{
  "id": "reynolds-number",
  "title": "Reynolds Number",
  "lang": "en",
  "source_lang": "en",
  "summary": "The Reynolds number is a dimensionless measure of inertial relative to viscous effects, used to compare fluid flows and characterize their regimes.",
  "category": "science",
  "subcategory": "physics",
  "type": "concept",
  "aliases": [
    "Reynolds’ number",
    "Reynolds's number"
  ],
  "generated_at": "2026-11-03",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/reynolds-number/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/reynolds-number/",
      "json": "https://aiwiki.page/zh/reynolds-number.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "The Reynolds number, usually written \\(\\mathrm{Re}\\), is a dimensionless quantity in fluid mechanics that measures the relative importance of inertial and viscous effects in a flow. Its standard form is\n\n\\[\\mathrm{Re}=\\frac{\\rho U L}{\\mu}=\\frac{UL}{\\nu},\\]\n\nwhere \\(\\rho\\) is fluid density, \\(U\\) is a characteristic speed, \\(L\\) is a characteristic length, \\(\\mu\\) is dynamic viscosity, and \\(\\nu=\\mu/\\rho\\) is kinematic viscosity. Small Reynolds numbers generally indicate that viscous effects strongly constrain motion; large Reynolds numbers indicate that inertia dominates over viscosity at the chosen scale, although viscosity can remain decisive near surfaces and at small scales.\n\nNamed after Osborne Reynolds, whose 1883 experiments clarified the transition between orderly and irregular pipe flow, the number is fundamental to pipe and channel design, aerodynamics, particle transport, biological locomotion, and numerical fluid dynamics. It helps identify flow regimes and establish similarity between experiments of different sizes. It is not, however, a universal switch between laminar and turbulent motion: its interpretation depends on geometry, disturbances, surface conditions, and the length and velocity used in its definition."
    },
    {
      "heading": "Definition, physical interpretation, and governing equations",
      "text": "The Reynolds number compares two characteristic terms in a fluid’s momentum balance. For a Newtonian fluid, whose viscous stress is proportional to the rate of deformation, the inertial term has characteristic magnitude \\(\\rho U^2/L\\), while the viscous term has magnitude \\(\\mu U/L^2\\). Their ratio is\n\n\\[\\frac{\\rho U^2/L}{\\mu U/L^2}=\\frac{\\rho UL}{\\mu}.\\]\n\nThis is a scaling comparison, rather than an exact ratio of two separately measurable forces in every flow. In particular, pressure forces may balance either inertia or viscosity, depending on the circumstances. The definition does not imply that pressure becomes negligible in either limit.\n\nAn equivalent interpretation compares timescales. The time required for fluid to travel a distance \\(L\\) is \\(t_a=L/U\\). The time for viscosity to diffuse momentum across that distance is \\(t_\\nu=L^2/\\nu\\). Consequently,\n\n\\[\\mathrm{Re}=\\frac{t_\\nu}{t_a}.\\]\n\nAt low Reynolds number, momentum diffusion acts rapidly compared with transport over the chosen distance. At high Reynolds number, a fluid parcel can traverse that distance before viscous diffusion smooths momentum variations across it. This interpretation connects the number directly to diffusion, rather than only to the informal contrast between “inertial forces” and “friction.”\n\nThe role of the number becomes explicit in the Navier–Stokes equations. For an incompressible fluid of constant density and viscosity, introduce dimensionless position, velocity, time, and pressure using \\(L\\), \\(U\\), \\(L/U\\), and \\(\\rho U^2\\). With starred symbols denoting dimensionless variables and body forces omitted, the equations become\n\n\\[\\frac{\\partial \\mathbf{u}^*}{\\partial t^*} +\\mathbf{u}^*\\cdot\\nabla^*\\mathbf{u}^* =-\\nabla^*p^* +\\frac{1}{\\mathrm{Re}}\\nabla^{*2}\\mathbf{u}^*, \\qquad \\nabla^*\\cdot\\mathbf{u}^*=0.\\]\n\nThe coefficient \\(1/\\mathrm{Re}\\) measures the strength of viscous diffusion relative to the inertial scaling. In creeping flow, a pressure scale \\(\\mu U/L\\) is often more useful; with that choice, the Reynolds number multiplies the inertial terms instead.