An Approximate Analytical Solution For The Baroclinic And Variable Eddy Diffusivity Semi-Geostrophic Ekman Boundary Layer

The WKB method has been used to develop an approximate solutionof the semi-geostrophic Ekman boundary layer with height-dependenteddy viscosity and a baroclinic pressure field. The approximate solutionretains the same simple form as the classical Ekman solution. Behavioursof the approximate solution are discussed for different eddy viscosityand the pressure systems. These features show that wind structure inthe semi-geostrophic Ekman boundary layer depends on the interactionbetween the inertial acceleration, variable eddy viscosity and baroclinicpressure gradient. Anticyclonic shear has an acceleration effect on theair motion in the boundary layer, while cyclonic shear has a decelerationeffect. Decreasing pressure gradient with height results in a super-geostrophicpeak in the wind speed profile, however the increasing pressure gradient withheight may remove the peak. Anticyclonic shear and decreasing the variableeddy viscosity with height has an enhanced effect on the peak.Variable eddy viscosity and inertial acceleration has an important role in thedivergence and vorticity in the boundary layer and the vertical motion at the top of the boundary layer that is called Ekman pumping. Compared to the constanteddy viscosity case, the variable eddy diffusivity reduces the absolute value ofEkman pumping, especially in the case of eddy viscosity initially increasing with height. The difference in the Ekman pumping produced by different eddy diffusivity assumptions is intensified in anticyclonic flow and reduced in cyclonic flow.     

变化涡动沾性系数的Ekman动量近似模式风场结构

在边界层动力学中，涡动粘性系数是影响边界层风场结构的一个重要参数。本文利用边界层动力学中的Ekman动量近似理论，给出了涡动粘性系数随高度缓变条件下的Ekman动量近似边界层模式解，着重讨论了边界层的风场结构、水平散度、垂直涡度以及边界层顶部的垂直速度。结果分析表明：与常值涡动粘性系数情况相比，在边界层低层随高度增加的涡动粘性系数可以导致低层边界层风速随高度迅速增加，即风速垂直切变增加，同时风速矢与地转风之间的夹角减小。惯性项作用可以导致上述作用在气旋性区域减小、而在反气旋性区域增大。随高度增加的涡动粘性系数导致水平散度绝对值、垂直涡度绝对值以及边界层顶部的垂直速度绝对值在气旋性区域减小，而在反气性旋区域增大。涡动粘性系数与惯性之间的非线性相互作用是边界层动力学中重要过程。     

The diurnal wind variation in a variable eddy viscosity semi-geostrophic Ekman boundary-layer model: Analytical study

Summary ?A time-dependent semi-geostrophic Ekman boundary-layer model (SG), including slowly varying eddy diffusivity with height and inertial term effects, is developed to investigate the diurnal wind variation in the planetary boundary layer (PBL). An approximate analytical solution of this model is derived by using the WKB method, which extends the Tan and Farahani (1998)’s solution by including the vertical variable eddy viscosity. The features of the diurnal wind variation in the PBL mainly depend on three factors: the latitude, horizontal momentum advection and eddy viscosity. The vertical variable eddy viscosity has little influence on diurnal wind variation in the PBL at the low latitude, however its effect may be exacerbated in the mid- and high latitudes. In comparing with the constant eddy viscosity case, the decreasing (increasing) with height eddy viscosity produces a large (small) maximum wind speed (MWS) in the PBL, however, the eddy viscosity that has a mid-layer peak in the vertical gives rise to a higher height of occurrence of MWS. For the boundary-layer wind structure, there is a singular point when the modified SG inertial oscillation frequency η equals the forcing frequency ω. The isotachs of boundary-layer wind speed have almost no tilt to left or right relative to time evolution and the occurrence time of the MWS is the earliest at the singular point. The feature will be enhanced in the decreasing with height eddy viscosity case and weakened in the eddy viscosity initially increasing with height case. Received April 6, 2001; accepted December 27, 2001     

Model evaluation of the atmospheric boundary layer and mixed-layer evolution

In the present study, an attempt is made to assess the atmospheric boundary-layer (ABL) depth over an urban area, as derived from different ABL schemes employed by the mesoscale model MM5. Furthermore, the relationship of the mixing height, as depicted by the measurements, to the calculated ABL depth or other features of the ABL structure, is also examined. In particular, the diurnal evolution of ABL depth is examined over the greater Athens area, employing four different ABL schemes plus a modified version, whereby urban features are considered. Measurements for two selected days, when convective conditions prevailed and a strong sea-breeze cell developed, were used for comparison. It was found that the calculated eddy viscosity profile seems to better indicate the mixing height in both cases, where either a deep convective boundary layer develops, or a more confined internal boundary layer is formed. For the urban scheme, the incorporation of both anthropogenic and storage heat release provides promising results for urban applications.     

