Wind Shear Velocity Research Articles

Analysis of unsteady flow over Offshore Wind Turbine in combination with different types of foundations

Environmental effects have an important influence on Offshore Wind Turbine (OWT) power generation efficiency and the structural stability of such turbines. In this study, we use an in-house Boundary Element (BEM)—panMARE code—to simulate the unsteady flow behavior of a full OWT with various combinations of aerodynamic and hydrodynamic loads in the time domain. This code is implemented to simulate potential flows for different applications and is based on a three-dimensional first-order panel method. Three different OWT configurations consisting of a generic 5 MW NREL rotor with three different types of foundations (Monopile, Tripod, and Jacket) are investigated. These three configurations are analyzed using the RANSE solver which is carried out using ANSYS CFX for validating the corresponding results. The simulations are performed under the same environmental atmospheric wind shear and rotor angular velocity, and the wave properties are wave height of 4 m and wave period of 7.16 s. In the present work, wave environmental effects were investigated firstly for the two solvers, and good agreement is achieved. Moreover, pressure distribution in each OWT case is presented, including detailed information about local flow fields. The time history of the forces at inflow direction and its moments around the mudline at each OWT part are presented in a dimensionless form with respect to the mean value of the last three loads and the moment amplitudes obtained from the BEM code, where the contribution of rotor force is lower in the tripod case and higher in the jacket case and the calculated hydrodynamic load that effect on jacket foundation type is lower than other two cases.

Optimal array of sand fences

Sand fences are widely applied to prevent soil erosion by wind in areas affected by desertification. Sand fences also provide a way to reduce the emission rate of dust particles, which is triggered mainly by the impacts of wind-blown sand grains onto the soil and affects the Earth’s climate. Many different types of fence have been designed and their effects on the sediment transport dynamics studied since many years. However, the search for the optimal array of fences has remained largely an empirical task. In order to achieve maximal soil protection using the minimal amount of fence material, a quantitative understanding of the flow profile over the relief encompassing the area to be protected including all employed fences is required. Here we use Computational Fluid Dynamics to calculate the average turbulent airflow through an array of fences as a function of the porosity, spacing and height of the fences. Specifically, we investigate the factors controlling the fraction of soil area over which the basal average wind shear velocity drops below the threshold for sand transport when the fences are applied. We introduce a cost function, given by the amount of material necessary to construct the fences. We find that, for typical sand-moving wind velocities, the optimal fence height (which minimizes this cost function) is around 50 cm, while using fences of height around 1.25 m leads to maximal cost.

Wave and wind effects on the oxygen transfer across an air‐water interface: An experimental study

ABSTRACTThe results of an experimental study on the oxygen transfer across the air‐water interface under the combined effect of wind and mechanically generated waves are presented in the paper. The experiments were carried out in a wind tunnel containing a water tank. From the analysis of the experimental data, a relationship for determining the gas transfer coefficient was obtained, and found that the mean square slope of the water waves, the wave celerity, the wind shear velocity, and the Schmidt number are the relevant variables involved in the phenomena. The results of this study can be considered as a contribution to the advancement of the phenomenon of reaeration in closed water bodies subjected to waves and wind, such as reservoirs and lakes.

