Planetary Boundary Layer Model Research Articles

An Evaluation of Scatterometer-Derived Oceanic Surface Pressure Fields

Abstract Oceanic surface pressure fields are derived from the NASA Quick Scatterometer (QuikSCAT) surface wind vector measurements using a two-layer similarity planetary boundary layer model in the midlatitudes and a mixed layer planetary boundary layer model in the tropics. These swath-based surface pressure fields are evaluated using the following three methods: 1) a comparison of bulk pressure gradients with buoy pressure measurements in the North Pacific and North Atlantic Oceans, 2) a least squares difference comparison with the European Centre for Medium-Range Weather Forecasts (ECMWF) surface pressure analyses, and 3) a parallel spectral analysis of the QuikSCAT and ECMWF surface pressure fields. The correlation coefficient squared between scatterometer-derived pressure fields and buoys is found to be R2 = 0.936. The average root-mean-square difference between the scatterometer-derived and the ECMWF pressure fields ranges from 1 to 3 hPa, depending on the latitude and season, and decreases after the assimilation of QuikSCAT winds in the ECMWF numerical weather prediction model. The spectral components of the scatterometer-derived pressure fields are larger than those of ECMWF surface analyses at all scales in the midlatitudes and only at shorter wavelengths in the tropics.

Evaluation of MM5 mesoscale model at local scale for air quality applications over the Swedish west coast: Influence of PBL and LSM parameterizations

The performance of MM5 mesoscale model (Version 3.6.3) using different planetary boundary layer (PBL) and land surface model (LSM) parameterizations is evaluated and compared using high temporal and spatial resolution GOTE2001 campaign data at local scale (a few kilometers) over the Greater Goteborg area along the Swedish west coast during 7–20 May 2001. The focus is on impact of PBL and LSM parameterizations on simulated meteorological variables important for air quality applications such as global radiation, diurnal cycle of near-surface air temperature and wind, diurnal cycle intensity, near-surface vertical temperature gradient, nocturnal temperature inversion, boundary layer height, and low-level jet (LLJ). The model performance for daytime and nighttime and under different weather conditions is also discussed. The purpose is to examine the performance of the model using different PBL and LSM parameterizations at local scale in this area for its potential applications in air quality modeling. The results indicate that the influence of PBL and LSM parameterizations on simulated global radiation, diurnal cycle of near-surface air temperature and wind speed, diurnal cycle intensity, vertical temperature gradient, nocturnal temperature inversion and PBL heights, which are critical parameters for air quality applications, is evident. Moreover, the intensity and location of LLJ are simulated well by all schemes, but there also exist some differences between simulated results by using different PBL and LSM schemes. Therefore, the choice of PBL and LSM parameterizations is important for MM5 applications to air quality studies.

Modification of surface winds near ocean fronts: Effects of Gulf Stream rings on scatterometer (QuikSCAT, NSCAT) wind observations

Modifications to surface winds by currents and sea surface temperature (SST) gradients near frontal boundaries of Gulf Stream rings are analyzed using satellite SST and scatterometer (NASA's Quick Scatterometer (QuikSCAT), NASA scatterometer (NSCAT)) wind observations. A component of scatterometer wind approximately equal and opposite to the surface current vector is observed and attributed to the fact that scatterometers detect relative motion of water and air. Warm‐core ring (WCR) SSTs act to destabilize the marine atmospheric boundary layer (MABL), increasing surface wind magnitude by 10–15% and decreasing veering angle by 5–15° relative to large‐scale mean winds. Cold‐core ring (CCR) SSTs cause impacts of similar magnitude and opposite sense. Magnitudes and directions of modifications are accounted for by MABL dynamics of a nonlinear planetary boundary layer model forced by air‐sea temperature differences. Wind modifications occur within tens of kilometers of SST fronts, implying a wind response timescale of order 1 hour. By contrast, uniformity of modified winds across the larger area within rings suggests the response time for the MABL to return to equilibrium downstream from a front exceeds 10 hours. Over WCRs, strong divergence (convergence) occurs on the upwind (downwind) side; curl is strongly negative (positive) to the right (left) side facing downwind. Opposite patterns are generally seen over CCRs. Divergence (curl) peaks where winds blow perpendicular (parallel) to SST fronts. SST image analysis indicates enhanced cloudiness occurs with downwind convergence over WCRs. Wind stress curl due to ring modifications causes dipolar Ekman pumping sufficient to influence ring translation and decay processes.