\n\nA Reynolds number has meaning only when its scales are specified. For a pipe, the conventional choices are mean axial speed and internal diameter. For an aircraft wing, they are often free-stream speed and chord length. For a sphere, the diameter and speed relative to the surrounding fluid are usual. Different choices can produce different numerical values for the same physical flow, so thresholds and empirical correlations must retain their original definitions.\n\nFor water near \\(20^\\circ\\mathrm{C}\\), \\(\\nu\\) is approximately \\(1.0\\times10^{-6}\\ \\mathrm{m^2\\,s^{-1}}\\). Flow at \\(1\\ \\mathrm{m\\,s^{-1}}\\) through a pipe of diameter \\(0.01\\ \\mathrm{m}\\) therefore has \\(\\mathrm{Re}\\approx10^4\\). Air near room temperature and atmospheric pressure has \\(\\nu\\approx1.5\\times10^{-5}\\ \\mathrm{m^2\\,s^{-1}}\\), giving a substantially lower Reynolds number at the same speed and length."
    },
    {
      "heading": "Historical development and the interpretation of transition",
      "text": "Osborne Reynolds’s 1883 paper on the circumstances determining whether water motion would be “direct or sinuous” established the experimental basis for the number’s modern use. His apparatus admitted water from a reservoir into a glass tube through a carefully shaped entrance. A thin stream of dye introduced near the tube axis made the motion visible. At sufficiently low flow rates, the dye remained a distinct filament; at higher rates, it became distorted and dispersed across the tube.\n\nReynolds varied flow speed, tube diameter, and water temperature. Temperature changed viscosity, allowing experiments to distinguish the effects of speed and size from those of fluid properties. The results showed that the transition could not be associated with a single dimensional velocity. Instead, a combination equivalent to \\(UD/\\nu\\) organized the observations. Earlier work, including that of George Gabriel Stokes, had already established the mathematical importance of comparing inertia and viscosity. Reynolds made that comparison central to an experimentally demonstrated change of flow regime.\n\nThe experiments also revealed that transition depended on disturbances. With a quiet supply and a smooth entrance, orderly flow could persist to higher speeds than when the incoming water was disturbed. This remains an essential qualification to the familiar engineering threshold. A critical Reynolds number is not necessarily an immutable material constant; it can describe a stability boundary, the survival of a disturbance, or a transition observed under a specified experimental protocol. These are related but distinct questions.\n\nModern terminology distinguishes laminar flow, in which fluid motion remains organized into smooth trajectories or layers, from turbulence, characterized by irregular, interacting motions across a range of scales. Laminar flow need not be steady: an orderly oscillating flow can be laminar. Conversely, a turbulent flow can have a mean velocity profile that is constant in time. Reynolds number therefore does not simply measure whether a flow changes with time.\n\nThe transition problem also depends strongly on geometry. Ideal fully developed flow in a straight circular pipe has no established finite critical Reynolds number for infinitesimal linear instability. Its observed transition instead involves finite-amplitude disturbances and nonlinear dynamics. Other flows, including some shear layers and boundary layers, can amplify small disturbances through linear instability under suitable conditions. There is consequently no single mechanism by which increasing Reynolds number produces turbulence.\n\nReynolds later helped establish the statistical treatment of turbulent motion by separating velocity into mean and fluctuating parts. The resulting additional momentum fluxes are now called Reynolds stresses. Although these share his name with the Reynolds number, they are different concepts: one describes turbulent transport in averaged equations, while the other characterizes the relative scales of inertia and viscosity."
