Skip to content

Errata in Southall et al. (2019)

This is the entry of the errata registry on Southall et al. (2019). The registry’s introduction says how each claim is checked against the printed page and what the status of an entry means.

Southall et al. (2019), Aquatic Mammals 45(2), Table 7 (impulsive peak SPL)

Section titled “Southall et al. (2019), Aquatic Mammals 45(2), Table 7 (impulsive peak SPL)”

Peer-reviewed journal paper, not a standard.

  • Location: Table 7 (printed p. 156), the impulsive-noise TTS and PTS onset criteria; the two in-air carnivore rows PCA and OCA.
  • The print: PCA TTS peak SPL 138 and PTS peak SPL 144; OCA TTS peak SPL 161 and PTS peak SPL 167 .
  • The problem: all four are typographical errors. The authors’ own errata (Aquatic Mammals 45(5), 569-572, DOI 10.1578/AM.45.5.2019.569) names all four on printed p. 569, “There are four typographical errors in Table 7 on page 156”, and reprints the corrected table on printed p. 570: PCA 155 and 161, OCA 170 and 176. The same errata also corrects the column headed “B” in Table 5 to the parameter b of Eq. (2), which it likewise calls a typographical error.
  • Evidence: the errata itself, which names each wrong value and its replacement, corroborated by the article’s own extrapolation rule. Note first what does not discriminate. The PTS peak = TTS peak + 6 dB rule of printed p. 155 is satisfied by the printed pair as well (, just as ), so it says nothing about which pair is right. Nor does the duplication visible in the printed rows, where the peak-SPL TTS entry equals that same row’s PTS-onset SEL entry (PCA 123 / 138 / 138 / 144 and OCA 146 / 161 / 161 / 167, reading TTS SEL, TTS peak, PTS SEL, PTS peak): for these two in-air rows that equality is forced by two rules the article states on printed p. 155, both adding 15 dB to the same base TTS SEL, so it would hold whatever the SEL values were. An earlier revision of this entry read that equality as the signature of a column slip; it is instead the printed table being internally consistent with the article’s own in-air method, which is what makes the errata the only thing that settles the matter.
    • Value. The corrected numbers are close to what the article’s extrapolation rule produces, with the caveat that the rule is not stated for these rows. Printed p. 155 sets the impulsive peak-SPL TTS onset of a group without direct data at the hearing threshold at the frequency of best sensitivity plus 159 dB, and restricts that rule explicitly to the in-water groups: “For other species groups in water (LF, SI, PCW, and OCW), 159 dB was added to the value of the hearing threshold at f₀”. It works the rule through for PCW: “Peak SPL TTS onset was estimated as 212 dB re 1 µPa (53 dB at f₀ + 159 dB)”. Evaluating the Table 2 group audiogram at the Table 4 reproduces the three in-water rows the errata does not touch (SI 219.6 against a published 220; PCW 212.5 against 212; OCW 226.1 against 226), which validates the rule where the article applies it. Extending it to the two in-air carnivore rows, which the article does not do, gives PCA at 2.3 kHz and OCA at 10 kHz, hence 154.4 and 170.4. Those reproduce the corrected 155 and 170 to within 0.6 dB and are 16 dB and 9 dB away from the printed 138 and 161, which is what makes them corroborating rather than confirming; note that 154.4 rounds to 154, not to 155, and an earlier revision of this entry claimed that it rounded to the corrected value.
    • A second, unrepaired inconsistency. Printed p. 155 states that for the in-air carnivores specifically “a nominal 15 dB offset is used … between the SEL-based TTS threshold and the peak SPL-based threshold”, which reproduces the printed 138 and 161 from the SEL column. That sentence, not the +159 dB rule, is the one the article’s own method applies to PCA and OCA. The errata resolves the conflict in favour of values consistent with the +159 dB rule, so it supersedes the sentence as well as the table; the sentence is left standing in the article. Verified on PDF page 31 (printed p. 155) and PDF page 32 (printed p. 156) of Southall et al. (2019), Aquatic Mammals 45(2), which carry the “in water (LF, SI, PCW, and OCW)” restriction, the 15 dB in-air offset in the same paragraph, both statements of the +6 dB rule, and the article’s Table 7 with the PCA row 123 / 138 / 138 / 144 and the OCA row 146 / 161 / 161 / 167. The errata is a publication of its own, Aquatic Mammals 45(5), 569-572, bound at the end of the copy the authors distribute: verified there on PDF page 109 (printed p. 569), which names all four values and their replacements, and PDF page 110 (printed p. 570), which reprints Table 7 with PCA 123 / 155 / 138 / 161 and OCA 146 / 170 / 161 / 176.
  • Library behaviour: the errata-corrected values are the ones implemented in bioacoustics/weighting.py, pinned by test_southall_table_7_errata_values_are_implemented, with the +159 dB rule itself checked against the audiogram in test_southall_impulsive_peak_spl_is_threshold_at_f0_plus_159_db for the in-water groups the article restricts it to and, separately and with the extrapolation labelled as such, for PCA and OCA.
  • Status: reported by the authors themselves (errata published 2019).