Wave motion induced by turbulent shear flows over growing Stokes waves

The recent analytical of multi-layer analyses proposed by Sajjadi et al. (J Eng Math 84:73, 2014) (SHD14 therein) is solved numerically for atmospheric turbulent shear flows blowing over growing (or unsteady) Stokes (bimodal) water waves, of low-to-moderate steepness. For unsteady surface waves, the amplitude $$a(t)\propto e^{kc_it}$$ , where $$kc_i$$ is the wave growth factor, k is the wavenumber, and $$c_i$$ is the complex part of the wave phase speed, and thus, the waves begin to grow as more energy is transferred to them by the wind. This will then display the critical height to a point, where the thickness of the inner layer $$k\ell _i$$ becomes comparable to the critical height $$kz_c$$ , where the mean wind shear velocity U(z) equals the real part of the wave speed $$c_r$$ . It is demonstrated that as the wave steepens further the inner layer exceeds the critical layer, and beneath the cat’s-eye, there is a strong reverse flow which will then affect the surface drag, but at the surface, the flow adjusts itself to the orbital velocity of the wave. We show that in the limit as $$c_r/U_*$$ is very small, namely, slow-moving waves (i.e., for waves traveling with a speed $$c_r$$ which is much less than the friction velocity $$U_*$$ ), the energy-transfer rate to the waves, $$\beta $$ (being proportional to momentum flux from wind to waves), computed here using an eddy-viscosity model, agrees with the asymptotic steady-state analysis in Belcher and Hunt (J Fluid Mech 251:109, 1993), and the earlier model in Townsend (J Fluid Mech 98:171, 1980). The non-separated sheltering flow determines the drag and the energy transfer and not the weak critical shear layer within the inner shear layer. Computations for the cases when the waves are traveling faster (i.e., when $$c_r>U_*$$ ) and growing significantly (i.e., when $$0<c_i/U_*$$ ) show critical shear layer forms outside the inner surface shear layer for steeper waves. Analysis, following Miles (J Fluid Mech 3:185; 1957, J Fluid Mech 256:427, 1993) and SHD14, shows that the critical layer produces a significant but not the dominant effect; a weak lee-side jet is formed by the inertial dynamics in the critical layer, which adds to the drag produced by the sheltering effect. The latter begins to decrease when $$c_r$$ is significantly exceeds $$U_*$$ , as has been verified experimentally. Over peaked waves, the inner layer flow on the lee side tends to slow and separate, which over a growing fast wave deflects the streamlines and the critical layer upwards on the lee side. This also tends to increase the drag and the magnitude of energy-transfer rate $$\beta $$ (from wind to waves). These complex results, computed with relatively simple turbulence closure model, agree broadly with DNS simulations in Sullivan et al. (J Fluid Mech 404:47, 2000). It is proposed, using an earlier study SHD14, that the mechanisms identified here for wave-induced motion contributes to a larger net growth of wind-driven water waves when the waves are non-linear (e.g., bimodal waves) compared with growth rates for monochromatic waves. This is because in non-linear waves, individual harmonics have stronger positive and weaker negative growth rates.

Modelling aeolian sand transport using a dynamic mass balancing approach

Knowledge of the changing rate of sediment flux in space and time is essential for quantifying surface erosion and deposition in desert landscapes. Whilst many aeolian studies have relied on time-averaged parameters such as wind velocity (U) and wind shear velocity (u⁎) to determine sediment flux, there is increasing field evidence that high-frequency turbulence is an important driving force behind the entrainment and transport of sand. At this scale of analysis, inertia in the saltation system causes changes in sediment transport to lag behind de/accelerations in flow. However, saltation inertia has yet to be incorporated into a functional sand transport model that can be used for predictive purposes. In this study, we present a new transport model that dynamically balances the sand mass being transported in the wind flow. The ‘dynamic mass balance’ (DMB) model we present accounts for high-frequency variations in the horizontal (u) component of wind flow, as saltation is most strongly associated with the positive u component of the wind. The performance of the DMB model is tested by fitting it to two field-derived (Namibia's Skeleton Coast) datasets of wind velocity and sediment transport: (i) a 10-min (10Hz measurement resolution) dataset; (ii) a 2-h (1Hz measurement resolution) dataset. The DMB model is shown to outperform two existing models that rely on time-averaged wind velocity data (e.g. Radok, 1977; Dong et al., 2003), when predicting sand transport over the two experiments. For all measurement averaging intervals presented in this study (10Hz–10min), the DMB model predicted total saltation count to within at least 0.48%, whereas the Radok and Dong models over- or underestimated total count by up to 5.50% and 20.53% respectively. The DMB model also produced more realistic (less ‘peaky’) time series of sand flux than the other two models, and a more accurate distribution of sand flux data. The best predictions of total sand transport are achieved using our DMB model at a temporal resolution of 4s in cases where the temporal scale of investigation is relatively short (on the order of minutes), and at a resolution of 1min for longer wind and transport datasets (on the order of hours). The proposed new sand transport model could prove to be significant for integrating turbulence-scale transport processes into longer-term, macro-scale landscape modelling of drylands.

