts confirms the disc continues to ten scale lengths, and a statistical background subtraction reproduces the spectroscopically determined luminosity function of Abell 2734, suggesting that offset fields can often be skipped.

What carries the argument

The load-bearing identity is the counts-in-cells variance formula $\sigma^2 = n\Omega + n^2\int d\Omega_1\,d\Omega_2\,\omega(\theta_{12})$, with $\omega(\theta) = A\theta^{-\delta}$. For a square cell the clustering integral reduces to 2.24 times a scale factor (about 1.53 for a circular cell), so the practical problem is fixing $n(m)$ and $A(m)$ in each band. GalaxyCount builds $n(m)$ by binning and cubic-spline fitting a large compilation of published number-count surveys, then applies a sigmoid completeness function $f = 1/(1+e^{a(x-b)})$ to remove the counts that a particular telescope, exposure, seeing, and detection threshold would miss. For $A(m)$ it offers either a linear fit to $\log A$ versus magnitude or an $\epsilon$-parameterised evolution model converted from the spatial correlation function; the $\epsilon=3$ option, which flattens $A$ at faint magnitudes, is the default used for the paper's headline results.

Extended reading notes

Core claim

The central claim is that one consistent framework—equation (1), in which the variance is the Poisson term $n\Omega$ plus a double integral over the angular correlation function $\omega(\theta)=A\theta^{-\delta}$—predicts the standard deviation of galaxy counts in deep multiband images once $n$ and $A$ are supplied as functions of magnitude. The authors demonstrate this against the Subaru Deep Field in the B, R, and I bands: using the measured SDF number counts as input, the model matches the observed total scatter, and a clustering-evolution model with parameter $\epsilon=3$, combined with a sigmoid completeness correction fitted to the SDF, reproduces both the total and the non-Poissonian parts of the variance. The same machinery, applied directly from compiled published counts, overpredicts faint counts unless completeness is folded in. The authors further report that the model places them within the upper quarter of offset-field realisations for the Abell 2734 luminosity function and that it accounts for the North–South Hubble Deep Field discrepancy as cosmic variance.

Load-bearing premise

The model's faint-end variance rests on an assumed behaviour of galaxy clustering at magnitudes fainter than any current measurement, so an incorrect extrapolation of the clustering amplitude would change the predicted uncertainty even if the number counts are correct.

Editorial extensions

If this is right

  • For observations where the number of galaxies is known a priori, the calculator gives the correct cosmic-variance uncertainty, so it can be used to quote errors for deep number-count measurements.
  • Statistical background subtraction with GalaxyCount matches the result obtained by subtracting a typical offset field in the Abell 2734 luminosity-function test, so blank-sky exposures may often be unnecessary.
  • The NGC 300 disc extends to about ten scale lengths; background levels high enough to force an earlier truncation are ruled out at high significance.
  • For surveys reaching $R \lesssim 26$ mag, the Poissonian error is within a factor of two of the true error only for areas smaller than about 10 square arcminutes; beyond that, cosmic variance dominates.
  • Published faint B-band number densities scatter more than cosmic variance predicts, so B-band statistical background subtraction is likely dominated by systematic uncertainties until the faint counts improve.

Reading between the lines

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

  • If the faint-end flattening of $A(m)$ is real, ultra-deep small-field surveys (the regime approaching JWST-era fields) will have larger background uncertainties than a log-linear extrapolation of $A$ would predict; choosing $\epsilon=3$ is effectively a conservative, variance-increasing default.
  • The same framework could be inverted: with well-measured completeness and counts, the excess scatter in counts-in-cells at faint magnitudes constrains $A(m)$ where direct measurements do not yet exist, turning the calculator into a probe of galaxy clustering at high redshift.
  • The completeness treatment's assumption that a single sigmoid shape scales across telescopes, filters, and seeing could be tested directly by comparing injection-recovery curves from different instruments; the paper's Figure 1 shows the slope $a$ varies from about 1.4 to 6.0 across detection strategies.
  • A star-count analogue of the calculator, which the paper lists as future work, would bring the same completeness-aware variance machinery to Milky Way structure studies and to surveys where stellar contamination must be modelled jointly with the galaxy background.
Original abstract