    },
    {
      "heading": "Pipe flow, ducts, and hydraulic resistance",
      "text": "For flow through a circular pipe, the conventional Reynolds number is\n\n\\[\\mathrm{Re}_D=\\frac{\\rho U_bD}{\\mu},\\]\n\nwhere \\(D\\) is internal diameter and \\(U_b=Q/A\\) is the bulk mean speed, obtained from volumetric flow rate \\(Q\\) and cross-sectional area \\(A\\). In ordinary engineering descriptions, flow below approximately \\(2{,}300\\) is usually classified as laminar, flow above approximately \\(4{,}000\\) as generally turbulent, and the interval between them as transitional. These are practical ranges, not exact universal boundaries. Strong disturbances can initiate turbulent structures near the lower range, while exceptionally quiet conditions can preserve laminar flow well above it.\n\nSteady, fully developed laminar flow of an incompressible Newtonian fluid in a straight circular pipe is described by the Hagen–Poiseuille law. With pipe radius \\(R\\), axial coordinate \\(z\\), and axial velocity \\(u(r)\\),\n\n\\[u(r)=-\\frac{1}{4\\mu}\\frac{\\mathrm{d}p}{\\mathrm{d}z}(R^2-r^2).\\]\n\nThe profile is parabolic, and the centerline speed is twice the bulk mean speed. For a pipe length \\(\\ell\\), the pressure drop is\n\n\\[\\Delta p=\\frac{32\\mu U_b\\ell}{D^2}.\\]\n\nUnder these assumptions, pressure loss is proportional to speed, rather than approximately proportional to speed squared.\n\nIn the Darcy–Weisbach equation,\n\n\\[\\Delta p=f_D\\frac{\\ell}{D}\\frac{\\rho U_b^2}{2},\\]\n\nthe Darcy friction factor for fully developed laminar circular-pipe flow is exactly\n\n\\[f_D=\\frac{64}{\\mathrm{Re}_D}.\\]\n\nThe factor must be distinguished from the Fanning friction factor, which is one quarter as large. Mixing the two conventions produces a fourfold error even when the Reynolds number is calculated correctly.\n\nFor turbulent pipe flow, resistance depends on both Reynolds number and wall roughness. A common representation is the Colebrook relation,\n\n\\[\\frac{1}{\\sqrt{f_D}} =-2\\log_{10}\\left( \\frac{\\varepsilon/D}{3.7} +\\frac{2.51}{\\mathrm{Re}_D\\sqrt{f_D}} \\right),\\]\n\nwhere \\(\\varepsilon\\) is an equivalent roughness height. In hydraulically smooth flow, viscous effects near the wall remain important to the resistance. In the fully rough limit, the friction factor becomes approximately independent of Reynolds number and is controlled mainly by relative roughness \\(\\varepsilon/D\\). A large bulk Reynolds number therefore does not eliminate the need to characterize the wall.\n\nFor noncircular ducts, a hydraulic diameter is often introduced:\n\n\\[D_h=\\frac{4A}{P_w},\\]\n\nwhere \\(P_w\\) is wetted perimeter. This supports useful comparisons, especially for turbulent duct flow, but it does not make all cross-sectional shapes equivalent. In laminar flow, the resistance coefficient retains an explicit dependence on shape and aspect ratio. Entrance conditions matter as well: a commonly used laminar estimate is \\(\\ell_e/D\\approx0.05\\mathrm{Re}_D\\), but this development-length relation applies only within its underlying assumptions."
