Estimating Marine Carbon Uptake in the Northeast Pacific Using a Neural Network Approach

Duke, Patrick J.; Hamme, Roberta C.; Ianson, Debby; Landschützer, Peter; Ahmed, Mohamed M. M.; Swart, Neil C.; Covert, Paul A.

doi:https://doi.org/10.5194/egusphere-2023-870

Preprints

https://doi.org/10.5194/egusphere-2023-870

Preprints

15 May 2023

| 15 May 2023

Estimating Marine Carbon Uptake in the Northeast Pacific Using a Neural Network Approach

Patrick J. Duke, Roberta C. Hamme, Debby Ianson, Peter Landschützer, Mohamed M. M. Ahmed, Neil C. Swart, and Paul A. Covert

Abstract. The global ocean takes up nearly a quarter of anthropogenic CO₂ emissions annually, but the variability of this uptake at regional scales remains poorly understood. Here we use a neural network approach to interpolate sparse observations, creating a monthly gridded seawater partial pressure of CO₂ (pCO₂) data product from January 1998 to December 2019, at 1/12° × 1/12° spatial resolution, in the Northeast Pacific open ocean. The data product (ANN-NEP; NCEI Record ID: BGSH2HNRP) was created from pCO₂ observations within the 2021 version of the Surface Ocean CO₂ Atlas (SOCAT), and a range of predictor variables acting as proxies for processes affecting pCO₂ to create non-linear relationships to interpolate observations at a spatial resolution four times greater than leading global products and with better overall performance. In moving to a higher resolution, we show that the internal division of training data is the most important parameter for reducing overfitting. Using our pCO₂ product, wind speed, and atmospheric CO₂, we evaluate air-sea CO₂ flux variability. On sub-decadal to decadal timescales, we find that the upwelling strength of the subpolar Alaskan Gyre, driven by large-scale atmospheric forcing, acts as the primary control on air-sea CO₂ flux variability (r² = 0.93, p < 0.01). In the northern part of our study region, divergence with atmospheric CO₂ is enhanced by increased local wind stress curl, enhancing upwelling and entrainment of naturally CO₂-rich subsurface waters, leading to decade-long intervals of strong winter outgassing. During recent Pacific marine heatwaves from 2013 on, we find enhanced atmospheric CO₂ uptake (by as much as 45 %) due to limited wintertime entrainment. Our product estimates long-term surface ocean pCO₂ increase at a rate below the atmospheric trend (1.4 ± 0.1 μatm yr⁻¹) with the slowest increase in the center of the subpolar gyre where there is strong interaction with subsurface waters. This mismatch suggests the Northeast Pacific Ocean sink for atmospheric CO₂ may be increasing.

Received: 01 May 2023 – Discussion started: 15 May 2023

Download & links

Preprint (PDF, 2167 KB)

Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
Preprint (2167 KB)

Supplement (892 KB)

Download & links

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Journal article(s) based on this preprint

27 Sep 2023

Estimating marine carbon uptake in the northeast Pacific using a neural network approach

Patrick J. Duke, Roberta C. Hamme, Debby Ianson, Peter Landschützer, Mohamed M. M. Ahmed, Neil C. Swart, and Paul A. Covert

Biogeosciences, 20, 3919–3941, https://doi.org/10.5194/bg-20-3919-2023,https://doi.org/10.5194/bg-20-3919-2023, 2023

Short summary

Patrick J. Duke et al.

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2023-870', Marine Fourrier, 02 Jun 2023
General comments
This paper by Duke and co-authors describes a neural network approach to interpolate observations into a gridded pCO₂ data product over the period 1998-2019 at 1/12° resolution specifically in the Northeast Pacific. After presenting the specific circulation and hydrographic features of their area, the authors detail the methodological choices they made in developing the NN and advise the reader on specific training steps to improve performance. They then go on to discuss the wider scientific applications of such a product.
Overall, the paper is very well written, nicely structured and easy to read. The figures could be larger to improve readability, but figures and legends are still clear. The level of detail in the methodological and NN development sections is much appreciated, as it is rare to find such detail in oceanographic research papers. In particular, the decision not to normalise all your inputs and to use dynamic provinces is very interesting. Also all the developments you have done on the internal division data ratio, which could be applied to many other uses.
A few specific comments remain, as detailed below:

Specific comments
It would be nice to further highlight the potential uses of your data product and/or your method outside the traditional straightforward pCO₂/FCO₂ observing community (i.e. modellers, determination of climate indices, ...).

