In order to determine the influence of various factors on flash flooding, six case studies during which flash flooding occurred across central North Carolina are examined: 1) 26 August 2002, 2) 11 October 2002, 3) 9-10 April 2003, 4) 16 June 2003, 5) 29 July 2003, and 6) 9 August 2003. Utilizing stream gage data from the United States Geological Survey combined with radar-estimated precipitation from the Weather Surveillance Radar-1988 Doppler (WSR-88D) KRAX near Clayton, NC, several statistical conclusions are drawn. These conclusions are based on relationships between the inputs ? rain rate and precipitation amount ? to the stream responses: the amount of time between when the stream began its rise and when the maximum stage was reached, the amount of time between the onset of precipitation and the initial response of the stream, the maximum stage reached, the change in height of the stream, and the rate of change of height of the stream. Results indicate that precipitation rate and amount tend to dominate the influence of stream response; however, in many situations, land-surface characteristics play an important role. The notable situations where precipitation rate and amount do not dominate are along the major rivers, in locations with sandy soils where infiltration is high, and in urban areas, where runoff occurs rapidly and streams thus respond quickly regardless of precipitation rate or amount. In addition, rain rate and precipitation amount do not necessarily have similar relationships with the stream response variables; rain rate has a stronger correlation with rate of change of stream rise, while precipitation amount has a stronger correlation with change in stream height. However, it is not enough to study rainfall rates and precipitation amounts if a flash flood warning is to be issued. The results of this research show that there is value and necessity in understanding the role of land-surface characteristics when determining if flash flooding is going to occur.

Ammonia has recently gained importance for its increasing atmospheric concentrations and its role in the formation of aerosols. Studies have shown increasing atmospheric concentration levels of NH3 and NH4+, especially in the regions of concentrated animal feeding operations. Atmospheric inputs of reduced nitrogen as ammonia and ammonium by dry and wet deposition may represent a substantial contribution to the acidification of semi natural ecosystems and could also affect sensitive coastal ecosystems and estuaries. The anaerobic lagoon and spray method, commonly used for waste storage and disposal in confined animal feeding operations (CAFO), is a significant source of ammonia emissions. An accurate emission model for ammonia from aqueous surfaces can help in the development of emission factors. Study of dispersion and dry deposition patterns of ammonia downwind of a hog farm will help us to understand how much ammonia gets dry deposited near the farm, and how remaining ammonia gets transported farther away. An experimental and modeling study is conducted of emissions, dispersion and dry deposition of ammonia taking one swine farm as a unit. Measurements of ammonia flux were made at 11 swine facilities in North Carolina using dynamic flow-through chamber system over the anaerobic waste treatment lagoons. Continuous measurements of ammonia flux, meteorological and lagoon parameters were made for 8-10 days at each farm during each of the warm and cold seasons. Ammonia concentrations were continuously measured in the chamber placed over the lagoon using a Thermo Environmental Instrument Incorporated (TECO) Model 17c chemiluminescnce ammonia analyzer. A similar ammonia analyzer was used to measure ammonia concentrations at selected locations on the farm. Barn emissions were measured using open-path Fourier transform infrared (OP-FTIR) spectroscopy. A 10 m meteorological tower was erected at each site to measure wind speed and direction, temperature, relative humidity and solar radiation. Data collected from field measurements made at hog waste lagoons in south eastern North Carolina, using the flow through dynamic chamber technique, were used to evaluate the Coupled Mass Transfer and Chemical Reactions model and Equilibrium model. Sensitivity analysis shows that ammonia flux increases exponentially with lagoon temperature and pH, but a linear increase was observed with an increase in total ammoniacal nitrogen (TAN). Ammonia flux also shows a nonlinear increase with increasing wind speed. Observed ammonia fluxes were generally lower in the cold season than in the warm season when lagoon temperatures are higher. About 41% of the Equilibrium model predictions and 43% of the Coupled model predictions are found to be within a factor of two of the observed fluxes. Several model performance statistics were used to evaluate the performance of the two models against the observed flux data. These indicate that the simpler Equilibrium model does as well as the Coupled model. The possible effects of the ?artificial? environment within the chamber, which is different from that in the ambient atmospheric conditions above the open lagoon surface, on the measured fluxes are also recognized. Actual layout of barns and lagoons on the farms was used to simulate dry deposition downwind of the farm. Dry deposition velocity, dispersion and dry deposition of ammonia were studied over different seasons and under different stability conditions. Dry deposition velocities were underpredicted by AERMOD when compared with observed dry deposition velocities. Dry deposition velocities were the highest under near neutral conditions and lowest under stable conditions. The highest deposition at short range occurs under nighttime stable conditions and the lowest deposition occurs during daytime unstable conditions. Significant differences in model predicted depositions over crop and grass surfaces are found under stable conditions. Wind orientation at the farm can also affect deposition of ammonia downwind of the farm.