    },
    {
      "heading": "Boundary layers, separation, and external flows",
      "text": "A large Reynolds number does not make viscosity uniformly negligible. At a stationary solid surface, the usual no-slip condition requires the fluid velocity to match that of the wall. Farther away, the fluid may move rapidly. The resulting steep velocity gradient confines strong viscous effects to a relatively thin boundary layer. Ludwig Prandtl’s 1904 boundary-layer theory explained how a nearly inviscid outer flow can coexist with a viscous region that controls wall stress, separation, and much of the resulting drag.\n\nFor steady laminar flow over a smooth flat plate aligned with a uniform stream, a local Reynolds number is defined by\n\n\\[\\mathrm{Re}_x=\\frac{U_\\infty x}{\\nu},\\]\n\nwhere \\(x\\) is distance from the leading edge. The Blasius solution for zero pressure gradient gives an approximate 99-percent boundary-layer thickness\n\n\\[\\delta_{99}\\approx\\frac{5x}{\\sqrt{\\mathrm{Re}_x}},\\]\n\nand a local skin-friction coefficient\n\n\\[C_{f,x}=\\frac{\\tau_w}{\\tfrac12\\rho U_\\infty^2} =\\frac{0.664}{\\sqrt{\\mathrm{Re}_x}}.\\]\n\nThese expressions show that a boundary layer becomes thinner relative to the distance traveled as Reynolds number increases. They apply to a particular laminar configuration, not to arbitrary bodies or pressure distributions.\n\nTransition on a flat plate is often associated with local Reynolds numbers of order several hundred thousand. Its actual location varies widely with free-stream disturbances, surface roughness, pressure gradient, vibration, and leading-edge geometry. A single quoted “critical Reynolds number” can therefore be misleading. On a wing, transition may also vary along the span, while the chord-based Reynolds number remains one fixed descriptor of the operating condition.\n\nBoundary-layer behavior strongly influences drag. A turbulent boundary layer normally has greater skin friction than a laminar one, but it also transports momentum more effectively toward the wall. It can therefore resist an adverse pressure gradient and remain attached farther downstream. On a bluff body, delayed separation may reduce pressure drag enough to offset increased skin friction. This is why producing turbulence can sometimes reduce total resistance rather than increase it.\n\nA smooth sphere provides a prominent example. Near a diameter-based Reynolds number of order \\(3\\times10^5\\), depending on surface condition and incoming turbulence, boundary-layer transition can move separation downstream and sharply lower the drag coefficient. This change is called the drag crisis. Surface roughness or dimples can shift the transition, which is central to the aerodynamic behavior of golf balls.\n\nFlow around a circular cylinder illustrates another limitation of a laminar–turbulent dichotomy. A periodic alternating wake, the Kármán vortex street, begins at a diameter-based Reynolds number of roughly 47 in the idealized unconfined case. The wake is unsteady but initially laminar. Its characteristic shedding frequency is expressed through the Strouhal number, \\(\\mathrm{St}=fD/U\\), showing how Reynolds number often operates alongside another dimensionless measure rather than describing the entire flow by itself."
    },
    {
      "heading": "Creeping flow, particles, and microscopic locomotion",
      "text": "When Reynolds number is much smaller than one, viscous effects dominate inertia on the chosen scales. For steady incompressible motion, neglecting inertia in the Navier–Stokes equations gives the Stokes equations:\n\n\\[-\\nabla p+\\mu\\nabla^2\\mathbf{u}+\\mathbf{f}=0, \\qquad \\nabla\\cdot\\mathbf{u}=0,\\]\n\nwhere \\(\\mathbf{f}\\) represents body force per unit volume. These equations are linear in velocity and pressure when geometry and material properties are fixed. Their solutions can therefore be superposed, a property that is absent from the full nonlinear equations.\n\nFor a rigid sphere of radius \\(a\\) translating slowly through an otherwise stationary, unbounded Newtonian fluid, the drag is\n\n\\[F_D=6\\pi\\mu aU.\\]\n\nThis is Stokes’s drag law. Using sphere diameter \\(d=2a\\) and the standard drag coefficient,\n\n\\[C_D=\\frac{F_D}{\\tfrac12\\rho U^2(\\pi d^2/4)},\\]\n\ngives\n\n\\[C_D=\\frac{24}{\\mathrm{Re}_d}.\\]\n\nThe increasing coefficient at decreasing Reynolds number does not mean the drag force diverges as speed approaches zero. The force decreases linearly with speed; the coefficient grows because its reference force decreases quadratically.\n\nBalancing this drag against net weight gives the terminal settling speed of an isolated small sphere:\n\n\\[U_t=\\frac{(\\rho_p-\\rho_f)gd^2}{18\\mu},\\]\n\nwhere \\(\\rho_p\\) and \\(\\rho_f\\) are particle and fluid densities. The result requires a sufficiently small particle Reynolds number, negligible interactions with other particles, and distant boundaries. It is used to interpret sedimentation, aerosols, and particle-sizing measurements. At very small gas-borne particle sizes, molecular slip can require corrections; at larger Reynolds numbers, inertial corrections or empirical drag relations are needed.