Mean Concentration Field Research Articles

On the errors and their mitigation in planar laser induced fluorescence

In planar laser induced fluorescence (PLIF), scalar concentration fields are obtained from fluorescence images of the flow where the scalar is ‘tagged’ with a suitable fluorescent dye and illuminated with a laser sheet. While the fluorescence occurs, the intensity of the laser sheet is attenuated along its passage through the dye field. This paper discusses the effect of correcting for the attenuation of the laser by the dye while obtaining the scalar concentration field from the fluorescent images of the flow. If the attenuation correction cannot be done, the paper discusses three possible calibration scenarios to minimise the error. It is shown that the error in ignoring the attenuation due to turbulent concentration fluctuations cannot be corrected by any means. However, the remaining error can be exactly corrected if it were possible to calibrate with the mean concentration field. It is shown that a uniform concentration field, centred around the mean field, greatly reduces the error when compared to having no attenuation correction but still is significantly more when compared to the use of the average field. Relying on these results, we propose an iterative technique which ideally achieves the accuracy obtainable when calibrating with the mean concentration field, but which requires no reference cell. In addition, the paper explores the removal striations, of a common artefact during PLIF image acquisition with origins in the image optics. A technique to remove the same is proposed and its advantages discussed when compared to available techniques.

Can We Accurately Quantify a Lateral Diffusivity from a Single Tracer Release?

Abstract Mixing along sloping isopycnals plays a key role in the transport and uptake of heat and carbon by the ocean. This mixing is quantified by a lateral diffusivity, which can be measured by tracking the lateral spreading of point release tracer patches. We present a definition for the area of a tracer patch, the time derivative of which provides the lateral diffusivity. To accurately estimate the diffusivity, an ensemble mean concentration field of many tracer release experiments is required. We use numerical experiments to quantify how accurately the “true” lateral diffusivity (obtained from the ensemble mean concentration field) can be estimated from a single tracer release experiment (one ensemble member). To simulate observational campaigns, we also estimate the diffusivity from a single tracer release that is spatially and/or temporally subsampled, quantifying how the error between the estimated diffusivity and the true diffusivity grows as this sampling resolution worsens. We perform these numerical experiments in a two-layer quasigeostrophic model of turbulent flow on a β plane, using an ensemble of 50 passive tracer release experiments, each initialized as a 2D Gaussian but with differing realizations of the turbulent flow. We find that the diffusivity estimates from the single tracer releases have a relative root-mean-square error (RMSE) of 1.43% from the true diffusivity. Subsampling a single tracer release experiment every 956 km increases the relative RMSE from the true diffusivity to 3.1%; also subsampling every 277 days raises this figure to 6.5%.

Study of turbulent dispersion of a concentration plume emitting from a line source over a rib‐roughened surface

Turbulent dispersion of a concentration plume emitting from a horizontal line source over a rib-roughened surface has been studied using direct numerical simulations (DNS). Four test cases have been compared to investigate the effects of the source elevation and location on the plume dispersion in a turbulent boundary layer. The transport process of the concentration is investigated in both physical and spectral spaces which includes analyses of statistical moments of the concentration field, decay rate, plume width, pre-multiplied spectra of the velocity and concentration fields, and the probability density function (PDF) of concentration fluctuations. It is observed that as the mean plume development enters the long-range dispersion stage, the decay rate of the mean concentration field begins to feature a constant slope of −3/2, while the vertical spread of the mean plume exhibits a constant slope of 1/3 in all four line source release cases in a ribbed wall flow. It is also observed that the profiles of both the mean concentration field and the RMS of concentration fluctuations exhibit a self-similarity pattern downstream of the line source in all four test cases. By comparing the characteristic length scales of the spanwise velocity and scalar fields, two distinct stages of the instantaneous plume development are observed, dominated by the turbulent convective and diffusive mechanisms. It is discovered that the transition from the turbulent convective stage to the turbulent diffusive stage occurs more rapidly for line sources placed near the wall and inside the cavities.