Have you considered comparing your FCO₂ with other products: e.g. SeaFlux (Fay et al., 2021)?

End of section 2.1 and Table 1: there seems to be some confusion here about what you have used for what. You mention the xCO₂ data produced by NOAA, but in Table 1 you cite Lanschutzer? As in the next sentence you mention the pCO₂ climatology by the same authors, restructuring the sentences if this is not a confusion would be very useful.

Table 1: for SST you kept the 1/20° resolution and did not aggregate to 1/12°, unlike the others?

Figure 2c: Is your validation data set representative enough? It does not cover your whole range?

End of section 2.5: You average the outputs of the 10 NN. You do this directly, but have you tried to give the median +/std as this is also useful information.

Line 210: "12.9 +:-1.1 µatm" : I'm not sure where this number comes from.

Line 332: flux densities as high as 3.6, but your figures end at a maximum of 3?

Lines 394-395: provide some insight into how.

In this section you compare an atmospheric xCO₂ with an increase in oceanic pCO₂. Can't you convert the atmospheric xCO₂ to pCO₂ to compare things in the same ranges/units?

pCO2ATM = [PT - (RH/100) × PH2O] × xCO2ATM

where PH2O is the water vapour pressure at atmospheric temperature for xCO2ATM (in atm) calculated according to Dickson et al. (2007), RH is the relative humidity (in %) and PT is the total atmospheric pressure (in atm). If some of these are missing, you can get them from products (SeaFlux mentioned above). Otherwise you are comparing things that are not directly comparable.

Line 437: detail how you got the vertical velocities with Ekman pumping

Technical corrections
Abstract: For readers not familiar with your particular area, it would be better to introduce the fact that your area is a sink earlier.

Line 58: "The estimated long-term trend in surface ocean pCO₂ appears to be increasing at less than the atmospheric rate." Less what? rate, amount of increase? Rewrite to clarify: "at a slower rate" or "less than the atmospheric rate of increase".

Line 119: fCO₂ has been converted to pCO₂: how, give equation.

Figure 3: Add some metrics to the graphs (from the text)

Line 150: "changing the internal division ratio", what did you end up using (see where you give more details).

Line 424: "could be partially tied to if" missing word/rewording

Figure 7: the black line is not really black.

Supplementary Figure 1 a&b: useful to superimpose bar charts of the whole dataset behind to further demonstrate how your subsampled data is representative.
Citation: https://doi.org/10.5194/egusphere-2023-870-RC1
- AC1: 'Reply on RC1', Patrick J. Duke, 22 Jul 2023
  
  Thank you Marine Fourrier for your time and careful consideration in the assessment of our manuscript. We are glad that the detailed description of the NN approach and transferability of the method development were received well. Attached we present a point-by-point response to comments.
  
  Citation: https://doi.org/10.5194/egusphere-2023-870-AC1
RC2:
'Comment on egusphere-2023-870', Anonymous Referee #2, 28 Jun 2023