An analytical study of diurnal wind-structure variations in the boundary layer and the low-level nocturnal jet

An analytical framework is proposed for studying variations in the diurnal wind structure in the planetary boundary layer (PBL) and the evolution of the low-level nocturnal jet. A time-dependent eddy-diffusivity coefficient corresponding to solar input is proposed, and an appropriate coordinate transformation ensures that mixing height varies continuously with ground heat-flux changes. The solution exhibits the receding character of the daytime PBL as evening approaches, thereby dividing the PBL into two regimes — the one just above the ground, representing the nocturnal boundary layer, and the region above it. It is assumed that inertial oscillations (IO) are triggered in the upper layer at about the time of sunset when the reversal in the direction of ground heat flux is felt in the upper layer. Two approaches are adopted to determine the characteristic features of IO and the evolution of the nocturnal low-level jet. The first one is based on the physical principle that release of horizontal momentum due to deviation from the geostrophic wind gives rise to the IO. The solution captures all the characteristic features of the IO, such as phase shift and decreasing amplitude of the IO with increasing height. According to this analysis the IO is triggered at a level as soon as the top of the receding boundary layer leaves that level. The solution is discontinuous with respect to the vertical coordinate. In the second approach we solve an initial-value problem to determine the solution in the upper layer, assuming that at about the time of sunset there is a rapid collapse of the daytime PBL to the steady, nocturnal boundary layer. The assumption is based on the mixing-height profiles prepared from climatological data collected at Delhi. The solution for the nocturnal boundary-layer regime is then obtained as a boundary-value problem. The solutions so obtained are continuous throughout the domain of interest and exhibit the characteristic features of an IO. The analysis leads to the conditions under which a low-level nocturnal jet is produced and provides quantitative estimates of the parameters, such as length of night, latitude, mixing height at sunset and nocturnal mixing height, that are conducive to the generation of a jet. The nocturnal wind profile produced by this approach compares well both with typical atmospheric data observed at Delhi and with output from a mesoscale numerical model. There is still some uncertainty related to the time of initiation of the IO as a function of latitude.     

Sodar observations of the atmospheric boundary layer over the ocean during ASTEX-91

A complex marine experiment was conducted in autumn 1991 on the research vessel Dmitry Mendelev in association with the Atlantic Stratocumulus Transition Experiment (ASTEX). A three-axis Doppler sodar designed at the Institute of Atmospheric Physics, Moscow, was used in this experiment. Total observation time was about 770 hours from 6 October to 23 November. Besides facsimile records illustrating spatial and temporal structure of the turbulence distribution in the atmospheric boundary layer (ABL), routine quantitative measurements of profiles of wind and echo-signal strength were taken. Some main characteristics of the ABL behavior over the ocean were revealed through an analysis of these data as well as the results of other kinds of measurements. An important peculiarity of the ABL observed between the Canary Islands and the Azores was the presence of diurnal variation of convective turbulence strength having a maximum between 04:00 and 07:00 LT. A similar diurnal variation was observed for low-level cloud cover. Occurrence of various types of thermal stratification and their diurnal variation were obtained. Comparison of elevated stable layers and low-level cumulus showed that the lower boundary of clouds correlates well with the height of the bottom of elevated inversion layers (at heights of 200–600 m). Canary and Cabo Verde observations showed that islands strongly affect the ABL structure. The strong effect of a surface water temperature gradient on the ABL stability was observed when crossing the Canary, Azores, and Labrador currents and the Gulf Stream.     

