Unpredictable variations in the ocean originate from both external atmospheric forcing and chaotic processes internal to the ocean itself, and are a crucial sink of predictability on interdecadal timescales. In a global ocean model, we present i.) an optimization framework to compute the most efficient noise patterns to generate uncertainty and ii.) a computationally inexpensive, dynamical method for attributing sources of ocean uncertainty to internal (mesoscale eddy-driven) and external (atmospherically driven) origins, sidestepping the more typical ensemble approach. These two methods are then applied to a range of metrics (heat content, volume transport, and heat transport) and time averages (monthly, yearly, and decadal) in the subtropical and subpolar North Atlantic. The optimal noise patterns create variability in integrated quantities of interest through features of the underlying circulation such as the North Atlantic Current and deep water formation regions. Meanwhile, noise forcing diagnosed from model representations of the actual climate system stimulates these theoretical patterns with various degrees of efficiency, ultimately leading to the growth of error. We reaffirm that higher frequency variations in meridional transports are primarily wind driven, while surface buoyancy forcing is the ultimately dominant source of uncertainty at lower frequencies. For year-averaged quantities in the subtropics, it is mesoscale eddies which contribute the most to oceanic uncertainty, accounting for up to 60% after 60 years of growth for volume transport at 25°N. The impact of eddies is greatly reduced in the subpolar region, which we suggest may be explained by overall lower sensitivity to small-scale noise there.

Abstract. Water mass ventilation provides an important link between the atmosphere and the global ocean circulation. In this study, we present a newly developed, probabilistic tool for offline water mass tracking. In particular, NEMOTAM, the tangent-linear and adjoint counterpart to the NEMO ocean general circulation model, is modified to allow passive-tracer transport. By terminating dynamic feedbacks in NEMOTAM, tagged water can be tracked forward and backward in time as a passive dye, producing a probability distribution of pathways and origins, respectively. To represent surface (re-)ventilation, we optionally decrease the tracer concentration in the surface layer and track this concentration removal to produce a ventilation record. Two test cases are detailed, examining the creation and fate of North Atlantic Subtropical Mode Water (NASMW) and North Atlantic Deep Water (NADW) in a 2∘ configuration of NEMO run with repeated annual forcing for up to 400 years. Model NASMW is shown to have an expected age of 4.5 years and is predominantly eradicated by internal processes. A bed of more persistent NASMW is detected below the mixed layer with an expected age of 8.7 years. It is shown that while model NADW has two distinct outcrops (in the Arctic and North Atlantic), its formation primarily takes place in the subpolar Labrador and Irminger seas. Its expected age is 112 years.

The role of ocean circulation in transforming surface forcing into interannual‐to‐multidecadal oceanic variability is an area of ongoing debate. Here, a novel method, establishing exact causal links, is used to quantitatively determine the role of ocean active and passive processes in transforming stochastic surface forcing into heat content variability. To this end, we use a global ocean model in which the dynamical response to forcing can be switched on (fully active) or off (purely passive) and consider the resulting effect on heat content variance. While passive ocean processes mainly control the surface variance (over 92%) in all basins, most regions show the importance of active processes at depth. This role is particularly important for full‐depth North Atlantic heat content, which we investigate further, highlighting signatures of the meridional overturning circulation in delaying the variance growth.

The role of ocean circulation in transforming surface forcing into interannual-to-multidecadal oceanic variability is an area of ongoing debate. Here, a novel method, establishing exact causal links, is used to quantitatively determine the role of ocean active and passive processes in transforming stochastic surface forcing into heat content variability. To this end, we use a global ocean model in which the dynamical response to forcing can be switched on (fully active) or o↵ (purely passive) and consider the resulting e↵ect on heat content variance. While the ocean passive processes mainly control the surface variance (over 92%) in all basins, most regions show the importance of active processes at depth. This role is particularly important for full-depth North Atlantic heat content, which we investigate further, highlighting signatures of the meridional overturning circulation.