the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 11 Sep 2024
Supernova effects on middle and upper atmospheric nitric oxide and stratospheric ozone
Abstract. We provide a quantitative test of the recent suggestion (Brunton et al., 2023) that supernovae could significantly disrupt planetary ozone layers through a multi-month flux of soft X-rays that produce ozone-destroying odd nitrogen (e.g. NO and NO2). Since soft X-rays do not directly penetrate down to the ozone layer, this effect would be indirect and require downward transport of NOx from the mesosphere. Mirroring previous studies of the indirect effects of energetic particle precipitation (EPP-IE), we call this the X-ray Indirect Effect (Xray-IE). We use the NCAR Thermosphere-Ionosphere-Mesosphere-Electrodynamics General Circulation Model (TIME-GCM) to simulate the production of NO and its transport into the stratosphere. We model the soft X-ray flux as if it were a multi-month long solar flare and use our previously developed solar flare model to simulate the soft X-ray enhancement. Our results yield significant enhancement in stratospheric odd nitrogen, most dramatically in the Southern Hemisphere. The most global effects are seen in the upper stratosphere at pressure surfaces between 1–3 hPa (about 42–48 km) consistent with previous observations of the EPP-IE. We then use a detailed stratospheric photochemistry model to quantify the effects of this NOx enhancement on ozone. Widespread ozone reductions of 8–15 % are indicated; however, because these are limited to the upper edges of the ozone layer, the effects on the ozone column are limited to 1–2 %. We thus conclude that the effects of a multi-month X-ray event on biologically damaging UV radiation at the surface is also likely to be small.
- Preprint
(1382 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on angeo-2024-15', Anonymous Referee #1, 23 Sep 2024
The article describes possible effects of enhanced X-ray fluxes on the Earth’s stratospheric ozone, and subsequently on biosphere, caused by a hypothetical supernova explosion. The authors extrapolate the effects of X-ray fluxes from documented solar flares and assume that such X-ray fluxes could persist for multiple days. The article is innovative and addresses important issues of the planetary habitability and possible space weather dangers. I have some questions regarding the numerical procedures used (in particular, the choice of nudging for the TIME-GCM simulations), as well as regarding the underlying physical assumptions. Also, I think the referencing of earlier works can be improved (see specific comments below).
Lines 93-94 and subsequently: The assumption that the effects of flares could be extrapolated in time needs extra justification. Is it assumed that the effects of increased X-ray flux would linearly accumulate in time? Should one also expect some qualitative, possibly non-linear, changes in NOx/ozone? One possibility I could see is due to a tidal feedback. Since atmospheric thermal tides are at least partly generated due to stratospheric ozone, the reduction in ozone would presumably change the tides and reduce the tidal mixing and NOx transport from mesosphere to stratosphere. Since tides have large amplitudes in middle atmosphere, their effects on transport/mixing could be substantial. In the current study, diurnal and semidiurnal tides in TIME-GCM are assumed to be given by GSWM (lines 202-203). Thus, the tides are presumably unaffected by ozone changes due to the flare. Does this mean that possible tidal feedback is disabled? Please clarify.
Lines 214-225: The authors chose to constraint the TIME-GCM simulations (winds and temperature fields) by MERRA-2 reanalysis instead of, e.g., NAVGEM-HA fields. Can they comment on the validity of MERRA-2 fields in the region of interest (55 and 75 km altitude)? Also, in their previous works the authors noted a great performance of NAVGEM-HA with respect to radar observations of middle atmosphere dynamics (e.g., McCormack et al., 2017). Thus, I am confused of why they chose to nudge the simulations to MERRA-2 dynamical fields instead of NAVGEM-HA.
Line 384-385: Can the authors clarify the claim about the only qualitative analysis? I thought that few other studies addressed the impact of solar flares on nitric oxide, including studies with ground instruments (e.g., Enell et al., 2008 and further references there).