\n\nMicroscopic swimming occurs in the same regime. A bacterium approximately \\(2\\ \\mu\\mathrm{m}\\) long moving at \\(20\\ \\mu\\mathrm{m\\,s^{-1}}\\) in water has a body-length Reynolds number of about \\(4\\times10^{-5}\\). Coasting is then ineffective: motion is determined primarily by the instantaneous balance of propulsive and viscous forces. Rotating flagella and traveling waves along flexible appendages provide effective propulsion without relying on substantial stored body momentum.\n\nA notable consequence is kinematic reversibility. In ideal Stokes flow, reversing prescribed boundary velocities reverses the fluid motion. Edward Purcell’s 1977 discussion of “life at low Reynolds number” popularized the scallop theorem: an ideal swimmer executing a reciprocal deformation with a single shape degree of freedom cannot achieve net propulsion in an unbounded Newtonian fluid under the theorem’s assumptions. Nonreciprocal shape cycles, traveling waves, and rotating helices avoid this restriction.\n\nSmall \\(UL/\\nu\\) alone does not justify every creeping-flow simplification. Rapid oscillation introduces an independent unsteady timescale, and fluid elasticity or nearby boundaries can alter behavior substantially. The ideal steady Stokes approximation requires that the neglected terms be small for the actual motion, not merely that a nominal Reynolds number be small."
    },
    {
      "heading": "Dynamic similarity and experimental scaling",
      "text": "The Reynolds number is a central result of dimensional analysis. For incompressible Newtonian flow around a rigid body, suppose the relevant quantities are force, density, viscosity, speed, and characteristic length, with geometry fixed. Dimensional reasoning then gives a relation of the form\n\n\\[C_D=\\Phi(\\mathrm{Re}),\\]\n\nwhere \\(C_D\\) is a dimensionless drag coefficient. If additional independent effects are present, the relation must include further dimensionless parameters. Dimensional analysis identifies the parameters but does not determine the function \\(\\Phi\\); experiments, theory, or numerical solutions are still needed.\n\nTwo flows are dynamically similar when their dimensionless governing equations and relevant boundary conditions agree. For otherwise comparable incompressible viscous flows, equal Reynolds numbers are therefore an essential condition. If a model has length \\(L_m\\) and a prototype has length \\(L_p\\), matching requires\n\n\\[\\frac{U_mL_m}{\\nu_m} =\\frac{U_pL_p}{\\nu_p}.\\]\n\nWith the same fluid at the same temperature, a one-tenth-scale model requires ten times the prototype speed. This can create practical difficulties involving power, structural loading, noise, and additional physical effects.\n\nWind tunnels address these difficulties by changing more than speed. Pressurization increases gas density and usually lowers kinematic viscosity; cooling can also raise achievable Reynolds numbers. Cryogenic tunnels use low-temperature gas to reproduce high-Reynolds-number conditions on models that fit within a test section. Water tunnels can achieve comparatively high Reynolds numbers at modest speeds because water’s kinematic viscosity is much lower than that of ordinary room-temperature air.\n\nMatching Reynolds number is not sufficient when compressibility matters. Aerodynamic similarity may also require the Mach number, \\(M=U/c\\), where \\(c\\) is the local speed of sound. Small models tested at increased speed can acquire compressibility effects absent from the prototype. In ship and open-channel testing, free-surface behavior is often governed by the Froude number, \\(U/\\sqrt{gL}\\). With the same fluid and gravitational acceleration, exact Reynolds and Froude similarity generally cannot both be achieved for models of different sizes.\n\nOther physical processes add their own requirements. Flows with a free interface may depend on the Weber number, which compares inertia with surface tension. Heat transfer introduces thermal diffusivity and the Prandtl number. Relative roughness, inlet turbulence, wall temperature, and geometric details may also have to be matched or accounted for. Equal Reynolds numbers do not compensate for different shapes or different boundary conditions.\n\nEven in nominally simple tests, the characteristic scales must be consistent. A wing Reynolds number based on chord cannot be directly compared with one based on span without conversion. Likewise, a diameter-based particle correlation cannot be used with radius while retaining the same numerical coefficients. Reliable scaling therefore includes not just a quoted Reynolds number but its definition, the fluid properties used, and the relevant conditions of the experiment."
