Size and shape of condensing steam jets and plumes
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Physics. The day in general physics: instruments, methods and results across the field.
Kai: We're looking at the paper: "Size and shape of condensing steam jets and plumes".
Title and authors: Kai: So, this paper is called "Size and shape of condensing steam jets and plumes," and it looks at how droplets form when you inject steam into air that already has condensation nuclei.
Mira: It tackles the coupled physics here, showing how phase transitions and turbulent mixing create these droplet-laden clouds in unsaturated air.
Lev: It’s interesting because it tries to build a model from scratch using both experiments and some mean-field theory to figure out the scaling laws for these clouds.
Kai: Exactly, and what they really establish are two different asymptotic regimes for the cloud size depending on where the cloud is located relative to the steam jet or that plume.
Mira: That means if you're near the source, one scaling applies, and if you move further out into a bigger plume, something else takes over.
Lev: So we’ve got two distinct ways to predict how big these clouds get based on their environment.
Kai: Right, and that leads directly to the core of what they found: in the momentum regime, the cloud dimensions scale with the injector radius Rs.
Mira: But when it gets into that buoyancy-dominated plume, things change completely, and then size follows a m squared over five scaling law based on how much steam mass flow rate you have injected.
Lev: That m to the power of five scaling is quite specific; it links the cloud size directly to the input steam flow rate.
Kai: And that’s where the experiment comes in, using a pressure cooker as a water vapor source and measuring droplet velocities and sizes with a Phase Doppler Anemometer.
Mira: The setup itself is quite controlled, housed in a chamber with specific temperature T infinity and water vapor density rho v infinity.
Lev: From an engineering standpoint, that setup gives them the hard data they need to test these theoretical scaling laws.
Kai: But the paper also goes into the governing physical processes, which is where it gets deeper than just measuring sizes.
Mira: They look at how turbulent mixing of water vapor and enthalpy interacts with those tiny condensation nuclei.
Lev: That two-way coupling between turbulence and microphysics is key; it means things like latent heat release modify buoyancy, and the droplet radiative properties affect the energy budget.
Kai: And they use a mean-field model to predict cloud persistence, assuming that inter-droplet air stays humidity saturated, which gives them a way to bridge those microphysical processes with the macroscale dynamics.
Title and authors: Mira: That’s really important because it moves beyond just describing what happens in the immediate vicinity of the jet and tries to predict how long these clouds actually last.
Lev: So they are linking the small-scale physics of droplet growth and mixing to a bigger picture of cloud evolution.
Kai: They derive some specific scaling laws for those hydrodynamic fields, like the centerline axial velocity w being proportional to r rho s over rho infinity Rs Us lambda z in the jet regime.
Mira: And then in that plume regime, the centerline axial velocity follows something like three B over two pi times rho infinity lambda squared z cubed.
Lev: Those scaling formulas are what you actually use to predict how far a cloud will travel or how big it will get under certain conditions.
Kai: The paper also looks at the transient condensation zone size, which they call zd.
Mira: They find that this size is governed by the competition between the microphysical relaxation time and the local turbulent advection time.
Lev: So it’s not just about one timescale; it's a competition between how fast evaporation and condensation happen versus how fast the flow moves them around.
Kai: This allows them to predict exactly when evaporation and condensation occur rapidly relative to the flow time, especially as the concentration of condensation nuclei changes.
Mira: That level of transient analysis helps you understand that initial phase where things are just forming.
Lev: And they also model an effective buoyancy flux B, which isn't just gravity; it includes physical contributions like the temperature excess of superheated vapor and latent heat released during condensation.
Kai: They argue this provides a more realistic description because it accounts for how turbulent mixing in the cloud behaves as if it were driven by a uniform effective buoyancy flux B.
Mira: That’s a clever way to simplify the turbulence while keeping the essential physical effects intact.
Lev: So, they’re essentially creating an averaged buoyancy effect that captures several complex physics happening simultaneously.
Kai: They also introduced a dimensionless threshold called phi that unifies the description of the cloud boundary across both flow regimes.