As a minor note, there are some inconsistencies in spelling. For example, it alternates between “X-ray”, “X ray” and “Xray”.
References
Enell et al., 2008, https://doi.org/10.5194/angeo-26-2311-2008
McCormack et al., 2017, https://doi.org/10.1016/j.jastp.2016.12.007Citation: https://doi.org/10.5194/angeo-2024-15-RC1 -
AC1: 'Reply on RC1', David Siskind, 10 Oct 2024
We thank the reviewer for his/her comments. Below, our responses are in bold-face/italic text.
The article describes possible effects of enhanced X-ray fluxes on the Earth’s stratospheric ozone, and subsequently on biosphere, caused by a hypothetical supernova explosion. The authors extrapolate the effects of X-ray fluxes from documented solar flares and assume that such X-ray fluxes could persist for multiple days. The article is innovative and addresses important issues of the planetary habitability and possible space weather dangers. I have some questions regarding the numerical procedures used (in particular, the choice of nudging for the TIME-GCM simulations), as well as regarding the underlying physical assumptions. Also, I think the referencing of earlier works can be improved (see specific comments below).
Response: We are a bit confused by the comment concerning the referencing. Of the two references given by the reviewer, we already cite one of them (McCormack et al., 2017). The other, as we argue below, is not relevant since it does not deal with nitric oxide- the main subject of this paper.
Lines 93-94 and subsequently: The assumption that the effects of flares could be extrapolated in time needs extra justification. Is it assumed that the effects of increased X-ray flux would linearly accumulate in time? Should one also expect some qualitative, possibly non-linear, changes in NOx/ozone? One possibility I could see is due to a tidal feedback. Since atmospheric thermal tides are at least partly generated due to stratospheric ozone, the reduction in ozone would presumably change the tides and reduce the tidal mixing and NOx transport from mesosphere to stratosphere. Since tides have large amplitudes in middle atmosphere, their effects on transport/mixing could be substantial. In the current study, diurnal and semidiurnal tides in TIME-GCM are assumed to be given by GSWM (lines 202-203). Thus, the tides are presumably unaffected by ozone changes due to the flare. Does this mean that possible tidal feedback is disabled? Please clarify.
Response: There are a couple of specific points here that should be corrected. First, the assumption is *not* that the effects can be extrapolated in time. Rather, we show that the buildup saturates after 10 days. After that, we still calculate the NO self-consistently. There is no extrapolation. And second, the tides are *not* given by GSWM- the text states that is the “standard” model of TIMEGCM and rather, here, we use MERRA-2 nudging. But rather, perhaps the larger concern that the reviewer is getting at, and a legitimate one, is that we’re assuming that the flood of descending NOx does not change the dynamics at all. That is clearly an assumption on our part and could be more clearly stated. We will add a couple of sentences at the end of section 2.1 to be clearer on this. We also note that we already commented on this assumption in the Discussion section with reference to Seppala et al.
Lines 214-225: The authors chose to constraint the TIME-GCM simulations (winds and temperature fields) by MERRA-2 reanalysis instead of, e.g., NAVGEM-HA fields. Can they comment on the validity of MERRA-2 fields in the region of interest (55 and 75 km altitude)? Also, in their previous works the authors noted a great performance of NAVGEM-HA with respect to radar observations of middle atmosphere dynamics (e.g., McCormack et al., 2017). Thus, I am confused of why they chose to nudge the simulations to MERRA-2 dynamical fields instead of NAVGEM-HA.