Entrainment and dilution in a fountain top

Experiments are carried out on a turbulent fountain to investigate the entrainment of ambient fluid and the dilution of scalar concentration in the fountain top (also referred to as the ‘cap’). The source Froude number ( $Fr_o$ ), defined based on the density difference between the source and the ambient fluids, is varied between 10 and 30, while the Reynolds number ( $Re_o$ ) is set to a minimum of 3000 to ensure a fully turbulent flow at the source. High-fidelity velocity and concentration measurements are obtained using particle image velocimetry and planar laser induced fluorescence techniques, respectively. The mean concentration field indicates that the cap is approximately hemispherical and its base is characterised by the local Froude number $Fr_z \approx 1.5$ . It is observed that the ratio of the entrained ( $Q_{top}$ ) volume flux in the fountain top and the volume flux supplied ( $Q_{in}$ ) at the base of the cap varies between 1.5 and 3.5 at different $Fr_o$ , exceeding the values of $Q_{top}/Q_{in} (= 0.5\text {--}0.8)$ for a fountain developing across a density interface (Lin & Linden, J. Fluid Mech., vol. 542, 2005, pp. 25–52). The present experimental results for $Q_{top}/Q_{in}$ are in excellent agreement with published numerical simulations of turbulent fountains at similar $Fr_o$ . Lastly, a robust metric to estimate the dilution of scalar in the fountain top has been proposed. The results clearly indicate that dilution is not equal to the entrainment ratio; however, the self-similarity of non-dimensional dilution profiles at different $Fr_o$ suggests that the phenomenological model for entrainment in the fountain top put forth by Debugne & Hunt (J. Fluid Mech., vol. 796, 2016, pp. 195–210), can also be effectively used to model the dilution of scalar concentration in the cap. Further, it is found that the variation of the dilution ratio is closely linked to the local Reynolds number ( $Re_z$ ) at the base of the cap. This is verified using an analytical model that describes the dependency of $Re_z$ on the local and source parameters – $Fr_z$ , $Re_o$ and $Fr_o$ .

Growth laws and self-similarity in the confined mixing zone of unstratified and strongly stably stratified isokinetic mixing-layers past a splitter plate

The velocity U and the mixing scalar field Θ of turbulent wake flows developing downstream of a slender splitter plate were studied experimentally for unstratified and strongly stably stratified conditions in a square channel with water as fluid. The experimental program comprises five different Reynolds numbers spanning Re0=10 000–60 000 for u0=u1=u2 (isokinetic conditions) between the two planar, initially separated streams and five relative density stratifications from Δρ=0 to 10%, since mixing studies for these stratification strengths are rather limited. This full-channel Reynolds number uses the hydraulic diameter dh=50 mm of the mixing section as the characteristic length scale. These experiments challenge the corresponding numerical calculations based on Reynolds-averaged Navier–Stokes approaches relying on the Boussinesq approximations of the first and second kind. The development of the mixing scalar field Θ was measured by two fundamentally different approaches: laser induced fluorescence (LIF) in the downstream x–y mid-plane and wire-mesh sensors (WMSs) in discrete lateral y–z planes. This allows for a direct comparison of both sensing techniques at locations where both planes intersect. For the unstratified conditions, it is shown that the downstream developing wake velocity deficit decay agrees well with the theoretical approach. The concentration fields based on LIF and WMS profiles of the mean and fluctuating (root mean square, RMS) data confirm self-similarity by introducing a mixing scalar Θ for both techniques. In the approximate limit, it is shown that the mean concentration field can be well described by an error function and the corresponding RMS data by a Gaussian profile in self-similar coordinates. By using the momentum thickness θm at the splitter plate as the normalization parameter, the downstream developing width of the mixing zone δ(x) becomes independent for Re0≥30 000. The mean and the RMS values of the concentration field Θ for the stably stratified experiments show self-similarity in the core of the concentration field but also a departure from the error function and Gaussian profile in the outer parts. Introducing a local gradient Richardson number Rig, it is shown that both the growth and growth-suppression of the mixing zone are well described by the Miles–Howard criterion for all the stratified experiments considered, even though this criterion was originally developed for shear layers.

