Abstract

This paper presents a study on air pollutant dispersion from an elevated accidental release from the space shuttle tower at the Kennedy Space Center in Florida under the influence of a stratified onshore flow. The temperature difference between land and ocean can generate a local sea-land circulation and a thermal internal boundary layer. Both play a significant role in the coastal dispersion. Results from a Gaussian dispersion model and those from numerical simulations show that the concentrations obtained from these two distinctly different methods are of the same order of magnitude and have similar patterns. Numerical simulations were performed by combining the Advanced Regional Prediction System with an Eulerian pollutant dispersion model. Numerical sensitivity experiments were conducted to investigate the effects of upwind stability, coastal topography, and calm wind condition. Numerical results also show that the dispersion pattern from a continuous release is significantly different from that of a finite release.

Abstract

Land-use patterns are a major factor that causes land surface heterogeneities, which in turn influence the development of mesoscale circulations. In the present study, effects of land-use patterns on the formation and structure of mesoscale circulations were investigated using the North Carolina State University mesoscale model linked with the soil–vegetation system. The Midwest type of low-level jet (LLJ) was successfully generated in the model simulation. Characteristics of the LLJ generated in the numerical experiments are consistent with observations. The results suggest that land surface heterogeneities could have significant impacts on the formation and the maintenance of the LLJ.

Abstract

Stomatal resistance (R _s ) calculation has a major impact on the surface energy partitioning that influences diverse boundary layer processes. Present operational limited area or mesoscale models have the Jarvis-type parameterization, whereas the microscale and the climate simulation models prefer physiological schemes for estimating R _s . The pivotal question regarding operational mesoscale models is whether an iterative physiological scheme needs to be adopted ahead of the analytical Jarvis-type formulation.

This question is addressed by comparing the ability of three physiological schemes along with a typical Jarvis-type scheme for predicting R _s using observations made during FIFE. The data used is typical of a C4-type vegetation, predominant in regions of high convective activity such as the semiarid Tropics and the southern United States grasslands. Data from three different intensive field campaigns are analyzed to account for vegetation and hydrological diversity.

It is found that the Jarvis-type approach has low variance in the outcome due to a poor feedback for the ambient changes. The physiological models, on the other hand, are found to be quite responsive to the external environment. All three physiological schemes have a similar performance qualitatively, which suggests that the vapor pressure deficit approach or the relative humidity descriptor used in the physiological schemes may not yield different results for routine meteorological applications. For the data considered, the physiological schemes had a consistently better performance compared to the Jarvis-type scheme in predicting R _s outcome. All four schemes can, however, provide a reasonable estimate of the ensemble mean of the samples considered. A significant influence of the seasonal change in the minimum R _s in the Jarvis-type scheme was also noticed, which suggests the use of nitrogen-based information for improving the performance of the Jarvis-type scheme. A possible interactive influence of soil moisture on the capabilities of the four schemes is also discussed. Overall, the physiological schemes performed better under higher moisture availability.

Abstract

Saturation pressure differences, a measure of parcel saturation, are calculated from upper-air soundings and compared to manual surface observations of cloudiness. The saturation pressure level p ^* (more commonly referred to as the lifted condensation level, LCL), can be calculated for each level in a sounding using the temperature and dewpoint temperatures. Thus, p ^* of an unsaturated air parcel is found by dry-adiabatic ascent to the pressure level where the parcel is just saturated. The difference between air parcel pressure and saturation pressure level defines the parcel saturation pressure difference. The mean saturation pressure difference between 1000 and 700, 700 and 400, and 400 and 300 mb is calculated and compared to the observed composite cloudiness for those layers. Results indicate that as the absolute value of saturation pressure difference decreases toward zero, the resulting ground observed composite cloud amount increases. However, the mean saturation pressure difference for high clouds ranges from 64 mb under clear skies to 16 mb for overcast conditions. This corresponds to relative humidities between 25% and 76%. Most previous studies do not indicate such large cloud amounts at these humidities. Three empirical relationships that define low, middle, and high clouds are developed based on one year of comparisons. These relationships are then tested on an independent dataset that include a wide variety of cloud cover conditions. Qualitative comparisons are made to manual observations of cloudiness and indicate that the relationships overall slightly overestimate the frequency of cloudiness. Cloudiness derived from the Visible-Infrared Spin Scan Radiometer (VISSR) Atmospheric Sounder (VAS) onboard the Geostationary Environmental Operational Satellite (GOES) 7 using the CO₂ slicing technique is also compared to surface observations. Results indicate that the satellite-derived cloudiness overestimates cloudiness compared with surface observations but is also very similar to the saturation pressure difference estimates.

