Tropical cyclone genesis potential across palaeoclimates

Koh, J. H.; Brierley, C. M.

doi:https://doi.org/10.5194/cp-11-1433-2015

Articles | Volume 11, issue 10

https://doi.org/10.5194/cp-11-1433-2015

© Author(s) 2015. This work is distributed under
the Creative Commons Attribution 3.0 License.

https://doi.org/10.5194/cp-11-1433-2015

© Author(s) 2015. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 11, issue 10

Research article

|

21 Oct 2015

Research article |

| 21 Oct 2015

Tropical cyclone genesis potential across palaeoclimates

J. H. Koh and C. M. Brierley

Download

Final revised paper (published on 21 Oct 2015)
Preprint (discussion started on 06 Feb 2015)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

SC C71: 'Question 1: Method clarification', Matthew Huber, 08 Mar 2015
- SC C87: 'Method clarification', Christopher Brierley, 13 Mar 2015
SC C72: 'Question 2: Choice of SSTs', Matthew Huber, 08 Mar 2015
- SC C88: 'Tropical Cyclone Feedbacks', Christopher Brierley, 13 Mar 2015
SC C74: 'Question 3: Frequency+intensity versus work done', Matthew Huber, 08 Mar 2015
- SC C90: 'Power Dissipation and/or Accumulated Cyclone Energy', Christopher Brierley, 13 Mar 2015
RC C75: 'Review of Koh and Brierley's paper on tropical cyclone genesis factors in paleoclimate simulations (reviewed by Robert Korty)', Robert Korty, 09 Mar 2015
- AC C91: 'Response to review', Christopher Brierley, 13 Mar 2015
  - SC C93: 'Normalizing the GPI', Robert Korty, 13 Mar 2015
    - SC C95: 'Further thought on Normalizing GPI', Robert Korty, 13 Mar 2015
RC C98: 'Review of Koh and Brierley', Timothy Merlis, 13 Mar 2015
RC C102: 'review', Matthew Huber, 13 Mar 2015

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Chris Brierley on behalf of the Authors (30 Jun 2015) Author's response Manuscript

ED: Referee Nomination & Report Request started (13 Jul 2015) by Gerrit Lohmann

RR by Matthew Huber (20 Jul 2015)

RR by Robert Korty (07 Aug 2015)

Suggestions for revision or reasons for rejection

This paper presents an analysis of the environmental conditions that support and sustain tropical cyclones in the most recent set of climate simulations of the Last Glacial Maximum and Middle Holocene, as well as the first analysis of such conditions in simulations of the Pliocene. The revised manuscript greatly clarifies the scope, techniques, and findings, and I enthusiastically recommend its publication. I am sorry if my review of the first draft appeared to harp incessantly on the distinction between cyclogenesis and genesis potential, but I was confused by what types of analysis had been conducted in that draft. I am pleased to find the revised presentation very clear and very readable, and I suspect it will be very interesting to a wide audience.

The broad stability of many of these environmental parameters across wildly differing climate states is perhaps the most striking result, though the variations between might still be important. That climate simulations of the Last Glacial Maximum yield tropical environments similarly favorable for genesis as do ones in the modern world---and that the much warmer Pliocene simulations do not yield massive poleward expansions of favorable regions---are fascinating results.

Below are a few areas in need of corrections as well as suggestions for further clarification.

Page 1, lines 8-11: I think two sentences were blended incompletely here. Remove periods surrounding “through analysis of” and replace “is calculated for” with “in”.

Page 1, line 20: Calculated would be better than simulated here.

Page 3, line 21: would add “deep-time” in front of palaeoclimates; there is some geologic data from the most recent millennia, and there are techniques in development that may uncover other sources of Holocene variability.

Page 3, line 30: “Korty et al. (2012a) observed higher genesis potential relative to the preindustrial era.” This is true in a globally integrated sense (though perhaps not robust, as the PMIP3 findings here show), but it would be helpful to point out here that results varied by basin (and model). For example, the ensemble mean in PMIP2 produced less favorable conditions in the Atlantic at LGM, while similar or better conditions in the western North Pacific.