Performance evaluation of approaches to predict sub-hourly peak odour concentrations

Abstract Atmospheric dispersion models are widely used to predict the potential for annoyance of odour emissions. It is well established that the predictive accuracy of these models is higher for the mean concentration. In reality, the concentration of a substance advected by a turbulent boundary layer flow often shows large fluctuations around its mean value. In the case of odorous gases, the mean concentration field only may thus hide information on the real impact on receptors, since such a process exhibits nonlinearity. Short-term odour concentrations have to be accounted for, thereby requiring additional knowledge on concentration fluctuations. Here, the main goal is to evaluate three selected approaches to predict sub-hourly odour peak concentrations. These approaches are the constant factor of 4, an empirical-based peak-to-mean procedure and the concentration-variance computation. A full-scale field dispersion experiment (Uttenweiler), for which concentration fluctuation measurements have been conducted, was designated for the investigation. In this respect, additional objectives of this work are to investigate the mean flow conditions encountered in the different dispersion trials, to examine the measured tracer fast-response concentration time series, and to characterise the performance of a Lagrangian particle dispersion model (LASAT) for the mean concentration. Several statistical indicators are used to compare predictions against observations. First, there is evidence, gained from evaluating the ultrasonic anemometer data, to show that not only neutral but also stable atmospheric conditions occurred during the trials. Second, the dispersion model showed an overall satisfactory performance under the given range of study conditions. A general bias towards underestimation was detected, with the dispersion model performing better for distances further away from the emission source. Third, for the concentration fluctuations, results are presented and discussed in terms of the fluctuation intensity, intermittency, peak-to-mean factors and probability density functions. Finally, regarding the three evaluated approaches, it is found that as the inherent complexity of the approaches grows, more accurate predictions of peak-to-mean factors Ψ90 are obtained. Namely, the concentration-variance computation approach performed best, with bias towards overestimating Ψ90. While the constant factor of 4 overestimated all Ψ90 observations, the empirical-based peak-to-mean procedure underestimated Ψ90 to a great extent. The results also confirm and reveal the advantages and shortcomings of each evaluated approach. The findings of this work have potential implications for future research and policymaking in this topic area.

Regional September Sea Ice Forecasting with Complex Networks and Gaussian Processes

Abstract Reliable predictions of the Arctic sea ice cover are becoming of paramount importance for Arctic communities and industry stakeholders. In this study pan-Arctic and regional September mean sea ice extents are forecast with lead times of up to 3 months using a complex network statistical approach. This method exploits relationships within climate time series data by constructing regions of spatiotemporal homogeneity (i.e., nodes), and subsequently deriving teleconnection links between them. Here the nodes and links of the networks are generated from monthly mean sea ice concentration fields in June, July, and August; hence, individual networks are constructed for each respective month. Network information is then utilized within a linear Gaussian process regression forecast model, a Bayesian inference technique, in order to generate predictions of sea ice extent. Pan-Arctic forecasts capture a significant amount of the variability in the satellite observations of September sea ice extent, with detrended predictive skills of 0.53, 0.62, and 0.81 at 3-, 2-, and 1-month lead times, respectively. Regional forecasts are also performed for nine Arctic regions. On average, the highest predictive skill is achieved in the Canadian Archipelago, Beaufort, Chukchi, East Siberian, Laptev, and Kara Seas, although the ability to accurately predict many of these regions appears to be changing over time.

A multi-city air pollution population exposure study: Combined use of chemical-transport and random-Forest models with dynamic population data

Cities are severely affected by air pollution. Local emissions and urban structures can produce large spatial heterogeneities. We aim to improve the estimation of NO2, O3, PM2.5 and PM10 concentrations in 6 Italian metropolitan areas, using chemical-transport and machine learning models, and to assess the effect on population exposure by using information on urban population mobility. Three years (2013–2015) of simulations were performed by the Chemical-Transport Model (CTM) FARM, at 1 km resolution, fed by boundary conditions provided by national-scale simulations, local emission inventories and meteorological fields. A downscaling of daily air pollutants at higher resolution (200 m) was then carried out by means of a machine learning Random-Forest (RF) model, considering CTM and spatial-temporal predictors, such as population, land-use, surface greenness and vehicular traffic, as input. RF achieved mean cross-validation (CV) R2 of 0.59, 0.72, 0.76 and 0.75 for NO2, PM10, PM2.5 and O3, respectively, improving results from CTM alone. Mean concentration fields exhibited clear geographical gradients caused by climate conditions, local emission sources and photochemical processes. Time series of population weighted exposure (PWE) were estimated for two months of the year 2015 and for five cities, by combining population mobility data (derived from mobile phone traffic volumes data), and concentration levels from the RF model. PWE_RF metric better approximated the observed concentrations compared with the predictions from either CTM alone or CTM and RF combined, especially for pollutants exhibiting strong spatial gradients, such as NO2. 50% of the population was estimated to be exposed to NO2 concentrations between 12 and 38 μg/m3 and PM10 between 20 and 35 μg/m3. This work supports the potential of machine learning methods in predicting air pollutant levels in urban areas at high spatial and temporal resolutions.