    },
    {
      "heading": "Turbulent scales and computational fluid dynamics",
      "text": "In turbulent flow, one Reynolds number describes only part of the scale structure. A large-scale Reynolds number,\n\n\\[\\mathrm{Re}_L=\\frac{u_LL}{\\nu},\\]\n\ncompares inertia and viscosity using a representative large-eddy velocity \\(u_L\\) and length \\(L\\). When this number is large, energy can pass through a broad range of motions before being dissipated by viscosity. The process is commonly described as an energy cascade, although the details depend on the flow and are not identical in every turbulent system.\n\nUnder the assumptions underlying Kolmogorov’s small-scale theory, the smallest dynamically important scales are estimated from kinematic viscosity and the mean kinetic-energy dissipation rate per unit mass, \\(\\varepsilon\\):\n\n\\[\\eta=\\left(\\frac{\\nu^3}{\\varepsilon}\\right)^{1/4}, \\qquad u_\\eta=(\\nu\\varepsilon)^{1/4}, \\qquad \\tau_\\eta=\\left(\\frac{\\nu}{\\varepsilon}\\right)^{1/2}.\\]\n\nTheir Reynolds number satisfies\n\n\\[\\frac{u_\\eta\\eta}{\\nu}=1.\\]\n\nThus a flow with an enormous large-scale Reynolds number still contains small motions for which viscous and inertial effects are comparable. Viscosity remains the mechanism converting kinetic energy into internal energy, even when its direct influence on the largest structures is weak.\n\nIf the dissipation rate scales as \\(\\varepsilon\\sim u_L^3/L\\), then\n\n\\[\\frac{L}{\\eta}\\sim\\mathrm{Re}_L^{3/4}.\\]\n\nA simple three-dimensional estimate therefore gives a spatial resolution requirement proportional to \\(\\mathrm{Re}_L^{9/4}\\) for resolving all scales in homogeneous turbulence. This is a scaling estimate, not a universal computational-cost formula. Time resolution, intermittency, domain size, and near-wall structures can make the actual requirements more demanding.\n\nThis growth explains the difficulty of computational fluid dynamics at high Reynolds number. Direct numerical simulation resolves the relevant turbulent motions without a turbulence closure, but becomes expensive as scale separation increases. Large-eddy simulation resolves larger motions while modeling the effects of unresolved smaller motions. Reynolds-averaged methods solve for statistical mean quantities and require models for turbulent stresses. These methods use the same molecular Reynolds number, but represent the unresolved physics differently.\n\nSpecialized Reynolds numbers describe particular parts of a turbulent flow. The Taylor-microscale Reynolds number is\n\n\\[\\mathrm{Re}_\\lambda=\\frac{u'\\lambda}{\\nu},\\]\n\nwhere \\(u'\\) is a specified root-mean-square velocity fluctuation and \\(\\lambda\\) is the Taylor microscale. In wall-bounded flow, the friction Reynolds number is\n\n\\[\\mathrm{Re}_\\tau=\\frac{u_\\tau\\delta}{\\nu}, \\qquad u_\\tau=\\sqrt{\\frac{\\tau_w}{\\rho}},\\]\n\nwhere \\(\\delta\\) is an appropriate outer length and \\(\\tau_w\\) is wall shear stress. These measures are not interchangeable with pipe or chord Reynolds numbers.\n\nNumerical calculations introduce a further distinction. A cell Reynolds number \\(U\\Delta/\\nu\\), based on mesh spacing \\(\\Delta\\), can indicate whether advection dominates molecular diffusion at the grid scale. It helps explain the behavior of discretization schemes, but it neither establishes the physical flow regime nor proves that a turbulent simulation is adequately resolved."