Page 5, line 22: “as the models have a hybrid vertical coordinate, the actual number of pressure levels used for the PI computation often differs” Were the data interpolated to standard pressure levels first, or were they entered into the algorithm with the precise pressures from the file? Doubt it matters for the result, but was confused by this sentence.

Page 5, line 24: Were data to at least 70 hPa available in all models? Resolution into the lower stratosphere can be important when calculating potential intensity. And a priori there is no guarantee that the tropopause will be at the same pressure level in a radically different climate state.

Page 5, line 24: “GPI is only calculated between 30 S and 30 N”. It’s fine to confine this to the tropics, but of course implicit in this is that high GPI does not expand poleward beyond this range. You could mention some evidence from your results that that is not the case here (e.g., in Figure 1e, potential intensity drops precipitously equatorward of 30 N in the Pliocene as it does in all other climates).

Page 6, line 13: insert “are” between “conditions” and “for”.

Page 6, line 20: Importantly Gray also noted the importance of atmospheric stability with respect to convection (buoyancy) in addition to SST in tropical cyclogenesis. Potential intensity is, in a sense, a marriage of these two thermodynamic parameters.

Page 6, line 29: “represents changes in cyclogenesis”…well, it’s really changes in favorability, which is assumed to be related to cyclogenesis.

Page 7, line 1: Emanuel and Nolan used relative humidity at a midtropospheric level rather than the moist entropy parameter chi. The latter replaced RH in later papers as it became clear that the 2004 formulation didn’t capture what was emerging from analysis of climate simulations of the 21st century. (Emanuel, K., R. Sundararajan, and J. Williams, 2008: Hurricanes and global warming: Results from downscaling IPCC AR4 simulations. Bull. Amer. Meteor. Soc., 89, 347-367.)

Page 8, line 23 (equation 2): Chi is defined as the first ratio, and it is approximately equal to the second because sb is approximately equal to s* wherever soundings are neutral with respect to moist convection. So the second equivalence sign should be replaced with a symbol indicating approximately equal to.

Page 10, lines 1-8: This is a great addition to the paper, and really helps clarify the scope of the findings. You could add reference to Camargo et al. (2014), who identified that the same four parameters in this formulation of the GPI provided the best fit between HiRAM’s environmental fields and the storms it simulated for future climate change. This provides some further justification for their use in climates other than the present. (S.J. Camargo, M.K. Tippett, A.H. Sobel, G.A. Vecchi, and M. Zhao, 2014. Testing the performance of tropical cyclone genesis indices in future climates using the HIRAM model. Journal of Climate, 27, 9171-9196, doi: 10.1175/JCLI-D-13-00505.1.)

Page 10, lines 13-14: “high values of potential intensity…are needed to induce deep tropospheric convection”. The cause and effect are backwards here. It isn’t that high values of PI are needed, but rather that they’re correlated with regions where deep convection is possible. There’s nothing magic about the number 55 m/s. (If you slightly modify Kerry Emanuel’s potential intensity algorithm so that it outputs the temperature and pressure at the level of convective outflow, you’ll see this divide explicitly.)

Page 11, line 5: year for Korty papers is 2012, not 2015.

Page 12, line 11: “calculated” rather than “simulated”

Page 12, lines 16-17: I don’t understand what the parenthetical information means here. Earlier versions of all 5 of these models were included in PMIP2 as well as PMIP3.

Page 14, line 3: insert “potential” after “generally stronger genesis”

Page 15, line 28: would rewrite this to say “…observed number of storms formed globally each year in the modern climate”

Page 16, lines 31-32: It is true that the October peak found in the PMIP2 ensemble has vanished here, but are July and August still lower at the Middle Holocene? And January appears higher (as it was PMIP2). Figure 10 is set up in a way that makes it hard to compare different climates with the presindustrial era control. One alternative to consider might be to revise this to feature just two panels: one showing the Northern Hemisphere cycle, the other the Southern Hemisphere cycle. Then all 3 paleoclimates could be featured as different colored bars side by side with the preindustrial era control’s annual cycle.