Low-cost sensors and microscale land use regression: Data fusion to resolve air quality variations with high spatial and temporal resolution

AbstractThe strong temporal and spatial gradients in NO2 concentrations frequently observed in urban microenvironments are very difficult to measure and model accurately. Recent developments in low-cost air quality instruments have led to improvement in the spatial coverage of time-resolved measurement, however interpolation is still needed to map pollutant concentrations and connect time-as well as space-dependent variations to urban design features. Here we propose a novel approach that uses a previously-described microscale land use regression (LUR) model to spatially interpolate data from a well-calibrated network of low-cost air quality instruments. We use a semiconducting oxide-based ozone sensor to provide a robust correction of the output of an electrochemical NO2 sensor for ozone interference. We characterise signal noise probably associated with meniscus fluctuations as a significant error source, that can be handled with appropriate signal averaging. The LUR model is used to provide high spatial resolution in the data set, whilst correlation with sensor measurements provides a time-dependent estimate associated with different land use types. Observations from the network of instruments showed marked variability in NO2 concentrations over short distances (on the scale of 100 m), with highest concentrations reached near bus stops, intersections and under shop awnings. This approach connects the complex time- and space-dependent variations to urban design features and is a promising way forward as a basis for objective spatial mapping of time-dependent mean concentration fields and local population exposure estimates.

Mathematical Modeling of the Mean Concentration Field in Random Stratified Structures with Regard for the Jumps of Sought Function on the Interfaces

We study the diffusion processes of an admixture in a two-phase stratified strip of randomly inhomogeneous structure with regard for the jumps of the concentration function and its derivative on the contact boundaries of the phases. A new representation of the operator of equation of mass transfer for the entire body is proposed. We formulate an equivalent integrodifferential equation whose solution is constructed in the form of a Neumann integral series. The obtained solution is averaged over the ensemble of phase configurations with uniform distribution function. It is shown that the computational formula for the mean concentration with explicit account of its jumps on the interfaces contains an additional term. It is demonstrated that the ratios of the diffusion coefficients, the concentration dependences of the chemical potentials in different phases, and their relationships affect the sign of this term. We find the ranges of parameters of the problem for which this term is negligibly small.

Using Low-Cost Air Quality Sensor Networks to Improve the Spatial and Temporal Resolution of Concentration Maps.

We present an approach to analyzing fine particulate matter (PM2.5) data from a network of “low cost air quality monitors” (LCAQM) to obtain a finely resolved concentration map. In the approach, based on a dispersion model, we first identify the probable locations of the sources, and then estimate the magnitudes of the emissions from these sources by fitting model estimates of concentrations to corresponding measurements. The emissions are then used to estimate concentrations on a grid covering the domain of interest. The residuals between model estimates at the monitor locations and the measured concentrations are then interpolated to the grid points using Kriging. We illustrate this approach by applying it to a network of 20 LCAQMs located in the Imperial Valley of Southern California. Estimating the underlying mean concentration field with a dispersion model provides a more realistic estimate of the spatial distribution of PM2.5 concentrations than that from the Kriging observations directly.

Large eddy simulation of pollutant dispersion in a naturally cross-ventilated model building: Comparison between sub-grid scale models

In the present article, the flow field and passive gaseous pollutant dispersion in a naturally cross-ventilated isolated single-zone model building have been investigated. The large eddy simulation (LES) approach has been applied together with three different sub-grid scale (SGS) models, namely, wall-adapting local eddy-viscosity (WALE), dynamic Smagorinsky-Lilly model (DSLM), and standard Smagorinsky-Lilly model (SSLM). The flow and passive scalar concentration fields have been compared with available experimental data. It is demonstrated that the three SGS models predict similar flow field. Also, it can be observed that the mean concentration field results obtained from the WALE sub-grid scale (SGS) model are in better agreement with the experimental data than those of DSLM and SSLM by about 8% and 5%, respectively. Moreover, the CPU time required for the computations by the WALE and SSLM models was almost 20% less than that of DSLM, making WALE a more suitable SGS model than the other two models for the prediction of flow and concentration fields in an isolated building. Furthermore, contributions of convective flux and turbulent diffusion flux to the pollutant transportation have been studied. It has been shown that although the convective flux is the main mechanism, the two fluxes have a significant influence on pollutant transportation and distribution.