Response: It would’ve been great had NAVGEM-HA been available for the period in question. It was not. Unlike MERRA-2, NAVGEM-HA is not an operational product, and only exists for select periods generally only covering several months with a time cadence needed to resolve higher frequency tides (i.e., semidiurnal tide) important in the mesosphere and thermosphere. But MERRA-2 is pretty good. We encourage the reviewer (and other readers) to compare Figure 21 of Gelaro et al (J. Clim, 2017) (which we already cite) with Figure 1 of Siskind et al. (JGR, 2010, http://dx.doi.org/10.1029/2010JD014114) which shows a NOGAPS-ALPHA (predecessor to NAVGEM-HA) simulation of the same period (the elevated stratopause of February 2006). It's clear that MERRA-2 does much better than MERRA in capturing this difficult-to-capture event. But ultimately, the fact that our baseline NOx descent and dispersion into the middle latitudes is so nicely simulated as compared with the MIPAS data presented by the Funke and Pettit papers that we cite is what demonstrates to us that our dynamics are adequately captured.
Line 384-385: Can the authors clarify the claim about the only qualitative analysis? I thought that few other studies addressed the impact of solar flares on nitric oxide, including studies with ground instruments (e.g., Enell et al., 2008 and further references there).
Enell et al. do not analyze nitric oxide data which is what we really meant. We will add the adjective “data” to line 384 so it reads “quantitative data analysis”. (Note, the reviewer said “qualitative”. That’s not what our text says). And actually, Enell don’t even show any modelled nitric oxide- just arbitrary scalings. And regarding "ground instruments"- (quoting the reviewer), MLT nitric oxide is not measurable via ground instruments.
As a minor note, there are some inconsistencies in spelling. For example, it alternates between “X-ray”, “X ray” and “Xray”.
We apologize for these inconsistencies; these will be completely corrected in final copy editing. But we note that Xray-IE is an acronym that we coined so we are free to spell it the way it is in the text.
References
As noted above, we already cite McCormack et al and do not feel Enell et al is relevant to nitric oxide analysis since they show no nitric oxide data nor do they compare any calculated nitric oxide with published results (their NO profiles are not calculations- they’re arbitrary scalings)
Enell et al., 2008, https://doi.org/10.5194/angeo-26-2311-2008
McCormack et al., 2017, https://doi.org/10.1016/j.jastp.2016.12.007Citation: https://doi.org/10.5194/angeo-2024-15-AC1 -
RC2: 'Reply on AC1', Anonymous Referee #1, 16 Oct 2024
Thank you for the clarifications. I think the paper has been improved and is generally suitable for publication, except that the referencing could be improved. My comment about Enell et al. study has been probably misunderstood. There was no claim that nitric oxide itself is measured by ground instruments. It was meant that Enell et al. addressed the same topic of flare impact and this should be acknowledged.
Citation: https://doi.org/10.5194/angeo-2024-15-RC2 -
AC3: 'Reply on RC2', David Siskind, 06 Dec 2024
As can be seen in our response to the other reviewer, we did add two new references that are more directly relevant to the work here. There are an enormous number of flare studies; we are trying to keep focused on middle atmosphere nitric oxide production
Citation: https://doi.org/10.5194/angeo-2024-15-AC3
-
AC3: 'Reply on RC2', David Siskind, 06 Dec 2024
-
RC2: 'Reply on AC1', Anonymous Referee #1, 16 Oct 2024
-
AC1: 'Reply on RC1', David Siskind, 10 Oct 2024
-
RC3: 'Comment on angeo-2024-15', Anonymous Referee #2, 02 Nov 2024
Review of the paper by Siskind et al.
The paper examines the impact of X-ray emission of nearby young supernova remnants (SNR). Using a realistic model of the middle atmosphere and lower thermosphere including photochemistry, the authors test for planet Earth the hypothesis brought up by Brunton et al. who found a possible threat to biospheres in the wider cosmic neighbourhood of such SNRs by the prolonged strong X-ray emission phase. The paper shows that the rough estimation of Brunton et al. for the lethal distance between a SNR and the impacted biosphere does not hold and that in the case of Earth its atmosphere effectively shields also against the threat by extended X-ray emission from young SNRs.
The paper is absolutely in time and is a very valuable contribution to the field of harsh environments for early life and existence of biospheres in stellar system with habitable planets. The methods applied are essentially sound, even when the authors deduce their esimation of the NOx input and related ozone loss from some kind of extrapolation.