This paper presents results on the rates and spatiotemporal dynamics of the coupled water flux (v) and nitrogen (N) fluxes (mainly fNO3 and fDON for nitrate and dissolved organic N) through a streambed in an agricultural watershed in North Carolina. Physical and chemical variables were measured at numerous points in the streambed of a 0.26 km reach: hydraulic conductivity (K) and head gradient (J), and concentrations of NO3- and other N species in the streambed groundwater, from which water (v=KJ) and N fluxes (e.g., fNO3=v[NO3-]) through the streambed were computed, mapped, and integrated in space. The result was a novel set of streambed maps of the linked variables (K, J, v, N concentrations and fluxes), showing their spatial variability and how it varied over a year (based on 7 bimonthly sets of maps). Mean fNO3 during the study year was 154 mmol m-2 day-1; this NO3- flux, together with that of DON (fDON = 17 mmol m-2 day-1) accounted for >99% of the total dissolved N flux through the streambed. Repeat measurements at the same locations on the streambed show significant temporal variability in fNO3, largely controlled by changes in v rather than changes in [NO3-]. One of the clearest and most temporally-persistent aspects of spatial variability was lateral variability across the channel from bank to bank. K and v had âcenter-highâ patterns (greater values in the center of the channel); this distribution of K (ultimately a reflection of sediment dynamics in the channel) apparently focuses groundwater discharge toward the center of the channel. The opposite âcenter-lowâ pattern was found for J, [NO3-], and (to a lesser extent) fNO3. Contrary to suggestions in some prior work, J was not a good index for v. fNO3 was characterized by localized zones of high and low values that changed in size and shape over time but remained in basically the same locations (the same was true of K, J, [NO3-], though less so for v), with 70% of NO3- flux occurring through about 38% of the streambed area. Lateral distributions of the physical hydrologic attributes (K, J, v) were highly symmetrical across the channel, while those of [NO3-] and fNO3 showed higher values on the right than left, likely a reflection of different N use on opposite sides of the stream. These and other results show the streambed-based approach taken here can offer a number of insights not possible with reach mass-balance approaches in which net exchange between a stream reach and surrounding groundwater is calculated using surface water data.

Frequent rainfall during the winter months in the Portland, Oregon region is associated with extratropical cyclones modified by the Coastal and Cascade Ranges. Operational WSR-88D radar observations from Portland, OR and upper-air soundings from Salem, OR over a 3-year period (2003-2006) from 1 November ? 31 March are used to determine a 3D climatology of winter storms. 84 % of the 117 storm events had a low-level wind direction from the south or southwest, between 158° - 248° azimuth. Stability varied between storms, with most storms being neutral to slightly stable. Wind direction was found to be more important in determining the geographic pattern of precipitation in the PNW. For S-SW flow storms, increasing the storm volume is primarily related to increasing precipitation frequency rather than precipitation areal coverage. Local maximum in precipitation frequency is seen typically at mid-windward slope rather than at the Cascade Range crest. 3D radar observations were also compared to MM5 output for the 2005-06 and 2006-07 winter seasons. Storms were grouped by their prevailing low-level wind direction and two individual cases (2005 Dec 29-31; 2006 Nov 6-7) to compare their radial velocity, precipitation frequency, and standard deviation of radial velocity. Errors were found in the standard deviation of Vr, with the model showing more variable wind speed and direction than the observations. The spatial pattern of precipitation frequency between the radar observations and model output were found to be similar, but the magnitudes were found to usually be larger in the model output.

