Erratum to: Sensitivity of multi-PMT optical modules in Antarctic ice to supernova neutrinos of MeV energy

1 Erratum to: Eur. Phys. J. C (2021) 81:1058 https://doi.org/10.1140/epjc/s10052-021-09809-y

We found an error in the code for calculating the CCSN detection range that led to double counting of signal events and thus to ranges that were too large by a factor of about 1.3. This results in the following corrections (old value $\rightarrow$ new value):

• The values for the detection ranges decrease as shown in the corrected Table 4. Figures  8, 9, 10 change accordingly. Correspondingly, the numbers change in:

  • Abstract: We find that exploiting temporal coincidences between signals in different photocathode segments, a $27\ {\mathrm {M}}_{\odot }$ progenitor mass CCSN can be detected up to a distance of ($341\,\hbox {kpc}\rightarrow 269\,\hbox {kpc}$) with a false detection rate of $0.01\,\hbox {year}^{-1}$ with a detector consisting of 10,000 sensors.

  • Section 4.2: The trigger condition ($m\ge 7$, $N_{\nu }\ge 7$) can be used to send supernova alerts with very high confidence (about one false detection per century), and identify CCSN at a distance of ($341\,\hbox {kpc}\rightarrow {269}\,\hbox {kpc}$) with $50\%$ probability.

  • Section 4.2: With a relaxed set of conditions of ($m\ge 7$, $N_{\nu }\ge 6$), SNe up to ($370\,\hbox {kpc}\rightarrow {291}\,\hbox {kpc}$) can be detected with less than one false CCSN detection per year.

  • Section 4.2: For example, for a number of detected events $N_\nu =5$ a background origin can be excluded at ${3.2}\,{\sigma }$, while at least a corresponding number of events will be detected in $50\%$ of cases from a $27\,{\mathrm {M}}_{\odot }$ CCSNe at a distance of ($407\,\hbox {kpc} \rightarrow {322}\,\hbox {kpc}$).

  • Section 4.2: If $N_\nu =7$ events with $m\ge 7$ are detected we obtain a ${4.9}\,{\sigma }$ confidence that such signal was not produced by background with a $50\%$ detection probability at (${341}\,\hbox {kpc}\rightarrow {269}\,\hbox {kpc}$) distance.

  • Conclusions: For a detector equipped with 10,000 sensors consisting of 24 3-inch photomultipliers, we find that CCSNe up to a distance of ($341\,\hbox {kpc}\rightarrow {269}\,\hbox {kpc}$) can be identified with $50\%$ probability with 0.01 false SN detection per year.

  • Conclusions: If the arrival time of CCSN neutrinos is known from an independent observation with $\delta t = 1\,\hbox {h}$, a $27\,{\mathrm {M}}_{\odot }$ CCSN at ($[407,341]\,\text {kpc}\rightarrow [322,269]\,\text {kpc}$) can be detected in $50\%$ of cases and with a [3.2, 4.9] $\sigma$ certainty that the signal was not produced by background.

• Change in the $5\sigma$ detection horizons, in case the arrival time of the burst is known exactly:

  • Section 4.2: The $5\sigma$ discovery horizon in this scenario reaches (${400}\,\hbox {kpc}\rightarrow {315}\,\hbox {kpc}$) for a $27\,{\mathrm {M}}_{\odot }$ CCSN using $m\ge 7$, and (${300}\,\hbox {kpc}\rightarrow {234}\,\hbox {kpc}$) for the $9.6\,{\mathrm {M}}_{\odot }$ model.

• To reach a detection of one CCSN about every decade doubling the number of modules, the necessary noise reduction changes from a factor $\sim 70$ to a factor $\sim 140$:

  • Abstract: Increasing the number of sensors to 20,000 and reducing the optical background by a factor of ($\sim 70 \rightarrow \ \sim 140$) expands the range such that a CCSN detection rate of ($0.1\rightarrow 0.08$) per year is achieved, while keeping the false detection rate at $0.01\,\hbox {year}^{-1}$.

  • Section 4.2: In contrast, doubling the number of modules installed would allow the false SN detection rate to be kept below $0.01\,\hbox {year}^{-1}$ while expecting ($1\rightarrow 0.8$) CCSN detection per decade if the radioactive noise within the glass vessel can be reduced by a factor of about ($70\rightarrow 140$).

  • Conclusions: Increasing the number of installed modules to 20,000 and using pressure vessels with significantly reduced optical background could extend the range such that one CCSN (per decade $\rightarrow$ every $\sim 12$ years) can be observed.

The conclusion from this work remains unchanged despite the reduced detection range: exploiting coincidences between detected photons within a segmented photosensor will significantly increase the sensitivity of sparsely instrumented neutrino telescopes to distant CCSNe.

Table 4 False CCSN detection rate and range of supernova detection (50% probability) for different values of m and $N_{\nu }$ (see trigger conditions in Section 4; $\Delta t_{\mathrm {coin}}=20\,\hbox {ns}$, $\Delta T_{{\mathrm {SN}}}={10}\,\hbox {s}$)

Fig. 8

Probability for the detection of a CCSN of $27\,\hbox {M}_\odot$ progenitor mass as a function of distance using the trigger conditions presented in Table 4

Fig. 9

Detection prospects for a CCSN whose time is known to within 1 h ($m\ge 7$, $\Delta t_{\mathrm {coin}}=20\,\hbox {ns}$, $\Delta T_{\mathrm {SN}} = 10\,\hbox {s}$). Left axis: Probability in $\sigma$ that the signal is not produced by background fluctuations. Right axis: Distance at which a $27\,{\mathrm {M}}_{\odot }$ progenitor mass CCSN is detected with $10\%$ (upper boundary), $50\%$ (middle mark) and $90\%$ probability (lower boundary), when at least $N_{\nu }$ detected events are required

Fig. 10

CCSN detection rate for hypothetical detectors with 10,000 (upper) and 20,000 (lower) mDOMs as a function of the false SN detection rate and a reduction in radioactive noise compared to standard mDOMs. The CCSN detection rates have been calculated using the estimated CCSNe population from [16] based on actual observations and scaled to the star formation rate