Characterising the marine carbon cycle in CMIP6

Recent ocean heat uptake has been unprecedented, with the last five years being the warmest in recorded history. Superimposed on this secular trend, human-induced warming of the oceans has also caused more persistent and intense periods of extreme sea surface temperature, termed ‘marine heatwaves’. Marine heatwaves can rapidly disrupt marine life and commercial fisheries as organisms are exposed to extreme conditions outside of their normal ranges. Recent high-profile marine heatwaves, most notably in the North Pacific, have triggered mass die-offs and shifts in ecosystem structure. Alongside this, climate-driven changes in marine biogeochemistry, particularly ocean deoxygenation and acidification, are imposing additional stress on marine ecosystems. Just as for heatwaves, more extreme shifts in biogeochemistry are an expected consequence of climate change, and individual cases of extreme, inhospitable low-oxygen or low-pH conditions are on the rise. However, the increased persistence of marine biogeochemical extremes in a changing climate remains underexplored by the scientific community, despite major consequences for ecosystem stability.

Plankton ecosystems continuously sequester ~1700 Pg of carbon from the atmosphere in the deep ocean via sinking particles of organic matter. The production of these particles is projected to decrease during the 21st century in response to warming-driven stratification of the surface ocean (Bopp et al., 2013). However, models are now including newly identified processes, such as temperature-dependent degradation, that influence how far these particles sink into the oldest deep ocean which in turns changes how long that carbon is stored for. However, this is not consistent across models. The impact of these new processes on future projections has yet to be quantified – do they change future projections of biologically sequestered carbon? Does the varied representation of sinking particles influence projections?

In this hackathon group we will use CMIP6 data (OMIP, DAMIP, scenarioMIP) to assess historical and projected future changes in marine biogeochemistry from timescales of daily extremes to long term (multidecadal) carbon storage.

Possible outcomes:

  1. Marine extreme events atlas: Mapping present and future co-occurrence of marine heatwaves + biogeochemical extremes (and their physical drivers – including stratification, eddy influence, atmospheric heatwaves). Specific focus will be given to high SST, low chlorophyll extremes.
  2. Future changes in biological carbon sequestration: Quantifying present and future carbon sequestration by marine biology across different models. Specific focus will be given to identifying the role of biological and physical drivers.

Possible methods:

  • (compound) extreme event statistics
  • climate attribution
  • analysis of physical drivers (eddy-tracking, Lagrangian framework)
  • downscaling or bias adjustment
  • analysis of biological drivers (export production, degradation of organic carbon)
  • quantifying carbon storage using apparent oxygen utilisation

Possible CMIP6 outputs to target:

  • experiments: piControl, historical, hist-NAT, hist-GHG, sspXXX
  • ocean outputs: thetao, so, mlotstmax, moltsmin, uo, vo, ssh (Omon); tos, sos (Oday)
  • ocnBiogeochem outputs: o2, o2sat, ph, co3satarag, co3satcal, expc, epc100, intpp, diss14cabio, dissoc (Omon); chlos and phycos (Oday)