A53A-0909
Turbulence Spectra in the Surface Layer with a Steady Surface Thermal Inversion
the EBEX-2000 (International Energy Balance Experiment, 2000, EBEX-2000) was carried out over a flood- irrigated cotton field with very strong evaporation and transpiration. And thus the latent heat flux took most part of the solar radiation and the sensible heat flux, which would directly heat the atmosphere, was very small and even became negative in mid-afternoon. Therefore, the thermal turbulence was suppressed and there always existed a surface thermal inversion during the observation. The temperatures measured at 8.7 m were always higher than that at 2.7 m, which further restrained the development of the turbulence in the lower part of the surface layer, and the turbulence exchanges for the momentum, energy and other were restrained too. Owing to strong action of the underlying surface, there is distinct wind shear, and the nearer the distance to the ground, the stronger the wind shear. Moreover, the surface thermal inversion makes the wind shear sustainable and stable. On the other hand, due to the strong blocking and friction action of the underlying surface, eddies would be strongly impacted when they came close to the ground, in particular for small eddies. That is to say, the nearer the distance to the ground, the stronger the influence of the ground on small eddies and the larger the range of eddy-size that can be directly influenced by the ground. Both the above factors contribute to the differences between the spectra at intermediate frequencies at the two heights: the horizontal power spectrum at 8.7 m does not obey -1 power law at intermediate frequencies, but it does at 2.7 m. The vertical power spectra at 8.7 m are somewhat flatter and broader at the spectral peak, while they are sharper and narrower at 2.7 m.
A53A-0910
Statistics of Volumes, Swept by Small Spheroidal Particles in a Turbulent Flow.
Collisions between non-spherical particles (ice crystals) give rise to formation of aggregates; collision of non- spherical crystals with cloud droplets is the main mechanism of graupel production. The rate of riming and that of ice-ice collisions is not well known even in a pure gravity case. Often these collisions take place in the regions of enhanced turbulence in cumulus clouds. In spite of its high importance, the problem of collisions of such particles in a turbulent flow is not yet solved. In this work we present novel method of collision kernels calculation between small (less than 30 mic) spheroid particles of different aspect ratios (both prolate and oblate). The collision kernel between two spheroids is defined in terms of velocity fluxes of the particle of one type relative to the particle of another type. In this study hydrodynamic interaction between particles is not taken into account, so that the collision kernel is the swept volume (hereafter, SV) of colliding particles. Scale analysis indicates that spatial and time characteristic scales of Lagrangian acceleration and turbulent shears are much larger then the scales determining particles collisions. The results of this analysis allows one to consider turbulent flow as a combination of small regions in which Lagrangian accelerations and shears can be considered frozen during the particles' approach and collision. The consequence of collisions may be then regarded as taking place at different independent values of these parameters. A large set of turbulent field realizations (acceleration/shear pairs) was generated using generators of shears and accelerations, reproducing probability distribution functions (PDF) at high Reynolds numbers and dissipation rates, as they were obtained in recent laboratory and theoretical studies. There was obtained approximate analytical solution of spheroid motion, valid for small Stokes numbers. This solution allowed us to find approximate probability distribution functions (PDF) of spheroid velocities (translation and angular) and orientations for any given realization of a turbulent field. Having in hand these PDFs, we were able to calculate analytically time series of SV for a given set of turbulent field realizations. Finally, PDF (histogram) of SV and the mean value were calculated from the time series. These results were obtained for a vide range of turbulent flow intensity (different Reynolds numbers and energy dissipations), from that corresponding to stratiform clouds up to deep cumulus clouds. The estimations were performed also for different values of aspect ratio (from a plate-like spheroid (aspect ratio 0.05) up to a needle-like one (aspect ratio 20)) and different particles sizes. The results manifest that: - PDF of SV differs significantly from Gaussian and the difference increases with turbulent flow intensity and particle non-sphericity; - turbulence magnifies SV up to several times comparing with the pure gravity case; the effect is enlarging with flow intensity and particle aspect ratio deviation from unity; - an influence of turbulence on SV becomes especially large for small particles (of order ) and particles of similar size.