Comments on the Synergism Between the Analytic Planetary Boundary-Layer Model and Remote Sensing Data

This paper is adapted from a presentation at the session of the European Geophysical Society meeting in 2002 honouring Joost Businger. It documents the interaction of the non-linear planetary boundary-layer (PBL) model (UW-PBL) and satellite remote sensing of marine surface winds from verification and calibration studies for the sensor model function to the current state of verification of the model by satellite data. It is also a personal history where Joost Businger had seminal input to this research at several critical junctures. The first scatterometer in space was on SeaSat in 1978, while currently in orbit there are the QuikSCAT and ERS-2 scatterometers and the WindSat radiometer. The volume and detail of data from the scatterometers during the past decade are unprecedented, though the value of these data depends on a careful interpretation of the PBL dynamics. The model functions (algorithms) that relate surface wind to sensor signal have evolved from straight empirical correlation with simple surface-layer 10-m winds to satellite sensor model functions for surface pressure fields. A surface stress model function is also available. The validation data for the satellite model functions depended crucially on the PBL solution. The non-linear solution for the flow of fluid in the boundary layer of a rotating coordinate system was completed in 1969. The implications for traditional ways of measuring and modelling the PBL were huge and continue to this day. Unfortunately, this solution replaced an elegant one by Ekman with a stability/finite perturbation equilibrium solution. Consequently, there has been great reluctance to accept this solution. The verification of model predictions has been obtained from the satellite data.

Nonlocal Convective PBL Model Based on New Third- and Fourth-Order Moments

Abstract The standard approach to studying the planetary boundary layer (PBL) via turbulence models begins with the first-moment equations for temperature, moisture, and mean velocity. These equations entail second-order moments that are solutions of dynamic equations, which in turn entail third-order moments, and so on. How and where to terminate (close) the moments equations has not been a generally agreed upon procedure and a variety of models differ precisely in the way they terminate the sequence. This can be viewed as a bottom-up approach. In this paper, a top-down procedure is suggested, worked out, and justified, in which a new closure model is proposed for the fourth-order moments (FOMs). The key reason for this consideration is the availability of new aircraft data that provide for the first time the z profile of several FOMs. The new FOM expressions have nonzero cumulants that the model relates to the z integrals of the third-order moments (TOMs), giving rise to a nonlocal model for the FOMs. The new FOM model is based on an analysis of the TOM equations with the aid of large-eddy simulation (LES) data, and is verified by comparison with the aircraft data. Use of the new FOMs in the equations for the TOMs yields a new TOM model, in which the TOMs are damped more realistically than in previous models. Surprisingly, the new FOMs with nonzero cumulants simplify, rather than complicate, the TOM model as compared with the quasi-normal (QN) approximation, since the resulting analytic expressions for the TOMs are considerably simpler than those of previous models and are free of algebraic singularities. The new TOMs are employed in a second-order moment (SOM) model, a numerical simulation of a convective PBL is run, and the resulting mean potential temperature T, the SOMs, and the TOMs are compared with several LES data. As a final consistency check, T, SOMs, and TOMs are substituted from the PBL run back into the FOMs, which are again compared with the aircraft data.