Abstract

Cloudiness derived from surface observations and the Geostationary Operational Environmental Satellite VISSR (Visible–Infrared Spin Scan Radiometer) Atmospheric Sounder (VAS) are compared with thermodynamic properties derived from upper-air soundings over the Gulf Stream locale during a developing winter storm. The Gulf Stream locale covers the United States mid-Atlantic coastal states, the Gulf Stream, and portions of the Sargasso Sea. Cloudiness is found quite frequently in this region. Cloud-top pressures are derived from VAS using the CO₂ slicing technique and a simple threshold procedure. Cloud-base heights and cloud fractions are obtained from National Weather Service hourly reporting stations. The saturation pressure differences, defined as the difference between air parcel pressure and saturation-level pressure (lifted condensation level), are derived from upper-air soundings. Collocated comparisons with VAS and surface observations are also made. Results indicate that cloudiness is observed nearly all of the time during the 6-day period, well above the 8-yr mean. High, middle, and low opaque cloudiness are found approximately equally. Furthermore, of the high- and midlevel cloudiness observed, a considerable amount is determined to be semitransparent to terrestrial radiation. Comparisons of satellite-inferred cloudiness with surface observations indicate that the satellite can complement surface observations of cloud cover, particularly above 700 mb.

Surface-observed cloudiness is segregated according to a composite cloud fraction and compared to the mean saturation pressure difference for a 1000–600-mb layer. The analysis suggests that this conserved variable may be a good indicator for estimating cloud fraction. Large negative values of saturation pressure difference correlate highly with clear skies, while those approaching zero correlate with overcast conditions. Scattered and broken cloud fractions are associated with increasing values of the saturation pressure difference. Furthermore, cloud fractions observed in this study are considerably higher than those reported in similar studies and by other cloud fraction formulations.

Abstract

Fields of cloudiness derived from the Geostationary Operational Environmental Satellite VISSR (Visible–Infrared Spin Scan Radiometer) Atmospheric Sounder are analyzed over the Gulf Stream locale (GSL) to investigate seasonal and geographical variations. The GSL in this study is defined as the region bounded from 31° to 38°N and 82° to 66°W. This region covers an area that includes the United States mid-Atlantic coast states, the Gulf Stream, and portions of the Sargasso Sea. Clouds over the GSL are found approximately three-quarters of the time between 1985 and 1993. However, large seasonal variations in the frequency of cloudiness exist. These seasonal variations show a distinct relationship to gradients in sea surface temperature (SST). For example, during winter when large SST gradients are present, large gradients in cloudiness are found. Clouds are observed least often during summer over the ocean portion of the GSL. This minimum coincides with an increase in atmospheric stability due to large-scale subsidence. Cloudiness is also found over the GSL in response to mesoscale convergence areas induced by sea surface temperature gradients. Geographical variations in cloudiness are found to be related to the meteorology of the region. During periods of cold-air advection, which are found most frequently in winter, clouds are found less often between the coastline and the core of the Gulf Stream and more often over the Sargasso Sea. During cyclogenesis, large cloud shields often develop and cover the entire domain.

Satellite estimates of cloudiness are found to be least reliable over land at night during the cold months. In these situations, the cloud retrieval algorithm often mistakes clear sky for low clouds. Satellite-derived cloudiness over land is compared with daytime surface observations of cloudiness. Results indicate that retrieved cloudiness agrees well with surface observations. Relative humidity fields taken from global analyses are compared with satellite cloud heights at three levels in the atmosphere. Cloudiness observed at these levels is found at relative humidities in the 75%–100% range but is also observed at humidities as low as 26%.

Abstract

This paper documents evidence of a diurnal variability in cloudiness over the Gulf Stream locale. The Gulf Stream locale (GSL) is defined as the region covering 31°–38°N, 82°–71°W. The Gulf Stream, which occupies a portion of the GSL, is a warm current of water that flows south to north along the east coast of the United States and provides conditions conducive for the development of cloudiness. Cloud heights derived from the GOES VISSR (Visible-infrared Spin Scan Radiometer) Atmospheric Sounder (VAS) are obtained and used to produce a 7-yr climatology of the diurnal variation in the frequency of low-, middle-,and high-level cloudiness. The climatology is segregated into summer and winter seasons.

Diurnal variations are found during the summer and winter. Satellite observations over land indicate a maximum in the frequency of low cloudiness during daytime and a minimum at night. In addition, high cloudiness is found to increase significantly late in the afternoon and evening. Over the Gulf Stream region, high cloudiness is found most frequently in the mid- to late morning hours. A midafternoon maximum in low cloudiness is found along the coastline of Georgia and South Carolina and north of the Gulf Stream east of Virginia. Nocturnal minimums in low cloudiness are reported in these regions. Results suggest that summertime low and high cloudiness over the GSL are related to prevalent convective activity. An analysis of the diurnally oscillating pattern of boundary layer convergence, derived from analyses from the National Meteorological Center's step coordinate model, indicates a strong relationship to the presence of high cloudiness. The strong correspondence between the timing of these two parameters suggests that atmosphere dynamics play a significant role in the diurnal cycle in high cloudiness.

In winter, when convective activity is suppressed there is less detectable response of the atmosphere to the 24-h solar cycle manifest in the diurnal variations of clouds. Nevertheless low- and midlevel cloudiness are found most frequently in the predawn hours, except over the Gulf Stream where low clouds exhibit an afternoon maximum and a nocturnal minimum. Surface observations of cloudiness support the diurnal variations reported by VAS.