We provide a consistent framework for estimating galaxy counts and variances in wide-field images for a range of photometric bands. The variances include both Poissonian noise and variations due to large scale structure. We demonstrate that our statistical theory is consistent with the counts in the deepest multiband surveys available. The statistical estimates depend on several observational parameters (e.g. seeing, signal to noise ratio), and include a sophisticated treatment of detection completeness. The JAVA calculator is freely available and offers the user the option to adopt our consistent framework or a different scheme. We also provide a summary table of statistical measures in the different bands for a range of different fields of view. Reliable estimation of the background counts has profound consequences in many areas of observational astronomy. We provide two such examples. One is from a recent study of the Sculptor galaxy NGC 300 where stellar photometry has been used to demonstrate that the outer disc extends to 10 effective radii, far beyond what was thought possible for a normal low-luminosity spiral. We confirm this finding by a reanalysis of the background counts. Secondly, we determine the luminosity function of the galaxy cluster Abell 2734, both through spectroscopically determined cluster membership, and through statistical subtraction of the background galaxies using the calculator and offset fields. We demonstrate very good agreement, suggesting that expensive spectroscopic follow-up, or off-source observations, may often be bypassed via determination of the galaxy background with GalaxyCount.

The compiled record

The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.

Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-09-23: 1 request, 8,794 input tokens.

What the abstract itself states, as probabilities: claims first 0.04 · states limitation 0.18 · human participants 0.01 · controlled comparison 0.22 · randomized assignment 0.01 · states quantitative result 0.85 · compares with prior results 0.94 · mentions code or data availability 0.97.

The abstract, sentence by sentence

arXiv listing abstract v2. Results and the main claim are underlined; .

We provide a consistent framework for estimating galaxy counts and variances in wide-field images for a range of photometric bands. The variances include both Poissonian noise and variations due to large scale structure. We demonstrate that our statistical theory is consistent with the counts in the deepest multiband surveys available. The statistical estimates depend on several observational parameters (e.g. seeing, signal to noise ratio), and include a sophisticated treatment of detection completeness. The JAVA calculator is freely available and offers the user the option to adopt our consistent framework or a different scheme. We also provide a summary table of statistical measures in the different bands for a range of different fields of view. Reliable estimation of the background counts has profound consequences in many areas of observational astronomy. We provide two such examples. One is from a recent study of the Sculptor galaxy NGC 300 where stellar photometry has been used to demonstrate that the outer disc extends to 10 effective radii, far beyond what was thought possible for a normal low-luminosity spiral. We confirm this finding by a reanalysis of the background counts. Secondly, we determine the luminosity function of the galaxy cluster Abell 2734, both through spectroscopically determined cluster membership, and through statistical subtraction of the background galaxies using the calculator and offset fields. We demonstrate very good agreement, suggesting that expensive spectroscopic follow-up, or off-source observations, may often be bypassed via determination of the galaxy background with GalaxyCount.

background contribution method result interpretation main claim

What the classifier reports for each sentence. The number is the probability it gave the label.
#sentencerolethe authors' own workstates the main conclusionstrength
s0 We provide a consistent framework for estimating galaxy counts and variances in wide-field images for a range of photometric bands. contribution 1.00 0.91 0.17 asserted 0.99
s1 The variances include both Poissonian noise and variations due to large scale structure. method 0.86 0.56 0.05 asserted 1.00
s2 We demonstrate that our statistical theory is consistent with the counts in the deepest multiband surveys available. result 0.94 0.91 0.22 asserted 0.91
s3 The statistical estimates depend on several observational parameters (e.g. seeing, signal to noise ratio), and include a sophisticated treatment of detection completeness. method 0.98 0.75 0.06 asserted 1.00
s4 The JAVA calculator is freely available and offers the user the option to adopt our consistent framework or a different scheme. contribution 0.91 0.77 0.06 asserted 1.00
s5 We also provide a summary table of statistical measures in the different bands for a range of different fields of view. contribution 0.97 0.90 0.05 asserted 1.00
s6 Reliable estimation of the background counts has profound consequences in many areas of observational astronomy. background 0.88 0.06 0.08 asserted 1.00
s7 We provide two such examples. contribution 0.90 0.59 0.03 asserted 0.90
s8 One is from a recent study of the Sculptor galaxy NGC 300 where stellar photometry has been used to demonstrate that the outer disc extends to 10 effective radii, far beyond what was thought possible for a normal low-luminosity spiral. background 0.69 0.08 0.05 asserted 1.00
s9 We confirm this finding by a reanalysis of the background counts. result 0.76 0.91 0.06 asserted 1.00
s10 Secondly, we determine the luminosity function of the galaxy cluster Abell 2734, both through spectroscopically determined cluster membership, and through statistical subtraction of the background galaxies using the calculator and offset fields. contribution 0.61 0.95 0.12 asserted 1.00
s11 We demonstrate very good agreement, suggesting that expensive spectroscopic follow-up, or off-source observations, may often be bypassed via determination of the galaxy background with GalaxyCount. interpretation 0.97 0.91 0.48 hedged possibility 0.97