The Impact of Land-Surface Parameter Properties and Resolution on the Simulated Cloud-Topped Atmospheric Boundary Layer

A method to simulate characteristics of wind speed in the boundary layer of tropical cyclones in an idealized manner is developed and evaluated. The method can be used in a single-column modelling set-up with a planetary boundary-layer parametrization, or within large-eddy simulations (LES). The key step is to include terms in the horizontal velocity equations representing advection and centrifugal acceleration in tropical cyclones that occurs on scales larger than the domain size. Compared to other recently developed methods, which require two input parameters (a reference wind speed, and radius from the centre of a tropical cyclone) this new method also requires a third input parameter: the radial gradient of reference wind speed. With the new method, simulated wind profiles are similar to composite profiles from dropsonde observations; in contrast, a classic Ekman-type method tends to overpredict inflow-layer depth and magnitude, and two recently developed methods for tropical cyclone environments tend to overpredict near-surface wind speed. When used in LES, the new technique produces vertical profiles of total turbulent stress and estimated eddy viscosity that are similar to values determined from low-level aircraft flights in tropical cyclones. Temporal spectra from LES produce an inertial subrange for frequencies \(\gtrsim \)0.1 Hz, but only when the horizontal grid spacing \(\lesssim \)20 m.     

Modelling Boundary-Layer Clouds With a Statistical Cloud Scheme and a Second-Order Turbulence Closure

We present a second-order turbulence model for the cloudy planetary boundary layer (PBL), which includes a statistical scheme of the sub-grid scale condensation. The model contains prognostic equations for the turbulent kinetic energy, total water, and liquid water temperature, the latter two being assumed to be conservative variables. Using these conservative thermodynamic variables the condensation process is formulated as a function of the departure of the total water from saturation and its variance. The computation of the variance requires second moment correlations which are modelled through the parameterization of the third-order moments using a convective mass-flux formulation. The inclusion of these third moments and new assumptions on heat flux transport lead to a nonlocal turbulence scheme with counter-gradient effects. The final form for the heat flux turns out to be a linearized version of a previously established result. For the statistical cloud formulation, a linear combination of a Gaussian and a positively skewed distribution function is used with a modified liquid water flux expression to account fornon-Gaussian behaviour.The effect of the turbulence scheme on the boundary-layer cloud structure is discussed and the performance of the model is tested by comparing it against the large eddy simulation (LES) of the undisturbed period of the Atlantic Stratocumulus Transition Experiment (ASTEX). The model is able to produce both mean and turbulent quantities that are in reasonable agreement with the LES output of ASTEX.     

A Comparison Between Modelled and Measured Mixing-Layer Height Over Munich

An attempt is made to correlate the mixing heights, derived from ceilometer and Sodar measurements, to those simulated by different atmospheric boundary-layer parameterization schemes. The comparison is performed at two sites (one suburban and one rural) close to Munich, Germany for two spring and two winter days. It is found that, under convective conditions, the mixing height determined, by both Sodar and ceilometer, corresponds to the middle or the top of the entrainment zone, respectively, as calculated from the eddy-viscosity profiles. Under stable conditions, the measured mixing height is related to the height where eddy viscosities attain their minimum values (Sodar) or to the height of residual mechanical turbulence (ceilometer). During a foehn case with weak turbulence, the measured mixing height from both Sodar and ceilometer is better inferred by considering the eddy-viscosity profiles during daytime and the height of the low-level jet during nighttime.     

Subcloud-layer kinematic and dynamic structures of a microburst-producing thunderstorm in Colorado determined from JAWS Dual-Doppler measurements

Some kinematic and dynamic structures of a microburst-producing storm in Colorado were investigated. Dual-Doppler data collected on 14 July, 1982 at 1647 MDT, during the Joint Airport Weather Studies (JAWS) project at Denver's Stapleton International Airport, were objectively analyzed to produce a three-dimensional wind field. The domain of interest had a horizontal dimension of 10 by 10 km centered on the microburst. Vertical velocities were computed by integrating the anelastic continuity equation downward from the storm's top with variational adjustment. Subsequently, fields of deviation perturbation pressure, density, and virtual temperature were retrieved from a detailed wind field using the three momentum equations.Results show that the microburst being investigated is embedded within a high-reflectivity region associated with heavy precipitation. A strong downflow impinges on the surface producing a stagnation mesohigh inside the microburst. This mesohigh is accompanied by mesolows in the strongest outflow regions, forming a pronounced horizontal perturbation pressure gradient outward from the high-pressure center. The outflow regions extend from the surface to approximately 1 km AGL with maximum divergence in excess of 10 ^–2 s^–1. Inclusion of friction in the pressure equation improves pressure recovery at all levels, especially in the atmospheric boundary layer (ABL). The microburst occurrence in the ABL enhances eddy transfer of momentum. Magnitudes of eddy viscosity and eddy stresses increase as a result of the microburst.     