A53A-0911
Thermodynamic Structure of a Grass Fire Plume
Fine scale measurements of temperature were made within a grass fire plume during the FireFlux experiment that was conducted near Houston, Texas in 2006. Fine-wire thermocouples were mounted on a 43 m guyed tower located within the experimental fuel bed. The plume temperatures were measured at 1 Hz s as the fire front passed the tower. Maximum plume temperatures measured were approximately 290 C at a height of 10 m AGL. Entrainment of ambient air was observed to occur from both the top and underneath the plume. The entrainment near the surface was associated with the formation of a horizontal vortex that formed in response to the fire front and strong wind shear at the fire-atmosphere interface. Further results to be presented include plume heating rates calculated from data obtained from the tower and an infrared thermal imaging camera.
A53A-0912
Investigation of Convective Initiation Along a Dryline Using Observations and Numerical Weather Prediction Model
A narrow zone of strong horizontal moisture gradient, known as a dryline, is frequently observed over portions of the Southern Great Plains of the United States. The dryline is a boundary separating warm, moist maritime air from the Gulf of Mexico and hot, dry continental air from southwest U.S. and northern Mexico. The dryline acts as a focus for severe convective storms, and often leads to flooding and tornadoes. Although most storms initiate at or near the dryline, the exact processes by which convection is triggered and the preferred location for convection along the dryline are not well understood. Because the underlying processes are highly nonlinear, current numerical weather prediction (NWP) models show poor skill in their ability to accurately forecast these events. In this research a non-convective dryline case over Oklahoma and Texas panhandle on 22 May 2002 was considered. Using extensive high spatial and temporal resolution observational data from the International H2O Project, a field campaign in 2002 (IHOP_2002), and the National Center for Atmospheric Research (NCAR) Weather Forecasting and Research (WRF) model moisture evolution and variability in the boundary layer is thoroughly analyzed and investigated. Performance of the model and the possible reason why the anticipated dryline on 22 May 2002 did not trigger convective storm over Homestead - OK area are discussed. Results of the observational analysis indicate that abundant moisture did not sustain over Homestead - OK area during 22 May 2002. Moreover, vertical structure of water vapor mixing ratio indicate that moisture was not deep enough for vertically moving air parcels due to the dryline convergence provide the necessary destabilization effect to support deep convection initiation during this period.
A53A-0913
Characteristics of cloud microphysical properties observed at the Cloud Physical Observation System (CPOS) site in Korea
The Cloud Physics Observation Site (CPOS) site was established in 2004 over the mountainous area in the eastern part of Korea to collect cloud physics data on a long term basis. The instruments installed included a Micro Rain Radar (MRR), a Forward Scattering Spectrometer Probe (FSSP), a Microwave radiometer (MWR), and a disdrometer. This study mainly focuses on the bright band and raindrop size distribution (RDSD) characteristics of precipitating clouds measured at the CPOS site for the three year period of 2004-2006. As a comparison we also analyzed the MRR data measured at a western coastal site in Korea, Hae-Nam, HN. First we classified the vertical equivalent reflectivity factor profiles from MRR into three types, low level rain (Type 1), rain with a distinct bright band (Type 2), and convective rain (Type 3). It was found that the mountainous site (CPOS) had more proportion of Types 1 and 2 profiles while Type 3 covered 51% of the coastal site (HN) precipitation, suggesting the prevalence of orographically induced stratiform precipitation at the CPOS and convective precipitation at HN. Furthermore, the bright band appeared to be thicker but the sharpness of the band is weaker for the CPOS Type 2 profiles than those of HN. It was argued that the contrast was due to the dominant growth and melting of unrimed snow particles at HN and the prevailing rimmed ice crystals with the supercooled drops at the CPOS in stratiform precipitation. Preliminary analysis showed that on average the RDSD of HN is larger than those of the CPOS. The difference in RDSD was prominent for the diameter greater than 2.5 mm, where the RDSD of HN was definitely larger than those of the CPOS. As discussed above, these results might be due to the differences in geographical settings between the CPOS and HN; HN was generally affected by convective precipitation and the CPOS has more occurrence of straitform precipitation. Further analyses of the date from other instruments will be presented at the conference
A53A-0914
Tank Experiments on the Influence of Rain on Air-Sea CO2-Exchange
In the frame of the national project WiSSCy (Impact of Wind, Rain, and Surface Slicks on Air-Sea CO2 Transfer Velocity â€" Tank Experiments) we have conducted laboratory experiments that were devoted to air-sea CO2 transfer in the presence of heavy artificial rain and wind. The experiments were performed in a 24-m- long and 1-m-wide linear wind-wave tank at wind speeds up to 8 m/s and rain rates up to 160 mm/h using slick- free and slick-covered freshwater. Rain was produced by 3000 hypodermic needles at a height of 4.5 m above the water surface. The rain generator covered an area of 2.3 m2, i.e. 10 % of the total surface area. After injecting CO2 into the tank's closed air channel, gas transfer from the air into the water was measured. Recent results of these experiments show evidence that heavy rain enhances the air-sea gas transfer at low wind speeds, whereas for wind speeds higher than 6 m/s no additional rain effect on gas transfer can be measured. The application of slicks to the water surface could be shown to result in a strong reduction of gas exchange. However, this effect was counteracted when slicks were combined with heavy rain. Fetch dependence and dynamics inside the water body were examined by analyzing water samples at different fetches and depths and measuring water current profiles using Acoustic Doppler Velocimetry. In order to investigate the impact of freshwater rain on seawater, similar experiments were performed in a 1m x 0.7m x 1.3m smaller tank for different rain rates and saline water.
