Changes in Ocean Heat Transport (OHT) have been associated with changes in climate over the oceans and continents. Signals such as the Atlantic Multidecadal Oscillation (an oscillation of the averaged surface temperature in the North Atlantic ocean), which drives anomalously dry/wet conditions over European, are thought to be primarily generated by basin wide fluctuations in the Atlantic Overturning Circulation and associated OHT. At the core of the link between OHT and anomalous climate conditions is a chain of events: changes in OHT result in changes in the Ocean Heat Content (OHC) over a broad range of depths, which in turn modifies the Sea Surface Temperature (SST). It is ultimately through this SST modification that the OHT changes can influence the atmospheric circulation. It has to be emphasized, however, that the OHT-OHC-SST chain is not systematic. In fact, it is likely that a large fraction of the OHT variability does only result in weak SST changes and thus little climate impact. Our current understanding of the link between OHT changes and climate variability is primarily based on statistical analyses of climate model simulations. Our proposal is that we need to understand better the mechanisms behind this link to be able to understand what the limited available observations can tell us about the ocean's impact on climate in real world, and whether or not our climate models are good approximations of the latter. Specifically, here are a few key questions that we wish to address: What makes an OHT change climatically important and potentially relevant to society? Is the OHT-OHC-SST chain properly represented in ocean and climate models? What are the implications for predictions of future climate change on the 1 to 10 year timescale of most relevance to many environmental policies of governments and businesses? To address this, we propose to explore the ocean interior dynamics that constrains the OHT-OHC-SST chain and its representation in climate prediction systems. An important innovation of the project is a methodology to evaluate the potential climate impact of OHT changes, to be used to interpret decade-long observations of the Meridional Overturning Circulation and associated OHT changes by the UK funded RAPID project and its successors. We will combine an analysis of subsurface ocean observations (Argo dataset) and ocean reanalysis products, a modeling approach with idealized models (to study physical processes) and realistic configurations (to link most readily with climate models and observations), and an evaluation of a global climate model and decadal prediction products. Besides contributing to the success of UK-funded projects (like RAPID mentioned above, but also the upcoming OSNAP), this research will also benefit the UK community in terms of evaluation and potential improvements of the UK climate models and prediction systems (UKESM1/NEMO, Met Office DePreSys).

The AMOC is a large-scale ocean circulation system composed of currents that carry warm, shallow water northwards and return cold, deeper water southwards. The AMOC is crucial in maintaining the relatively mild winter climate of Northwest Europe. A shutdown of the AMOC would strongly impact European temperature and precipitation variability. Because of the AMOC's role in regulating the global climate system, several direct ocean observing programmes have been put in place to monitor the AMOC. In the North Atlantic, the two main observing arrays are the Rapid Climate Change Program (RAPID) and the Overturning in the Subpolar North Atlantic Program (OSNAP) at 26N and 50-60N, respectively. While these programmes have transformed our understanding of changes in the AMOC, they are limited to single lines of latitude and have relatively short lifespans (two decades at most). These constraints prevent us from being able to understand changes on long (decadal to centennial) timescales or understand how the AMOC is connected across the latitudes where we don't have direct measurements. Further, the maintenance of the observing arrays is costly and there is no backup system in place in the event of instrumentation failure or loss. To overcome the limitations of the RAPID and OSNAP observing arrays, the oceanographic community has sought alternative solutions for monitoring the AMOC using cost-effective observing systems, like existing satellite and autonomous ocean robot data that have high spatial and temporal coverage. These alternative solutions for monitoring the AMOC have recently been trialled at a few places in the North Atlantic. At the same time, advances in machine learning and modelling methods are starting to prove useful for monitoring the AMOC from indirect measurements. The AMOC monitoring at RAPID and OSNAP will soon achieve 20- and 10- years worth of continuous measurements, respectively. The combination of these AMOC records with the recent developments in alternative AMOC monitoring methods means that the oceanographic community now has the tools in place to dramatically improve our ability to understand the AMOC across the North Atlantic Ocean over long time periods. Thus, MEZCAL will combine computational advances with the recent proven alternative methods for monitoring the AMOC to extend the coverage of AMOC observations across the North Atlantic and deliver a new framework that will make a step change in our understanding of AMOC variability. This project will also provide recommendations for how to build a sustainable AMOC monitoring system moving forward.

