Correction to “Impacts of climate change, ozone recovery, and increasing methane on surface ozone and the tropospheric oxidizing capacity”


  • Olaf Morgenstern,

  • Guang Zeng,

  • N. Luke Abraham,

  • Paul J. Telford,

  • Peter Braesicke,

  • John A. Pyle,

  • Steven C. Hardiman,

  • Fiona M. O'Connor,

  • Colin E. Johnson


This article corrects:

  1. Impacts of climate change, ozone recovery, and increasing methane on surface ozone and the tropospheric oxidizing capacity Volume 118, Issue 2, 1028–1041, Article first published online: 16 January 2013

1 Introduction

In the paper “Impacts of climate change, ozone recovery, and increasing methane on surface ozone and the tropospheric oxidizing capacity,” Olaf Morgenstern et al. (Journal of Geophysical Research – Atmospheres, 118, 1028–1041, doi:10.1029/2012JD018382, 2013) found that tropospheric photolysis rates were largely unaffected by stratospheric ozone changes and that other effects, e.g., due to differences in cloud cover or the sea ice albedo effect, were more important. Recently, an error has been discovered in the model setup used by Morgenstern et al. [2013]: Actually, photolysis rates in their model had not been based on the correct, interactive ozone field; instead, an ozone climatology had been used which did not respond to any differences in forcing between the simulations. Despite this error, the present-day reference presented by Morgenstern et al. [2013] compared well to observations; e.g., the background climatology of total column ozone in their Unified Model - UK Chemistry and Aerosols Module (UMUKCA) model was quite realistic, surface ozone for present-day conditions compared reasonably to observations, and the methane lifetime in UMUKCA was close a recent multimodel result [Voulgarakis et al., 2013]. Nonetheless, the model's tropospheric chemistry did not quite realistically respond to ozone recovery. Other feedbacks, particularly stratosphere-troposphere exchange and the warming and moistening of the troposphere under climate change, were adequately captured. Telford et al. [2013] and studies derived from that paper are not affected by this issue.

Here we revisit the key messages from the previous study, now correctly accounting for the impact of ozone changes on photolysis. To varying degrees, all figures in Morgenstern et al. [2013] are affected by the programming error; revised versions of all 10 figures are therefore included here. In the interest of brevity, we will only discuss here three of them which show substantial differences. We also revise the tropospheric ozone budget terms.

2 The Near-Surface Ozone Photolysis Rate

Figure 6 shows differences in the near-surface photolysis rate of O3+hν→O(1D)+O2 between pairs of new simulations, now correctly accounting for ozone column changes (Figure 4). For this diagnostic, Morgenstern et al. [2013] found hardly any significant differences apart from those caused by receding sea ice and changes in cloud cover (cf. their Figure 6). Using the revised model, Figures 6c, 6d, 6f, and 6h—all contrasting ozone recovery (OR) with non-ozone recovery simulations—now indicate significant reductions of the O3→ O(1D) photolysis rate due to ozone recovery [Voulgarakis et al., 2013]. In the tropics, increasing methane leads to a further decrease of this photolysis rate in a stratospheric feedback (Figures 6c and 6d, contrast high-methane with low-methane ozone recovery simulations). Global warming, with unchanged chemical forcing (simulation climate change (CC)), causes a decrease in total column ozone and an increase in the photolysis rate in the tropics (Figure 6b); in the high Arctic, substantial summertime decreases in the photolysis rate are mainly due to sea ice loss [Morgenstern et al., 2013]. In the combined simulation using all forcings (ozone recovery and climate change (ORCC), Figure 6f), in the extratropics photolysis rates are set to decrease due to increased ozone columns under ozone recovery and sea ice loss, whereas in the tropics there is a cancellation of decreasing photolysis rates due to methane increases (which drive stratospheric and tropospheric ozone increases; Figures 6c and 6d), and photolysis rate increases/ozone decreases associated with “global warming” (Figure 6b).

Figure 1.

Colors: 10-year, zonal, and monthly-mean total column ozone (in Dobson Units, DU) in REF. Contours: TOMS/SBUV climatology.

Figure 2.

(top) Colors: 10-year, zonal-, and monthly-mean ozone (ppmv) in REF, in (left) February and (right) September. Black contours: NIWA vertically resolved ozone database, averaged over 1995–2004. White contours: Ozone climatology after Fortuin and Langematz [1995]. (bottom) Percentage bias of modelled zonal- and monthly-mean ozone, relative to the climatologies, in February and September.

