Pith. sign in

REVIEW 3 major objections 5 minor 23 references

Atmospheric pressure dependance of HAWC scaler system

T0 review · 3 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

Pith's one-line read The HAWC scaler rates' 12-hour oscillation is a barometric signal, giving a per-PMT pressure coefficient of about $-0.34\,\%/\mathrm{hPa}$ and a quantitative malfunction flag for bad PMTs.

desk verdict Solid detector-calibration extension with a useful PMT-health diagnostic, but the 2-cpd barometric isolation needs a control before the health cuts are relied on. read the letter →

arxiv 1908.07473 v1 pith:RREMGWGQ submitted 2019-08-20 astro-ph.HE astro-ph.IM

classification astro-ph.HEastro-ph.IM
keywords HAWCscalerratespressurecoefficientatmospherictideFFTnarrow-bandfilterPMThealthmonitoringsolarmodulationcosmic-raysecondaries
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

Ground-based cosmic-ray data mix true solar-driven modulations with weather effects, and HAWC's scaler rates are no exception; separating them is the necessary step before the detector can be used for space-weather studies. This paper tries to establish that the 12-hour (2 cycles per day) oscillation seen in every PMT rate is a barometric signal, produced by the atmospheric pressure tide at the HAWC site. Using a narrow FFT filter centered on 2 cycles per day, the paper isolates that component and finds a near-perfect anti-correlation between pressure and rate, giving a pressure coefficient $\beta_P$ of about $-0.34\,\%/\mathrm{hPa}$ for single-PMT rates and stable month-to-month values for the multiplicity rates. Because this coefficient should be the same for every PMT, the paper further claims that PMTs whose fitted value departs by more than $3\sigma$ are malfunctioning and should be excluded from solar-modulation analysis. If correct, the result turns a nuisance correction into a self-calibrating quality check for all 1180 PMTs and 295 tanks.

What carries the argument

The load-bearing object is the 12-hour solar atmospheric tide at the HAWC site, visible as a dominant 2-cycle-per-day peak in both barometer and scaler-rate power spectra. The paper's tool is a narrow-band frequency filter $W(f)$ with central frequency $f_c=2$ cycles per day and $\Delta f=0.01$: full acceptance from $1.99$ to $2.01$ cycles per day, a sinusoidal roll-off to zero by $1.98$ and $2.02$, and zero outside. Applied to the FFT of pressure and rates and followed by an inverse FFT, the filter reconstructs only the barometric oscillation with zero baseline, so a direct fit of rate against pressure yields $\beta_P$ for each of the 1180 PMTs without contamination from solar diurnal anisotropy or other low-frequency modulations.

What would settle it

Apply the same narrow FFT filter to a multi-day stretch in which the barometer has no 12-hour pressure tide, such as sustained stormy weather; if the reconstructed rate still oscillates at 2 cycles per day with comparable amplitude, the scaler-rate component is not dominantly barometric, and the beta estimate would be contaminated.

Watch

Extended reading notes

Core claim

The central claim is that the 2-cycle-per-day component of the HAWC TDC-scaler rates is dominated by atmospheric pressure, and that this fact can be exploited without a long calibration run. The paper filters both pressure and rate data in the Fourier domain with the passband $1.98$-$2.02$ cycles per day, transforms back, and observes a clean 12-hour anti-correlation. Fitting $R(P)=R(P_m)\exp(\beta_P\,\Delta P)$ and its linear approximation gives $\beta_P=-0.337\,\%/\mathrm{hPa}$ (exponential) and $-0.343\,\%/\mathrm{hPa}$ (linear) for the average single-PMT rate for September-November 2016, with multiplicity rates varying by multiplicity. The paper then argues that since the pressure response of the atmosphere is independent of the detector, the spread of $\beta_P$ across PMTs should be small; with $\sigma=1.29\times10^{-2}\,\%/\mathrm{hPa}$ it classifies PMTs and tanks as good, within $3\sigma$-$5\sigma$, or above $5\sigma$, and retains the first group for solar-modulation studies.

Load-bearing premise

The argument assumes that at 2 cycles per day the only significant source of oscillation in the scaler rates is atmospheric pressure, so the narrow FFT filter removes solar and temperature contributions along with the barometric signal.