What kind of paper the classifier reports this is: method or tool 1.00 · evidence from analysis of existing data 0.90.

What was examined, and what was not

sources read
  • arXiv listing abstract v2
not examined
  • full text
  • figures and tables
  • references and citing sentences
  • relations to other papers
levels run
  • P abstract
not run
  • S sections
  • T body sentences
  • C citations
  • R cross-paper relations

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. Every number in the measurements table is the authors', checked by code. Record as JSON

Figures
Figure 1. Completeness functions compiled from the literature and our own data, all translated to have 50 per cent completeness at zero magnitude. The points show the data from: the Subaru Deep Field B band galaxy counts (open circles, Kashikawa et al. 2004); the author’s own GMOS g band observations of a red￾shift 0.7 cluster of galaxies (open squares); the author’s own IRIS2 K band observations of the same field (open triangles); ISAAC 1.19µm narrow band filter data of the Chandra Deep Field South (closed circles, Cuby et al. 2007); NACO AO cor￾rected observations (closed squares, Cresci et…
Figure 3. The effect of the throughput on the resulting number counts and standard deviation for 18 6 B 6 27 in a 5×5 ′ area, for a 595 minute exposure, with SNR=3, Γ ≈ 1”, on an 8m aperture telescope. The solid line shows the number counts and the shaded area the 1σ uncertainty on those counts. amplitude of the angular correlation function is the same in both observations, through the equation, σ2 = p f 2σ 2 1 + N1f(1 − f), (5) which may be combined with equation 4, to scale the results presented in
Figure 4. The expected number, standard deviation, and stan￾dard deviation divided by square root of the number (i.e. the non-Poissonian component of the standard deviation) of galax￾ies in surveys of different magnitude limits, with a 5′ × 5 ′ FOV. The thick black line is for the total number of galaxies, the dot￾ted and dashed lines include estimates for the incompleteness of the observations for exposure times of 595, 300 and 10 minutes, with SNR=3, Γ ≈ 1”, with an 8m aperture telescope with 36% throughput and deviate from the thick black line in that order. The functional dependence of A…
Figure 5. The modelled star counts from Robin et al. (2003) compared to galaxy counts compiled from the literature shown by the points. ξ(r) =  r0 r γ . (6) The scaling of ξ(r) with redshift, z, may be parametrized as (Groth & Peebles 1977), ξ(r, z) = h(z)  r0 r γ , (7) where h(z) = (1 + z) −(3+ǫ) , (8) where r is the proper distance and ǫ describes the evolution of the clustering, with ǫ = ( 0 – clustering fixed in proper co-ordinates −1.2 – clustering fixed in comoving co-ordinates 0.8 – prediction of linear theory. (9) In all cases with ǫ > −1.2 the comoving galaxy correla￾tion length…
Figure 6. The amplitude of the angular correlation function as a function of median apparent magnitude, compiled from the sources listed in appendix A2. The models described in section 2.3 are shown. In most cases a linear model fits the data very well. c 0000 RAS, MNRAS 000, 000–000
Figure 8. The effect of the window function on the standard deviation for a constant area of 25 square arcminutes. R ≈ F606 − 0.1 (16) I ≈ F814. 2.5 The effect of the window function GalaxyCount allows a choice of window functions so dif￾ferent detector shapes and illumination patterns may be modelled. The choices are square, circular, rectangular and elliptic. Square and circular windows are treated separately from rectangular and elliptic window functions for the pur￾pose of increasing the speed of the calculations. In