Upgrading the coupled vegetation boundary layer model HIRVAC by new soil water and interception modules

The coupled atmospheric boundary layer model HIRVAC (High Resolution Vegetation Atmosphere Coupler) combines a 1.5 dimensional planetary boundary layer model (developed at the Humboldt University Berlin) and the mechanistic photosynthesis module PSN6 (University of Bayreuth) as well as a soil water module. This multi layer model includes a vertically structured vegetation and covers a range from 0-2 km height. The model layers are coupled via the turbulent exchange using a K-approach. Recent work aimed to include several new modules to enable model calculations for realistic conditions. Therefore rainfall and a multi layer interception and a soil water module were included. Further the light extinction through the canopy was upgraded by including sunfleck dynamics for calculating a more realistic radiation distribution within and below canopy. The upgraded model version was run for the Anchor Station in Tharandt where various measurements for comparison with model results are available. The obtained results demonstrate that periods of several days can be modelled successfully. A good model performance was found for both, soil water and interception. The modelled energy fluxes finally confirm the reliability of the overall model performance.

A new Planetary Boundary Layer model and its application to heavy rain modeling

A new Planetary Boundary Layer (PBL) model (namely MY-4 in this paper) was developed with reference to Mellor-Yamada’s Level 4 turbulent closure concept. Having been coupled with a meso-scale model MM5, MY-4 was utilized to simulate a heavy-rain process, which took place over South China during June 8–9, 1998. Its model outputs indicated that the rainfall process was well captured in terms of its intensity and geographical distribution. More importantly, in comparison with MM5’s original boundary layer models, MY-4 made the following improvements: (1) MY-4 not only simulated the major weather systems like low vortexes and low-level jets more accurately, thus improving the general weather pattern of the rainfall process, but it was also capable of restraining the occurrence of false rainfall centers with maximum precipitation amounts exceeding 160 mm. (2) Having been compared with the wind profile observed at Hong Kong, it revealed that MY-4 could reproduce the wind speed fluctuations in a short time scale reasonably well, which was not yet achieved in the original PBL models in MM5. Furthermore, a more detailed comparative study was made on the results simulated by MY-4 and a PBL model (which is based on Mellor-Yamada’s Level 2.5 concept) respectively. It showed that the contributions of turbulences generated by the two categories of PBL models to the wind fields at a lower atmosphere were increasingly different even in the first 1–2 hours of integrations. As the analysis demonstrated, under the nonlinear interactions within the meso-scale MM5 model, it was the turbulences in the boundary layer that had the most important impacts on the final model outputs and MY-4 seemed to better reflect this turbulent process, hence leading to the aforementioned improvements.

Baroclinic Planetary Boundary-Layer Model for Neutral and Stable Stratification Conditions

The temperature and wind profiles in abaroclinic atmospheric boundary layer (ABL) are investigated.Assuming stationary conditions, the turbulent state in the ABL forstable and neutral conditions is uniquely determined by the Rossbynumber, the external stratification parameter and two externalbaroclinic parameters. A simple two-layer baroclinic model isdeveloped. It consists of a surface layer (SL) and overlyingEkman-type layer. The system of dynamic and heat transfer equations isclosed using K-theory. In the SL the turbulent exchangecoefficient is consistent with the results of similarity theorywhile in the Ekman layer it is assumed constant. The universalfunctions in the resistance, heat and humidity transfer laws arededuced from the analytical solutions for the wind and temperatureprofiles. The solutions of the ABL resistance laws for theinternal ABL parameters, necessary for the calculations of the ABLprofiles, are approximated in terms of the external ABLparameters. Favourable agreement of model results with theavailable experimental data is demonstrated.

THE SCATTEROMETER AS A PRESSURE SENSOR

The UW Planetary Boundary Layer (PBL) model is used to derive surface pressure fields from scatterometer surface measurements. After a decade of comparison with NMC wind products, it is clear that the combination of the scatterometer and the PBL model yield superior information about the global pressure fields. The pressure fields are used to correct the scatterometer winds where they are contaminated by rain or poorly retrieved due to antenna geometry. The new winds are compared to buoy wind measurements. The scatterometer-derived surface pressure fields are assessed with buoy measurements and NWS analyses (NCEP, ECMWF and MM5 in the Gulf of Alaska). The gradients are systematically compared to buoy-measured pressure gradients. The model has been used to initialize a GCM with Geostrophic (pressure gradient) winds, yielding superior prognoses to initialization with surface winds. Original and corrected scatterometer winds are being assimilated in a NWF model.