Experimental and numerical study of Sharp’s shadow zone hypothesis on sand ripple wavelength

Abstract Despite advances in understanding processes of sand transport by saltation and reptation, which are involved in the formation of sand ripples, the mechanisms that determine the linear dependence of ripple spacing on wind speed and their relative importance are yet unknown. In a pivotal study, Sharp (1963) proposed that this linear dependence arises from the scaling of the ripples’ shadow zone – the part of the ripple devoid of particle impacts – with the wind speed. Here, we test this hypothesis by integrating wind tunnel experiments with numerical simulations of saltation. Specifically, we measured the effective shadow zone by using sand traps designed for this purpose and found a linear relationship between the shadow zone and the wind shear velocity, consistent with Sharp’s hypothesis. However, contrary to what Sharp assumed, we found that the shadow zone is not completely screened from particle impacts, which as indicated by numerical simulations is due to the wide distribution of impact angles. Nonetheless, the shadow zone can be one of the major mechanisms contributing to the linear increase of the ripple wavelength with wind speed at the nonlinear growth stage of the ripples where merging events between small ripples take place. However, for the initial stage of ripple development, when the ripple dimension is small, other mechanisms can be dominant, such as the recently suggested resonant saltation trajectory (Duran et al., 2014).

Diffusional mass transfer coefficient at the water–sediment interface for wind-induced flow in very shallow lagoons

Very shallow lagoons that are a few centimeters deep are common in the arid Andes of Northern Chile, Argentina, Bolivia and Peru. The dynamics of these lagoons are dominated by the water–sediment interface (WSI) and strong afternoon winds. Although many studies have examined the diffusional mass transfer coefficients (k t ) of open channel flows, estimates for wind-induced flows are still unknown. The aim of this article is to propose and validate an analytical expression for computing k t at the WSI for wind-induced flow. The laboratory measurements were conducted in a wind tunnel with a water tank of variable depth located at its downwind end. Natural muddy sediments were placed in the middle of the tank so that the dissolved oxygen (DO) was consumed in the sediments. The diffusional mass transfer coefficient that characterizes the DO uptake in the sediment was obtained from DO micro-profiles measured with an OX-25 Unisense microelectrode. Water velocity profiles were measured with a 2D side-view Sontek acoustic doppler velocimetry (ADV), and the wind shear velocity was computed based on wind velocity profiles that were measured with an Extech hot-wire anemometer. A total of 16 experiments were conducted with different water depths and wind shear stresses. The constants required by the model were determined from these experiments, and the analytical expression was successfully validated by the laboratory observations. The analytical expression obtained for computing k t was also validated with field observations that were conducted in October, 2012, in Salar del Huasco, Northern Chile (20.274° S, 68.883° W, 3800 m above sea level). The comparison between the observed and predicted values of k t provides a determination coefficient of r 2 = 0.48 and a p value < 0.01. The results show that the value of k t for wind-induced flow is proportional to the wind shear velocity and the inverse of the Reynolds number of the wind-induced current.

Aeolian dynamics of beach scraped ridge and dyke structures

Where urban areas are situated close to a beach, sand dunes act as protection from flooding and erosion. When a dune has been removed or damaged by erosion, dune, ridge or dyke re-building using heavy machinery, a process known as beach scraping, is a common method of restoration. Following construction, natural accretion of sediment on the backshore is preferable as it facilitates sustained natural dune building, growth of vegetation, and habitat creation and reduces the need for further beach scraping.This study investigates the near surface flow and transport potential for three artificial structure designs: a single ridge, a double ridge and a dyke. The three shapes contained an identical volume of sand and were preceded by 50m of beach at an angle of 3°. A computational fluid dynamic model (CFD) was created for each scenario to calculate wind flow and shear velocity from 4 different wind directions at 22.5° intervals from 0° (onshore) to 67.5°. From this data sediment flux was predicted along a two dimensional transect for each of the scenarios.For all structures, shear velocity on the beach and stoss slope decreased as incident wind direction became more oblique; conversely shear velocity in the lee of the crest increased. A reduction in shear velocity at the foot of each structure also occurred and appears related to stoss slope, with the greatest reduction at the toe of the dyke structure (stoss slope 34°) and the least before the single ridge (stoss slope 17°).Specifically the results suggest that the double ridge structure is the most resilient to aeolian erosion. Shear velocity reduction on the back beach is comparable to the dyke and sediment flux from the stoss slope of the double ridge structure may become trapped in the swale between the two ridges encouraging sediment deposition, thus reducing sediment transport beyond the dunes and backshore. Although the dyke structure underwent the greatest reduction in shear velocity on the back beach it experienced substantial sediment flux at the crest and along the top of the structure, making it susceptible to erosion during a strong wind event. The highest sediment transport rate was calculated at the crest of the single ridge, and the single ridge structure also created the smallest reduction of shear velocity on the back beach, thus making it less desirable than the double ridge.