Figure 9. Comparison of the model to the SDF R band. The model number counts have been calculated for an ǫ = 3 model. The left hand panels show the data matched exactly to the SDF number counts. The middle panels show a model with number counts calculated from number density surveys (see appendix A1). The right hand panels show a model with number counts calculated from published surveys, but corrected for the incompleteness of the SDF observations. The top panels show the SDF mean number counts and the associated stand deviation, as open points with error bars. The grey area represents the…
Figure 10. The number density of galaxies vs. magnitude, with B on the left and R on the right. The grey shaded curves show the number density and 1σ uncertainty expected from cosmic variance. The hatched curves show the cosmic variance as measured in a survey with the the area of the Hubble Deep Field, and subsequently extrapolated to 1 deg2 . in the outer disc of NGC 300, a late-type spiral in the Sculp￾tor group. They used the Gemini Multi-object Spectrograph (GMOS) on the Gemini South 8m telescope in exceptional conditions (0.6” FWHM seeing). At a 3σ point source detec￾tion threshold of r…
Figure 11. The luminosity profile of NGC 300. The top panel shows the raw counts as the open symbols, with the associated error (calculated as the square root of the counts), the solid line represents the expected contribution to the raw counts from back￾ground galaxies, and the shaded region represents the uncertainty on this number as predicted by GalaxyCount. In the bottom panel the closed symbols represent the corrected star counts, and the combined errors. spectroscopic LF. Since incompleteness has been accounted for in the GalaxyCount derived model we also tested the latter technique.…
Figure 13. The best fitting Schechter functions as described in the text. The contours show the 68 and 90 per cent confidence limits of the spectroscopic LF. i.e. it accurately reflects the expected average result that would be obtained from subtracting a large number of offset fields. Thus GalaxyCount provides an accurate and efficient way to estimate the contribution of foreground and back￾ground galaxies to cluster LFs. In many cases expensive spectroscopic follow-up or off-source background observa￾tions may be by-passed and replaced with background esti￾mates from GalaxyCount combined…
Figure 14. Histogram of the ˜χ 2 difference between the statis￾tically subtracted LFs and the spectroscopic LF. The value for GalaxyCount is overlaid and is in the upper 25 per cent of the offset fields. 4.3 Comparison to the Hubble deep fields The Hubble Deep Field North (HDF-N) has higher galaxy number counts than the Hubble Deep Field South (HDF￾S, Metcalfe et al. 2001)
Figure 15. The difference between the Hubble deep fields N (squares) and S (circles), compared to the expected standard de￾viation (1σ=grey shaded area, 2σ=hatched area). pleteness is not an issue the predicted number counts are reliable. The problem may be alleviated with future, more pre￾cise determinations of the faint end number counts. Until such data is available it is up to the user to choose wisely the number counts which he or she wishes to use and to be aware of the possible systematic uncertainties introduced at faint magnitudes. The ǫ = 3 model was found to predict the standard…
Figure 16. The factor by which the total standard deviation dominates over the Poissonian error, σ/√ N, as a function of R band limiting magnitude for various field sizes (labelled in the plot in units of square arcminutes). ǫ is required to fit the faint end of the A scaling relation (McCracken et al. 2000b; McCracken et al. 2001; Coil et al. 2004), but note that these models cannot fit both the bright and the faint ends of the relation simultaneously. A precise determination of the value of A at faint magnitudes awaits future measurements. We remind the reader, however, that the models used…

Showing 13 of the paper's 16 figures. See all figures in the PDF ↗ Figures reproduced from arXiv:astro-ph/0602573 by the authors.

References

Discussion

Disagree with the review, or know something it missed? Leave a comment, highlight any passage above to comment on it, or challenge this review formally.