Editorial extensions

If this is right

  • Pressure-corrected R1 shows a drastically reduced 2-cycle-per-day peak while other frequency amplitudes are nearly unchanged, so the remaining data are cleaner inputs for solar-modulation studies.
  • PMTs and tanks outside the $3\sigma$ interval of the $\beta_P$ distribution can be rejected as malfunctioning, giving a quantitative, automated quality cut for 1180 PMTs and 295 tanks.
  • The $\beta_P$ values for September, October, and November 2016 are consistent within statistics, so the correction can be applied month by month without retuning.
  • Multiplicity rates M2, M3, and M4 receive the same correction, allowing tank-level analyses to use the same pressure removal.

Reading between the lines

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

  • A testable extension is to check whether the $3\sigma$ outliers in $\beta_P$ coincide with PMTs flagged by independent diagnostics such as gain drift or dark-noise rate; the paper does not report such cross-validation.
  • The same frequency-isolation recipe should transfer to other ground-based detectors at tropical or equatorial latitudes where the 12-hour pressure tide is strong, though the paper only demonstrates it for HAWC.
  • The paper leaves open the possibility that temperature or a solar-diurnal harmonic also contributes at 2 cycles per day; correlating the filtered rate component with simultaneous temperature and interplanetary indices would quantify that contamination.
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.

Referee Report

3 major / 5 minor

Summary. The paper proposes a method to isolate the atmospheric-pressure modulation in HAWC TDC-scaler rates by exploiting the strong 12-hour (2 cycles per day) periodicity of pressure at the HAWC site. A narrow-band FFT filter around 2 cpd is applied to both the pressure and rate time series, the filtered data are transformed back to the time domain, and the slope of the filtered rate versus filtered pressure yields a pressure coefficient beta_P. The mean beta_P for single-PMT rates is about -0.34%/hPa, with similar values for multiplicity rates. The paper then uses the distribution of per-PMT beta_P values to classify PMTs as healthy or malfunctioning, selecting only PMTs within a 3-sigma range for solar modulation studies. The central claim is that the 2-cpd component of the scaler rates is dominated by pressure, so the filtered signal isolates barometric effects.

Significance. If the causal attribution to pressure is correct, the method provides a simple, data-driven way to estimate beta_P and a potentially useful PMT health diagnostic for HAWC and similar detectors. The paper is a conference proceedings that directly addresses a practical need for solar modulation studies. The manuscript is honest about the data and methods, and the tables of monthly beta_P values are useful. However, the central claim relies on an untested spectral-coincidence argument, and the health classification inherits that assumption, so the significance is limited until the attribution is validated.

major comments (3)
  1. [Section 3, Eq. (3.1) and text after it] The statement that the narrow-band filter W(f) extracts 'only the 2 cpd barometric effects' is an assumption, not a demonstrated result. The paper shows that both the pressure and rate power spectra exhibit a peak at 2 cpd, but spectral coincidence does not establish that no other physical process contributes coherent power at that frequency. Atmospheric temperature tides and solar semidiurnal cosmic-ray anisotropy are known 2-cpd modulations; if present, they would be absorbed into the fitted beta_P and bias its value. Because beta_P is subsequently used as the PMT health gate in Section 5, this untested causal attribution is load-bearing for the paper's main applications.
  2. [Section 4, Fig. 9] The demonstration that pressure correction 'reduced drastically' the 2-cpd peak is a consistency check of the fit, not an independent validation. The beta_P used for the correction is estimated from the same filtered 2-cpd band that the correction then removes, so a reduction of that band is expected by construction. The paper does not quantify the residual 2-cpd power after correction or compare it with the level expected from statistical fluctuations. Without such a comparison, the claim that pressure accounts for the bulk of the 2-cpd rate modulation is not supported.
  3. [Section 5, Fig. 10 and Table 2] The PMT health classification assumes that all healthy PMTs have a common beta_P and that any deviation is due to malfunction, but the paper does not test this against independent diagnostics. For example, beta_P might correlate with PMT gain, operating voltage, or other known characteristics, and electronics or readout differences could produce apparent dispersion in beta_P without indicating a detector fault. To make the health classification convincing, the authors should validate it by showing that flagged PMTs exhibit independent signs of malfunction (e.g., abnormal pulse shapes, rate discontinuities, or poor performance in other analyses) rather than relying on beta_P alone.
minor comments (5)
  1. [Title and Abstract] The word 'dependance' is misspelled in the title and abstract; it should be 'dependence'.
  2. [Section 3] The text contains a garbled phrase 'Fast Fourier TransformÂt' (around the FFT definition); this appears to be an encoding artifact and should be corrected.
  3. [Section 3, Eq. (3.1)] The filter parameters fc and Delta f are introduced without explicitly stating the units of Delta f; the text later says Delta f = 0.01, but it would be clearer to state that this is in cycles per day.
  4. [Section 4, Fig. 9 caption] The caption does not explain which color corresponds to before and after correction in the power spectrum panel; the text mentions red and blue but the caption relies on the reader to infer the order.
  5. [Section 5, text after Table 2] The phrase 'The smilar selection process' contains a typo; it should be 'similar'.