I only have a few general comments:
1. The authors use the TIME-GCM for their study. The authors state themselves that the model in this configuration is not able to simulate elevated stratosphere events, which are key for strong NOx intrusions in the NH, and probably underestimates downward transport inside the polar winter vortex in the lower mesophere in general. The authors comment that comparisons with MIPAS data show good agreement for midlatitudes instead, but these airmasses see the Sun also during winter and NOx here has a limited lifetime. Somehow related, NOx transported below the lower boundary is lost in their model. But during summer with the change of the circulation this NOx is brought to the middle stratosphere again where it could contribute to ozone loss. Can the authors give an estimation of this contribution which is lost in the model?
In addition, the second, strong case does not run through and the authors must rely on some reasonable extrapolation. Despite the authors state this clearly, perhaps other models like WACCM are obviously better suited to study such events if the shortwave photolysis part developed by Siskind would have been implemented there.
2. Despite the rather large energy input, the modelled impact on the ozone layer is small. Typical particle events connected to magnetic storms show hemispheric power values of a few hundred GigaWatts for several hours which corresponds to energy input around 50 J/m2. Obviously, the energy spectrum of the ionizing radiation or particles is most important for the impact in the stratosphere. Brunton et al. speculate (section 3.1) that the spectrum of young SNRs may be significantly harder than assumed in this paper. This would result in deeper penetration into the middle atmosphere and possibly could strongly enhance longlived NOx in the stratosphere as then NOx could bridge the gap of photolysis loss descending NOx faces in spring. The authors should point to that critical uncertainty. Of course, observations in the extended spectral range are needed to further reduce this uncertainty, but also some sensitivity model study would be helpful. In this context, a
3. The authors should estimate the total amount of NOx brought into the atmosphere, not just for the MLT. For comparison, particle events like the Halloween storm show inputs of about 2 Gmol of NOx, and a recent study by Reddmann et al. https://doi.org/10.35097/1104, modelling an extreme SPE/storm event with an input of 30 Gmol also show limited impact on the ozone layer, in line with this study.
4. The position of the source in ecliptic plane may be not the position with maximum impact. Could the authors try to have a source at the celestial poles? This would put the NOx enhancement deeper into the atmosphere due to the vertical infall and could bring NOx to also the stratosphere.
5. The results of this paper is at strong odds with the result of Brunton et al.. For further studies it would be helpful, if the authors could explain why the results of their study and that of Brunton et al. differ so much. A possible reason for the differences could be the hard X-ray spectrum part. Can you compare your spectrum with Fig. 1 of Ejzak et al?
6. The labeling of most figures does not confirm with standard. Often the y-axis shows the unit but not the quantitity or vs versa; how to show units (with brackets or without) is not consistent.
Minor comments and typos:Title: "Supernova effects" is rather unspecific. Perhaps "No threat to the ozone layer by X-ray luminous SNs" ?
L14 "planetary": The paper only deals with the Earth, so perhaps "ozone layers of Earth-like planets"
L24 "most global": "strongest global"?
L35 "these": delete
L88 "Thus our": delete "our"
L117 "For our purposes ...": This is too general
L156: "our spectrum" replace "our"
L158ff: "well covered with modern spectra" "suggests that .... agrees" : please reformulate to be more concise
L232: Kp = 3 (L235) still shows some particle ionization.
L236: 30°
L276: Michelson
L289: "descent" is purely dynamic. You mean the amount which descends.
L348, Fig. 5: Why not showing diffs?
L361, "1.0": unit?
Citation: https://doi.org/10.5194/angeo-2024-15-RC3 - AC2: 'Reply on RC3', David Siskind, 06 Dec 2024
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 252 | 91 | 43 | 386 | 7 | 7 |
- HTML: 252
- PDF: 91
- XML: 43
- Total: 386
- BibTeX: 7
- EndNote: 7
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1