Figure 3.

(symbols) Observed monthly- and multiannual-mean surface ozone at selected stations. The data are from the World Data Center for Greenhouse Gases (WDCGG) and cover the years 1995–2004 unless indicated otherwise. (lines) Monthly- and multiannual-mean ozone in REF at the same stations. The model has been sampled at the elevation of the station or the lowest model level, whichever is higher. At sea level the model surface level has a thickness of 20 m.

Figure 4.

Difference in zonal-, monthly-, and multiannual-mean total ozone column between the pairs of simulations as indicated in the titles, in DU. ‘+’ symbols indicate that the two time series do not have a significantly different mean.

Figure 5.

Difference in zonal- and annual-mean ozone between the pairs of simulations as indicated in the titles, in ppmv.

Figure 6.

Differences in zonal-, monthly-, and multiannual mean rates of photolysis of ozone to form O(1D), in 10−3 s−1 at 20 m, for the pairs of simulations as indicated in the titles. Stippling indicates that differences are insignificant at the 95% confidence level, using Student's t-test. HM = high methane. CC = climate change (i.e., increased radiative forcing due to long-lived greenhouse gases). OR = ozone recovery (i.e., reduced halogen loading for chemistry). REF = year-2000 reference [see Table 1 of Morgenstern et al., 2013].

3 The Hydroxyl Radical

A comparison of Figure 7 with Figure 7 of Morgenstern et al. [2013] suggests that allowing for the stratospheric photolysis feedback has mostly increased the sensitivity of tropospheric OH to the applied forcings. For example, there is a larger decrease in OH due to increased methane if the impact of methane on photolysis rates is taken into account (Figures 7a and 7c). Increasing methane increases total column ozone (Figure 4), which causes decreases in the O3 → O(1D) photolysis rate (section 2), which in turn reduces production of OH via O(1D) + H2O → 2 OH. Figure 7d of Morgenstern et al. [2013] showed seasonal increases in OH in the extratropics due to ozone recovery; these were due to increases in stratosphere-troposphere exchange (STE). With interactive photolysis rates, these increases largely disappear (Figures 7d and 7h) due to a cancellation of the STE effect with reduced photochemical production of OH.

Figure 7.

Same as Figure 6, but for the surface to 10 km OH column, in 1012 molecules cm−2.

Figure 8.

Same as Figure 1, but for the surface to 10 km OH column, in 1012 molecules cm−2, in REF.

Figure 9.

Differences in monthly- zonal, and multiannual-mean surface ozone (at 10 m) between pairs of simulations as indicated, in ppbv. Stippling indicates insignificance.

4 Surface Ozone

Figure 10 shows the surface ozone differences for June-July-August. A comparison with Figure 10 of Morgenstern et al. [2013] indicates that the photolysis feedback increases surface ozone relative to the present-day reference for the ozone-recovery simulations (Figures 10c, 10d, 10f, and 10h). In particular, the surface ozone increase due to stratospheric ozone recovery alone (Figure 10d) now is significant throughout most of the northern midlatitudes, whereas Morgenstern et al. [2013] found it was not. Consequently, the all forcings simulation (ORCC, Figure 10f) indicates larger increases in northern midlatitude ozone than previously reported. The increased sensitivity to methane changes relates to the larger decrease in OH production under high-methane conditions found above, which is associated with increases in surface ozone in most areas of the globe (Figures 10a and 10c).

Figure 10.

Differences in seasonal- and multiannual-mean surface ozone (at 10m) in northern summer between pairs of simulations as indicated, in ppbv. Dots (stippling) indicate insignificant differences.

5 The Tropospheric Ozone Budget

Table 2 shows the revised ozone budget terms and the methane lifetime. A comparison with Table 2 of Morgenstern et al. [2013] indicates that changes in modeled photolysis rates increase the methane lifetime. This is likely because the bias in the tropical ozone column of around 20 Dobson units (cf. Figure 1) now reduces OH production and consequently increases the methane lifetime. Morgenstern et al. [2013] also had a bias in tropical column ozone, but this erroneously did not affect the production of tropospheric OH. The sensitivity of the methane lifetime to the methane abundance is substantially increased if the photolysis feedback is accounted for. Morgenstern et al. [2013] found an increase of the methane lifetime of 1.2 years for increasing methane from 1.76 to 2.4 ppmv at the surface, whereas here we find an increase of 2.2 years (for the high-methane (HM) simulation, relative to reference (REF)). Ozone recovery alone (simulation ozone-recovery, low-methane (ORLM)) increases the methane lifetime by 0.7 years relative to REF, so is a lesser effect, probably because ozone recovery affects the tropics less than the extratropics but most methane is lost in the tropics. Climate change (CC) alone decreases the lifetime (as found before). For the ozone lifetime we also find a much larger sensitivity to both methane increases and ozone recovery, indicating that coupling of tropospheric and stratospheric ozone chemistry is much more important than suggested by Morgenstern et al. [2013]. We calculate a methane lifetime enhancement factor inline image, which is considerably larger than the 0.32 previously reported by Morgenstern et al. [2013] and also considerably larger than reported by Prather et al. [2001] and Fiore et al. [2009]. We plan to investigate this further in a separate paper.

