The performance of different detrending methods in removing the low-frequencycontribution to the calculation of turbulent fluxes is investigated. The detrending methodsare applied to the calculation of turbulent fluxes of different scalars (temperature, ultrafineparticle number concentration, carbon dioxide and water vapour concentration), collectedat two different measurement sites: one urban and one suburban. We test and compare theperformance of filtering methodologies frequently used in real-time and automated procedures(mean removal, linear detrending, running mean, autoregressive filter) with the resultsobtained from a reference method, which is a spectral filter based on the Fourier decompositionof the time series. In general, the largest differences are found in the comparison betweenthe reference and the mean-removal procedures. The linear detrending and running-meanprocedures produce comparable results, and turbulent-flux estimations in better agreementwith the reference procedure than those obtained with the mean-removal procedure. Thebest agreement between the running mean and the spectral filter is achieved with a timewindow of 15min at both sites. For all the variables studied, average fluxes calculated usingthe autoregressive filter are increasingly overestimated for a time constant ? compared withthat obtained using the spectral filter. The minimization of the difference between the twodetrending methods is achieved with a time constant of 120 s, with similar behaviour observedat both sites.

In low-wind regimes (wind speed less than 1.5 m s-1) the nocturnal boundary layer is still inadequately understood. In such conditions, turbulence is weak and often intermittent, whereas dynamics and total fluxes are often driven by submeso motions. As a consequence, the momentum, mass and energy transfers are poorly represented by dispersion models. In low-wind conditions, an important fraction of submeso motions is represented by meandering modes, which can be detected through the Eulerian autocorrelation functions of horizontal wind components and temperature. Such an approach has been proven to be reliable. However, a deeper insight could be useful for understanding the phenomenon, especially when complex or multiple mesoscale motions simultaneously develop. In this work, Eulerian autocorrelation functions and the Morlet continuous wavelet transform were used to investigate an 8 h nocturnal period characterized by the coexistence of horizontal meandering and vertical oscillations, rarely observed elsewhere. This 'nice' episode represents a good case-study to compare the two used methodologies and to investigate the phenomenology of the two submeso phenomena simultaneously occurring in a low-wind-speed regime. The two methodologies identified the same time-scale for the detected meandering structures over the whole period. Moreover, the wavelet analysis: (i) was able to discriminate horizontal meandering and gravity waves, which simultaneously developed in the second part of the analysed period; (ii) showed that, in the investigated case, the horizontal meandering was not triggered by gravity waves which appeared later in the night; (iii) highlighted how both gravity waves and meandering can contribute to an increase of the vertical turbulent energy and fluxes, confirming the crucial role of submeso structures in the turbulence production during low-wind regimes in stable conditions.

Meandering is generally defined in terms of largevariation of the wind direction due to a complex mix of motions on scales between the main turbulent eddiesand the smallest mesoscale motions. There is not a general consensus on the physical causes of the motionsresponsible of the wind meandering during low-wind speed conditions. They include internal gravity waves,quasi-2D pancake motions, cold air drainage, solitons, vortices with either a horizontal axis or a vertical axis.In this work we present the analysis of wind and temperature data measured with two sonic anemometers ina low-wind stably stratified night observed during the Urban Turbulent Project (Torino, Italy). An originalapproach to estimate the meandering time scales of the wind velocity and temperature using two complementarymethodologies is proposed.In the literature the meandering time-scale is evaluated fitting the Eulerian auto-correlation functions of the windvelocity with an oscillating theoretical behaviour on hourly datasets. First we extend this method consideringthe dependence of the time-scale on the dataset length considering longer datasets (1, 2, 3, 4 hours) and then wecompare these results with a wavelet analysis. The continuous wavelet transform based on the Morlet basis isused to detect and characterize the time-scale of the wavelike oscillations both in the wind velocity and in thetemperature signals. Moreover cross-wavelet spectra are used to identify the nature of the wavy patterns in orderto discriminate the presence of gravity waves.The wavelet analysis corroborates the results obtained with the auto-correlation functions and opens newpromising perspectives for the study of the meandering phenomenon.