Mira: This is useful because it allows for consistent modeling no matter whether you are looking at the momentum-dominated or the buoyancy-dominated flow.
Lev: So they’ve found a way to make one single framework work for both scenarios, which simplifies things a lot when you want to compare results.
Kai: And they showed this unification by plotting the ratio of cloud length to jet length against the ratio of plume length to jet length, and seeing how it collapses onto a single master curve.
Title and authors: Mira: That kind of collapse shows that their different scaling laws are consistent under that one unified description.
Lev: It’s a strong piece of evidence that their two regimes aren't totally separate phenomena but parts of a continuous picture governed by this threshold phi.
Kai: They also looked at evaporation altitude prediction using an exponential decay model for isolated droplet swarms.
Mira: This accounts for Lagrangian persistence, meaning isolated clusters can survive before turbulent stretching and folding events mix them with the sub-saturated environment.
Lev: That’s a more nuanced approach than just calculating a simple geometric length scale because it respects how droplets move through the flow field over time.
Kai: So, they are looking at both the geometric size and the temporal survival of these clusters.
Mira: It really highlights that timing is as important as spatial extent in these processes.
Lev: What this means for us is that if you’re trying to design a system, whether it's a lab experiment or something bigger, you need to account for both the size scaling and the lifetime of those structures.
Kai: So, to wrap up on "Size and shape of condensing steam jets and plumes," these findings establish concrete scaling laws based on location within the plume—momentum versus buoyancy—and they provide a mean-field model for cloud persistence governed by enthalpy and humidity mixing.
Mira: It gives us a quantitative framework for how these clouds evolve from their formation zone all the way into the larger plume.
Lev: And what this means is that we can start using these formulas to predict cloud length scales in atmospheric simulations or even in more controlled experimental setups, provided you know which regime you are in.
Kai: It’s a lot of detail, but it really shows how turbulence and microphysics are coupled together when dealing with phase transitions.
Mira: It connects the small-scale droplet behavior to the larger cloud dynamics we see up higher up.
Lev: For real hardware, this means that when we model fluid dynamics or complex chemical reactions in a flow field, we have these established scaling limits to work within rather than just guessing how fast things will evolve.
Kai: So it’s about moving from just seeing what happens to understanding the underlying mathematical structure that governs the whole process.
Mira: It’s a solid piece of work for anyone interested in how fluid dynamics and thermodynamics interact on a small scale.
Lev: We're ready to see what these scaling laws can tell us about larger atmospheric phenomena next time.
The paper's summary: Kai: So we’ve seen how they set up their lab experiment using that pressure cooker and the Phase Doppler Anemometer, right? And now we’re looking at what they actually found about the size of those clouds.
Mira: They established two distinct scaling regimes for cloud size depending on where the cloud is located relative to the steam jet or that plume. It means if you’re in one zone, one set of rules applies, and if you move into another zone, entirely different scaling takes over.
Lev: That distinction between momentum-dominated and buoyancy-dominated zones is really important for us when we think about how to model the flow on real hardware. It suggests that a single set of equations won't work for the whole setup.
Kai: Exactly. In the momentum regime, they found that cloud dimensions scale with the injector radius, which is just a fixed size thing related to where you inject your steam.
Mira: But then when it gets into that buoyancy regime, they found a specific m to the power of five scaling law for the cloud size based on how much steam mass flow rate you inject. That links the physical input directly to the resulting cloud size in that larger plume area.
Lev: So it’s not just one simple formula; you have two different ways to predict the scale depending on whether you’re near the nozzle or way out in the plume, and that makes sense for capturing how a system evolves.
Kai: And they modeled cloud persistence by looking at how enthalpy and humidity mixing interacts. They assumed that inter-droplet air stays humidity saturated, which helps them connect those tiny microphysical events to the larger dynamics of the whole cloud structure.
Mira: That bridging of microphysics and macroscale dynamics is what makes this work for predicting how long these clouds actually last before they dissipate into the surrounding air.
Lev: If you’re trying to build a system, knowing that you need to account for both the size scaling and that lifetime prediction is crucial because you can't ignore either factor in your simulation or hardware design.
