Sea Surface Roughness Dependence on Ocean Wave Parameters through Large Eddy Simulation with Local Subfilter Wave Drag

This study develops a local, scale-invariant subfilter wave drag model for Large Eddy Simulations to characterize how specific ocean wave parameters influence sea surface roughness and momentum flux within the Marine Atmospheric Boundary Layer, demonstrating that these relationships extend beyond simple monotonic dependencies between wind speed and surface stress.

The Gist

Imagine the ocean and the sky as two giant, invisible dancers constantly pushing and pulling against each other. The wind blows over the waves, creating friction that slows the wind down and speeds up the waves. This "dance" is crucial for everything from predicting the weather to knowing how much power a wind farm can generate.

However, figuring out exactly how they interact is incredibly hard. The ocean isn't just flat; it's covered in waves of all sizes, from tiny ripples to massive swells. Trying to simulate every single ripple on a computer is like trying to count every grain of sand on a beach while running a marathon—it takes too much computing power.

Here is what this paper does, broken down into simple concepts:

1. The Problem: The "Blind Spot" in Computer Simulations

Scientists use supercomputers to simulate the air above the ocean (called the Marine Atmospheric Boundary Layer). To make the math work, they divide the air into a 3D grid of boxes (like a giant Rubik's cube).

• The Big Waves: If a wave is bigger than the box, the computer can see it and calculate how it pushes the air.
• The Small Waves: If a wave is smaller than the box, the computer can't "see" it. It's a blind spot.

In the past, scientists tried to fix this blind spot by using a "global average." Imagine trying to describe the texture of a rug by saying, "On average, it feels like sandpaper." This works okay if the whole rug is the same, but it fails if one corner is silk and the other is burlap. The old models assumed the "roughness" of the ocean was the same everywhere, all the time.

2. The New Solution: A "Local" and "Slippery" Model

The authors created a new way to handle those invisible, tiny waves. They introduced two key improvements:

• The "Local" Approach: Instead of saying "the whole ocean is rough," their model says, "This specific spot is rough because of these specific tiny waves, but that spot over there is smooth." It calculates the drag (friction) based on what is happening right there, right now.
• The "Slip" Factor: This is the clever part. The authors realized that tiny waves don't just sit still; they move. Sometimes they move with the wind, sometimes against it.
  • Analogy: Imagine you are walking on a moving walkway at an airport. If the walkway is moving in the same direction you are walking, you feel less resistance (you "slip" along). If it's moving against you, you have to work harder.
  • The new model accounts for this "slip." It calculates the friction based on the speed difference between the wind and the tiny waves. If the waves are moving fast with the wind, they create less drag. If they are slow or moving the other way, they create more drag.

3. The Experiment: Changing the Dancers

To test this new model, the researchers ran simulations where they could change the wind and the waves independently. Usually, in nature, strong winds create big waves, so they are linked. But in the computer, the researchers could say, "Let's have a gentle breeze but huge, steep waves," or "Let's have a hurricane-force wind but tiny ripples."

They found two distinct things happen:

1. Changing the Wind: When they just made the wind stronger, the relationship between wind speed and friction went up in a predictable, straight line. It was like turning up the volume on a speaker; the sound gets louder, but the song stays the same.
2. Changing the Waves: When they changed the shape and size of the waves (making them steeper or taller) without changing the wind, the friction changed in a completely different way. It was like switching from a smooth jazz song to a heavy metal track; the "feel" of the interaction changed entirely, even if the volume (wind speed) stayed the same.

4. Why This Matters (According to the Paper)

The paper shows that we cannot just use one simple formula to describe how the wind and ocean interact. The old formulas assumed that if you know the wind speed, you know the friction. This new research proves that the shape of the waves matters just as much as the wind speed.

• For Offshore Wind: If you are building wind turbines, knowing exactly how the waves are "slipping" against the wind helps predict exactly how much power the turbines will generate.
• For Climate Models: It helps scientists understand how energy moves between the ocean and the atmosphere more accurately.

The Bottom Line

The authors built a smarter "virtual ruler" to measure the friction between wind and waves. By realizing that tiny waves move and that friction changes from spot to spot, they created a model that is more accurate than the old "one-size-fits-all" approach. They proved that to understand the dance between the sky and the sea, you have to watch the specific steps of the dancers, not just guess the average move.

Technical Summary

Technical Summary: Sea Surface Roughness Dependence on Ocean Wave Parameters through Large Eddy Simulation with Local Subfilter Wave Drag

Problem Statement
Characterizing the Marine Atmospheric Boundary Layer (MABL) requires accurately modeling the coupling between ocean waves and the turbulent atmosphere. This coupling governs momentum exchange, which is critical for global climate modeling, ocean current dynamics, and offshore wind energy applications. Current computational approaches often rely on simplified representations of wave drag, such as constant static roughness parameters (e.g., the Charnock relation) or global dynamic coefficients. These methods fail to capture the spatial and temporal variability of wave effects, particularly when wind and wave conditions are not in equilibrium. Furthermore, field data exhibits significant scatter in momentum flux relationships that cannot be explained solely by wind speed, suggesting that wave parameters (such as significant wave height and peak frequency) play an independent and complex role in determining surface stress.