Rainfall mechanism over the rain-shadow region of north peninsular India

Rainfall mechanism over the rain-shadow region of north peninsular India during the summer monsoon season has been investigated using dynamic, thermodynamic, cloud microphysics and cloud dynamic (CMCD) forcings. Daily rainfall data has been used to understand rainfall variability. Daily ECMWF wind data for the period 2009–2011 have been used to study the wind divergence, shear and vertical velocity profiles. Daily thermodynamic parameters from upper air soundings of Hyderabad (17.448°N, 78.381°E), have been examined. Aircraft data have been used to study CMCD parameters. The divergence is found between surfaces to 850 hPa level whereas the convergence is at 850 hPa, which comes down to surface level during presence of low pressure systems. The divergence is observed at low (700 hPa), mid (600–300 hPa) and upper (250–150 hPa) tropospheric levels. In consequence of these divergence structures, three types of cloud systems viz. shallow, congestus and deep are developed with bases just above 850 hPa and tops at corresponding three divergent levels. The vertical profiles of relative humidity observed by radiosonde data have been analyzed to get the frequency distribution of shallow, congestus and deep clouds. The highest frequency observed is that of congestus clouds. The thermodynamic structure shows dry surface level and warm and moist middle troposphere with tongues of dry air and multilevel inversions which have been attributed to advection of aerosol-rich dry air. The aircraft observations showed high aerosol concentrations from surface to 5 km, polluted clouds with cloud droplet effective radius smaller than that required for collision–coalescence process. The factors which are responsible for causing low rainfall over the rain-shadow area have been identified.

Out-of-phase relationship between tropical cyclones generated locally in the South China Sea and non-locally from the Northwest Pacific Ocean

Tropical cyclones (TCs) in the South China Sea (SCS) are either generated locally or formed non-locally in the Northwest Pacific Ocean (NWP) and entered into the SCS. Here, it is found that local TCs vary out-of-phase with non-local TCs in summer. That is, if fewer (more) TCs enter into the SCS from the NWP, more (fewer) TCs are generated over the SCS. Further analyses show that variability of the western Pacific subtropical high (WPSH) is responsible for the out-of-phase relationship. As the WPSH shifts eastward (westward), fewer (more) non-local TCs formed in the NWP can enter into the SCS because TCs are recurved northward (moved westward). Due to the eastward (westward) movement of the WPSH, positive (negative) low-level vorticity anomalies, weak (strong) vertical wind shear between the upper and lower troposphere, and anomalous upward (downward) motion in the middle troposphere are induced in the northern SCS. The changes in the relative vorticity, vertical wind shear and vertical velocity are favorable (unfavorable) for the local TC genesis in the SCS. These variations of non-local and local TCs result in an out-of-phase relationship between TCs formed locally in the SCS and non-locally in the NWP.

Ensemble-Based Error and Predictability Metrics Associated with Tropical Cyclogenesis. Part I: Basinwide Perspective

Abstract Several metrics are employed to evaluate predictive skill and attempt to quantify predictability using the ECMWF Ensemble Prediction System during the 2010 Atlantic hurricane season, with an emphasis on large-scale variables relevant to tropical cyclogenesis. These metrics include the following: 1) growth and saturation of error, 2) errors versus climatology, 3) predicted forecast error standard deviation, and 4) predictive power. Overall, variables that are more directly related to large-scale, slowly varying phenomena are found to be much more predictable than variables that are inherently related to small-scale convective processes, regardless of the metric. For example, 850–200-hPa wind shear and 200-hPa velocity potential are found to be predictable beyond one week, while 200-hPa divergence and 850-hPa relative vorticity are only predictable to about one day. Similarly, area-averaged quantities such as circulation are much more predictable than nonaveraged quantities such as vorticity. Significant day-to-day and month-to-month variability of predictability for a given metric also exists, likely due to the flow regime. For wind shear, more amplified flow regimes are associated with lower predictive power (and thereby lower predictability) than less amplified regimes. Relative humidity is found to be less predictable in the early and late season when there exists greater uncertainty of the timing and location of dry air. Last, the ensemble demonstrates the potential to predict error standard deviation of variables averaged in 10° × 10° boxes, in that forecasts with greater ensemble standard deviation are on average associated with greater mean error. However, the ensemble tends to be underdispersive.