Three-Dimensional Structure and Low-Level Features of the North Pacific Storm Track from 1999 to 2005

A storm track is a region in which synoptic eddy activities are statistically most prevalent and intense. At daily weather charts, it roughly corresponds to the mean trajectories of cyclones and anticyclones. In this paper, the recent QuikSCAT （Quick Scatterometer） satellite sea winds data with a 0.5°×0.5° horizontal resolution, and the NCEP （National Centers for Environmental Prediction） 10-m height Gaussian grid wind data and pressure-level reanalysis data, are employed to document the spatial structure of the North Pacific storm track in winter （January） and summer （July） from 1999 to 2005. The results show that in winter the North Pacific storm track is stronger, and is located in lower latitudes with a distinct zonal distribution. In summer, it is weaker, and is located in higher latitudes. Based on the horizontal distributions of geopotential height variance at various levels, three-dimensional schematic diagrams of the North Pacific storm track in winter and summer are extracted and presented. Analyses of the QuikSCAT wind data indicate that this dataset can depict the low-level storm track features in detail. The double storm tracks over the Southern Oceans found by Nakamura and Shimpo are confirmed. More significantly, two new pairs of low-level storm tracks over the North Pacific and the North Atlantic are identified by using this high-resolution dataset. The pair over the North Pacific is focused in this paper, and is named as the ＂subtropical storm track＂ and the ＂subpolar storm track＂, respectively. Moreover, statistical analyses of cyclone and anticyclone trajectories in the winters of 1999 to 2005 reveal as well the existence of the low-level double storm tracks over the North Pacific.     

Interpreting Reynolds stress from perspective of eddy geometry

Geometric features in oceanic mesoscale eddies such as tilt and anisotropy can influence the properties of the Reynolds stress that provides feedback between the eddies and the background flow. By regarding an eddy as a wave, previous studies have parameterized the Reynolds stress based on the equivalence in the tilt angle between the phase of the eddy stream functions and the variance ellipse for the Reynolds stress (RS-ellipse). However, the wave assumption cannot predict the anisotropy of the RS-ellipse, and also largely simplifies the eddy geometry, which would naturally be an ellipsoid rather than a wave. The present study explores the shape relation between elliptical eddies and the RS-ellipse, by mathematically reformulating the Reynolds stress based on the eddy shape. The new formula reveals that the shape relation is regulated by the horizontal extent of the occurrence probability distribution (PDF) of the eddy, and that the shape of the eddy and RS-ellipse are identical at the place of maximum PDF when the horizontal scale of the PDF is sufficiently larger than the size of the eddy. A similar tendency is found in eddies detected by satellite altimetry in the Kuroshio Extension jet region. A detailed analysis of the PDF in this region shows that the tilts of the eddies are likely to be consistent with the destabilization effect on the jet, suggesting a strong relation between the eddy geometry and the jet's stability in this region. These findings may open a path toward a new method to parameterize the Reynolds stress with the background state, exploiting the shape equivalence between the eddies and the RS-ellipse.     

Optimal estimation of Eddy viscosity and friction coefficients for a Quasi‐three‐dimensional numerical tidal model

Abstract

It is shown that the parameters in a quasi‐three‐dimensional numerical tidal model can be estimated accurately by assimilation of data from current meters and tide gauges. The tidal model considered is a semi‐linearized one in which advective nonlinearities are neglected but nonlinear bottom friction is included. The parameters estimated are the eddy viscosity, bottom friction coefficient, water depth and wind drag coefficient, the first three of which are allowed to be position‐dependent. The adjoint method is used to construct the gradient of a cost function defined as a certain norm of the difference between computed and observed current and surface elevations. On the basis of a number of tests, it is shown that very effective estimation of the nodal values of the parameters can be achieved using the current data either alone or in combination with elevation data. When random errors are introduced into the data, the estimated parameters are quite sensitive to the magnitude of the errors, and in particular the eddy viscosity is unstably sensitive. The sensitivity of the viscosity can be stabilized by incorporating an appropriate penalty term in the cost function or alternatively by reducing the number of estimated viscosity values via a finite element approximation. Once stabilized, the sensitivity of the estimates to data errors is significantly reduced by assimilating a longer data record.     