Summary:
Duke and coauthors use a two-step cluster–regression method to map surface partial pressure of carbon dioxide (pCO₂) in the Northeast Pacific Ocean. Their approach is novel in that they grid pCO₂ observations and produce pCO₂ maps with high spatial resolution (1/12°); in doing so, they offer insightful observations about optimal model parameters and regional driving factors of CO₂ flux variability. This work represents not only a useful product for investigating surface carbonate chemistry in the Northeast Pacific (NEP), but a valuable roadmap for increasing the spatial resolution of observation-based surface ocean pCO₂ products.
This manuscript is very well-written and clear to follow. I was especially impressed with the analysis surrounding the training of artificial neural networks with progressively finer resolution, and the critical nature of the training/evaluation data split in these instances. The examination of driving factors of CO₂ uptake variability and the effects of marine heatwaves is interesting and will be beneficial for researchers seeking a region-wide carbonate chemistry context for the NEP. I detail a few general and line-specific suggestions below, but overall support the acceptance of this manuscript.
General suggestions:
The conclusion that the training data to internal evaluation data ratio should be optimized and likely increased toward finer resolution grids will be extremely valuable as global-scale observation-based pCO₂ data products with finer than 1° resolution are beginning to be produced. In that context, it may be helpful to expand upon the statement at the end of section 3.4 that this result “creates a precedent for stepping to a higher resolution product with nearly no loss in performance”. How might you envision that higher resolution step being taken at a global scale? What are some important considerations and potential pitfalls when taking this approach beyond the NEP? Any thoughts about increases to the temporal resolution?
One limitation of the validation performed here is that the statistical metrics represent the ability of the ANN-NEP procedure to estimate pCO₂ only at the spatiotemporal grid cells where observations are available. This may mask location-specific seasonal biases, especially at high latitudes where wintertime observations are likely not as plentiful. In lieu of a comprehensive model simulation experiment to evaluate these unquantified biases, this consideration may warrant some discussion in section 3.2 or elsewhere.
A figure displaying the most frequent occurrence of each SOM province over the timeseries would be informative. As an additional suggestion for future work: to reduce the discontinuities at the borders of biogeochemical provinces it would be interesting to explore soft clustering approaches in addition to hard clustering like SOMs. Soft clustering approaches provide probabilities for each clustered grid cell, which can be used as weights to blend pCO₂ predictions across clusters.
Line-by-line comments:
Line 85: It would be valuable to articulate why the coastal ocean was excluded in this study.
Lines 142–143: It isn’t immediately clear why choosing not to normalize predictor data implicitly weights the SOM predictors toward the pCO₂ climatology. Is it related to the relative range of each chosen predictor?
Lines 151–152: I don’t understand what is meant by “we introduced each predictor variable again after deseasonalizing”. Can this be explained more clearly?
Lines 280–281: Very interesting and insightful conclusion!
Lines 455–456: It would be good just to clarify in this sentence that “stepping to a significantly higher spatial resolution” refers to a higher resolution “than typical observation-based pCO₂ products (1/4° or 1° resolution)” or something along those lines.

Citation: https://doi.org/10.5194/egusphere-2023-870-RC2
- AC2: 'Reply on RC2', Patrick J. Duke, 22 Jul 2023
  
  Thank you for your time and careful consideration providing feedback on our manuscript. We appreciate your encouragement regarding the potential of our study to serve as a template for global products aiming to achieve higher spatial resolution. Attached we have addressed your comments in a point-by-point manner.
  
  Citation: https://doi.org/10.5194/egusphere-2023-870-AC2

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2023-870', Marine Fourrier, 02 Jun 2023
General comments
This paper by Duke and co-authors describes a neural network approach to interpolate observations into a gridded pCO₂ data product over the period 1998-2019 at 1/12° resolution specifically in the Northeast Pacific. After presenting the specific circulation and hydrographic features of their area, the authors detail the methodological choices they made in developing the NN and advise the reader on specific training steps to improve performance. They then go on to discuss the wider scientific applications of such a product.
Overall, the paper is very well written, nicely structured and easy to read. The figures could be larger to improve readability, but figures and legends are still clear. The level of detail in the methodological and NN development sections is much appreciated, as it is rare to find such detail in oceanographic research papers. In particular, the decision not to normalise all your inputs and to use dynamic provinces is very interesting. Also all the developments you have done on the internal division data ratio, which could be applied to many other uses.
A few specific comments remain, as detailed below:

Specific comments
It would be nice to further highlight the potential uses of your data product and/or your method outside the traditional straightforward pCO₂/FCO₂ observing community (i.e. modellers, determination of climate indices, ...).

Have you considered comparing your FCO₂ with other products: e.g. SeaFlux (Fay et al., 2021)?

End of section 2.1 and Table 1: there seems to be some confusion here about what you have used for what. You mention the xCO₂ data produced by NOAA, but in Table 1 you cite Lanschutzer? As in the next sentence you mention the pCO₂ climatology by the same authors, restructuring the sentences if this is not a confusion would be very useful.

Table 1: for SST you kept the 1/20° resolution and did not aggregate to 1/12°, unlike the others?

Figure 2c: Is your validation data set representative enough? It does not cover your whole range?