A53A-0915
Global Sea Salt Modeling Using a Coupled Microphysical and Climate Model and Its Comparison with Observations
The indirect effect of anthropogenic aerosols on climate change is modulated by the abundance of natural aerosols. We model the global loading and size-distribution of sea salt aerosol (SSA) using a coupled microphysical/climate model based on the NCAR community atmospheric model (CAM) and the University of Colorado / NASA Community Aerosol and Radiation Model for Atmospheres (CARMA). Production, transport, vertical diffusion, particle growth and removal processes are represented in the model. Sea salt particles with 80% relative humidity radius ranging from 0.01 to greater than 50um are generated using a combined sea salt source function based on Andreas [1998], Gong [2003] and Martensson [2003]. The friction wind speed under different stability conditions is utilized to drive the production of SSA. Particle dry deposition and wet deposition are calculated on a size-resolved basis. The mass prediction is compared with sea salt mass data from a network of coastal sites that monitored aerosol composition [Savoie and Prospero, 1977] and INDOEX, ACE-Asia and NEAQS shipboard observations. The optical properties of size-resolved SSA are also compared with MODIS/MISR satellite data and AERONET sun photometer data at remote oceanic sites.
A53A-0916
Improvement of the parameterization of the initial cloud droplet size distribution in a convective cloud model with detailed microphysics
In a previous study, Yang and Yum (2007) investigated the CCN effects on the cloud and precipitation developments under different thermodynamic conditions, using the Takahashi's 2D convective cloud model with detailed microphysics. Two different maximum CCN concentrations, maritime (300 cm-3) and continental (1000 cm-3), were used as input CCN data and the model was run on eight thermodynamic conditions obtained from observational soundings. Generally, a decrease in CCN number concentration enhanced cloud and precipitation developments but the CCN effects also critically depended on the given thermodynamic conditions. However, the Takahashi model has two major problems - one is that the initial droplet distribution is prescribed and the other is the excessive latent heat release to accommodate the generation of initial cloud droplets, especially when CCN concentrations are extremely high. Here we implement an improved method to parameterize the nucleation process, where the newly activated droplet distribution is determined by the supersaturation: as the supersaturation becomes higher, activated droplet concentration becomes higher but the size distribution is shifted toward the smaller sizes. Unlike the model runs with the original setting that showed more intense cloud development for the extreme continental CCN (5000 cm-3) than for the continental CCN for some thermodynamic conditions, the model runs with the improved parameterization of the nucleation process produced weaker convective clouds and slower onset of precipitation for the extreme continental clouds than for the continental as well as maritime clouds. As before, the CCN effects are more prominent in warm clouds than in cold clouds. The improved method will be further highlighted and the contrasts of the cloud and precipitation developments for the three CCN concentrations will be closely examined at the conference. Yang, Hee-Jung, and Seong Soo Yum, 2007, Effects of prescribed initial cloud droplet spectra on cloud and precipitation developments under different thermodynamic conditions: a modeling and observational study. Atmos. Res., doi:10.1016/j.atmosres.2007.04.004