A global monthly climatology of total alkalinity: a neural network approach

Broullón, Daniel; Pérez, Fíz F.; Velo, A.; Hoppema, Mario; Olsen, Are; Takahashi, Taro; Key, Robert M.; Tanhua, Toste; González‐Dávila, Melchor; Jeansson, Emil; Kozyr, Alex; Heuven, Steven M. A. C. van

doi:10.5194/essd-11-1109-2019

Cited by 41 publications

(70 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A feed-forward neural network was configured to compute TCO2 throughout the entire global ocean and to create a global climatology based on the good results previously obtained with this method in similar studies (e.g., Broullón et al, 2019). Briefly, a neural network of this type (Fig.…”

Section: Neural Network Designmentioning

confidence: 99%

“…The method used here is equivalent to that fully described by Broullón et al (2019) for AT. In addition to the target variable (TCO2 instead of AT), the main changes in the present study compared to that of Broullón 135 et al (2019) are the inclusion of the input variable "year", accounting for the anthropogenic increase of the TCO2 pool, and the use of the pCO2 database from LDEO (Takahashi et al, 2017) in addition to the extended GLODAPv2.2019 (Olsen et al 2019) to enable more robust TCO2 estimates in the surface ocean.…”

Section: Neural Network Designmentioning

confidence: 99%

“…Similar to Broullón et al (2019), the neural networks were trained using the Levenberg-Marquardt method (Levenberg, 1944;Marquardt, 1963) through the trainlm function (detailed in Beale et al, 2018) in…”

Section: Neural Network Designmentioning

confidence: 99%

“…Ocean CO2 Atlas (SOCAT; Bakker et al, 2016)) and the great ability of the neural networks to interpolate as shown in other climatology studies about CO2 system variables (Landschützer et al, 2014;Broullón et al, 2019), show the appropriateness of this approach for generating a global monthly climatology covering more than the surface ocean.…”

mentioning

confidence: 96%

“…https://doi.org/10.5194/essd-2020-37 In the present study, we introduce the use of neural networks for going one step further in the design of a climatology. We have generated a climatology of TCO2 with a resolution consistent with that of the climatology of AT of Broullón et al (2019): horizontal resolution of 1ºx1º, 102 depth levels between 0 and 5500 m and a monthly (0-1500 m) and annual (1550-5500 m) temporal resolution. The availability of global databases containing variables of the seawater CO2 system with more and more data (e.g.,…”

mentioning

confidence: 99%

See 4 more Smart Citations

A global monthly climatology of oceanic total dissolved inorganic carbon: a neural network approach

Broullón¹,

Pérez²,

Velo³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Anthropogenic emissions of CO2 to the atmosphere have modified the carbon cycle for more than two centuries. As the ocean stores most of the carbon on our planet, there is an important task in unraveling the natural and anthropogenic processes that drive the carbon cycle at different spatial and temporal scales. We 25 contribute to this by designing a global monthly climatology of total dissolved inorganic carbon (TCO2) which offers a robust basis in carbon cycle modeling but also for other studies related to this cycle. A feedforward neural network (dubbed NNGv2LDEO) was configured to extract from the Global Ocean Data Analysis Project version 2.2019 (GLODAPv2.2019) and the Lamont-Doherty Earth Observatory (LDEO) datasets the relations between TCO2 and a set of variables related to the former's variability. The global 30 root-mean-squared error (RMSE) of mapping TCO2 is relatively low for the two datasets (GLODAPv2.2019: 7.2 µmol kg -1 ; LDEO: 11.4 µmol kg -1 ) and also for independent data, suggesting that the network does not overfit possible errors in data. The ability of NNGv2LDEO in capturing the monthly variability of TCO2 was testified through the good reproduction of the seasonal cycle in ten time-series stations spread over different regions of the ocean (RMSE: 3.6 to 13.1 µmol kg -1 ). The climatology was 35 obtained by passing through NNGv2LDEO the monthly climatological fields of temperature, salinity and oxygen from World Ocean Atlas 2013, and phosphate, nitrate and silicate computed from a neural network fed with the previous fields. The resolution is 1º x 1º in the horizontal, 102 depth levels (0-5500m) and monthly (0-1500 m) to annual (1550-5500 m), and it is centered in the year 1995. The uncertainty of the climatology is low when compared with climatological values derived from measured TCO2 in the largest 40 time-series stations. Furthermore, a computed climatology of partial pressure of CO2 (pCO2) from a previous climatology of total alkalinity and the present one of TCO2 supports the robustness of this product through the good correlation with a widely used pCO2 climatology (Landschützer et al., 2016). Our TCO2 climatology is distributed through the data repository of the Spanish National Research Council (CSIC; http://dx.doi.org/10.20350/digitalCSIC/10551, Broullón et al., 2020).45 65proportions of the chemical species of TCO2 (Kleypas and Langdon, 2000; Orr et al., 2005). These changes in seawater chemistry framed in the ocean acidification process can negatively influence various processes