An evaluation of a Lagrangian stochastic model for the assessment of odours

In this work, we propose and evaluate a one-particle Lagrangian stochastic model for the assessment of the concentration-fluctuation variance, which is often used to predict odour hours for example. The main challenge is to provide a suitable parameterisation for the concentration-variance dissipation time scale. While the particle mass is conserved along the particle trajectory, the concentration variance dissipates. The concentration variance in a Lagrangian particle model evolves according to the same Langevin equation used for the simulation of the turbulent dispersion resulting in a mean concentration field. The variance dissipation is taken into account via a decay term with a finite time scale. We herein discuss and compare different parameterisations of the variance-dissipation time scale against observations from a field experiment. It was found that best results are obtained prescribing the ratio between the concentration-variance dissipation time scale and the Lagrangian time scale, which depends on the travel time.

Prediction of flow and dispersion in cross–ventilated buildings

The present paper presents a detailed computational analysis of flow and dispersion in a generic isolated single–zone buildings. First, a grid generation strategy is discussed, that is inspired by a previous computational analysis and a grid independence study. Different turbulence models are appliedincluding two-equation turbulence models, the differential Reynolds Stress Model, Detached Eddy Simulation and Zonal Large Eddy Simulation. The mean velocity and concentration fields are calculated and compared with the measurements. A satisfactory agreement with the experiments is not observed by any of the modelling approaches, indicating the highly demanding flow and turbulence structure of the problem.

Microfluidic ballistic regime for the generation of linear gradients inside a capillary column: Proof-of-concept and application to the miniaturized acid-base volumetric titration

This work details a simple and original approach for the generation of linear gradients inside straight cylindrical microchannels such as a capillary column. The concept takes advantage of an oft-overlooked regime of dispersion of flowing liquids inside narrow channels: the ballistic regime. The ballistic regime is a pure convective regime and is produced by imposing a high velocity flow in a pre-filled capillary thus limited diffusion takes place. This is obtained by forcing the injection of a plug of solution on a short time scale t, much shorter than t<110×D/r2, D is the diffusion coefficient and r the capillary radius. The result is a stretched solution of a given length or depth of penetration, inside the capillary column. This leads to a linear mean concentration field through the mixing zone forming a linear gradient. In miniaturized systems, this transient regime is followed by mainly radial diffusion of the solution inside the capillary due to the short characteristic diffusion time of narrow channels. A convection-diffusion simulation was used to model the gradient formed under this ballistic regime. A specific experimental prototype set-up was designed to investigate this ballistic regime and the formation of a linear gradient of titrant NaOH solution inside a capillary tubing of 500 µm inner diameter and 35-cm total length pre-filled with nitric acid solution. With this prototype, the linear gradient was then pushed in a non-ballistic regime over a confocal fluorescence point detection system in order to measure the fluorescence emission of a fluorescent dye added to the solutions. Considering strong acid-base reaction, fluorescein was used due to its strong fluorescence dependency with pH near its pKa, i.e 6.4. A first set of experiments was realized to demonstrate the validity of such an approach and to determine the optimal condition for the formation of a linear gradient over 300 mm of the 350-mm capillary length. An injection pressure of 1000-mbars over 0.75 s was chosen. The first result was the stability of the system in its ability to produce reproducible linear gradients. As further proofs of feasibility, samples of different nitric acid concentrations were titrated with a 0.25 M NaOH solution. The result was rapid and reproducible titration curves obtained with a fully automated system that consumes less than approximately 70 µL of sample solution.

Effects of a Fence on Pollutant Dispersion in a Boundary Layer Exposed to a Rural-to-Urban Transition

Simultaneous particle-image velocimetry and laser-induced fluorescence combined with large-eddy simulations are used to investigate the flow and pollutant dispersion behaviour in a rural-to-urban roughness transition. The urban roughness is characterized by an array of cubical obstacles in an aligned arrangement. A plane fence is added one obstacle height h upstream of the urban roughness elements, with three different fence heights considered. A smooth-wall turbulent boundary layer with a depth of 10h is used as the approaching flow, and a passive tracer is released from a uniform line source 1h upstream of the fence. A shear layer is formed at the top of the fence, which increases in strength for the higher fence cases, resulting in a deeper internal boundary layer (IBL). It is found that the mean flow for the rural-to-urban transition can be described by means of a mixing-length model provided that the transitional effects are accounted for. The mixing-length formulation for sparse urban canopies, as found in the literature, is extended to take into account the blockage effect in dense canopies. Additionally, the average mean concentration field is found to scale with the IBL depth and the bulk velocity in the IBL.

