Estimates of CO<sub>2</sub> fluxes over the city of Cape Town, South Africa, through Bayesian inverse modelling

Nickless, Alecia; Rayner, Peter J.; Engelbrecht, Francois; Brunke, Ernst-Günther; Erni, Birgit; Scholes, Robert J.

doi:https://doi.org/10.5194/acp-18-4765-2018

Articles | Volume 18, issue 7

https://doi.org/10.5194/acp-18-4765-2018

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

https://doi.org/10.5194/acp-18-4765-2018

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

Articles | Volume 18, issue 7

Research article

|

09 Apr 2018

Research article |

| 09 Apr 2018

Estimates of CO₂ fluxes over the city of Cape Town, South Africa, through Bayesian inverse modelling

Alecia Nickless, Peter J. Rayner, Francois Engelbrecht, Ernst-Günther Brunke, Birgit Erni, and Robert J. Scholes

Download

Final revised paper (published on 09 Apr 2018)
Supplement to the final revised paper
Preprint (discussion started on 25 Jul 2017)
Supplement to the preprint

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC1: 'Review of "Estimates of CO2 fluxes over the City of Cape Town, South Africa, through Bayesian inverse modelling"', Anonymous Referee #1, 19 Sep 2017
- AC1: 'Response to interactive comment by Anonymous Referee 1 on “Estimates of CO2 fluxes over the City of Cape Town, South Africa, through Bayesian inverse modelling” by Alecia Nickless et al.', Alecia Nickless, 30 Nov 2017
RC2: 'Review Nickless et al.', Anonymous Referee #2, 20 Oct 2017
- AC2: 'Response to interactive comment by Anonymous Referee 2 on “Estimates of CO2 fluxes over the City of Cape Town, South Africa, through Bayesian inverse modelling” by Alecia Nickless et al.', Alecia Nickless, 30 Nov 2017

Peer-review completion

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

AR by Alecia Nickless on behalf of the Authors (12 Jan 2018) Author's response Manuscript

ED: Referee Nomination & Report Request started (20 Jan 2018) by Martin Heimann

RR by Anonymous Referee #2 (06 Feb 2018)

RR by Anonymous Referee #3 (14 Feb 2018)

Suggestions for revision or reasons for rejection

The responses provided by the authors have addressed the main issues raised by the reviewers. It is important to note here that the authors have produced new figures and results to support their responses. The only answer that lacks some clarity in the response is related to the first point raised by the referee. The discussion related to the large errors in NEE from CABLE seems to explain the problem at the Hangklip site. It is less obvious at the Robben Island site where fossil fuel dominates the observed variability. My main concern, directly related to the reviewer’s comment, is about the improvement after inversion, better illustrated in the latest manuscript by the residuals shown in Figures 4 and 5. The original residuals look as large as the actual variability at the site. However, the inversion seems to correct most of the initial mismatch, including the high-frequency variability. To the extent that the inversion should be able to reduce the mismatches, the actual improvement corresponds to a high degree of confidence in the system. While the mean fluxes should clearly match the observations better, it is quite surprising to note the absence of peaks after inversion. One could worry about the degree of freedom in the system, allowing for these corrections after the optimization of the fluxes. The reviewer was concerned by the prior mismatches, comparable to the signals themselves, being reduced to an excellent match.

Your response here should focus on demonstrating that the observed improvement in the residuals is actually physically realistic, and not a consequence of being over-confident in your inversion system. The reference to a companion paper is welcome but additional details in your prior error structures, or at least the degree to which your system can adjust to observed mismatches, would be a convincing argument to the observed improvement. The reference to Michalak et al. (2005) demonstrates that your errors are reasonable and in agreement with the observed mismatches. But one could argue that such an improvement after inversion may be the consequence of a high DFS, with innovations in space and time allowing to adjust the mixing ratios with too much confidence, as described in the paper: “The inversion was able to reduce uncertainty of the total flux within a pixel by up to 97.7%”. This discussion should be added in the same section, to understand if such an error reduction is realistic.

Hide

ED: Publish subject to minor revisions (review by editor) (15 Feb 2018) by Martin Heimann

AR by Alecia Nickless on behalf of the Authors (25 Feb 2018) Author's response Manuscript

ED: Publish as is (07 Mar 2018) by Martin Heimann

AR by Alecia Nickless on behalf of the Authors (16 Mar 2018) Manuscript

Download

Article (7178 KB)
Full-text XML

Short summary

Carbon dioxide emissions and uptake were estimated for Cape Town, South Africa. We placed two high-precision analysers in lighthouses located on either end of Cape Town (Robben Island and Hangklip). The Cape Point GAW station provided background measurements. We were able to improve the agreement between modelled and observed concentrations, relative to initial estimates provided. This methodology could potentially be scaled up to provide monitoring and verification of city emissions.

Estimates of CO2 fluxes over the city of Cape Town, South Africa, through Bayesian inverse modelling

Download

Interactive discussion

Peer-review completion

Estimates of CO₂ fluxes over the city of Cape Town, South Africa, through Bayesian inverse modelling