Methodology
The authors developed and implemented a new subfilter wave drag model within a Large Eddy Simulation (LES) framework using the NGA solver. The methodology involves three primary components:

1. Wave Spectrum Parameterization: Ocean waves are represented by a one-dimensional Equilibrium wave spectrum. The wave field is defined by significant wave height ($H_s$) and peak wavenumber ($k_p$), with the wind friction velocity ($u_*$) incorporated into the energy prefactor to allow for independent variation of wind and wave parameters in the simulations.
2. Resolved Wave Drag: For wave modes larger than the LES grid spacing ($\Delta x$), a phase-resolving drag force is applied to the bottom-most grid cell. This force is calculated based on the relative velocity between the wind and the wave phase speed, distinguishing between wind waves (slower than wind) and swell waves (faster than wind).
3. Local Dynamic Subfilter Drag Model: For unresolved wave modes (smaller than $\Delta x$), the authors propose an updated subfilter drag model. Unlike previous global dynamic models, this formulation introduces a local dynamic parameter ($\alpha_w$) and a slip factor ($B$).
   • The slip factor accounts for the relative velocity between the wind and subfilter waves ($B = \frac{u_b - c_{sf}}{u_b + c_{sf}}$), where $c_{sf}$ is a characteristic subfilter wave phase speed derived from the wave energy spectrum.
   • The dynamic parameter $\alpha_w$ is calculated locally at every grid point and time step by equating the total drag at the grid filter scale and a larger test filter scale (typically $2\Delta x$). This ensures the model is scale-invariant and adapts to local spatial heterogeneity.

Key Contributions

• Novel Subfilter Model: The paper presents a local, scale-invariant subfilter wave drag model that incorporates a slip factor to account for the dispersion relation of gravity waves. This addresses the limitation of previous models that assumed universal drag coefficients or neglected the relative velocity between wind and unresolved waves.
• Scale Invariance Validation: An a priori analysis demonstrates that the inclusion of the slip factor renders the dynamic parameter $\alpha_w$ significantly more scale-invariant across varying grid resolutions compared to models without the slip factor or with different weighting schemes.
• Independent Parameter Sweep: The study utilizes a computational framework that does not prescribe a global friction velocity (using constant mass flux instead). This allows for the independent variation of bulk wind velocity ($U_{bulk}$), significant wave height ($H_s$), and peak wavenumber ($k_p$), isolating the specific influence of wave parameters on MABL dynamics.

Results
The authors conducted a parameter sweep of full-scale MABL simulations and compared the results to aggregated field data (Edson et al., 2013; Romero et al., 2010) and existing parameterizations (COARE 3.5, Charnock).

• Wind vs. Wave Effects: The simulations reveal that changing the bulk wind velocity results in variations in friction velocity ($u_*$) and wind speed at 10m ($U_{10}$) that follow a trajectory roughly parallel to the COARE 3.5 fit. In contrast, varying wave parameters (specifically the dimensionless wave steepness $k_p H_s$) causes the data points to move across the COARE 3.5 relationship.
• Roughness Dependence: The equivalent global roughness length ($z_0$) is found to be largely insensitive to changes in bulk wind velocity but highly sensitive to wave conditions. Specifically, increasing wave steepness ($k_p H_s$) leads to a significant increase in drag and roughness.
• Limitations of Universal Parameterizations: The results indicate that the relationship between $u_*$ and $U_{10}$ cannot be reduced to a single universal parameterization (such as a fixed Charnock coefficient). Instead, surface drag is a function of two distinct variables: wind speed and characteristic wave steepness.
• Local Dynamics: The local dynamic parameter $\alpha_w$ exhibits spatial variations of approximately one order of magnitude within the domain, driven by local wave conditions and wind velocity, highlighting the necessity of a local rather than global approach.

Significance
The paper claims that accurately representing the MABL requires accounting for both wind and wave effects independently, rather than assuming a fixed relationship driven solely by wind speed. The developed local subfilter wave drag model provides a computationally efficient tool to simulate these dynamics without requiring the resolution of the full wave spectrum. This capability is significant for improving predictions in scenarios where wind and waves are not in equilibrium, such as in offshore wind farms where turbine wakes create sudden changes in wind speed, or in global climate models where fluctuating winds may not match the wave state. The study demonstrates that the interaction between wind and waves is more complex than previously captured by standard parameterizations, necessitating dynamic, local modeling approaches.