    },
    {
      "heading": "Environmental flows, complex fluids, and limits of interpretation",
      "text": "The Reynolds number spans an extraordinary range in natural systems. A river with representative speed \\(1\\ \\mathrm{m\\,s^{-1}}\\), depth \\(1\\ \\mathrm{m}\\), and water-like kinematic viscosity has a depth-based Reynolds number of about \\(10^6\\). Its bulk motion is strongly inertial and usually turbulent. Yet narrow spaces between sediment grains and thin layers adjacent to surfaces may have much smaller local Reynolds numbers. The same system can therefore contain inertia-dominated transport, viscous sublayers, and nearly creeping motion around sufficiently small objects.\n\nIn groundwater flow, the relevant scale is commonly associated with pore or grain size rather than the dimensions of an entire aquifer. Reynolds numbers based on these scales can be small even when water travels over kilometres. This helps explain the importance of viscous resistance in porous media. Definitions vary, however, because superficial velocity, pore velocity, grain diameter, and pore length are not equivalent choices. Criteria for departure from linear pressure–flow behavior must be interpreted with the definition used to establish them.\n\nLarge oceanic and atmospheric flows generally have very high molecular Reynolds numbers, but rotation and stratification can dominate their organization. The Coriolis effect, buoyancy, and stable density layering introduce additional timescales and dimensionless parameters. Large Reynolds number does not imply that every direction or scale is equally turbulent. An effective “eddy viscosity” used in a large-scale model represents unresolved transport; it is not the molecular viscosity that appears in the fundamental Reynolds number.\n\nFor non-Newtonian fluids, viscosity may depend on shear rate, deformation history, or elastic response. A Reynolds number calculated using one nominal viscosity can consequently be ambiguous. For a power-law model,\n\n\\[\\tau=K|\\dot{\\gamma}|^{n-1}\\dot{\\gamma},\\]\n\na generalized inertial-to-viscous ratio scales as\n\n\\[\\mathrm{Re}_{\\mathrm{gen}}\\sim \\frac{\\rho U^{2-n}L^n}{K}.\\]\n\nSpecific engineering definitions include additional numerical factors chosen for a particular geometry and correlation. Such a generalized Reynolds number must be used with the corresponding constitutive assumptions rather than substituted indiscriminately into Newtonian formulas.\n\nRapidly varying flows require independent treatment of time. In pulsatile pipe flow, for example, the Womersley number\n\n\\[\\alpha=R\\sqrt{\\frac{\\omega}{\\nu}}\\]\n\ncompares unsteady inertia with viscous diffusion across the radius. Two flows can have the same mean Reynolds number but very different velocity profiles and phase relationships if their oscillation frequencies differ. Similarly, viscoelastic flows require measures of material relaxation relative to deformation time; elastic instabilities can occur even when ordinary inertial effects are weak.\n\nAt sufficiently small scales or low gas pressures, the continuum assumption itself may fail. The Knudsen number, the ratio of molecular mean free path to a characteristic length, then becomes important, and slip or kinetic descriptions may be required. Low Reynolds number does not guarantee continuum behavior: the two parameters address different physical questions. Reynolds number characterizes an inertia–viscosity balance within a specified description of the fluid, while the validity of that description depends on additional molecular, material, geometric, and temporal conditions."
    }
  ],
  "links": [
    "fluid-mechanics",
    "density",
    "viscosity",
    "pressure",
    "momentum",
    "diffusion",
    "navier-stokes-equations",
    "water",
    "turbulence",
    "boundary-layer",
    "drag-force",
    "heat",
    "prandtl-number",
    "river",
    "groundwater",
    "aquifer",
    "coriolis-effect",
    "non-newtonian-fluid"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}