Table 2. Tropospheric Ozone Budget Terms, in Tg(O3)/yra
  1. a

    aP = chemical production by RO2 + NO, with R = H, CH3, C2H5, etc.). L = chemical loss via H2O + O(1D) → 2 OH, HO2 + O3 → OH + 2 O2, and OH + O3 → HO2 + O2. DD = dry deposition. STE = inferred stratospheric flux (L + DD − P). B = annual-mean burden (Tg(O3)). inline image (days). The global methane lifetime (years) is denoted with inline image. The methane lifetime is calculated taking into account all chemical (including stratospheric) sinks and assuming a soil sink of 30 Tg/yr (as assumed by Prather et al. [2001] and Stevenson et al. [2006]). Percentage differences (given in brackets) are relative to REF. S06 = multimodel mean and standard deviation in Table 5 [Stevenson et al., 2006]. TAR = Prather et al. [2001].

P (Tg(O3)/a)5110 ± 606342037224025 (8%)3998 (7%)3759 (1%)3927 (6%)4067 (9%)
L4668 ± 727347032813517 (7%)3525 (7%)3333 (2%)3493 (6%)3657 (11%)
DD1003 ± 200770734794 (8%)814 (11%)762 (4%)719 (−2%)801 (9%)
STE552 ± 168770294284 (−3%)341 (16%)337 (14%)288 (−2%)390 (33%)
B (Tg(O3))344 ± 39300256324 (27%)334 (30%)315 (23%)287 (12%)277 (8%)
inline image (days)22.3 ± 22423.027.1 (18%)27.7 (20%)27.7 (20%)24.5 (7%)22.4 (−3%)
inline image (years)8.67 ± 1.328.411.313.5 (19%)13.7 (21%)12.0 (6%)11.1 (−2%)12.9 (14%)

Our tropospheric ozone burden, in REF, is low compared to literature estimates [e.g., Stevenson et al., 2006]. This is consistent with relatively low chemical production and loss, and particularly with low STE [Morgenstern et al., 2013]. Chemical turnaround would increase if isoprene chemistry was included [Telford et al., 2013]; this would also reduce the methane lifetime. The low STE appears to be a peculiarity of this version of the model [Morgenstern et al., 2013].

6 Discussion and Conclusions

Due to an error in model setup, Morgenstern et al. [2013] did not interactively couple ozone to photolysis. Correcting this issue modifies some of the previous results. The main changes are:

  1. Photolysis rates are substantially affected by the impact that the three anthropogenic developments studied here (stratospheric ozone recovery, the chemical impact of methane increases, and climate change) have via associated changes in tropospheric UV.

  2. Both ozone recovery and methane increases reduce OH production via their impact on tropospheric UV.

  3. Our previously reported increases in extratropical OH caused by increased stratosphere-troposphere exchange under ozone recovery are now largely canceled by decreased photochemical production of OH.

  4. Surface ozone increases, due to both stratospheric ozone recovery and methane increases, are now larger than previously reported and significant throughout much of the Northern Hemisphere during summer.

We find a considerably stronger influence of stratospheric ozone changes onto the tropospheric ozone budget and particularly onto the methane lifetime than reported previously.

Table 2 of Morgenstern et al. [2013] suggested that many effects added linearly. Here we find some considerable nonlinearities introduced by the impact on photolysis rates. For example, the tropospheric ozone burden B is substantially larger under all the perturbation experiments than in the reference (REF), but the increase is smallest in the all forcings simulations (ORCC).

Overall, the impact of stratospheric ozone changes on UV penetration, and hence, tropospheric photochemistry needs to be included in models for an accurate prediction of changes to the tropospheric oxidizing capacity.


The data used in this paper can be obtained by request from the lead author.