show abstract

Section: Neural Network Designmentioning

confidence: 99%

Section: Neural Network Designmentioning

confidence: 99%

Section: Neural Network Designmentioning

confidence: 99%

mentioning

confidence: 96%

mentioning

confidence: 99%

See 3 more Smart Citations

A global monthly climatology of oceanic total dissolved inorganic carbon: a neural network approach

Broullón¹,

Pérez²,

Velo³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Fingerprint of climate change on Southern Ocean carbon storage

Wright

Quéré

Mayot

et al. 2022

Preprint

View full text Add to dashboard Cite

The Southern Ocean plays a critical role in the uptake, transport and storage of carbon by the global oceans. It is the ocean's largest sink of CO2, yet it is also among the regions with the lowest storage of anthropogenic carbon. This behaviour results from the unique combination of high winds driving the upwelling of deep waters and the subduction and northwards transport of surface carbon. Here we identify the indirect effect of climate-related changes in ocean conditions relative to the direct effect of anthropogenic changes in atmospheric CO2 on the reorganisation of carbon in the Southern Ocean using a combination of modelling and observations. We show that the effect of climate variability and climate change on the storage of carbon in the Southern Ocean is nearly as large as the effect of anthropogenic CO2 during the period 1998-2018 compared with a climatology around the year 1995. We identify a distinct climate fingerprint in dissolved inorganic carbon (DIC), with elevated DIC concentration in the surface ocean that reinforces the anthropogenic CO2 signal, and reduced DIC concentration in the subsurface ocean that offsets the anthropogenic CO2 signal. The fingerprint is strongest at lower latitudes (30°S-55°S). This fingerprint could serve to monitor the highly uncertain evolution of carbon within this critical ocean basin, and better identify its drivers.

show abstract

New and updated global empirical seawater property estimation routines

Carter

Fassbender

Sharp

et al. 2021

Limnology & Ocean Methods

View full text Add to dashboard Cite

We introduce three new Empirical Seawater Property Estimation Routines (ESPERs) capable of predicting seawater phosphate, nitrate, silicate, oxygen, total titration seawater alkalinity, total hydrogen scale pH (pH T ), and total dissolved inorganic carbon (DIC) from up to 16 combinations of seawater property measurements. The routines generate estimates from neural networks (ESPER_NN), locally interpolated regressions (ESPER_LIR), or both (ESPER_Mixed). They require a salinity value and coordinate information, and benefit from additional seawater measurements if available. These routines are intended for seawater property measurement quality control and quality assessment, generating estimates for calculations that require approximate values, original science, and producing biogeochemical property context from a data set. Relative to earlier LIR routines, the updates expand their functionality, including new estimated properties and combinations of predictors, a larger training data product including new cruises from the 2020 Global Data Analysis Project data product release, and the implementation of a first-principles approach for quantifying the impacts of anthropogenic carbon on DIC and pH T . We show that the new routines perform at least as well as existing routines, and, in some cases, outperform existing approaches, even when limited to the same training data. Given that additional training data has been incorporated into these updated routines, these updates should be considered an improvement over earlier versions. The routines are intended for all ocean depths for the interval from 1980 to $2030 c.e., and we caution against using the routines to directly quantify surface ocean seasonality or make more distant predictions of DIC or pH T .

show abstract

A global monthly climatology of total alkalinity: a neural network approach

Cited by 41 publications

References 49 publications

A global monthly climatology of oceanic total dissolved inorganic carbon: a neural network approach

A global monthly climatology of oceanic total dissolved inorganic carbon: a neural network approach

Fingerprint of climate change on Southern Ocean carbon storage

New and updated global empirical seawater property estimation routines

Contact Info

Product

Resources

About