Calculating surface ocean pCO2 from biogeochemical Argo floats equipped with pH: An uncertainty analysis

Williams, Nancy L.; Juranek, L. W.; Feely, Richard A.; Johnson, Kenneth S.; Sarmiento, Jorge L.; Talley, Lynne D.; Dickson, Andrew G.; Gray, Alison R.; Wanninkhof, Rik; Russell, J. L.; Riser, Stephen C.; Takeshita, Yuichiro

doi:10.1002/2016gb005541

Cited by 139 publications

(201 citation statements)

References 42 publications

Supporting

Mentioning

191

Contrasting

Unclassified

Order By: Relevance

“…This makes a strong case for autonomous sampling platforms to the Southern Ocean's winter sampling gap. The SOCCOM float project may soon yield such measurements with pHderived estimates of pCO 2 (Russell et al, 2014;Johnson et al, 2017;Williams et al, 2017). Secondly, the model used to generate the synthetic data may not be representative of the Southern Ocean.…”

Section: Socat Biasesmentioning

confidence: 99%

Empirical methods for the estimation of Southern Ocean CO2: support vector and random forest regression

Gregor

Kok

Monteiro³

2017

Biogeosciences

View full text Add to dashboard Cite

Abstract. The Southern Ocean accounts for 40 % of oceanic CO 2 uptake, but the estimates are bound by large uncertainties due to a paucity in observations. Gap-filling empirical methods have been used to good effect to approximate pCO 2 from satellite observable variables in other parts of the ocean, but many of these methods are not in agreement in the Southern Ocean. In this study we propose two additional methods that perform well in the Southern Ocean: support vector regression (SVR) and random forest regression (RFR). The methods are used to estimate pCO 2 in the Southern Ocean based on SOCAT v3, achieving similar trends to the SOM-FFN method by Landschützer et al. (2014). Results show that the SOM-FFN and RFR approaches have RMSEs of similar magnitude (14.84 and 16.45 µatm, where 1 atm = 101 325 Pa) where the SVR method has a larger RMSE (24.40 µatm). However, the larger errors for SVR and RFR are, in part, due to an increase in coastal observations from SOCAT v2 to v3, where the SOM-FFN method used v2 data. The success of both SOM-FFN and RFR depends on the ability to adapt to different modes of variability. The SOM-FFN achieves this by having independent regression models for each cluster, while this flexibility is intrinsic to the RFR method. Analyses of the estimates shows that the SVR and RFR's respective sensitivity and robustness to outliers define the outcome significantly. Further analyses on the methods were performed by using a synthetic dataset to assess the following: which method (RFR or SVR) has the best performance? What is the effect of using time, latitude and longitude as proxy variables on pCO 2 ? What is the impact of the sampling bias in the SOCAT v3 dataset on the estimates? We find that while RFR is indeed better than SVR, the ensemble of the two methods outperforms either one, due to complementary strengths and weaknesses of the methods. Results also show that for the RFR and SVR implementations, it is better to include coordinates as proxy variables as RMSE scores are lowered and the phasing of the seasonal cycle is more accurate. Lastly, we show that there is only a weak bias due to undersampling. The synthetic data provide a useful framework to test methods in regions of sparse data coverage and show potential as a useful tool to evaluate methods in future studies.

show abstract

Section: Socat Biasesmentioning

confidence: 99%

Empirical methods for the estimation of Southern Ocean CO2: support vector and random forest regression

Gregor

Kok

Monteiro³

2017

Biogeosciences

View full text Add to dashboard Cite

show abstract

“…With the float's capability to measure pH and utilization of existing algorithms for predicting total alkalinity, pCO 2 can be calculated from the collected observations [Williams et al 2017] and compared to underway observations. The uncertainty range for these calculated pCO 2 values is estimated to be ±11 µatm and results from 155!…”

Section: !mentioning

confidence: 99%

“…The uncertainty range for these calculated pCO 2 values is estimated to be ±11 µatm and results from 155! various components of the calculation procedure [Williams et al 2017]. pCO 2 estimates from autonomous floats such as the SOCCOM floats have not been included in the SOCAT database because they do not report original surface water CO 2 data.…”

Section: !mentioning

confidence: 99%

“…In addition, empirical algorithms for alkalinity must be used to estimate pCO 2 from float-based pH [Carter et al 2016;Williams et al 2016]. A thorough analysis of all sources of uncertainty estimated the total uncertainty in float-based pCO 2 to be as high as 11 µatm, although crossover comparisons with underway ship data collected close to the first float profile found a mean difference of only 3.7 µatm [Williams et al 2017]. However, it is expected that the sensors will drift or require…”

Section: !mentioning

confidence: 99%

See 1 more Smart Citation

Utilizing the Drake Passage Time-series to understand variability and change in subpolar Southern Ocean pCO2

