A13B-01
Diagnosing low-level wind-moisture-precipitation relationships along tropical convective margins
Analysis of the margins of tropical convection zones is performed using low-level wind, moisture, and precipitation data from models and observations, with an emphasis on identifying relationships among these quantities. A simple theoretical prototype, discussed in prior work, provides intuition about how changes in low-level wind affect moisture and, in turn, deep convective activity. Here, diagnostics suggested by the prototype are applied to the study of synoptic-scale and monthly-mean fluctuations along the South Pacific and South Atlantic Convergence Zones. We suggest that such diagnostics may be useful for model-observation and model-model intercomparison, refinement of model convective parameterizations, and understanding sources of model biases.
A13B-02
Moisture Sources for Central America: A Lagrangian analysis Method Application
Location of Central America in a region characterized by the influence of important climate modulator elements, Central American particular climate patterns and role of tropical moisture on global hydrological cycle motivate the present study on moisture sources. Central American main moisture sources are analyzed under the implementation of a Lagrangian type analysis. The FLEXPART dispersion model is applied with data from the European Center for Medium Range Weather Forecast (ECMWF) for the 2000-2004 period as input for the FLEXPART model runs. In order to analyse the source-receptor relationship backward and forward runs were performed, tracking and integration of changes in specific moisture along its trajectory, for the set of 1 298 801 particles released is used to determine (E-P) contributions for the interest region. Obtained results allow the identification of the main sources of the moisture reaching Central America as well as the evaluation of the spatial evolution of the moisture. The presence of a primary moisture source for Central America is identified over the Caribbean Sea and a secondary moisture source seems to be present near the Equatorial Pacific region, the case of the Gulf of Mexico is not presented. The dominance of the Caribbean Sea region as a moisture source for Central America is presented and the role of the Intra Americas Seas Low Level Jet as a transport mechanism is qualitatively ensured from the results presented and the seasonal behavior of both structures and considering the moisture transport evolution patterns, (E-P) was quantified in order to show the differences in the contribution from the distinct sources while reaching the same region for the ten days backward and forward to obtain the ten days mean moisture gains and losses. The role of local topography is discussed in order to explain the observed differences of the contribution of the secondary source for Central America. A comparison with Eulerian analysis results is presented and the obtained results are discussed and related to recent isotopes analysis performed in the region for a similar period. Main results on the source- receptor relationship and main seasonal patterns are presented with an analysis on the locally quantified (E-P) as well as the main mechanisms involved and further analysis purposes.
A13B-03
Design of a Vector-Vorticity Dynamical Core on a Hexagonal Grid (Hex-VVDC)
We are developing a Global Cloud-Resolving Model (GCRM) based on the vector-vorticity dynamical core (VVDC) that predicts three-dimensional vorticity. The VVDC in an anelastic framework has been originally designed by Jung and Arakawa and used in a Cartesian grid. As the first step in developing the GCRM, we have constructed an intermediate model with the VVDC using a hexagonal grid on a planar quasi-rectangular horizontal domain, which we call the hexagonal VVDC (Hex-VVDC). The purpose of developing the Hex-VVDC is to test and verify the discretization of equations of the VVDC on a hexagonal grid before applying them to the geodesic hexagon/pentagon grid that will be used for the GCRM. It is found that having twice as many as cell walls and three times as many as cell corners than the cell centers in a hexagonal grid produces a computational mode in horizontal and vertical components of vorticity and horizontal velocity fields, which the dynamics cannot control. We have designed a special scheme to treat this computational mode. With this treatment, the Hex-VVDC seems to work well. Currently, we are performing further tests with the Hex-VVDC and also continuing the application of the dynamical core to the geodesic hexagon/pentagon grid. In this presentation, we will discuss the computational design of the Hex-VVDC and show simulated results primarily focusing on vertical transports of vorticity and momentum by convection.
A13B-04
Examining how land surface effects modulate rainfall in the eastern Amazon Basin
In the Amazon, it is important to apportion rainfall by storm type. In the eastern Amazon (approximately from Belém to Santarém) rainfall associated with large instability lines produces nearly half of the total, and this is complemented by that produced by rainfall from local convective systems. Our recent observational studies in the indicate that the relative importance of the nocturnal squall lines is exaggerated if one relies solely on data from the climate stations along the Amazon River channel. River breezes inhibit convective rainfall near the main channel, but in some areas river proximity effects lead to enhanced nocturnal rainfall of squall origin. Moreover, enhanced rainfall to the north of the Amazon main channel could be the result of orographic uplift. In this study we complement a limited climatological study of instability lines with two mesoscale model (Brazilian version of RAMS, B-RAMS) case studies to examine the effects of topography and river proximity on rain producing mechanisms in the eastern Amazon Basin. Two numerical experiments were done to examine the relative importance of these two rain-producing mechanisms in the region. In each, three nested grids were used. Results from the prototype simulation for the propagating squall line were compared with GOES images, NCEP reanalyses, and data from the LBA-ECO surface station network near Santarém (approximately 55°W). In this case we also examined the role of topography on squall line development by performing a sensitivity test of the case study squall development with and without topography. The locally-dominated convection study was based on a case of slack easterlies during cold frontal penetration into the western Amazon region.