Global Pressure Fields from Scatterometer Winds

A method is presented for computing global surface pressure fields from satellite scatterometer winds. Pressure gradients are estimated using a two-layer similarity planetary boundary layer model in the midlatitudes and a mixed-layer model in the Tropics. A global pressure field is then fit to the pressure gradients by least squares optimization. A series of surface pressure fields calculated from SeaWinds-on-QuikSCAT (Quick Scatterometer) measurements are compared with numerical weather analyses and buoy measurements. Surface pressure observations in the tropical oceans are scarce and come largely from ships of opportunity. At present no buoy in the Atlantic Ocean and only 10 buoys in the Pacific Ocean have pressure sensors. The method presented here suggests that 0.58-resolution maps of sea surface pressure can be readily retrieved from available satellite remote sensing data every 12 h in near‐real time. It is shown that these fields are at least of comparable quality to the ECMWF analyses.

Improvement of the K-profile Model for the Planetary Boundary Layer based on Large Eddy Simulation Data

Modifications of the widely used K-profile model of the planetary boundary layer (PBL), reported by Troen and Mahrt (TM) in 1986, are proposed and their effects examined by comparison with large eddy simulation (LES) data. The modifications involve three parts. First, the heat flux from the entrainment at the inversion layer is incorporated into the heat and momentum profiles, and it is used to predict the growth of the PBL directly. Second, profiles of the velocity scale and the Prandtl number in the PBL are proposed, in contrast to the constant values used in the TM model. Finally, non-local mixing of momentum was included. The results from the new PBL model and the original TM model are compared with LES data. The TM model was found to give too high PBL heights in the PBL with strong shear, and too low heights for the convection-dominated PBL, which causes unrealistic heat flux profiles. The new PBL model improves the predictability of the PBL height and produces profiles that are more realistic. Moreover, the new PBL model produces more realistic profiles of potential temperature and velocity. We also investigated how each of these three modifications affects the results, and found that explicit representation of the entrainment rate is the most critical.

Stability‐induced modification of sea surface winds over Gulf Stream rings

Satellite‐borne scatterometer and infrared data collected over Gulf Stream warm and cold core rings are used to study the effect of the sea‐air temperature difference on the wind speed over rings. The observed acceleration of the wind over warm core rings and deceleration over cold core rings is found to be consistent with that predicted by the planetary boundary layer model of Brown and Foster [1994]. In addition it is shown that the distance over which the winds respond to an ocean surface temperature step is short (≤25km) while the distance over which the marine boundary layer responds to a surface temperature step is long (≥175km).

Direct Observations of Atmospheric Boundary Layer Response to SST Variations Associated with Tropical Instability Waves over the Eastern Equatorial Pacific*

Tropical instability waves (TIWs), with a typical wavelength of 1000 km and period of 30 days, cause the equatorial front to meander and result in SST variations on the order of 1°–2°C. Vertical soundings of temperature, humidity, and wind velocity were obtained on board a Japanese research vessel, which sailed through three fully developed SST waves from 140° to 110°W along 2°N during 21–28 September 1999. A strong temperature inversion is observed throughout the cruise along 2°N, capping the planetary boundary layer (PBL) that is 1–1.5 km deep. Temperature response to TIW-induced SST changes penetrates the whole depth of the PBL. In response to an SST increase, air temperature rises in the lowest kilometer and shows a strong cooling at the mean inversion height. As a result, this temperature dipole is associated with little TIW signal in the observed sea level pressure (SLP). The cruise mean vertical profiles show a speed maximum at 400–500 m for both zonal and meridional velocities. SST-based composite profiles of zonal wind velocity show weakened (intensified) vertical shear within the PBL that is consistent with enhanced (reduced) vertical mixing, causing surface wind to accelerate (decelerate) over warm (cold) SSTs. Taken together, the temperature and wind soundings indicate the dominance of the vertical mixing over the SLP-driving mechanism. Based on the authors' measurements, a physical interpretation of the widely used PBL model proposed by Lindzen and Nigam is presented.