Circularity Check

2 steps flagged · score 6.0 of 10

The 2-cpd filter output is labeled 'barometric' by construction, and the fitted beta_P is then used to 'confirm' that the correction removes the 2-cpd peak.

  1. fitted input called prediction [Section 4, 'Pressure correction', Fig. 9; Eqs. (3.2)-(3.3)]
    "It is clear from the figure that the amplitude of 2 cpd was reduced drastically which was mainly due to the pressure component, whereas the amplitudes corresponding to other frequencies were almost unaffected."

    beta_P was obtained by regressing the 2 cpd filtered R1 on the 2 cpd filtered pressure in Section 3 (Eqs. 3.2-3.3). Subtracting beta_P times pressure from R1 therefore minimizes, by construction, the residual power in that same 2 cpd band. The 'drastic' reduction of the 2 cpd peak is the least-squares objective of the fit, not an independent confirmation that the removed component was generated by pressure. Any coherent non-pressure 2 cpd signal, such as a semidiurnal temperature tide or solar semidiurnal anisotropy, would be absorbed into beta_P and removed as well.

  2. self definitional [Section 3, 'Estimation of pressure coefficient beta_P', Eq. (3.1)]
    "To extract only the 2 cpd barometric effects from the data sets we used a narrow-band filter W(f)."

    The filter output is labeled 'barometric effects' before any test of exclusivity. The paper later concludes that the 2 cpd peak is 'mainly due to the pressure component' (Section 4), but that conclusion is already assumed in the definition of the filter target. Any 2 cpd signal, whatever its physical origin, is selected by W(f) and attributed to pressure, so the causal claim is built into the choice of the passband rather than derived from the data.

full rationale

The beta_P estimate itself is a legitimate least-squares regression of the 2 cpd filtered rate on the 2 cpd filtered pressure, so the numerical value -0.34 %/hPa has independent content and could be checked against storm data or known muon pressure coefficients. The circularity is in the causal validation, not in the fit. Section 3 defines the narrow-band filter output as 'only the 2 cpd barometric effects,' so any non-pressure 2 cpd modulation is attributed to pressure by construction. Section 4 then shows the 2 cpd peak is 'reduced drastically' after correction; since beta_P was fitted to minimize the residual between the same two filtered time series, that reduction is the expected outcome of the regression, not independent evidence for the pressure origin. The three-month consistency demonstrates stability of the fit but does not test the causal attribution. The PMT-health classification in Section 5 assumes beta_P uniformity and interprets outliers as malfunction without external malfunction records; this is an unsupported assumption and a correctness risk, but it is not a construction-level circularity. The self-citations to refs. [6,7] for 'similar filters' are not load-bearing because the filter W(f) is fully specified in Eq. (3.1) of this paper. Overall, partial circularity: the central coefficient has independent content, but the paper's demonstration that the 2 cpd component is pressure reduces to the filter definition and the fitted correction.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The analysis introduces no new physical entities. It depends on several domain assumptions about the atmospheric tide, the uniformity of PMT response, and the representativeness of the pressure measurement. The fitted pressure coefficient is the measured quantity itself, while the free parameters listed here are the user-chosen analysis settings that could affect the result.