A13B-05
Observed Self-Similarity of Precipitation Regimes Over the Tropical Oceans
Recently developed retrieval algorithms based on Tropical Rainfall Measuring Mission (TRMM) observations now provide the opportunity to study not only latent heating profiles over the tropics, but radiative heating profiles as well. Using these new products in conjunction with radar data from the TRMM Precipitation Radar (PR) from 2007, we developed an algorithm that classifies precipitation regimes based on the ensemble of storm top heights present at a given time over a specified grid-size, operating under the assumption that precipitation regimes are mixed in terms of cloud types. We apply the analysis separately to adjacent geographic basins spanning the tropical oceans, and interestingly, three statistically self-similar precipitation regimes emerge consistently within each region, the characteristics of which are largely independent of the basin under investigation. Whereas geographic basins such as the tropical west Pacific and tropical east Pacific are quite dissimilar in time-mean rainfall and vertical latent heating profiles, we find that when a particular regime is observed in one basin, it is quantitatively similar to the same regime observed in the other basin in terms of the multi-modal distribution of storm top heights, rainfall rate distributions, and multi-modal vertical profiles of latent and radiative heating. These results can be used in evaluating the vertical diabatic heating profiles and cloud type distributions corresponding to specific precipitation regimes in model simulations beyond a geographic temporal-mean comparison.
A13B-06
FREE THERMAL CONVECTION INSIDE A STABLY STRATIFIED FLUID:A STUDY BY MEANS OF THREE DIMENSIONAL PARTICLE TRACKING VELOCIMETRY
Free thermal convection refers to the motion of vertical turbulent plumes or domes, which can occur when, an initially in-rest stratified fluid, is submitted to buoyancy forces, caused by a permanent perturbation associated to a heat transfer mechanism. When a fluid, in equilibrium, is stably stratified the external forcing can produce an unstable configuration ensuing the increasing in amplitude of internal waves, and, if it has strength enough, it can definitely erode the stratification, involving an increasing thickness of fluid volume. The entrainment phenomenon justifies the penetrative feature of convection and causes the growth of a convective boundary layer of well mixed fluid (Convective Mixing Layer) against the adjacent stable stratified layer. The non-steady phenomenon of penetrative convection in a stably stratified fluid has been reproduced in laboratory employing a tank filled with water and subjected to heating from below. The goal in the experiment is predicting the convective boundary layer growth as a function of initial and boundary conditions and describing the fate of a tracer dissolved in the fluid phase. The motivations of the research are mostly related to its connections to environmental topics. In nature the dynamics of penetrative convection influences the transport and mixing features of stratified fluids, playing a fundamental role in characterizing and forecasting the distribution of chemical species, with implication for water or air quality in the upper oceans and lakes or in the lower troposphere. When studying turbulent convective phenomenon, dispersion is mostly due to transport by large organized structures while molecular diffusion can be neglected. The knowledge of the horizontal and vertical extension of the structures dominating the flow field appears to be mandatory. In order to better understanding and likely describing the evolution of turbulent structures inside the convective layer, a fully three dimensional experimental technique is required. The equipment employed is suitable for simultaneously providing temperatures inside the domain through thermocouples and Lagrangian particle trajectories obtained by using a 3D-PTV technique. The combined use of a vertical array of thermocouples and 3D-PTV allows, simultaneously, profiling temperature and the 3D velocity components. A properly calibrated stereoscopic system of three monochrome 25 fps CCD cameras has been employed. The combination of image and object space based information is applied to establish the spatio-temporal correspondences between particle position of consecutive time steps, resulting in the reconstruction of 3D trajectories. The vertical dimension of convective structures is associated to the mixing layer height, detected both employing temperature data and statistics of the velocity field. On the other hand, the spatial correlation of the velocity field, providing the plume horizontal dimension, allows the horizontal extension of the mixing region to be determined. This information coupled to the knowledge of the mixing layer height allows the spatial extension of the convective region to be fully described.
A13B-07
Convective Sources of Gravity Waves from US High Resolution Radiosonde Data
Expressions derived from gravity wave polarization relations indicate the energy of vertical velocity fluctuations, VE, should be more sensitive to higher frequency waves than the kinetic energy of the horizontal velocities, KE, or the potential energy of the temperature fluctuations, PE. It is demonstrated that these different gravity wave energies derived from high vertical-resolution radiosonde soundings have different annual variations, with VE in the troposphere having clear summer mid-latitude maxima that are not seen in KE or PE in the troposphere or in any of these energies in the lower stratosphere. Furthermore, it is shown that VE 00Z-12Z differences in the troposphere have the same summer pattern as that implied by the diurnal variations in convective activity over the US. In contrast to mid-latitudes, there is a clear signature of VE responding to deep convection in the tropics in both the troposphere and lower stratosphere. This is interpreted with the aid of GROGRAT simulations of gravity wave propagation. Large values of VE during winter seem to be largely a result of spontaneous adjustment.
A13B-08
Atmospheric Wind Spectral Gap: Comparison between Fourier and Wavelet Spectra
The wind speed spectrum, observed in atmospheric measures, shows a local maximum at temporal scales of the order of more than one day, and another of at temporal scales of less than an hour. There is a gap between them that indicates that energy does not flow from large to smaller scales. Reynolds averaging, which is a fundamental concept to evaluate turbulent transports from the theoretical point of view, can be applied to observational tests. A set of measurements taken near the city of Rio Grande, in a very homogeneous and plan, with a fecth of more than one kilometer in all directions, was used in this study. The measures of wind speed were performed with a triaxial sonic anemometer with frequency of 20 Hz at a height of 8 m, and the period from 18 to 23 October 2006 was selected. The tower it is located on the banks of an irrigation canal and it is located between two lakes, the Mirim in the south, and Patos, to the north, and the Atlantic Ocean in the east. The Fourier transform was applied to the data and the result was submitted to an average for the frequency bands. There is a maximum period of about 30 hours, with a partial second maximum around 10 hours, a minimum around 1 hour and a third maximum around 20 minutes. This interval of time was also obtained in another study, which indicated that this was the interval most appropriate for performing means. The variance of the wavelet transform identified the same maxima. However, obtaining these later results are easier to interpret, even though the algorithm for calculating it contains the Fourier transform and, therefore, has the same restrictions.