End of section 2.5: You average the outputs of the 10 NN. You do this directly, but have you tried to give the median +/std as this is also useful information.

Line 210: "12.9 +:-1.1 µatm" : I'm not sure where this number comes from.

Line 332: flux densities as high as 3.6, but your figures end at a maximum of 3?

Lines 394-395: provide some insight into how.

In this section you compare an atmospheric xCO₂ with an increase in oceanic pCO₂. Can't you convert the atmospheric xCO₂ to pCO₂ to compare things in the same ranges/units?

pCO2ATM = [PT - (RH/100) × PH2O] × xCO2ATM

where PH2O is the water vapour pressure at atmospheric temperature for xCO2ATM (in atm) calculated according to Dickson et al. (2007), RH is the relative humidity (in %) and PT is the total atmospheric pressure (in atm). If some of these are missing, you can get them from products (SeaFlux mentioned above). Otherwise you are comparing things that are not directly comparable.

Line 437: detail how you got the vertical velocities with Ekman pumping

Technical corrections
Abstract: For readers not familiar with your particular area, it would be better to introduce the fact that your area is a sink earlier.

Line 58: "The estimated long-term trend in surface ocean pCO₂ appears to be increasing at less than the atmospheric rate." Less what? rate, amount of increase? Rewrite to clarify: "at a slower rate" or "less than the atmospheric rate of increase".

Line 119: fCO₂ has been converted to pCO₂: how, give equation.

Figure 3: Add some metrics to the graphs (from the text)

Line 150: "changing the internal division ratio", what did you end up using (see where you give more details).

Line 424: "could be partially tied to if" missing word/rewording

Figure 7: the black line is not really black.

Supplementary Figure 1 a&b: useful to superimpose bar charts of the whole dataset behind to further demonstrate how your subsampled data is representative.
Citation: https://doi.org/10.5194/egusphere-2023-870-RC1
- AC1: 'Reply on RC1', Patrick J. Duke, 22 Jul 2023
  
  Thank you Marine Fourrier for your time and careful consideration in the assessment of our manuscript. We are glad that the detailed description of the NN approach and transferability of the method development were received well. Attached we present a point-by-point response to comments.
  
  Citation: https://doi.org/10.5194/egusphere-2023-870-AC1
RC2:
'Comment on egusphere-2023-870', Anonymous Referee #2, 28 Jun 2023

Summary:
Duke and coauthors use a two-step cluster–regression method to map surface partial pressure of carbon dioxide (pCO₂) in the Northeast Pacific Ocean. Their approach is novel in that they grid pCO₂ observations and produce pCO₂ maps with high spatial resolution (1/12°); in doing so, they offer insightful observations about optimal model parameters and regional driving factors of CO₂ flux variability. This work represents not only a useful product for investigating surface carbonate chemistry in the Northeast Pacific (NEP), but a valuable roadmap for increasing the spatial resolution of observation-based surface ocean pCO₂ products.
This manuscript is very well-written and clear to follow. I was especially impressed with the analysis surrounding the training of artificial neural networks with progressively finer resolution, and the critical nature of the training/evaluation data split in these instances. The examination of driving factors of CO₂ uptake variability and the effects of marine heatwaves is interesting and will be beneficial for researchers seeking a region-wide carbonate chemistry context for the NEP. I detail a few general and line-specific suggestions below, but overall support the acceptance of this manuscript.
General suggestions:
The conclusion that the training data to internal evaluation data ratio should be optimized and likely increased toward finer resolution grids will be extremely valuable as global-scale observation-based pCO₂ data products with finer than 1° resolution are beginning to be produced. In that context, it may be helpful to expand upon the statement at the end of section 3.4 that this result “creates a precedent for stepping to a higher resolution product with nearly no loss in performance”. How might you envision that higher resolution step being taken at a global scale? What are some important considerations and potential pitfalls when taking this approach beyond the NEP? Any thoughts about increases to the temporal resolution?
One limitation of the validation performed here is that the statistical metrics represent the ability of the ANN-NEP procedure to estimate pCO₂ only at the spatiotemporal grid cells where observations are available. This may mask location-specific seasonal biases, especially at high latitudes where wintertime observations are likely not as plentiful. In lieu of a comprehensive model simulation experiment to evaluate these unquantified biases, this consideration may warrant some discussion in section 3.2 or elsewhere.
A figure displaying the most frequent occurrence of each SOM province over the timeseries would be informative. As an additional suggestion for future work: to reduce the discontinuities at the borders of biogeochemical provinces it would be interesting to explore soft clustering approaches in addition to hard clustering like SOMs. Soft clustering approaches provide probabilities for each clustered grid cell, which can be used as weights to blend pCO₂ predictions across clusters.
Line-by-line comments:
Line 85: It would be valuable to articulate why the coastal ocean was excluded in this study.
Lines 142–143: It isn’t immediately clear why choosing not to normalize predictor data implicitly weights the SOM predictors toward the pCO₂ climatology. Is it related to the relative range of each chosen predictor?
Lines 151–152: I don’t understand what is meant by “we introduced each predictor variable again after deseasonalizing”. Can this be explained more clearly?
Lines 280–281: Very interesting and insightful conclusion!
Lines 455–456: It would be good just to clarify in this sentence that “stepping to a significantly higher spatial resolution” refers to a higher resolution “than typical observation-based pCO₂ products (1/4° or 1° resolution)” or something along those lines.