free parameters (4)
  • Filter half-width delta_f in W(f) = 0.01 cpd
    Chosen by hand in Section 3; no robustness scan over delta_f is shown, yet it determines which frequencies are attributed to pressure.
  • Pressure bin width = 0.1 hPa
    Used in Section 3 to bin IFFT pressure-rate pairs before fitting; no sensitivity check is reported.
  • Outlier cutoff in sigma = 3 sigma, with 2 sigma and 5 sigma also shown
    Section 5 classifies PMTs as good or malfunctioning using the 3 sigma threshold; this choice is not derived from data or external validation.
  • Linear versus exponential fit choice = Linear method chosen
    Section 4 chooses linear over exponential based on mean chi-square values without reporting degrees of freedom or parameter correlations.
assumptions (5)
  • domain assumption The 12-hour oscillation at the HAWC site is a solar atmospheric tide with no significant non-pressure contribution at 2 cpd.
    Invoked in Section 3 when the filter W(f) is said to select 'only the 2 cpd barometric effects'; no test for solar or temperature contributions at that frequency is given.
  • domain assumption TDC-scaler rates respond to pressure according to R(P)=R(Pm) exp(beta_P delta_P) with a single beta_P per PMT.
    Equation 3.2/3.3 is chosen empirically and only first-order linearized; no model comparison to alternative functional forms is presented.
  • domain assumption Pressure modulation is a purely physical atmospheric phenomenon, so all PMTs should exhibit the same percentage pressure coefficient regardless of gain and quantum efficiency.
    Section 2 states that percentage variations are the same for every PMT, and Section 5 uses this to classify deviations as malfunctions. The claim is not validated against external PMT health records.
  • domain assumption The pressure measured by the single HAWC barometer represents the air column over every PMT and tank.
    Section 2 states that pressure is measured every minute with a barometer; no correction for spatial pressure gradients is discussed.
  • standard math Standard FFT and IFFT narrow-band filtering preserve the phase and amplitude of the 2 cpd component.
    The method relies on standard Fourier analysis; no formal proof is given, but this is accepted tooling in signal processing.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Atmospheric pressure dependance of HAWC scaler system." pith.science (2026). https://pith.science/paper/RREMGWGQ

@misc{pith2026190807473,
  author       = {Pith},
  title        = {Pith review of: Atmospheric pressure dependance of HAWC scaler system},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RREMGWGQ}},
  note         = {Machine review of arXiv:1908.07473}
}
abstract

The variation in atmospheric pressure is due to changes in mass of the air column above, which in turn resembles the density variation of atmosphere and will affect the decay of secondary particles of cosmic rays. The ground based cosmic ray detectors observe pressure dependent variation in their flux. The High Altitude Water Cherenkov (HAWC) gamma ray observatory is a great detector of secondary particles because of its high altitude, high uptime, and large area (including total photo-cathode area), which makes the HAWC scaler system an ideal instrument for solar modulation studies. Although, in order to perform these studies it is necessary to isolate and remove the atmospheric modulations. The observed rate in each PMT has signatures of both the solar and atmospheric modulations, which makes it difficult to measure the pressure coefficient ($\beta_P$ ). The pressure at the HAWC site shows a periodic behavior ($\sim$ 12 hours), which also reflects in the scalar rates. This periodic property was used to isolate the pressure modulation and $\beta_P$ were estimated with accuracy. Since the pressure dependence is a physical phenomenon, the estimated coefficients for PMTs should be identical, any deviation from this can be due to malfunction of the PMT. This make this method a useful tool to identify the malfunctioning PMTs and help us to isolate them from the analysis. In this analysis we are presenting the method of estimation of the pressure coefficients and its usage to correct the HAWC scalar data to make it suitable for the solar modulations studies.

Figures

Figures reproduced from arXiv: 1908.07473 by the authors.

