Land/Atmosphere Interactions
The goal of this research is to improve the numerical representation of the energy and water exchanges occurring at the land-air interface. Modeling these exchange processes involves state-of-the-art, interdisciplinary representations of soil physics, hydrology, vegetation physiology, radiation exchange, turbulence and boundary layer theory, and non-precipitating cloud physical and radiative processes. This research is critical for climate models and data assimilation systems, since they are most sensitive to errors in surface fluxes. This motivated the development of the PLACE model (Parameterization for Land-Atmosphere-Cloud Exchange), a detailed-interactive process model that contains a partly cloudy atmospheric boundary layer and underlying heterogeneous land surfaces. PLACE has been implemented into both cloud and regional-scale models for experiments related to major international field programs. An intercomparison of two dozen models has demonstrated that PLACE is one of the more stable and accurate models.
The PLACE annual mean root zone soil wetness (dimensionless) for the year 1987 is shown in Figure7.Soil wetness
is the ratio of the computed soil water content to the soil storage capacity, and is strongly correlated with the surface latent heat
flux through transpiration from plants and evaporation from bare soil. Soil water content is a function of atmospheric forcing,
vegetation cover, and soil texture. Fixed and time varying 2D fields of these parameters were obtained from the Global Energy
Cycle Experiments International Satellite Land-Surface Climatology Project (GEWEX ISLSCP) CD collection.
Simulations of Convective Systems
A key component of the hydrological cycle is the precipitation that falls from convective cloud systems in the Tropics. Tropical precipitation accounts for about two-thirds of the global precipitation and its associated latent heat release is the major energy source driving the hydrological cycle and the large-scale (global) circulation of the atmosphere. Clouds associated with convective systems also play a major role in the Earth's radiation budget. The modeling research tasks are to (1) study cloud processes in support of major NASA missions, such as TRMM and EOS; (2) evaluate and improve the representation of clouds and cloud systems in climate and global circulation models; and (3) understand and quantify the cloud-radiation forcing and its dynamical feedback over tropical and midlatitude regions. The results from this research link directly to the GEWEX Cloud System Study (GCSS) because of their contribution towards improved representation of convective processes and their interaction with radiative processes in GCMs and climate models. A state-of-the-art microphysical scheme was developed that implemented the following improvements: (1) calculation of number concentrations of ice crystals, snow, graupel, and frozen drops/hail; (2) differentiation between source mechanisms for graupel and frozen drops/hail (because of the crucial differences in their characteristic fall speeds and densities;) and (3) improved representation of primary ice nucleation mechanisms, aggregation, ice-multiplication processes, and complex conversion processes between different hydrometeor species. Organized squall systems in widely contrasting large-scale environments were successfully simulated in the GCE model using the new scheme. Simulated radar and hydrometeor struct ures were much more consistent with the observations, when compared to those produced using other microphysical parameterizations (Figure 8). It was found that the rate of ice initiation and subsequent laciation of convection was strongly dependent on the strength of the updrafts at middle levels. Weak updrafts over the tropical oceans glaciated rapidly as a result of rimed ice residing for long periods of time in a region between -3oC and -8oC, where numerous ice splinters are produced.
The GCE model has been used to simulate a TOGA COARE convective system 
(Figure 9), and the results are in good agreement with radar observations. The 4-D data sets from this GCE model
simulation have been used for testing and improving the performance of multi-frequency passive-microwave and spaceborne radar algorithms for TRMM.
Water budgets of cloud systems simulated in a variety of environments over the tropical oceans indicate that warm rain processes are dominant
in the convective region. Several definitions of precipitation efficiency (PE) were investigated using GCE model results. Storms
that had erect updrafts and formed in moist ambient conditions were the most efficient.
Regional-scale Modeling
Regional-scale Modeling
This research improves our understanding of the dynamical and physical processes governing the formation, maintenance, and dissipation of various types of mesoscale circulation systems that produce precipitation. Branch scientists have placed a strong emphasis upon combined theoretical, numerical, and observational investigations of the three-dimensional structures, evolution, and dynamics of midlatitude frontal rainbands, tropical cyclones, super cloud clusters (SCCs) and their associated westerly wind bursts (WWB). The impact of these mesoscale circulation systems upon global weather patterns and regional climate variations can be studied using a regional scale model (MM5) for long-time integrations. MM5 has been used to study SCCs that developed over the west Pacific warm pool region. The model results are in good agreement with observations, namely an explosive development of the SCCs and their associated WWBs that propagate over the equatorial warm pool region (Figure 10).
The MM5 model-derived winds, fresh water fluxes and radiative
budgets associated with WWBs can be used as the input to an ocean mixed-layer model. In this way we can assess the impact
of the warm sea surface temperature scenario and its impact on climate change.
MM5 has been used to successfully simulate the development of a tropical cyclone (Hurricane Florence, 9-10
September 1988). Model sensitivity to initial conditions, four-dimensional assimilation of satellite-derived rainfall rates, and
cumulus parameterization scheme was demonstrated. The results suggest that better (enhanced) initial conditions will more
successfully simulate the rapid development of the hurricane. The results also suggest that the dynamic data assimilation of
satellite-derived rainfall rates is very sensitive to the cumulus parameterization scheme (i.e. Betts-Miller and Kuo-Anthes) used
in MM5.
Visualization of model results using Vis5D and GEMPAK greatly enhanced the perception of the 3-D dynamics and the
structure of the Hurricane Florence. A Vis5D snapshot is shown in Figure 11. Theisosurfaces of potential vorticity, cloud water
(not shown) and rain water (light green) clearly illustrate the 3-D structure of a precipitating system over the Gulf of Mexico.
Trajectories released from various levels vividly describe the spiral nature of the flow near the center of the hurricane (red,
yellow) and the location of the upper level jet (blue).
Atmospheric Transport and Chemistry
The purpose of this research is to understand the relationships between convective systems and the tropospheric distribution of trace species. Clouds, precipitation and water vapor play important roles in the distributions of tropospheric ozone and related trace gases. Our specific goals are to use numerical models to quantify the vertical transport effects of clouds and organized convection on important trace gases (i.e., ozone [O3], carbon monoxide [CO], hydrocarbons [HC], and nitrogen oxides [NOx]). A direct linkage between the GCE model and a chemistry model is currently being initiated. Both GCE and MM5 model-generated wind fields were used to redistribute the initial concentrations of CO (measured by aircraft), which were assumed to act as conserved tracers during the period of convective mixing. The degree of vertical redistribution (or overturning) of a trace constituent is indicative of the intensity of the convection, and reflects the transport structure responsible for the mixing
(Figure 12, upper right panel). These modeling results clearly indicated that man-made
environmental change (i.e., biomass burning) in the lower troposphere could only be transported to the upper troposphere and
lower stratosphere by individual deep convective cells associated with mesoscale precipitation events (Figure 12, two panels
on left). Significant downward transport from the mid-troposphere to lower-troposphere can also take place due to convective
and mesoscale downdrafts (Figure 12, lower right panel).
For futher information consult Modeling Group Homepage.