An Analytical Study of the Diurnal Variations of Wind in a Semi-geostrophic Ekman Boundary Layer Model

A time-dependent semi-geostrophic Ekman boundary-layer model based on the geostrophic momentum approximation is used to study the diurnal wind variation in the planetary boundary layer (PBL) and the evolution of the low-level nocturnal jet (LLJ). The coefficient of eddy viscosity varies periodically with time, varies linearly with height in the surface layer and is constant above the surface layer. The influence of horizontal advection of momentum on the diurnal wind variation in the PBL, the development of inertial oscillations (IOs) and the formation of the LLJ are examined.In comparison with the Ekman solutions, the diurnal wind variation in semi-geostrophic Ekman boundary-layer dynamics has the following features: (1) the phase angle of the diurnal wind wave shifts with height, the rate of shifting is increased in anticyclonic regions and decreased in cyclonic regions, (2) the time of occurrence of the low-level maximum wind speed is later in anticyclonic regions and earlier in cyclonic regions, (3) the height of occurrence of the maximum wind speed is higher in the anticyclonic and lower in cyclonic regions, (4) the wind speed maximum and the amplitude of the diurnal wind variation are larger in anticyclonic and smaller in cyclonic regions, (5) the period of IOs is larger in anticyclonic regions and smaller in cyclonic regions, (6) anticyclonic vorticity is conducive to the generation of LLJ in the PBL. These features are interpreted by means of the physical properties of semi-geostrophic Ekman boundary-layer dynamics and inertial oscillation dynamics.     

Boundary-layer structure near the cold front of a marine cyclone during “ERICA”

The boundary layer in the warm sector of a moderately deepening winter cyclone during the Experiment on Rapidly Intensifying Cyclones over the Atlantic (ERICA) is studied near the cold front. Data from the National Center for Atmospheric Research Electra research aircraft are used to examine mean and turbulence quantities. The aircraft data and supplemental data from ships, drifting buoys and moored buoys reveal an equivalent-barotropic pressure field. The area is found to be dominated by gradients in temperature and in turbulent fluxes, with changes occurring over 100 km horizontally being comparable to changes over 350 m vertically. The horizontal components of the gradients are found to be a maximum in a direction perpendicular to the front. Cross-sections perpendicular to the front are used to illustrate boundary-layer structure. Profiles of wind speed, stress, wind direction and stress direction are estimated from an Ekman model that is modified to take into account the equivalent-barotropic pressure field. Comparison of profiles from the model to the aircraft-measured data show reasonable agreement far from the front (100 km) when the model uses a constant eddy viscosity of approximately 6 kg m^–1 s^–1. Near the front there is less agreement with the model. Profiles of turbulent fluxes of momentum, heat and latent heat are divergent, with along-wind momentum flux negative and decreasing upward, cross-wind momentum flux positive and increasing upward, and heat flux and latent heat flux small, positive and decreasing upward. Far from the front, the turbulent kinetic energy budget shows that dissipation balances shear production. However, near-front behavior has an imbalance at low altitude, with shear production appearing as a TKE sink.     

A one dimensional wind profile model with stability and eddy viscosity estimations from sodar data

In a one-dimensional wind profile model, methods for eddy viscosity and stability estimations from sodar data have been evaluated with soundings. For eddy viscosity parameterization the ageostrophic method and mixing-length theory have been investigated. Three methods for estimating the static stability have been evaluated; a wind profile adjustment method, gravity wave analysis of sodar backscatter and flux profile functions for windspeed and standard deviation of vertical wind-speed. The wind profile model with variable momentum flux (VF) with height shows better results than an earlier constant flux model (Bergström, 1986). The VF model can be used for extending the sodar profile up to 1500 m.     