Kai: They also used a dimensionless threshold called phi to unify how they describe the cloud boundary across both those different flow regimes. It means there’s one way to look at it regardless of whether it’s momentum-dominated or buoyancy-dominated, which is pretty neat for consistency.
Mira: And they even showed this unification by plotting the ratio of cloud length against jet length and comparing it to the ratio of plume length against jet length, and seeing them both collapse onto a single master curve. That really proves that their two different scaling laws are consistent under that unified description.
Lev: That collapse is strong evidence that these regimes aren't completely separate; they’re just different behaviors governed by this threshold phi, which simplifies things for modeling the whole process together.
Kai: So, in short, this paper gives us concrete scaling laws based on location within the plume and a mean-field model for cloud persistence driven by enthalpy and humidity mixing. It’s a quantitative way to track how these clouds evolve from their formation zone all the way into the larger plume.
Mira: This gives us a framework we can actually start using to predict cloud length scales in atmospheric simulations or even in more controlled experimental setups, provided you know which regime you are operating in.
Lev: For hardware designers, this means we have these established limits to work within instead of just guessing how fast things will evolve when you're modeling fluid dynamics or complex chemical reactions.
Kai: It really shows how turbulence and microphysics couple together when dealing with phase transitions, connecting the small-scale droplet behavior to the larger cloud dynamics up higher up.
Mira: It’s a solid piece of work for anyone interested in how thermodynamics and fluid dynamics interact on a small scale.
Lev: We can start using these formulas to predict cloud length scales in atmospheric simulations or even in more controlled experimental setups, provided you know which regime you are in.
The paper's improvements: Tom: So we’re shifting gears now to what the authors themselves suggest for making this kind of cloud dynamics study better, right?
Kai: They suggest improving how we predict that transient condensation zone size, which they call z sub d. They say it depends on a competition between the microphysical relaxation time and the local turbulent advection time.
Mira: That’s interesting because it moves us beyond just measuring sizes in one spot; it tells us when evaporation and condensation happen fast compared to how fast the flow actually moves those things around.
Lev: For real hardware, that means we need to figure out these two timescales precisely so we know if our simulation is catching the right physical process at that moment.
Kai: They also modeled an effective buoyancy flux B that accounts for several physical contributions, like the temperature excess of superheated vapor and the latent heat released during condensation. It’s a more realistic description because it shows how turbulent mixing behaves as if it were driven by this uniform effective buoyancy flux B.
Mira: That approach is clever because it simplifies the turbulence itself but keeps all those essential physics—the heat, the mass change—in play when you're modeling the cloud evolution.
Lev: If we can get a better grasp on that effective buoyancy flux, it’s much more realistic for our error correction models because we’re dealing with complex fluid dynamics there.
Kai: They introduced a dimensionless threshold called phi to unify the description of the cloud boundary across both flow regimes. This is useful because it allows for consistent modeling no matter whether you are looking at momentum-dominated or buoyancy-dominated flow.
Mira: That unification is key because it makes comparing results from different experimental setups much more straightforward, so we don't have to worry about switching models constantly.
Lev: If the model can handle that threshold phi consistently, then the overhead for running a complex simulation on real hardware becomes predictable rather than just a guess at where the failure point will land.
Kai: They also refined evaporation altitude prediction using an exponential decay model for isolated droplet swarms. This accounts for Lagrangian persistence, meaning isolated clusters survive before turbulent stretching and folding events mix them with the sub-saturated environment.
Mira: That’s more nuanced than just calculating a simple geometric length scale because it respects how droplets actually move through the flow field over time, which is where the actual survival happens.
Lev: Lagrangian persistence is something we definitely need to account for if we’re looking at how particles disperse in a fluid; it adds a layer of temporal complexity that makes the simulation more accurate.
Kai: So, they are looking at both the geometric size and the temporal survival of these clusters, showing that timing is as important as spatial extent in these processes.
Mira: This whole set of improvements moves us toward a much more robust framework for understanding how these clouds evolve from their formation zone all the way into the larger plume.