This proposal seeks to understand how a prodigious 1,250 km runout submarine sediment avalanche (turbidity current) was triggered on 14th January 2020, by the largest flood in 50 years along the Congo River. This submarine flow broke two seabed telecommunication cables that underpin data traffic to West Africa causing the internet to slow from Nigeria to South Africa. These submarine cables had not previously broken in the last 20 years. This flow also caused a series of oceanographic moorings to surface, placed along Congo Submarine Canyon by a NERC project (NE/R001952). Cable breaks and surfaced moorings show that this remarkable flow ran out for over 1,200 km, as measured along the canyon axis. Moreover, the flow continuously self-accelerated, such that it reached front speeds of >8 m/s, some 1,150 to 1,250 km from its source at the mouth of the Congo River. This is the longest runout turbidity current yet monitored in action, and the only monitored flow to continuously self-accelerate for over a thousand kilometres. It is important to understand how such powerful and very long runout turbidity currents are triggered, especially for hazards to strategic seabed cables, including cable routes that are planned for 2020-21 off West Africa. The January 14-16th submarine flow is not associated with an earthquake, and it occurred during a period of low wave heights. However, it does coincide with an extreme flood of 80,000 m3s-1 observed in December 2019 along the Congo River. It is thus also important to determine how the frequency of submarine flows will be effected by future climate and hydrological changes in the Congo Basin. Here we seek to understand how this exceptional river flood triggered a thousand kilometre submarine flow, by conducting a detailed survey of the Congo River mouth. We will use the geomorphology of that river-to-submarine-canyon transition to understand how the offshore flow was triggered by the river flood, for example by mapping landslide scars, or testing a hypothesis that river bedload was driven over a single steep avalanche face. This is an urgency grant because evidence of how the Jan 2020 flow was triggered (e.g. seabed failure scarps) will be buried or wiped-out by the next peak discharge of the Congo River in Oct 2020. There are extremely few direct measurements of the most powerful turbidity currents that run out for hundreds to thousands kilometres to the deep ocean, and the few measurements available previously produced step changes in understanding. Indeed, there has only been one previously directly-measured turbidity current on this scale, which is the Grand Banks event in 1929 that broke all ~20 cables across the N. Atlantic. The Grand Banks event ran out for over 800km, but decelerated from 19 m/s to 3 m/s, rather than continuously accelerating as in the Jan 2020 event. Moreover, the Jan 2020 event already has much more detailed measurements from the timing of offshore moorings, with further data to come via recovery of these moorings and 12 OBS (with hydrophones and geophones) on a NERC cruise. This Jan 2020 event is thus a rare and extremely valuable opportunity to understand how far large-scale flows operate, linked to exceptional river floods with much longer (50-100 year) recurrence intervals. The main gap in our understanding of the Jan 2020 event is what happened at the river mouth, and this is key for predicting flow frequency and links to climate change. The geomorphology of the river to canyon-head transition is currently unknown. For example, UKHO bathymetric charts mainly use data collected in the 1890s. Here we will use swath multibeam echosounder systems to survey the river to canyon transition at much higher resolution and in three dimensions, thereby documenting its geomorphology in unprecedented detail. Past work shows how a single multibeam bathymetric survey can produce major insights into turbidity currents triggering at river mouths.