Page 18, lines 1-10: I think this discussion of the differences between the two series of LGM simulations is very helpful.

Page 31, line 13: Figure 3 caption still calls this “cyclone genesis” and needs to be updated to say “genesis potential”

Page 31, line 25: Figure 5 caption likewise still calls this “cyclone genesis difference” when it is “genesis potential difference”

Hide

RR by Timothy Merlis (18 Aug 2015)

Suggestions for revision or reasons for rejection

Review of revised submission of Koh & Brierley

The authors have addressed my major concerns from the initial submission: primarily the language of writing about genesis indices as though it was actual genesis. The authors have made a nice contribution to the literature concerning tropical cyclone genesis indices in changed climates. I recommend publication following the correction of some mistakes I found when reading the revised manuscript.

p.1 first sentence of abstract
p. 3 ambiguous dichotomy of LGM being an icehouse climate vs. "current greenhouse climate". I agree the strength of the greenhouse effect was different, but the two climate states fall on a continuum; this continuum is central to the analysis.
p.3 "unavailability of 6hr data". Even if you had high frequency data, the explicitly simulated TCs at the resolution of the coupled models are probably non-existent or not worth looking at carefully because they are so far from Earth's TCs.
p.5 Line 8: Seems like you should directly reach out to the people who ran the HadCM3 simulations to check if there are undocumented differences.
p.5 L25: for latitudinal averages over the tropics, it would be good to specify if land regions are excluded from these
p.6 L9 missing word in first sentence of section 3
p. 8 L7 "pouch" I think this idea goes back at least to Dunkerton et al. 2009, rather than Tory et al. 2012
Eqn. 2: the second equality uses the "defines" triple equal sign symbol, but the text that follows is clear that this is an approximation/assumption for an quasi-equilibrium stratification.
p.9 L14 "naturally occurring cycles" is "naturally" used for "observed". I'm confused by the word choice would "unnatural" be a numerical simulation of a TC?
p. 9 relationship between CAPE and PI has been carefully discussed in a recent paper by S. Garner in JAS. I recommend citing it because it is really the most careful presentation of what the variables are that go into the formula and their interpretation.
p. 10 L21 "can" -> "cannot" (Vecchi et al. 2008)?
p. 11 I think it's important to note in section 4.2 that GPIs are optimized to capture the spatial variation and seasonality of TC genesis in observations.
p. 13 L28 redundant to have the minus sign following the word "cooling"
p. 16 is Pliocene change "significant"? What do authors, e.g., Camargo, think of 10% changes in GPI the context of future climate?
p. 17 L21 "Increased TC genesis is observed during the mid-Holocene..." This is the type of language that was problematic in the first submission. There is no TC genesis observed or simulated here. Please search for other remaining instances of this.
p. 19 climate misspelled L22
p. 23 I think it's "Matt Huber" with two 't's, but I could be wrong.

Hide

ED: Publish subject to minor revisions (review by Editor) (03 Sep 2015) by Gerrit Lohmann

AR by Chris Brierley on behalf of the Authors (23 Sep 2015) Author's response Manuscript

ED: Publish as is (27 Sep 2015) by Gerrit Lohmann

AR by Chris Brierley on behalf of the Authors (05 Oct 2015)

Short summary

Here we diagnose simulated changes in large-scale climate variables associated with the formation of tropical cyclones (i.e. hurricanes and typhoons). The cumulative potential for storm formation is pretty constant, despite the climate changes between the Last Glacial Maximum and the warm Pliocene. There are, however, coherent shifts in the relative strength of the storm regions. Little connection appears between the past behaviour in the five models studied and their future projections.