Prediction of sea water levels using wind information and soft computing techniques

Large variations of sea water levels are a matter of concern for the offshore and coastal locations having shallow water depths. Safety of maritime activities, and properties, as well as human lives at such locations can be ensured by using the accurately predicted water levels. Harmonic analysis is traditionally employed for tide predictions, but often the values of predicted tides and observed (measured) water levels are not identical. The difference between them is called sea level anomaly. This can be attributed to non-inclusion of meteorological parameters as an input for tide prediction. Therefore other prediction techniques become necessary. The earlier studies on sea level predictions indicate better efficiency of alternate techniques such as Artificial Neural Network (ANN) and Genetic Programming (GP), and that most researchers have used sea level time series as model inputs. Present work predicts sea levels indirectly by predicting sea level anomalies (SLAs) using hourly local wind shear velocity components of the present time and up to the previous 12h as inputs at four stations near the USA coastline with the techniques of GP and ANN. The error measures and graphs indicate that predictions are satisfactory.

The effect of roughness elements on wind erosion: The importance of surface shear stress distribution

AbstractRepresentation of surface roughness effects on aeolian sediment transport is a key source of uncertainty in wind erosion models. Drag partitioning schemes are used to account for roughness by scaling the soil entrainment threshold by the ratio of shear stress on roughness elements to that on the vegetated land surface. This approach does not explicitly account for the effects of roughness configuration, which may be important for sediment flux. Here we investigate the significance of roughness configuration for aeolian sediment transport, the ability of drag partitioning approaches to represent roughness configuration effects, and the implications for model accuracy. We use wind tunnel measurements of surface shear stress distributions to calculate sediment flux for a suite of roughness configurations, roughness densities, and wind velocities. Roughness configuration has a significant effect on sediment flux, influencing estimates by more than 1 order of magnitude. Measured and modeled drag partitioning approaches overestimate the predicted flux by 2 to 3 orders of magnitude. The drag partition is sensitive to roughness configuration, but current models cannot effectively represent this sensitivity. The effectiveness of drag partitioning approaches is also affected by estimates of the aerodynamic roughness height used to calculate wind shear velocity. Unless the roughness height is consistent with the drag partition, resulting fluxes can show physically implausible patterns. These results should make us question current assessments of the magnitude of vegetated dryland dust emissions. Representing roughness effects on surface shear stress distributions will reduce uncertainty in quantifying wind erosion, enabling better assessment of its impacts and management solutions.

Numerical modeling of the wind flow over a transverse dune

Transverse dunes, which form under unidirectional winds and have fixed profile in the direction perpendicular to the wind, occur on all celestial objects of our solar system where dunes have been detected. Here we perform a numerical study of the average turbulent wind flow over a transverse dune by means of computational fluid dynamics simulations. We find that the length of the zone of recirculating flow at the dune lee — the separation bubble — displays a surprisingly strong dependence on the wind shear velocity, u*: it is nearly independent of u* for shear velocities within the range between 0.2 m/s and 0.8 m/s but increases linearly with u* for larger shear velocities. Our calculations show that transport in the direction opposite to dune migration within the separation bubble can be sustained if u* is larger than approximately 0.39 m/s, whereas a larger value of u* (about 0.49 m/s) is required to initiate this reverse transport.

Mapping and characterization of small-scale aeolian structures on Mars: An example from the MSL landing site in Gale Crater

A new set of methodologies, which allow a simple and fast mapping and characterization of small-scale aeolian structures on Mars is introduced in this work. We follow an object-based approach in which the bedform crestlines are automatically mapped and characterized.From the methodology validation, we conclude that the quality of the obtained results is comparable with human-produced photointerpretations. We show that the accuracy associated with the measurement of mean trends from the automatically mapped patterns is less than 10°.Through the analysis of two areas located near the MSL landing site in Gale Crater, we explore some of the possibilities that the automatic mapping technique enables. Namely, for multitemporal surveys and ripple pattern analysis.We demonstrate how the mapped ripple patterns can be used to assess local wind orientations, and we analyze some examples that illustrate the diversity of wavelength spatial distributions that can be found on Mars. We try to relate these pattern wavelength variations with the possible local influence of granulometry and wind shear velocity.