Application of Surface-Adjusted GOES Low-Level Cloud-Drift Winds in the Environment of Atlantic Tropical Cyclones. Part II: Integration into Surface Wind Analyses

The Cooperative Institute for Meteorological Satellite Studies at the University of Wisconsin—Madison recently (1997 season) began providing real-time Geostationary Operational Environmental Satellite (GOES) low-level cloud-drift winds in the vicinity of tropical cyclones on an experimental basis to the National Oceanic and Atmospheric Administration's (NOAA) Hurricane Research Division (HRD). The cloud-drift winds are derived from sequential high-resolution GOES visible channel imagery. These data were included in many of HRD's real-time tropical cyclone surface wind objective analyses, which were sent to NOAA's National Hurricane Center and the Central Pacific Hurricane Center on an experimental basis during the 1997–2001 hurricane seasons. These wind analyses were used to support the forecasters' tropical cyclone advisories and warnings. The satellite wind observations provide essential low-level coverage in the periphery of the tropical cyclone circulation where conventional in situ observations (e.g., ships, buoys, and Coastal-Marine Automated Network stations) are often widely spaced or nonexistent and reconnaissance aircraft do not normally fly. Though winds derived from microwave channels on polar orbiting satellites provide valuable surface wind data for HRD surface wind analyses, their swath coverage and orbital passes are limited spatially and temporally. GOES low-level visible (GLLV) winds offer nearly continuous spatial and temporal coverage in the western Atlantic and eastern Pacific basins. The GLLV winds were extrapolated to the surface using a planetary boundary layer model developed at HRD. These surface-adjusted satellite data were used in real-time surface wind analyses of 1998 Hurricane Georges, as well as in poststorm analyses of 1996 Hurricane Lili and 1997 Tropical Storm Claudette. The satellite observations often helped to define the spatial extent of the 17.5 m s−1 (34 kt) surface wind radii and also redefined the 25.7 m s−1 (50 kt) wind radius for one case. Examples of the impact of these data on real-time hurricane surface wind fields provided to the NHC will be discussed.

Wind structure in an intermediate boundary layer model based on Ekman momentum approximation

A quasi three–dimensional, intermediate planetary boundary layer (PBL) model is developed by including inertial acceleration with the Ekman momentum approximation, but a nonlinear eddy viscosity based on Blackadar’s scheme was included to improve the theoretical model proposed by Tan and Wu (1993). The model could keep the same complexity as the classical Ekman model in numerical, but extends the conventional Ekman model to include the horizontal accelerated flow with the Ekman momentum approximation.

A Gradient Wind Correction for Surface Pressure Fields Retrieved from Scatterometer Winds

Given a field of geostrophic winds and at least one pressure observation, a pressure field can be computed. If the winds are in reality gradient winds, then a correction must be applied to calculate the actual geostrophic winds. Here a method is proposed for including a gradient wind correction in the retrieval of geostrophic winds from Quick Scatterometer (QuikSCAT) surface measurements with a planetary boundary layer model. This correction translates into a better estimate of the corresponding surface pressure fields. The scheme is assessed by comparing these pressure fields to buoy measurements in the Gulf of Alaska and to radiosonde measurements in Hurricane Floyd. The gradient wind correction has a curvature component and a time-dependent component. Their relative magnitude is evaluated.

Comparison of Planetary Boundary Layer Model Winds with Dropsonde Observations in Tropical Cyclones

The values of surface winds simulated by the University of Washington (UW) two-layer similarity planetary boundary layer (PBL) model are compared with National Oceanic and Atmospheric Administration Hurricane Research Division global positioning system dropsonde observations and the surface wind analyses of a numerical weather prediction model. These three wind products compare fairly well at moderate wind speeds, away from the center of the storms where the coarse resolution of the numerical model is not a major factor. In the very high wind regime, the UW PBL model winds match the dropsonde observations fairly well, which is consistent with the unique characteristic of the PBL model being able to account for the nonlinear effects of organized large eddies. These eddies transport momentum and heat fluxes more efficiently than the smaller-scale, local turbulence can, leading to simulations of higher winds with mesoscale variability.

