Interfacial Spectral Memory as a State Variable for Finite-Depth Salt-Finger Exchange Journal: Deep-Sea Research Part I
Sriram P. Kalathoor · 2026
Read Interfacial Spectral Memory as a State Variable for Finite-Depth Salt-Finger Exchange Journal: Deep-Sea Research Part I by Sriram P. Kalathoor (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2606-26045.txt.
Introduction
Sriram Kalathoor studies salt fingers, the narrow sinking plumes that mix heat and salt between ocean layers. Standard theory judges an interface by local conditions alone. Kalathoor argues the interface also remembers its past. Waves, stirring, and earlier mixing leave it rough. That inherited roughness, stored as a spectrum of bumps, may steer how fingering develops.
Three-dimensional direct simulations test the claim. Every control setting stays fixed: density ratio, diffusivity ratio, layer depth, and more. Only the starting roughness spectrum changes between runs. Broad, low-mode roughness yields the largest total salt exchange. It also brings the earliest contact across the layer depth. Fine-scale roughness stays local and weaker. Mixed roughness delays scale transfer but stays rich in scalar structure across repeat runs. The results support treating spectral memory as a genuine state variable for such exchange.
Catalog record
| CREATOR | Sriram P. Kalathoor |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 48 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2606.26045
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Interfacial Spectral Memory as a State Variable for Finite-Depth Salt-Finger Exchange Journal: Deep-Sea Research Part I Sriram P. Kalathoor Abstract Thermohaline interfaces in the ocean are often treated through local double-diffusive favorability, yet finite interfaces can also inherit roughness from prior waves, stirring, intrusions, and earlier mixing events. Such inherited geometry can matter because salt fingering does not develop from a flat abstract surface in many geophysical settings. We use controlled three-dimensional direct simulations to test whether the spectral state of a finite rough interface changes the pathway by which salt-finger activity develops between adjacent layers. The density ratio, diffusivity ratio, Prandtl number, interface thickness, roughness amplitude, domain, resolution, and analysis window are held fixed; only the imposed roughness spectrum and, for one pair, the realization are changed. Broad low-mode memory produces the largest cumulative salt exchange and the earliest finite-depth contact. High-annulus memory remains localized and intermediate-scale dominated. Mixed memory produces delayed scale transfer and scalar-rich structure that is robust in integrated exchange and broad-memory measures across a second realization, while local plume timing and probe amplitudes remain realization-sensitive. The simulations therefore support treating interfacial spectral memory as an additional state variable for finite-depth double-diffusive exchange, complementary to local thermodynamic descriptors. 1 Introduction Salt fingering occurs when warm, salty water overlies cooler, fresher water, so that heat diffuses faster than salt and the stratification becomes unstable to narrow vertical exchange. The mechanism appears in the salt-fountain thought experiment of 24 and in the thermohaline convection theory of 22 . It remains important because it provides a route for vertical salt and heat transport in stratified water columns. Reviews by 28 and 17 place salt fingering within the broader double-diffusive literature, and 11 gives a modern theoretical synthesis. Practical descriptions often begin with local stability or profile measures such as density ratio and Turner angle ( 13 ; 31 ) . Flux measurements, tracer structure, and mixing estimates then connect those local indicators to transport in the ocean interior ( 2 ; 14 ) . Local-gradient measures are necessary; finite interfaces also carry geometric state. Ocean interfaces are not always flat surfaces waiting for an instability to act on them. They may carry finite-amplitude roughness inherited from internal waves, shear, intrusions, earlier double-diffusive events, or nearby turbulent motions. Two interfaces can have the same local double-diffusive favorability and the same mean layer contrast while presenting very different geometric states to the subsequent fingering dynamics. A broad coherent displacement, a short-wave rough interface, and a multi-scale rough interface are not equivalent initial states, even if they share the same density ratio. The central question is whether that inherited geometric state is dynamically disposable or physically consequential. Local thermodynamic parameters would be sufficient for early exchange if the interface roughness spectrum were erased quickly. In contrast, a spectrum that conditions whether activity remains localized, transfers into broad scales, or reaches finite-depth regions becomes part of the physical state of the interface. For interpreting