While occurrences of wavelike motion in the stable boundary layer due to the presence of a significant restoring buoyancy force are rarely disputed, their modalities and interaction with turbulence remain a subject of active research. In this work, the characteristics of gravity waves and their impact on flow statistics, including turbulent fluxes, are presented using data collected above an Antarctic Ice sheet during an Austral Summer. Antarctica is an ideal location for exploring the characteristics of gravity waves because of persistent conditions of strong atmospheric stability in the lower troposphere. Periods dominated by wavelike motion have been identified by analysing time series measured by fast response instrumentation. The nature and characteristic of the dominant wavy motions are investigated using Fourier cross-spectral indicators. Moreover, a multi-resolution decomposition has been applied to separate gravity waves from turbulent fluctuations in case of a sufficiently defined spectral gap. Statistics computed after removing wavy disturbances highlight the large impact of gravity waves on second order turbulent quantities including turbulent flux calculations.

In this work the performances of three different tests widely used in scientific community for assessing turbulent flow stationarity have been tested and compared. The stationarity tests have been applied to turbulent fluxes of different scalars (temperature, carbon dioxide and water vapour concentration, ultrafine particle number concentration), collected over a wide range of surface roughness conditions, from almost smooth surfaces to surfaces with different degrees of complexity: suburban, urban and forested canopies. The application of the selected tests to the scalar and particle turbulent fluxes measured above the different experimental sites has frequently shown an ambiguous identification of non-stationarity in turbulent flux records, regardless of the surface conditions and of the transported quantity. The potential impact of surface characteristics, transported scalar, atmospheric stratification and micrometeorological conditions on test performances has been investigated. Finally a new operative stationarity index obtained by the combination of two tests has been proposed to assess the second-order stationarity of turbulence time series. (C) 2014 Elsevier B.V. All rights reserved.

Expressions for the logarithmic variations of the normalized turbulent longitudinal velocity ((u(2p)) over bar (+))(1/p) with normalized distance z/delta from a boundary for high-order (p) moments in the intermediate region of wall bounded flows characterized by thickness delta are derived. The ansatz that ((u(2p)) over bar (+))(1/p) variation in ln(z/delta) originates from a compound effect of random sweeping and -1 power-law scaling in the longitudinal velocity spectrum E-u(k) is discussed, where k is the wavenumber. Using velocity time series sampled above a uniform ice sheet, an E-u(k) similar to k(-1) scaling is confirmed for k z < 1 and k delta > 1. The data were then used to analyze assumptions required for the utility of the random sweeping decorrelation (RSD) hypothesis connecting the k-1 power-law with log-scaling in ((u(2p)) over bar (+))(1/p). It has been found out that while the RSD hypothesis is operationally applicable to scales associated with attached eddies bounded by k z < 1 and kd > 1, significant interactions among high-order turbulent velocity and velocity increments lead to the conclusion that the RSD hypothesis cannot be exactly valid. Its operational utility stems from the observations that some of the interaction terms among the high-order velocity and velocity increments act in opposite directions thereby canceling their additive effects in RSD. Published by AIP Publishing.

Turbulence within the stable boundary layer (SBL) remains a ubiquitous feature of many geophysical flows, especially over glaciers and ice-sheets. Although numerous studies have investigated various aspects of the boundary layer motion during stable atmospheric conditions, a unified picture of turbulent transport within the SBL remains elusive. In a strongly stratified SBL, turbulence generation is frequently associated with interactions with sub-meso scale motions that are often a combination of gravity waves (GWs) and horizontal modes. While some progress has been made in the inclusion of GW parameterisation within global models, description and parameterisation of the turbulence-wave interaction remain an open question. The discrimination between waves and turbulence is a focal point needed to make progress as these two motions have different properties with regards to heat, moisture and pollutant transport. In fact, the occurrence of GWs can cause significant differences and ambiguities in the interpretation of turbulence statistics and fluxes if not a priori filtered from the analysis.In this work, the characteristics of GW and their impact on turbulent statistics were investigated using wind velocity components and scalars collected above an Antarctic Ice sheet during an Austral Summer. Antarctica is an ideal location for exploring the characteristics of GW because of persistent conditions of strongly stable atmospheric stability in the lower troposphere. Periods dominated by wavy motions have been identified by analysing time series measured by fast response instrumentation. The GWs nature and features have been investigated using Fourier cross-spectral indicators. The detected waves were frequently characterised by variable amplitude and period; moreover, they often produced non-stationarity and large intermittency in turbulent fluctuations that can significantly alter the estimation of turbulence statistics in general and fluxes in particular. A multi-resolution decomposition based on the Haar wavelet has been applied to separate gravity waves from turbulent fluctuations in case of a sufficiently defined spectral gap. Statistics computed after removing wavy disturbances highlight the large impact of gravity waves on second order turbulent quantities. One of the most impacted parameters is turbulent kinetic energy, in particular in the longitudinal and lateral components. The effect of wave activity on momentum and scalar fluxes is more complex because waves can produce large errors in sign and magnitude of computed turbulent fluxes or they themselves can contribute to intermittent turbulent mixing. The proposed filtering procedure based on the multi-resolution decomposition restored the correct sign in the turbulent sensible heat flux values. These findings highlight the significance of a correct evaluation of the impact of wave components when the goal is determining the turbulent transport component of mass and energy in the SBL.