Fay¹,

Lovenduski²,

McKinley³

et al. 2017

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…The spread of these con di tions in the ba sin de pended not only on the wa ter tem per a ture and salin ity, but also on the in ten sity of hy dro ther mal and diagenetic pro cesses (Brener, 1984;Tucker, 1991;Katz et al, 2001). These pro cesses as so ci ated with the re lease of vol ca nic and biogenic gases and flu ids (CO 2 , NH 3 and SiO 2 ) also fa voured the for ma tion of min er als (phosphorite, sid er ite) which con tributed to cal cium uti li za tion and acid i fi ca tion Wil liams et al, 2017).…”

Section: Introductionmentioning

confidence: 92%

Environmental stress in the northern Tethys during the Paleogene: a review of foraminiferal and geochemical records from the Polish Outer Carpathians

Olszewska

Szydło

2017

View full text Add to dashboard Cite

Olszewska, B., Szyd³o, A., 2017. En vi ron men tal stress in the north ern Tethys dur ing the Paleogene: a re view of foraminiferal and geo chem i cal re cords from the Pol ish Outer Carpathians. Geo log i cal Quar terly, 61 (3): 682-695, doi: 10.7306/gq.1369 Dur ing the Paleogene, the area of the north ern Tethys was con trolled by a tur bid ity sys tem stim u lated by di a strophic and geodynamic pro cesses. These fac tors con trib uted to the dis per sion and rapid ox y gen a tion of or ganic de bris. Its ac cu mu lation was a con se quence of stag nant bot tom wa ter con di tions that pe ri od i cally oc curred in the ba sin. In these pe ri ods, in tense de com po si tion in ten si fied by hy dro ther mal and diagenetic pro cesses was as so ci ated with ox y gen con sump tion and the release of green house gases, which led to hypoxia and acid i fi ca tion. These phe nom ena in ten si fied by ther mal and den sity strat i fi ca tion had a ma jor im pact on the struc ture, evo lu tion and dis tri bu tion of biota. Stress as so ci ated with rap idly chang ing con di tions in duced by sed i men tary pro cess and upwelling re sulted in the dom i nance of forms that col o nized most sed i ments (Glomospira, Ammodiscus, Recurvoides, Rzehakina) and sur face wa ters (Guembelitria, Chiloguembelina, Globanomalina, Globigerina, Cassigerinella, Catapsydrax). At the time, foraminifera were lim ited to low-di ver si fied eutrophic as sem blages or were mainly re placed by si li ceous phytoplankton (radio lar ians and di a toms). Geo chem i cal data con firm that en vi ron men tal cri ses in the Paleogene ba sin took place un der chang ing ther mal con di tions that re flect global events (KTBE, PTME, EEOC and TTE). Ther mal stress fa vour ing the for ma tion of cer tain min er als or rocks oc ca sion ally oc curred dur ing the Paleocene to Eocene (sid er ite, phos phates) and dom i nated in the Early Oligocene (sil ica).

show abstract

Calculating surface ocean pCO₂ from biogeochemical Argo floats equipped with pH: An uncertainty analysis

Cited by 139 publications

References 42 publications

Empirical methods for the estimation of Southern Ocean CO<sub>2</sub>: support vector and random forest regression

Empirical methods for the estimation of Southern Ocean CO<sub>2</sub>: support vector and random forest regression

Utilizing the Drake Passage Time-series to understand variability and change in subpolar Southern Ocean pCO<sub>2</sub>

Environmental stress in the northern Tethys during the Paleogene: a review of foraminiferal and geochemical records from the Polish Outer Carpathians

Contact Info

Product

Resources

About

Calculating surface ocean pCO2 from biogeochemical Argo floats equipped with pH: An uncertainty analysis

Cited by 139 publications

References 42 publications

Empirical methods for the estimation of Southern Ocean CO&lt;sub&gt;2&lt;/sub&gt;: support vector and random forest regression

Empirical methods for the estimation of Southern Ocean CO&lt;sub&gt;2&lt;/sub&gt;: support vector and random forest regression

Utilizing the Drake Passage Time-series to understand variability and change in subpolar Southern Ocean pCO&lt;sub&gt;2&lt;/sub&gt;

Environmental stress in the northern Tethys during the Paleogene: a review of foraminiferal and geochemical records from the Polish Outer Carpathians

Contact Info

Product

Resources

About

Calculating surface ocean pCO₂ from biogeochemical Argo floats equipped with pH: An uncertainty analysis

Empirical methods for the estimation of Southern Ocean CO<sub>2</sub>: support vector and random forest regression

Empirical methods for the estimation of Southern Ocean CO<sub>2</sub>: support vector and random forest regression

Utilizing the Drake Passage Time-series to understand variability and change in subpolar Southern Ocean pCO<sub>2</sub>