Lev: We can start using this improved version to predict cloud length scales in atmospheric simulations or even in more controlled experimental setups, provided we know which regime we are in.
Conclusion: Kai: So, to wrap up "Size and shape of condensing steam jets and plumes," we established that cloud size scaling depends entirely on whether you are in the momentum regime or the buoyancy regime within that plume.
Mira: And they showed that by using a dimensionless threshold called phi, you can use one single description for the whole cloud boundary, which is really useful for consistent modeling across different flow conditions.
Lev: It’s solid work because it gives us a way to predict those length scales in atmospheric simulations based on the input steam flow rate and where you are in space.
Kai: They also provided a mean-field model for cloud persistence governed by enthalpy and humidity mixing, which helps us understand how long these clouds actually last before they vanish.
Mira: That means we can bridge the gap between those small microphysical processes of condensation and the larger dynamics we see on a bigger scale.
Lev: For our error correction work, knowing that overhead lands in predictable ways based on these scaling laws is helpful because you aren't guessing at the worst-case scenario.
Kai: Overall, this paper gives us a quantitative framework for how these clouds evolve from their formation zone all the way into the larger plume using those established scaling laws.
Mira: It’s a solid piece of work connecting turbulent mixing and microphysics in phase transitions.
Lev: We can start using these formulas to predict cloud length scales in atmospheric simulations or even in more controlled experimental setups, provided we know which regime we are in.
N. Fourcade, F. Poydenot, P. Langlois, F. Barbe, N. Sinnathamby, P. Claudin, B. Andreotti
Laboratoire de Physique de l’Ecole Normale Sup´erieure, UMR 8023 PSL-SU-UPCit´e-CNRS · Physique et M´ecanique des Milieux H´et´erog`enes, UMR 7636 ESPCI-PSL-SU-UPCit´e-CNRS
physics.flu-dyn, cond-mat.soft, physics.ao-ph
Submitted: 2026-09-13
Updated: 2026-11-16
Comments: Published version, 6 pages, 4 figures. Supplemental Material as ancillary file
Journal ref: Phys. Rev. Lett. 137, 154001 (2026)
DOI: 10.1103/g8x4-l1q7
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Importance score: 92/100
The gist: Droplet-laden cloud dynamics emerge from coupled phase transitions and turbulent mixing, establishing scaling laws for steam-injected clouds in unsaturated air seeded with condensation nuclei The…
Key concepts
- Momentum Regime
- This regime describes cloud size when it is close to the steam jet. In this area, the cloud's dimensions scale directly with the radius of the injector ($R_s$). It indicates that near the source, airflow momentum dictates how large a droplet cluster can grow before other forces take over.
- Buoyancy Regime
- This regime applies to clouds extending into the plume where buoyancy drives growth. In this area, cloud size scales with $m^{2/5}$, where $m$ is the injected steam mass flow rate. This shows that as the plume moves away from the source, its size is controlled by how much steam is being injected.
- Two Asymptotic Regimes
- The paper identifies two distinct zones in the cloud's development: a momentum-dominated zone near the injector and a buoyancy-dominated plume. Understanding these regimes allows researchers to accurately model and predict cloud size under different physical conditions, linking microphysics to macroscale dynamics.