Turbulence structure in the very stable boundary layer (SBL) is characterized by complex interactions between the static stability of the air and non-turbulent processes that govern the mechanical generation of turbulence. Submeso motions, on scales from meters to few kilometres, can complicate the turbulence behaviour and can modulate the turbulent fluxes through the production of intermittent mixing events related to localized flow acceleration. Submeso motions can take a variety of forms including gravity waves, density currents, drainage flows. An important fraction of them is represented by horizontal meandering modes, particularly when the large-scale flow is weak; however their characteristics and their impact on turbulent transport are yet poorly studied and understood.The aim of this work is the investigation of the characteristics of horizontal meandering motions and of their role in the turbulence productions in a very SBL, through the analysis of wind velocity components and temperature collected above an Antarctic Ice sheet during an Austral Summer. For the analysis two different methodologies have been used: the first, specifically implemented for identifying meandering motions, based on the evaluation of Eulerian auto-correlation functions (EAFs); and the second based on a Morlet continuous wavelet transform. Such combined approach enables a clear identification of periods interested by horizontal meandering motions and the determination of their characteristic time scales.The selected periods are further analysed by using a complementary wavelet approach based on the multi- resolution flux decomposition (MRD) useful for detecting the scales of motions responsible for the shear generation of turbulence and for the flux variability. The use of the MRD methodology allows us to gain more insight on understanding the mechanism of turbulence generation by submeso motions on different time-scales and their role in influencing the turbulent transport in very stable low-wind conditions.The analysis of the selected periods though the combination of the different methodologies of analysis allows the understanding of the conditions that trigger the meandering activation; finally, the comparison of data collected at three different measurement levels allows the study of the flow coupling at different heights within a very SBL.

A 140-m micrometeorological tower provides detailed observations on the vertical structure of mean and turbulent fields of meteorological variables at a coastal region in southeastern Brazil, besides revealing the extent to which a nearby power plant affects the local atmospheric boundary layer.A 140-m micrometeorological tower has been operating since August 2016 at 4 km from the coastline and 250 m from a thermal power plant that releases heat from its 20-m stacks in southeastern Brazil. The measurements include 11 level of turbulence observations, as well as 10 levels of slow-response temperature and humidity. The observed atmospheric structure is largely dependent on the wind direction with respect to the power plant. When winds blow from the plant to the tower, the air layer between 20 and 60 m of the atmosphere may be warmed by as much as 2 C. In these circumstances there are events when the emissions pass directly by the tower. They allow the analysis of turbulence structures of thermal plumes generated from the plant heat release in comparison with those generated by the surface heating. In the more common case of winds blowing from the tower to the plant, the observations allow a detailed description of the local atmospheric boundary layer. During the day, vertical profiles of turbulent quantities and of their spectral distributions show a cycle controlled by interactions between the land and oceanic surfaces, such as a thermal internal boundary layer. At night, there is a systematic tendency of progressive stabilization throughout the period, suitable for the analysis of the boundary layer transition from weakly to very stable conditions. The data also grant the inference of detailed vertical profiles of turbulent diffusion coefficients directly from observations.

The coexistence of wavelike submeso motions and anisotropic intermittent turbulence in a night time stable boundary layer is investigated. Submeso motions of different characteristics and amplitudes interact with each other. These interactions may lead to intermittent turbulence production that alters the turbulent structure of the stable boundary layer. On the other hand, the production and transfer of turbulence affect the delicate balance of submeso motions. In this work sonic anemometer data collected at eleven levels in south-eastern Brazil have been used to study a case of a nocturnal boundary layer in a coastal site. The absence of forcing at the synoptic scale allows the development of a breeze circulation on which a low level jet of moderate intensity (4 m s-1) and low height (about 50 m) takes place. The jet evolution is coupled with dirty waves, while its full development is associated with gravity waves driven by a strong vertical temperature gradient. The layer centred at the low-level jet nose is characterized by horizontal meandering and very weak turbulence intensity. The air far below and above the low-level jet maximum experiences bursts of significant increase of the turbulence intensity, showing a three-layer structure. The oscillations of the wind velocity horizontal components exhibit the same frequency of the temperature oscillations, suggesting that the presence of an adequate temperature horizontal gradient is one of the fundamental driver of the meandering phenomenon. The considered night has been studied by means of the Eulerian auto-correlation functions for the detection of the meandering hours and their oscillation time-scales and by means of the continuous Morlet wavelet function for the detection of the gravity waves and the characterization of their spatial time scales and temporal evolution.

