March 2023

Frontiers in Ocean Observing: Emerging Technologies for Understanding and Managing a Changing Ocean

The mCDR deployments will necessarily modify the ocean,

but intended and unintended ecological impacts are poorly

known. This will require monitoring ecological variables in

at least the upper 1,000 m as well as deeper, including at

the seafloor where benthic systems could be affected. Side

effects would be assessed through modeling studies of the

impacts of biogeochemical changes on marine ecosystems.

Unacceptable ecological side effects, acute or chronic, may

lead to the termination of an mCDR deployment.

MODEL VALIDATION, EXCLUSION,

AND ADDITIONALITY

To achieve these three objectives, model simulations will

be run, with and without mCDR. In order to fulfill this role,

simulations will have to capture many ongoing changes

in the ocean that include those due to climate change,

the hysteresis effects from climate change, the effects

of other mCDR (and terrestrial CDR), and the effects of

emissions reductions on the ocean carbon cycle. It will

require advanced modeling capabilities that could effec-

tively simulate the state of the coupled physical and bio-

geochemical ocean and its changes under the different

mCDR scenarios. Such models pose many scientific and

technological challenges that impede the development of

Digital Twins of the Ocean (DTO). The DTO will combine

next- generation ocean modeling, artificial intelligence, and

high- performance computing to create digital replicas of

the ocean that are regularly informed and improved with

observations. Some of the observations will be used in the

models, and others will be kept for model validation.

Extensive validation of these models and their improve-

ments (e.g.,  optimization of model parameters) will be

needed well in advance of and during mCDR deploy-

ments. Confirming that the simulations are consistent

with observations will require the initiation of monitoring

a long time before any mCDR deployment. During this pre-

deployment period, no mCDR could be undertaken in the

mCDR- intended region.

However, there may be cases where model validation

could not be undertaken well in advance of the mCDR

deployments, for example, where a deployment would

take place without prior, long- term consultation with the

authority to which monitoring would be reported. In such

a case, the results of model simulations run with and with-

out mCDR could be compared with observations made in

the mCDR- deployment region and one or more control

regions (with long- term time- series observations) where

conditions would be comparable to those in the selected

mCDR region and where no mCDR of any type would

be deployed. Of course, control regions will eventually

become “contaminated” by the spread of DIC—via ocean

circulation—from mCDR deployed elsewhere (e.g.,  Boyd

and Bressac, 2016), but they could be used for model vali-

dation before this occurs and even after.

Failure to implement either pre- deployment periods

or control regions would make attribution impossible

and therefore compromise the monitoring of all mCDR

deployments in a given ocean region. In addition, an open

registry or metadatabase of the mCDR pilot studies and

deployments would be very useful in this context. It would

provide information, in particular for modelers, on the

location and depth of each activity and key information

on its technical aspects (e.g., for alkalinization, the mineral

type, timing, and amount of alkalinity added).

Furthermore, we assume here the desirable exclusion

principle, whereby the deployment of one type of mCDR

in a given ocean region excludes the possibility of deploy-

ing other types there. This principle stems from the like-

lihood that multiple- type deployments in a given region,

especially as mCDR deployments aim to sequester carbon

at the gigaton scale,1 would make attribution of individ-

ual deployments impossible (Boyd and Bressac, 2016).

Exclusion is also important because multiple-type deploy-

ments could potentially cause interactive side effects that

were not anticipated by single-type mCDR pilot studies.

Pre- deployment periods or control regions—in situ or

in models—are also needed to assess the additionality of

mCDR deployments, defined in IPCC (2022) as: “The prop-

erty of being additional. Mitigation is additional if the

greenhouse gas emission reductions or removals would

not have occurred in the absence of the associated policy

intervention or activity.” Thus, additionality is the require-

ment that the net increase in the air- to- sea CO2 flux due to

an mCDR deployment (i.e., based on detection and attribu-

tion) exceeds the flux in the absence of this mCDR.

Monitoring for additionality will be especially challeng-

ing as the change in net carbon flux into the ocean and the

magnitude of carbon sequestration caused by an mCDR

deployment will be very small compared to natural air-

sea carbon fluxes and the magnitude of the ocean carbon

sink. In addition, the models will need to take into account

the inherent uncertainties of field measurements. This will

pose challenges for both observation and modeling.

1 The intended gigaton-scale magnitude of mCDR deployments would be much larger than existing multiple overlapping uses and perturbations.

For example, global marine capture fisheries and aquaculture harvested 112 Mt of animals and 36 Mt of seaweeds (fresh weight) in marine waters

in 2020 (FAO, 2022), which represented ~0.02 GtC yr–1.

Made with Publuu - flipbook maker