observed interfaces, this distinction is practical as well as theoretical. Measurements can often detect interface thickness, roughness scale, scalar complexity, or local plume passage more readily than they can reconstruct full three-dimensional transport. The unresolved problem is how those observable interface properties relate to exchange pathways. Prior work establishes several pieces of this problem. Classical stability studies identify the growth-rate setting for fingering-favorable gradients ( 27 ; 15 ) . Finite-length and secondary instability analyses show that fingers need not remain infinitesimal local modes ( 3 ; 7 ) . Collective effects and layer-formation theories then connect fingering motions to larger-scale organization ( 23 ; 10 ) . Numerical studies have followed three-dimensional salt fingers from secondary instability to chaotic convection ( 20 ) , quantified small-scale fluxes and large-scale instabilities ( 26 ) , and connected fingering dynamics to spontaneous layer formation ( 21 ) . Observations in stratified water masses show that these processes can matter outside idealized theory ( 16 ; 14 ) , while staircase studies describe persistent layered structures in double-diffusive systems ( 8 ; 9 ) . Existing studies leave open how a finite rough interface itself influences the subsequent exchange. We test whether finite-amplitude spectral memory in the interface changes the pathway and outcome of exchange before making a mature staircase or universal flux-law statement. We isolate one question: for the same local double-diffusive parameters and finite-depth geometry, does the inherited interface spectrum change the exchange pathway? In this controlled system, broad low-mode memory creates a fast, broad connection, high-annulus memory remains localized and intermediate-scale dominated, and mixed memory produces delayed scale transfer with rich scalar structure. A second mixed realization preserves the mixed response in integrated and spectral-memory measures while local timing and probe amplitudes vary. All comparisons are made at fixed density ratio, interface thickness, roughness amplitude, and background state. 2 Background and Physical Gap Classical salt-finger theory identifies the basic instability mechanism and the role of differential diffusion. When the vertical gradients of temperature and salinity have the fingering-favorable sign, heat diffuses across a displaced parcel faster than salt, allowing salinity anomalies to maintain buoyancy differences and drive narrow vertical motions. The density ratio, diffusivity ratio, and Prandtl number summarize this local setting in both classical and multicomponent descriptions ( 29 ; 11 ) . Those parameters remain central here; all four integrations use the same values. Holding them fixed isolates whether the finite interface from which the instability develops contains additional information. Studies of observed water columns have extended the classical theory in several directions. Microstructure measurements connect salt fingering to thermocline mixing and water-mass modification ( 16 ; 14 ) . Differential-diffusion reviews and Turner-angle climatologies connect the same physics to tracer structure and profile classification ( 2 ; 31 ) . Arctic staircase observations show that double-diffusive interfaces can persist as measurable oceanic structures ( 25 ; 18 ) . Permanent staircases in the Tyrrhenian Sea provide a complementary Mediterranean example ( 1 ) . Together, these studies establish that salt fingering can move beyond an infinitesimal local instability and reorganize the scalar and velocity fields. It also motivates the practical need to relate measurable quantities, such as gradients, layer thicknesses, rough interfaces, and local plume passage, to exchange consequences. A large fraction of the quantitative literature is naturally organized around homogeneous or locally idealized gradients. This organization is powerful because it isolates the double-diffusive instability itself. For example, 20 followed three-dimensional salt fingers from secondary instability to chaotic convection, while 26 quantified small-scale fluxes and large-scale fingering instabilities. Staircase-forming simulations by 21 further demonstrated how idealized fingering systems can reorganize into layered states. However, a finite ocean interface is a geometrical object, not merely a pointwise gradient. It has thickness, displacement, roughness, lateral scale content, and a history. Internal waves can corrugate the interface; shear and intrusions can strain it; previous mixing can leave residual roughness; nearby layer deformation can create finite-amplitude departures from a flat state. Such features can be measured or inferred in field contexts, but they are not usually treated as state variables in salt-finger exchange estimates. The gap lies in the finite interface before mature staircase formation. The simulations prescribe finite-amplitude roughness states and follow how the same salt-fingering system uses