Figure 1
Figure 1. The top four panels show the observed HAWC TDC-scaler data R1 and the multiplicity rates RM2, RM3, RM4 respectively from top to bottom. Rates of a few example PMTs and tanks are shown in different colors. The bottom￾most panel shows the ambient pressure at HAWC site. The GCRs reaching the top of the atmo￾sphere interact with atmospheric nuclei and pro￾duce an increasing flux of secondary particles as they propagate … view at source ↗
Figure 2
Figure 2. FFT spectrum of pressure at the top panel and TDC scaler rate R1 at bottom panle, for the month of October 2016 [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 4
Figure 4. Top panel shows the pressure variation at HAWC site from 2 October 00 hr from 8th October 00 hr, 2016. the bottom panel shows the IFT of the 12 hour periodic nature of pressure during the same time [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗
Figures from the paper (3 more)
Figure 6
Figure 6. Figure 6: IFFT data of pressure and R1 in local time domain, folded to a 24 hour format, top panel is of pressure and bottom one for TDC scaler rate R1 [PITH_FULL_IMAGE:figures/full_fig_p005_6.png]
Figure 8
Figure 8. Figure 8: Distribution of the pressure coefficient βP for all PMTs for the months of September, October, and November. the plots in left side is for exponential method, and that in right sides are by linear approximation method. Mean value of distribution is given along with the…
Figure 9
Figure 9. Figure 9: Top panel shows the R1, the one in red is R1 before correction and in blue is the same after pressure correction. The bottom panel shows the power spectra of R1, before pressure correction in red and after correction in blue [PITH_FULL_IMAGE:figures/full_fig_p007_9.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

23 extracted references · 23 canonical work pages

  1. [1]

    Forbush, J

    S.E. Forbush, J. Geophys. Res. 59 (1954) 525

  2. [2]

    Burlaga, L. F., et. al, 1985, J. Geophys. Res. 90, 12027. doi:10.1029/JA090iA12p1202

  3. [3]

    Subramanian, et al., Astron

    P. Subramanian, et al., Astron. Astrophys. 494 (2009) 1107

  4. [4]

    Arunbabu, et al., Astron

    K.P. Arunbabu, et al., Astron. Astrophys. 555 (2013) A139

  5. [5]

    Arunbabu, et al., Astron

    K.P. Arunbabu, et al., Astron. Astrophys. 580 (2015) A41

  6. [6]

    Mohanty, et al., Pramana J

    P.K. Mohanty, et al., Pramana J. Phys. 81 (2013) 343

  7. [7]

    Mohanty et al., Astropart

    P.K. Mohanty et al., Astropart. Phys. 79 (2016) 23

  8. [8]

    Arunbabu et al., Astropart

    K.P. Arunbabu et al., Astropart. Phys. 94 (2017) 22

Show all 23 references
  1. [9]

    Greisen, Ann

    K. Greisen, Ann. Rev. Nucl. Sci. 10 (1960) 63

  2. [10]

    Olive, et al., Chinese Phys

    K.A. Olive, et al., Chinese Phys. C 38 (2014) 090001

  3. [11]

    Grieder, Cosmic Rays at Earth, Elsevier Science, (2001)

    P.K.F. Grieder, Cosmic Rays at Earth, Elsevier Science, (2001)

  4. [12]

    Achard, et al., Phys

    P. Achard, et al., Phys. Lett. B 598 (2004) 15

  5. [13]

    of the 33^ rd International Cosmic Ray Conference

    Alejandro Lara, 2013, Proc. of the 33^ rd International Cosmic Ray Conference. Rio DE Janeiro, Brasil

  6. [14]

    Lindzen R. S., Ann. Rev. Planet Sci. (1979), 7

  7. [15]

    Carrasco E., et al., Mon. Not. R. Astron. Soc. (2009), 398, 407-421

  8. [16]

    Mohanty, et al., Proceedings of the 34^ th International Cosmic Ray Conference

    P.K. Mohanty, et al., Proceedings of the 34^ th International Cosmic Ray Conference. The Hague, The Netherlands , POS(ICRC2015)042

  9. [17]

    http://root.cern.ch/root/html/TVirtualFFT.html

  10. [18]

    Harris, Proc

    F.J. Harris, Proc. IEEE 66 (1978) 51

  11. [19]

    S., et al., 2014, Sol

    Potgieter, M. S., et al., 2014, Sol. Phys., 289, 391

  12. [20]

    Giacalone, J., & Jokipii, J. R. 1999, ApJ, 520, 204

  13. [21]

    Candia, J., & Roulet, E. 2004, J. Cosmol. Astropart. Phys., 10, 007

  14. [22]

    H., et al., 2003, ApJ, 590, 53

    Matthaeus, W. H., et al., 2003, ApJ, 590, 53

  15. [23]

    2010, ApJ, 720, 127

    Shalchi, A. 2010, ApJ, 720, 127

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.