On the Effect of Regular Waves on Inclined Negatively Buoyant Jets

The target of this paper is to measure the modifications that regular waves induce on the geometrical features and dilution of inclined negatively buoyant jets. In order to achieve this aim, we have carried out a set of experiments in a wavemaker-equipped flume, by measuring the concentration fields via light-induced fluorescence, a non-intrusive and full-field image analysis technique. The wave and jet parameters were selected in order to simulate the case of a typical discharge of brine, from a desalination plant, into the Mediterranean Sea, and compare it to a reference case, i.e., the same jet discharging into a stagnant water body. The mean concentration fields were obtained, as well as the geometrical features and dilution of the jets. The three main effects of waves on inclined negatively buoyant jets are the bifurcation (i.e., the separation in two branches), the rotation of the point of maximum height and the oscillation of the impact point around a fixed position different from the stationary one, and the reduction in size of the sea region interested by the discharge; this last effect increases with the wave period. As a consequence, under waves with high period and amplitude, the dilution of inclined negatively buoyant jets tends to decrease.

Impact of roof height non-uniformity on pollutant transport between a street canyon and intersections

This paper presents an extension of our previous wind-tunnel study (Nosek et al., 2016) in which we highlighted the need for investigation of the removal mechanisms of traffic pollution from all openings of a 3D street canyon. The extension represents the pollution flux (turbulent and advective) measurements at the lateral openings of three different 3D street canyons for the winds perpendicular and oblique to the along-canyon axis. The pollution was simulated by emitting a passive gas (ethane) from a homogeneous ground-level line source positioned along the centreline of the investigated street canyons. The street canyons were formed by courtyard-type buildings of two different regular urban-array models. The first model has a uniform building roof height, while the second model has a non-uniform roof height along each building's wall. The mean flow and concentration fields at the canyons' lateral openings confirm the findings of other studies that the buildings' roof-height variability at the intersections plays an important role in the dispersion of the traffic pollutants within the canyons. For the perpendicular wind, the non-uniform roof-height canyon appreciably removes or entrains the pollutant through its lateral openings, contrary to the uniform canyon, where the pollutant was removed primarily through the top. The analysis of the turbulent mass transport revealed that the coherent flow structures of the lateral momentum transport correlate with the ventilation processes at the lateral openings of all studied canyons. These flow structures coincide at the same areas and hence simultaneously transport the pollutant in opposite directions.

Comparison of atmospheric modelling systems simulating the flow, turbulence and dispersion at the microscale within obstacles

Three different modelling techniques to simulate the pollutant dispersion in the atmosphere at the microscale and in presence of obstacles are evaluated and compared. The Eulerian and Lagrangian approaches are discussed, using RAMS6.0 and MicroSpray models respectively. Both prognostic and diagnostic modelling systems are considered for the meteorology as input to the Lagrangian model, their differences and performances are investigated. An experiment from the Mock Urban Setting Test field campaign observed dataset, measured within an idealized urban roughness, is used as reference for the comparison. A case in neutral conditions was chosen among the available ones. The predicted mean flow, turbulence and concentration fields are analysed on the basis of the observed data. The performances of the different modelling approaches are compared and their specific characteristics are addressed. Given the same flow and turbulence input fields, the quality of the Lagrangian particle model is found to be overall comparable to the full-Eulerian approach. The diagnostic approach for the meteorology shows a worse agreement with observations than the prognostic approach but still providing, in a much shorter simulation time, fields that are suitable and reliable for driving the dispersion model.

Evaluation of high-order concentration statistics in a dispersing plume

In this paper we analyze the concentration higher-order moments of a dispersing plume in a neutral wind tunnel boundary layer. They are evaluated through a fluctuating plume model in which the plume barycenter is directly calculated from the measured mean concentration field. The reliability of the model is firstly verified by comparing the simulated and measured moments up to the fourth order. Then the relationships between higher and lower-order moments are investigated seeking for a proper scaling of the fourth, fifth and sixth-order normalized moments. We found that a general behavior appears when they are expressed as a function of the third-order moments. In this case each order calculated or measured at different distances collapses on the same curve.