A scheme for improving scatterometer surface wind fields

A method is presented for improving QuikSCAT surface wind fields. The University of Washington Planetary Boundary Layer model is used to retrieve a surface pressure field from any swath of QuikSCAT surface wind vectors. An alternate set of surface wind vectors is computed from the newly calculated pressure field. The latter can be smoothed and the process can be iterated. New surface wind vectors can be calculated where ambiguity removal fails and where measurements are missing. The present methodology preserves boundary layer dynamics and is an improvement over a statistical filter.

A physical‐model‐based, field‐wise and self‐contained algorithm for removing directional ambiguities of ocean surface winds retrieved from scatterometer measurements

An algorithm is introduced to remove the directional ambiguities in ocean surface winds measured by scatterometers, which requires scatterometer data only. It is based on two versions of PBL (planetary boundary layer) models and a low‐pass filter. A pressure field is first derived from the median‐filtered scatterometer winds, is then noise‐filtered, and is finally converted back to the winds, respectively, by an inverted PBL model, a smoothing algorithm, and a PBL model. The derived wind field is used to remove the directional ambiguities in the scatterometer data. This new algorithm is applied to Hurricane Eugene and produces results comparable to those from the current standard ambiguity removal algorithm for NASA/JPL SeaWinds project, which requires external numerical weather forecast/analyses data.

On detectors used in solar spectrographs: CAO Wen-da, JI Kai-fan & SONG Qian Yunnan Observatory, CAS, Kunming 650011

Impacts of meteorological modeling in the planetary boundary layer (PBL) and urban canopy model (UCM) on the vertical mixing of pollutants are studied. Concentrations of gaseous chemical species, including ozone (O3) and nitrogen dioxide (NO2), and particulate matter over Paris and the near suburbs are simulated using the 3-dimensional chemistry-transport model Polair3D of the Polyphemus platform. Simulated concentrations of O3, NO2 and PM10/PM2.5 (particulate matter of aerodynamic diameter lower than 10 μm/2.5 μm, respectively) are first evaluated using ground measurements. Higher surface concentrations are obtained for PM10, PM2.5 and NO2 with the MYNN PBL scheme than the YSU PBL scheme because of lower PBL heights in the MYNN scheme. Differences between simulations using different PBL schemes are lower than differences between simulations with and without the UCM and the Corine land-use over urban areas. Regarding the root mean square error, the simulations using the UCM and the Corine land-use tend to perform better than the simulations without it. At urban stations, the PM10 and PM2.5 concentrations are over-estimated and the over-estimation is reduced using the UCM and the Corine land-use. The ability of the model to reproduce vertical mixing is evaluated using NO2 measurement data at the upper air observation station of the Eiffel Tower, and measurement data at a ground station near the Eiffel Tower. Although NO2 is under-estimated in all simulations, vertical mixing is greatly improved when using the UCM and the Corine land-use. Comparisons of the modeled PM10 vertical distributions to distributions deduced from surface and mobile lidar measurements are performed. The use of the UCM and the Corine land-use is crucial to accurately model PM10 concentrations during nighttime in the center of Paris. In the nocturnal stable boundary layer, PM10 is relatively well modeled, although it is over-estimated on 24 May and under-estimated on 25 May. However, PM10 is under-estimated on both days in the residual layer, and over-estimated on both days over the residual layer. The under-estimations in the residual layer are partly due to difficulties to estimate the PBL height, to an over-estimation of vertical mixing during nighttime at high altitudes and to uncertainties in PM10 emissions. The PBL schemes and the UCM influence the PM vertical distributions not only because they influence vertical mixing (PBL height and eddy–diffusion coefficient), but also horizontal wind fields and humidity. However, for the UCM, it is the influence on vertical mixing that impacts the most the PM10 vertical distribution below 1.5 km.