The scaling laws of the vertical (Fwc) and longitudinal (Fuc) velocity-scalar cospectra within the inertial subrange are explored using dimensional arguments and a sim- plified cospectral budget in the canopy sublayer above three distinct forested ecosystems. The cospectral budget was shown to be consistent with plausible scaling laws originating from dimensional considerations. Using the analytical solution to the novel cospectral bud- get, it was shown that Fwc(k) and Fuc(k) are governed by the linear superposition of two terms that scale as k-2/3-? and k-?, where k is the wavenumber, -? is the exponent of the velocity spectrum, and ?(>= 7/3) depends on the ratio of the similarity constants for the pressure-scalar covariance and the flux transport terms. It was also demonstrated that, when the magnitude of the mean scalar concentration gradient is large, the k-2/3-? term dominates the velocity-scalar cospectral budget. For such a case, correcting for biases emerging from high frequency losses in eddy-covariance scalar flux measurements can be readily formulated by using the measured velocity spectral exponent in the inertial subrange.

Osservatorio Climatico-Ambientale di I-AMICA a Lecce: attività e prospettive di un centro d'eccellenza al servizio del territorio

The eddy covariance is the most direct, efficient and reliable method to measure the turbulent flux of a scalar (Baldocchi, 2003). Required conditions for high-quality eddy covariance measurements are amongst others stationarity of the measured data and a fully developed turbulence. The simplest method for obtaining the fluctuating components for covariance calculation according to Reynolds averaging rules under ideal stationary conditions is the so called mean removal method. However steady state conditions rarely exist in the atmosphere, because of the diurnal cycle, changes in meteorological conditions, or sensor drift. All these phenomena produce trends or low-frequency changes superimposed to the turbulent signal. Different methods for trend removal have been proposed in literature; however a general agreement on how separate low frequency perturbations from turbulence has not yet been reached. The most commonly applied methods are the linear detrending (Gash and Culf, 1996) and the high-pass filter, namely the moving average (Moncrieff et al., 2004). Moreover Vickers and Mahrt (2003) proposed a multi resolution decomposition method in order to select an appropriate time scale for mean removal as a function of atmospheric stability conditions. The present work investigates the performance of these different detrending methods in removing the low frequency contribution to the turbulent fluxes calculation, including also a spectral filter by a Fourier decomposition of the time series. The different methods have been applied to the calculation of the turbulent fluxes for different scalars (temperature, ultrafine particles number concentration, carbon dioxide and water vapour concentration). A comparison of the detrending methods will be performed also for different measurement site, namely a urban site, a suburban area, and a remote area in Antarctica. Moreover the performance of the moving average in detrending time series has been analyzed as a function of the averaging windows for different scalars in different measurement campaigns.

In the nocturnal boundary layer episodes of horizontal wind meandering are frequent. These episodes are characterised by low-wind regimes (wind speed less than 1.5 m s-1) in which submeso motions drive the wind dynamics and turbulence is weak and often intermittent. The inception of the meandering phenomenon as well as the interaction between turbulence and the submeso oscillations are still poorly understood.In this work we study the vertical evolution of the wind meandering by analysingnight-time anemometric data. The observations were carried on at a coastal site in Espirito Santo state, south-eastern Brazil from august to November 2016. The turbulent data, divided in hourly series, were collected in a 140-m tower designed to provide micrometeorological observations with high vertical resolution and deep coverage of the lower portion of the atmospheric boundary layer. Particularly, turbulence observations of the wind components and temperature are carried at 11 vertical levels. The tower has been deployed next to a natural gas power plant, at 3 km from the ocean. The terrain is generally flat for an area of 30 km from the tower, where moderate hills exist.The meandering timescale at each level is evaluated through the Eulerian Autocorrelation Functions of the horizontal wind velocity components and temperature, while the interactions between the different scales of motions is studied using the multi-correlation analysis. Thus the vertical evolution of meandering and time scales structure can be studied.