Abstract

Stomatal resistance (R _s ) forms a pivotal component of the surface energy budget and of the terrestrial biosphere–atmosphere interactions. Using a statistical–graphical technique, the R _s -related interactions between different atmospheric and physiological variables are resolved explicitly from observations made during the First ISLSCP (International Satellite Land Surface Climatology Project) Field Experiment (FIFE). A similar analysis was undertaken for the R _s parameterization schemes, as used in the present models. Three physiological schemes (the Ball–Woodrow–Berry, Kim and Verma, and Jacobs) and one operational Jarvis-type scheme were evaluated in terms of their ability to replicate the terrestrial biosphere–atmosphere interactions.

It was found that all of the R _s parameterization schemes have similar qualitative behavior for routine meteorological applications (without carbon assimilation). Compared to the observations, there was no significant difference found in employing either the relative humidity or the vapor pressure deficit as the humidity descriptor in the analysis. Overall, the relative humidity–based interactions were more linear than the vapor pressure deficit and hence could be considered more convenient in the scaling exercises. It was found that with high photosynthesis rates, all of the schemes had similar behavior. It was found with low assimilation rates, however, that the discrepancies and nonlinearity in the interactions, as well as the uncertainties, were exaggerated.

Introduction of CO₂ into the analysis created a different dimension to the problem. It was found that for CO₂-based studies, the outcome had high uncertainty, as the interactions were nonlinear and the schemes could not converge onto a single interpretive scenario. This study highlights the secondary or indirect effects, and the interactions are crucial prior to evaluation of the climate and terrestrial biosphere–related changes even in the boundary layer perspective. Overall, it was found that direct and indirect effects could lead the system convergence toward different scenarios and have to be explicitly considered for atmospheric applications at all scales.

Abstract

Variable vegetation cover is a possible trigger for convection, especially in semiarid areas due to differential surface forcing. A two-dimensional numerical model with explicit cloud physics and a detailed vegetation parameterization scheme is used to investigate the role of vegetation differences in triggering convective cloud formation. The ground surface in all simulations includes two irrigated vegetation areas with a dry steppe in the center of the domain. The effects of atmospheric stability, ambient moisture profile, and horizontal heating scale are investigated.

Atmospheric stability controls the growth of convective circulations. Thermal circulations form at the interfaces between the vegetated areas and the dry steppe. In the more stable environment, two distinct convective cells persist; they merge into one cell in the less stable cases. The existence of low-level moisture controls the timing and persistence of clouds that form. An interesting result is the earlier dissipation of clouds in less stable cases, as greater mixing with drier air from aloft leads to the dilution of the cloud water. Since the largest thermal forcing exists at the interfaces, length of the steppe interacts with the stability to control the merger of the cells. The two cells merge quickly into one for narrow horizontal heating. For the widest heating scale studied, no merger occurs.

Abstract

Current land surface schemes used for mesoscale weather forecast models use the Jarvis-type stomatal resistance formulations for representing the vegetation transpiration processes. The Jarvis scheme, however, despite its robustness, needs significant tuning of the hypothetical minimum-stomatal resistance term to simulate surface energy balances. In this study, the authors show that the Jarvis-type stomatal resistance/transpiration model can be efficiently replaced in a coupled land–atmosphere model with a photosynthesis-based scheme and still achieve dynamically consistent results. To demonstrate this transformative potential, the authors developed and coupled a photosynthesis, gas exchange–based surface evapotranspiration model (GEM) as a land surface scheme for mesoscale weather forecasting model applications. The GEM was dynamically coupled with a prognostic soil moisture–soil temperature model and an atmospheric boundary layer (ABL) model. This coupled system was then validated over different natural surfaces including temperate C4 vegetation (prairie grass and corn field) and C3 vegetation (soybean, fallow, and hardwood forest) under contrasting surface conditions (such as different soil moisture and leaf area index). Results indicated that the coupled model was able to realistically simulate the surface fluxes and the boundary layer characteristics over different landscapes. The surface energy fluxes, particularly for latent heat, are typically within 10%–20% of the observations without any tuning of the biophysical–vegetation characteristics, and the response to the changes in the surface characteristics is consistent with observations and theory. This result shows that photosynthesis-based transpiration/stomatal resistance models such as GEM, despite various complexities, can be applied for mesoscale weather forecasting applications. Future efforts for understanding the different scaling parameterizations and for correcting errors for low soil moisture and/or wilting vegetation conditions are necessary to improve model performance. Results from this study suggest that the GEM approach using the photosynthesis-based soil vegetation atmosphere transfer (SVAT) scheme is thus superior to the Jarvis-based approaches. Currently GEM is being implemented within the Noah land surface model for the community Weather Research and Forecasting (WRF) Advanced Research Version Modeling System (ARW) and the NCAR high-resolution land data assimilation system (HRLDAS), and validation is under way.