FAIR v1.3: a simple emissions-based impulse response and carbon cycle model

Smith, Christopher J.; Forster, Piers M.; Allen, Myles; Leach, Nicholas; Millar, Richard J.; Passerello, Giovanni A.; Regayre, Leighton A.

doi:https://doi.org/10.5194/gmd-11-2273-2018

Articles | Volume 11, issue 6

https://doi.org/10.5194/gmd-11-2273-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/gmd-11-2273-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 11, issue 6

Model description paper

|

18 Jun 2018

Model description paper |

| 18 Jun 2018

FAIR v1.3: a simple emissions-based impulse response and carbon cycle model

Christopher J. Smith, Piers M. Forster, Myles Allen, Nicholas Leach, Richard J. Millar, Giovanni A. Passerello, and Leighton A. Regayre

Download

Final revised paper (published on 18 Jun 2018)
Supplement to the final revised paper
Preprint (discussion started on 07 Dec 2017)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

SC1: 'Code Availability', Lutz Gross, 22 Dec 2017
- SC2: 'Code availability', Christopher Smith, 10 Jan 2018
RC1: 'Review of FAIR v1.1', William Collins, 09 Jan 2018
- AC2: 'Response to reviewer #1', Christopher Smith, 08 Mar 2018
RC2: 'Review of Smith et al., FAIR 1.1', Brian O'Neill, 01 Feb 2018
- AC1: 'Response to reviewer #2', Christopher Smith, 08 Mar 2018

Peer-review completion

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

AR by Anna Mirena Feist-Polner on behalf of the Authors (23 Mar 2018) Author's response

ED: Referee Nomination & Report Request started (23 Mar 2018) by Volker Grewe

RR by William Collins (06 Apr 2018)

Suggestions for revision or reasons for rejection

Second review of Smith et al.

The authors have given detailed and thoughtful responses to my first review. I just have two main outstanding points.

As I stated before it is dangerous to hard-wire conversions from land CO2 emissions to albedo forcing, and from aircraft NOx to contrails. While these (by design) give the correct response to the RCP8.5 scenario, this model is likely to be taken up and (the authors should hope) used by many diverse groups for different purposes and different scenarios. It is very likely that this hard-wiring will persist out in the community even when the authors have updated their own version. As an example the RCP6.0 scenario has strong negative land CO2 emissions which presumably will give incorrect land albedo forcing. The SSP2.6 and 1.9 also have negative land emissions. If the input data is land use area then at least any future user is forced to think about where this data is coming from.

To respond to the author’s responses on methane lifetime:

1. If using the Holmes et al. parameters gives a slightly different result to MAGICC (see also their Fig 7) that doesn’t seem an issue as there is no reason to assume MAGICC is the “truth”. I agree with RCP8.5 there are two opposing effects – methane increases the lifetime (+29%), climate decreases the lifetime (-17%), but that might not be the same in all scenarios e.g. where climate increases but methane doesn’t or vice-versa.

2. The relationship has been shown (in theory, and in models) to hold for methane doubling in many papers, particularly by Michael Prather and Chris Holmes.

3. Applying a simple methane feedback won’t work so it is not surprising it doesn’t reproduce the historical or future without extreme natural emissions. Over the historical the methane self-feedback effect will be offset by increases in NOx emissions and changes in climate, Naik et al. 2013 ACP find these effects more or less cancel. In the future, as above Holmes et al. find climate change offsets more than half of the methane self-feedback in RCP8.5.

Specific comments:

Page 2, line 13: Do you really mean different responses to “forcings” here (i.e. different efficacies) or do you mean different responses to emissions?

Page 2, line 15: Gasser et al. 2017 “Accounting for the climate–carbon feedback in emission metrics” Earth Syst. Dynam. have quantified the effect of non-CO2 agents on the carbon cycle.

Page 2, line 23: I should have been clearer in my comment on this the first time round, the Joos parameter certainly do include feedbacks for both biospheric carbon uptake and temperature (see AR5 section 8.7.1.4 and above Gasser paper).

Page 3, line 16: It is not obvious whether the methane is already accounted for in the carbon budgets (Boucher et al. 2009). See my further comment on page 6.

Page 3, line 28 & page 4, line 23: Either use “molar mixing ratio” instead of “concentration”, or specifically state that the units are mol mol-1. While it is generally fine to use “concentration” colloquially, here specific equations are being presented and the reader needs to know whether the results are in kg m-3 or mol mol-1.

Page 5, lines 19-25. If you are going to mention the perturbation lifetime, I think you should explicitly say why it is not appropriate (changes in emissions of NOx & VOCs, changes in climate). Naik et al. 2013 ACP find the methane lifetime is essentially unchanged over the historical period due to compensating terms (models range from -15% to +12%).

Page 6, lines 1-2. The exponential decay of methane has no affect at all on the double counting as this decay is to CO2. Even if the decay was instantaneous each mole of CH4 would end up as a mole of CO2.

Page 8, line 29-30: The semi-direct effect applies (almost) entirely to BC in IPCC. No significant semi-direct effect has been quantified for the scattering aerosols. Therefore your BC coefficient should be scaled down (and certainly not up by 1.37) until you get a forcing of -0.45, and all other coefficients unchanged.

Section 2.2.9: It is very dangerous to have a conversion from CO2 emissions to land-use hard-wired into FAIR – for instance in RCP6.0 land use CO2 emissions are strongly negative (I don’t know why). It would be much safer to have land use area as an input. RCP data can then be pre-processed to create a land use area input. As the authors note the latitudinal dependence is very large. The new SSPs provide land use area change for different cover types.

Hide

RR by Brian O'Neill (22 Apr 2018)

ED: Publish subject to minor revisions (review by editor) (28 Apr 2018) by Volker Grewe

AR by Christopher Smith on behalf of the Authors (08 May 2018) Author's response Manuscript

ED: Publish as is (11 May 2018) by Volker Grewe

Short summary

FAIR v1.3 is a simple Python-based climate model emulator. It takes emissions of greenhouse gases and aerosol and ozone precursors to calculate radiative forcing and temperature change. It includes a simple representation of carbon cycle feedbacks due to temperature and accumulated carbon uptake. Large ensembles can be run with minimal computational expense for any user-specified emissions pathway. We produce such an ensemble using the RCP emissions datasets.