Terminology
Summary
Droplet-laden cloud dynamics emerge from coupled phase transitions and turbulent mixing, establishing scaling laws for steam-injected clouds in unsaturated air seeded with condensation nuclei The cloud size exhibits two asymptotic scalings depending on whether it is confined to the initial momentum-dominated zone or extends into the buoyancy-dominated plume
Experimental Setup and Measurements
The laboratory experiment characterizes the droplet-laden region generated by injecting a steam jet into unsaturated air containing CCNs The setup uses a sealed pressure cooker as a water vapor source and measures the mass flow rate through a cylindrical nozzle of radius Rs Phase Doppler Anemometer (PDA) is used to measure droplet velocities and size distributions The experiment is housed in a 100 m3 climatic chamber with controlled temperature T∞ and water vapor density ρ∞v
Scaling Regimes of Cloud Size
Two asymptotic regimes arise depending on the cloud's location relative to the jet or plume In the momentum regime, cloud dimensions scale with the injector radius Rs In the buoyancy regime, size follows a ˙m2/5 scaling, where ˙m is the injected steam mass flow rate
Governing Physical Processes
The life cycle of cloud droplets is governed by the turbulent mixing of water vapor, enthalpy, and CCNs This two-way coupling between turbulence and microphysics governs cloud dynamics and climate impact Latent heat release modifies buoyancy, droplet radiative properties affect energy budgets, and drag forces alter airflow dynamics The mean-field model assumes inter-droplet air remains humidity-saturated, predicting cloud persistence governed by enthalpy and humidity mixing
Self-Similar Asymptotics and Scaling Laws
The hydrodynamic fields transition between two asymptotic regimes: a jet regime where momentum flux is conserved, and a plume regime where it grows monotonically with z driven by a constant buoyancy flux B The centerline axial velocity scales as w(z) = r ρs / ρ∞ RsUs λz in the jet regime In the plume regime, w(z) = 3B / (2πρ∞λ2z)3 in Eq.
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Improved prediction of cloud size under different flow regimes by utilizing two asymptotic regimes:
In the momentum regime, cloud dimensions scale with the injector radius
andIn the buoyancy regime, size follows a m2/5 scaling.
This allows for accurate sizing predictions based on whether the cloud is in theinitial momentum-dominated zone of the jet or extends into the buoyancy-dominated plume.
-
Enhanced understanding of cloud persistence by modeling
cloud persistence governed by enthalpy and humidity mixing
through mean-field analysis. This enables predicting howinter-droplet air remains humidity-saturated
and how this influences the overall evolution, bridgingmicrophysical processes and macroscale dynamics.
-
Development of a quantitative framework for atmospheric cloud evolution by using scaling laws derived from laboratory experiments. The system can now predict cloud length scales using formulas like
l j = rrho inf / rho s Rs λΦ
in the jet regime andl p = 2rho inf mdot squared / (3pi2βρ3sλ4gΦ3/5
in the plume regime. -
Precise determination of the transient condensation zone size zd by analyzing the competition between timescales, where
The spatial extent zd of the near-injector transient condensation zone is governed by the competition between the microphysical relaxation time and the local turbulent advection time.
This allows for accurate modeling of whenevaporation and condensation occur rapidly relative to the flow time
as CCN concentration changes. -
Modeling of effective buoyancy flux B that accounts for physical contributions like
the temperature excess of the superheated vapor, the latent heat released during condensation, and the lower molar mass of water vapor compared to dry air.
This provides a more realistic description of howturbulent mixing in the cloud behaves as if driven by a uniform effective buoyancy flux B.
-
Improved prediction of cloud boundary conditions by using a dimensionless threshold Φ that
unifies the description of the cloud boundary across both flow regimes.
This allows for consistent modeling regardless of whether the flow is momentum-dominated or buoyancy-dominated, as shown by howplotting l/l j as a function of the scaling ratio l p/l j collapses the experimental data onto a single master curve.
-
Refinement of evaporation altitude prediction by using an
exponential decay
model for isolated droplet swarms. This accounts for Lagrangian persistence, whereisolated droplet clusters survive before turbulent stretching-folding events mix them with the sub-saturated environment,
providing a more nuanced understanding than simple geometric length scales.
Related papers
- Data-driven impeller model for efficient large eddy simulations of metastable von K'arm'an flows
- Numerical study of nitrogen oxides chemistry during plasma assisted combustion in a sequential combustor
- Emergent Transfer of a Physics Foundation Model from Simulation to Laboratory Turbulence
- Early onset of secondary shear instability in Kelvin-Helmholtz braids at high Reynolds number
- Rotational covariance, continuous spectrum and non-modal growth of three-dimensional planar weak viscous shock layers
- Deep Learning-based Algebraic Reynolds Stress Closures for RANS Simulations of Turbulent Flows