The apparent roughness of a sand surface blown by wind from an analytical model of saltation

We present an analytical model of aeolian sand transport. The model quantifies the momentum transfer from the wind to the transported sand by providing expressions for the thickness of the saltation layer and the apparent surface roughness. These expressions are derived from basic physical principles and a small number of assumptions. The model further predicts the sand transport rate (mass flux) and the impact threshold (the smallest value of the wind shear velocity at which saltation can be sustained). We show that, in contrast to previous studies, the present model's predictions are in very good agreement with a range of experiments, as well as with numerical simulations of aeolian saltation. Because of its physical basis, we anticipate that our model will find application in studies of aeolian sand transport on both Earth and Mars.

Fetch-limited growth of wind waves

[1] The wave growth mechanisms proposed by Miles (1957) and Phillips (1957) more than 50 years ago, in combination with the concept of the “limiting spectrum,” are used to obtain mathematical solutions to the problem of fetch-limited wavefields. Explicit expressions for (1) the spectra, (2) the significant wave height, and (3) the peak frequency in terms of the wind shear velocity and the fetch are given.

The influence of coastal morphology on wind dynamics

An analysis of the dynamics of wind velocity along the Baltic coast of Lithuania is presented, based on data collected during field experiments in the summer, fall and winter of 1999-2001 and 2007-2009 at several sites (Būtingė, Saipiai, Smiltynė, Juodkrantė, Pervalka and Nida). The locations were chosen in order to encompass a wide spectrum of beach and dune ridge morphology. The relationship between wind velocity dynamics and coastal morphology was established, based on measurements of the slope angle, height and shape of the dune crest, as well as measurements of the morphology of the area behind the foredune ridge. On the basis of a comparison of near-surface wind velocity patterns, shear velocity (U*) and surface roughness length (z0) were calculated. It was determined that U* decreases from the middle of the beach towards the foredune toe, then increases towards the crest of the foredune and decreases down the lee slope. A direct correlation exists between U* and the stoss slope inclination, and the relative height of the foredune. Surface roughness length also increases from the beach towards the foredune crest.

Sand transport by wind on complex surfaces: Field studies in the McMurdo Dry Valleys, Antarctica

Field experiments on boundary layer winds and sand transport in the Victoria Valley, Antarctica have provided information on the spatial and temporal variability of sand transport across coarse sand surfaces covered with isolated roughness elements (rock clusters). Five sand transport events, with a duration ranging between 9.6 and 22 h, were studied. Sand transport threshold velocity was determined using the time faction equivalence method and varied between 0.30 and 0.35 m s−1. Mean rates of sand transport ranged from 0.008 to 0.072 g cm s−1. Sand transport was variably intermittent, with continuous saltation (intermittency (γ) = 1) occurring from 11% to as much as 31% of the time. For all events, there is a close temporal correspondence between intermittency values, shear velocity (u*, m s−1) and the ratio u*/u*t (u*t is threshold shear velocity), sand flux, and sand transport intensity as measured by piezoelectric saltation sensors. Although the roughness density of this site is low, partitioning of shear stress between the rock clusters and the intervening sand surface is important to the spatial pattern of sand transport and acts to increase the mean threshold wind shear velocity or shear stress for sediment transport by approximately 1.2 times compared to a bare sand surface. The effective local threshold wind shear velocity, however, varies by up to 6 times from the smoother sand areas to the most protected areas of the rock clusters. As a result, continuous and widespread sand transport will only occur at this site when the overall wind shear velocity exceeds 0.58 m s−1 (1.6 × u*t minimum).

Measurements of the aeolian sand transport saturation length

The wavelength at which a dune pattern emerges from a flat sand bed is controlled by the sediment transport saturation length, which is the length needed for the sand flux to adapt to a change in wind strength. The influence of the wind shear velocity on this saturation length and on the subsequent dune initial wavelength has remained controversial. In this paper, we present direct measurements of the saturation length performed in a wind tunnel experiment. Complementary to this, initial dune wavelengths are measured under different wind conditions — in particular after storms. Using the linear stability analysis of dune formation, it is then possible to deduce the saturation length from field data. Both direct and indirect measurements agree that the saturation length is almost independent of the wind strength. This result supports the idea that grain inertia is the dominant dynamical mechanism limiting sediment transport saturation on dunes.