Citation: https://doi.org/10.5194/egusphere-2023-870-RC2
- AC2: 'Reply on RC2', Patrick J. Duke, 22 Jul 2023
  
  Thank you for your time and careful consideration providing feedback on our manuscript. We appreciate your encouragement regarding the potential of our study to serve as a template for global products aiming to achieve higher spatial resolution. Attached we have addressed your comments in a point-by-point manner.
  
  Citation: https://doi.org/10.5194/egusphere-2023-870-AC2

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload

ED: Publish subject to minor revisions (review by editor) (27 Jul 2023) by Julia Uitz

AR by Patrick J. Duke on behalf of the Authors (31 Jul 2023) Author's response Author's tracked changes Manuscript

ED: Publish as is (15 Aug 2023) by Julia Uitz

AR by Patrick J. Duke on behalf of the Authors (18 Aug 2023) Manuscript

Journal article(s) based on this preprint

27 Sep 2023

Estimating marine carbon uptake in the northeast Pacific using a neural network approach

Patrick J. Duke, Roberta C. Hamme, Debby Ianson, Peter Landschützer, Mohamed M. M. Ahmed, Neil C. Swart, and Paul A. Covert

Biogeosciences, 20, 3919–3941, https://doi.org/10.5194/bg-20-3919-2023,https://doi.org/10.5194/bg-20-3919-2023, 2023

Short summary

Patrick J. Duke et al.

Supplement

https://doi.org/10.5194/egusphere-2023-870-supplement

Patrick J. Duke et al.

Viewed

Total article views: 494 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
378	100	16	494	29	3	4

HTML: 378
PDF: 100
XML: 16
Total: 494
Supplement: 29
BibTeX: 3
EndNote: 4

Views and downloads (calculated since 15 May 2023)

Month	HTML	PDF	XML	Total
May 2023	119	38	2	159
Jun 2023	118	14	5	137
Jul 2023	74	23	6	103
Aug 2023	35	16	1	52
Sep 2023	32	9	2	43

Cumulative views and downloads (calculated since 15 May 2023)

Month	HTML	PDF	XML	Total
May 2023	119	38	2	159
Jun 2023	118	14	5	137
Jul 2023	74	23	6	103
Aug 2023	35	16	1	52
Sep 2023	32	9	2	43

Viewed (geographical distribution)

Total article views: 497 (including HTML, PDF, and XML) Thereof 497 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 27 Sep 2023

Download

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Preprint (2167 KB)
Metadata XML

Short summary

The ocean is both impacted by climate change and helps mitigate its affects through taking up carbon from the atmosphere. We used a machine learning approach to investigate what controls open ocean carbon uptake in the Northeast Pacific. Marine heatwaves that lasted 2–3 years increased uptake, while the Alaskan Gyre controlled uptake over 10-year time periods. The trend from 1998–2019 suggests carbon uptake in the Northeast Pacific open ocean is increasing.


Total:	0
HTML:	0
PDF:	0
XML:	0