Dynamics of the planetary boundary layer with fine structure in the large scale background field

Summary On the basis of Wu and Blumen's work (1982) on the geostrophic momentum approximation (GMA) in the planetary boundary layer (PBL) and Tan and Wu (1992, 1994) on the Ekman momentum approximation (EMA) in the PBL, some improvements about the eddy exchange coefficientK, the advective inertial force and the lower boundary condition of the PBL are developed in this paper: (1) apply theK which is a gradually varying function of height instead of a constant value in the Ekamn layer, and introduce a surface layer; (2) take the effect of the vertical advective inertial force into account; (3) the solution technique is extended from level terrain to orographically formed terrain. Under the condition of the equilibrium among four forces (the pressure — gradient force, Coriolis force, eddy viscous force and inertial force including horizontal and vertical advective inertial forces), we have obtained the analytical solutions of the distributions of the wind and the vertical velocity. The computation of an individual example shows that: (1) both the wind velocity near surface and the angle between which and the non-viscous wind are more consistent with usual observations than that of Wu and Blumen (1982); (2) comparing with the horizontal advective inertial force, the vertical advective inertial force can not be neglected, when the orography is considered, the effect of the latter is even more important than the former.With 3 Figures     

Meso-beta scale numerical simulations of terrain drag-induced along-stream circulations Part I: Midtropospheric frontogenesis

Summary The problem of along-stream ageostrophic frontogenesis is studied by employing a numerical model at meso-alpha and meso-beta scales in simulations of the downstream circulations over the Front Range of the Rocky Mountains. Three-dimensional real data simulations at these two scales of motion are used to diagnose the transition from semigeostrophic cross-stream frontogenesis accompanying a propagating baroclinic upper-level jet streak to midtropospheric along-stream ageostrophic frontogenesis. This along-stream ageostrophic frontogenesis results from the perturbation of the jet streak by the Rocky Mountain range. The case study represents an example of internal wave dynamics which are forced by the drag of the Rocky Mountains on a strong jet streak in the presence of a low-level inversion.The simulation results indicate that, unlike semi-geostrophic frontogenesis, a front (which is alligned perpendicular to the axis of the jet stream) may form when significant adiabatic heating occurs within a stratified shear flow over horizontal length scales shorter than the Rossby radius of deformation. The mechanism responsible for the frontogenesis is the growth of the divergent along-stream wind velocity component which becomes coupled to the front's along-stream pressure gradient force. This nonlinear interaction produces hydrostatic mesoscale frontogenesis as follows: 1) vertical wind shear in the along-stream plane strengthens resulting in the increasingly nonuniform vertical variation of horizontal temperature advection as the ageostrophic wind component grows in magnitude downstream of the meso-scale terrain-induced adiabatic heating, 2) increasing along-stream differential vertical motions (i.e., along-stream thermally indirect circulation with warm air sinking to the west and cold air rising to the east) tilt the vertical gradient of isentropes into the horizontal as the vertical temperature gradient increases due to the previous process in proximity to horizontal gradients in the along-stream component of the ageostrophic wind, 3) as tilting motions act to increase the along-stream horizontal temperature gradient, the along-stream confluence acts to nonuniformly increase the along-stream frontal temperature gradient which increases the along-stream pressure gradient force resulting in further accelerations, ageostrophy, and frontal steepening as part of a scale contraction process.The evolution of the aforementioned processes results in the three-dimensional hydrostatic frontogenesis accompanying the overturning of isentropic surfaces. These adjustments act to turn air parcels to the right of the southwesterly geostrophic wind vector at successively lower atmospheric levels as the scale contraction continues. This simulated along-stream front is verified from diagnostic analysis of the profiler-derived temperature and wind fields.With 17 Figures     

地转动量近似下边界层中风场的调整

本文利用地转动量近似,并假设气压场为定常的圆形涡旋和初始风场不满足四力平衡(气压梯度力、科里奥利力、湍流粘性力和半地转惯性力)的条件下,求解了正压边界层中风场向气压场调整的初边值问题,得到了一些初步结论。本工作为利用四力平衡下的风速分布来诊断预报边界层风场提供了理论依据。     

On gradient transport turbulence models for stably stratified shear flow

The gradient transport model for stably stratified horizontal shear flow in which eddy diffusivity and viscosity are assumed to depend on the gradient Richardson number, Ri, is augmented with terms representing a finite adjustment time of the exchange coefficients. Barenblatt et al. (J. Fluid Mech., 253: 341–358, 1993) showed that using such a model, initial value problems for the formation of a stepwise structure of the buoyancy distribution are well posed. The model proposed is analysed taking into account the interaction between buoyancy and velocity fields. A condition for the formation of steps is derived from a linear stability analysis. Numerical computations show that a realistic stepwise finestructure develops, provided linear instability is allowed on a finite interval of Ri only.