Human activities have caused atmospheric CO2 levels to increase dramatically, but their growth has been slowed by the oceans absorbing approximately one quarter of this anthropogenic carbon (Canth). Globally, the North Atlantic Ocean stores the highest quantities of Canth, due to local CO2 uptake from the atmosphere, and large-scale ocean currents, particularly the Atlantic Meridional Overturning Circulation (AMOC) delivering waters high in Canth to northern locations where they cool, get denser and sink to great depths away from contact with the atmosphere. Models project that the size of this carbon sink will reduce in the coming decades despite continued atmospheric CO2 increases, as surface warming increases stratification, decreases CO2 solubility, and AMOC weakening slows the transport of dense waters to depth. However there is substantial model spread regarding flux peak, and decline timing. The same models show a large range in ocean carbon transports, often related to AMOC representation. The balance between air-sea fluxes and ocean transports to North Atlantic Canth accumulation is thus not well constrained both now and into the future, and subject to large uncertainties. Previous observational studies have attempted to quantify the contributions of these processes to Canth accumulation in order to assist with model verification and validation. However, it is not currently possible to directly measure anthropogenic air-sea CO2 fluxes - they are chemically identical to those with 'normal', non-human-derived CO2. And while they can be calculated indirectly from trans-ocean basin decadal repeat cruises, this approach is subject to large uncertainties. It is thus impossible to constrain why fluxes (or carbon transports) vary on shorter timescales, or how they interact with the AMOC. For this we require frequent estimates of ocean transports combined with frequent estimates of how quickly carbon concentrations are increasing in the ocean. This project will look to do precisely that. Firstly, we will generate new high-resolution estimates of Canth transports across the subtropical and subpolar boundaries of the North Atlantic, relying on the outputs from the RAPID (10day) and OSNAP (monthly) mooring arrays. At RAPID, we will extend to 2024 the 2004-2013 time-series we published in 2021 and that identified a stable, northward Canth transport that was highly variable over all time scales (weekly, monthly, seasonally, annually, interannually), and highly correlated to the AMOC. We will collect new sub-seasonal water samples in Florida Straits, at the western boundary. The waters that flow through the Straits represent the vast majority of the upper, northward-flowing part of the overturning circulation but we don't currently account for any variability in water mass characteristics (chemical or otherwise) in the transport calculation there, so are not fully characterising the AMOC:carbon coupling. We'll generate a novel Canth transports time-series for 2014-2022 at the OSNAP, identifying how it co-varies with AMOC, and RAPID carbon transports. We'll track the changing interior (anthropogenic) carbon signal using novel, publicly-available datasets based on ship and autonomous platform data. Combined, we'll form a North Atlantic budget with transports at the southern and northern boundaries, and evolving concentrations in the interior. The residual will represent Canth entering (or leaving) through the surface - the air-sea flux. The contributions of air-sea fluxes and ocean circulation to regional carbon accumulation will be determined, better understanding how, with AMOC, they work together to store carbon. The calculation scheme, its components and transport/air-sea flux/AMOC relationships will be tested in earth system models, before observations are compared to simulation outputs. Our findings will help improve the accuracy of climate models, which is crucial for predicting the effects of climate change.

The Arctic region is undergoing dramatic changes, in the atmosphere, ocean, ice and on land. The Arctic lower atmosphere is warming at more than twice the rate of the global average, the Arctic sea ice and Greenland Ice Sheet melt have accelerated in the past 30 years. Notable observed changes in the ocean include the freshening of the Beaufort Gyre, and 'Atlantification' of the Barents Sea and of the Eastern Arctic Ocean. Such profound environmental change is likely to have implications across the globe - it is often said, "What happens in the Arctic doesn't stay in the Arctic". Past work has indicated that Arctic amplification can, in principle, affect European climate and extreme weather, but a clear picture of how and why is currently lacking. The 2019 Intergovernmental Panel on Climate Change (IPCC) Special Report on Oceans and Cryosphere concluded "changes in Arctic sea ice have the potential to influence midlatitude weather, but there is low confidence in the detection of this influence for specific weather types". ArctiCONNECT brings together experts in climate dynamics, polar and subpolar oceanography, and extreme weather, in order to transform understanding of the effects of accelerating Arctic warming on European climate and extreme weather, through an innovative and integrative program of research bridging theory, models of varying complexity, and observations. It will (i) uncover the atmospheric and oceanic mechanisms of Arctic influence on Europe; (ii) determine the ability of state-of-the-art climate models to simulate realistic Arctic-to-Europe teleconnections; and (iii) quantify and understand the contribution of Arctic warming to projected changes in European weather extremes and to the hazards posed to society.