Two low-level convectively-induced turbulence (CIT) events east of the Appalachian Mountains are investigated utilizing observations, satellite, radar, and numerical simulations. Both events had an inordinate amount of low-level turbulence reported, but one event had more than twice as many severe or greater reports. The events were compared?to include the 72 hours leading up to the turbulence reports?and similarities and differences at the various scales from the synoptic to meso-alpha, meso-beta, meso-gamma, and microscale are noted. The case of weaker turbulence featured a meridional wave pattern with ridging over the East Coast and a single upper-level jet closely coupled with the large-scale frontal system. The stronger turbulence case possessed a zonal wave pattern with a vortex over eastern Canada and both a polar jet and subtropical jet. These differences are reflected in the low-level temperature and potential vorticity patterns and affected the hydraulic structures as well?with the stronger turbulence environment more prone to a blocking-type regime. Hydrostatic mountain waves were observed for both events. Stronger cross-mountain flow combined with a strong low-level leeside inversion resulted in a more vigorous mountain wave with a stronger downstream isentropic upfold (mid-level cold pool) in the stronger turbulence event. This mid-level cold pool was deformed by the large-scale jet resulting in a mid-level cold front (downstream from the surface cold front), surface pressure rises to the lee of the Allegheny Mountains, and ultimately a surface cold surge (edgewave) that merges with warm air from the south. The phasing of the mid-level cold pool and the convergence with the northerly cold surge and southerly warm air results in kata-frontogenesis and cellular convection that transits the severe turbulence location in space and time. Convection in the weaker turbulence case was lineal in structure and tied to the large-scale cold anafront. Vorticity, enstrophy, turbulent kinetic energy, and Richardson number analyses indicated maxima were lineal in structure and upstream from the convection in the weaker case and arc-like in appearance and downstream from convection in the stronger case. A turbulence index was formulated based on three-dimensional vorticity (enstrophy), vertical wind shear, and static stability.

The main task of this study focuses on studying the effect of wave-current interaction on currents, storm surge and inundation as well as effects of depth-induced wave breaking, wind field and current on waves by using numerical models. The results show that it is important to incorporate the wave-current interaction effect into coastal circulation, storm surge and inundation models. At the same time, it should consider effects of depth-induced wave breaking, wind field, currents and sea surface elevation in prediction of waves. Specially, we found that: (1) Depth-induced wave breaking plays an important role in wave field in shallow water areas; (2) To more properly model the hurricane induced wave field, it is important to consider the asymmetric structure of the hurricane wind field, the changes in the hurricane translation time history, and the incorporation background wind field into hurricane wind field; (3) For SWH, it will be decreased when waves propagate in the following current direction. On the other hand, current will increase the SWH when waves propagate countercurrent direction. The change of wave propagation direction after crossing Gulf Stream depends not only on refraction, but also on others (e.g. trap waves effect). (4) It is important to introduce wave-current effects into any storm surge and inundation prediction modeling system. Specially, the consideration of wave-induced wind stress, bottom shear stress, and 3-D radiation stress in storm surge and inundation modeling can significantly improve the correctness of the prediction.

On 30 March 2006 a convective episode occurred featuring isolated supercells, a mesoscale convective system (MCS) with parallel stratiform (PS) precipitation, and an MCS with leading stratiform (LS) precipitation. These three distinct convective modes occurred simulataneously across the same region in eastern Kansas. Multi-modal events are especially challenging for forecasters given the wide range of severe weather threats that accompany the different modes. In order to better understand the mechanisms that govern such events, this study examined the 30 March 2006 episode through a combination of an observation-based case study and numerical simulations. From the results of this study we conclude that, for this event, localized environmental variations were largely responsible for the eventual convective mode, with the method of storm initiation having only limited effects. The resultant mode was very sensitive to both the environmental thermodynamic and shear profiles, as variations in either led to different convective modes within the numerical simulations. Finally, we conclude that while the individual modes each developed within an environment distinctly favorable for that mode, they were able to persist in close proximity to one another due to a "middle ground" environment permissive of all three. Strong vertical shear and moderate instability led to the development of supercells in western Oklahoma and similarly strong shear oriented parallel to a surface dryline coupled with dry air in the middle and upper levels led to the development of the PS linear MCS in central Kansas. Meanwhile, moderate wind shear coupled with high instability and strong linear forcing led to the development of the LS MCS in eastern Kansas. Without this linear forcing, the moderate shear environment was supportive of both linear and isolated supercell modes, resulting in the storms that moved into this region maintaining their original organization.

For many species such as the blue crab, Callinectes sapidus, successful estuarine recruitment to juvenile nursery habitats is dependent on the biophysical processes experienced during dispersal of the early life stages. The goal of this study was to determine how blue crab primary (postlarval) and secondary (early juvenile) dispersal occurs within a predominately wind-driven estuary, Pamlico Sound, North Carolina, USA. We (1) characterized circulation patterns in Pamlico Sound during the fall blue crab recruitment months over two consecutive years using current meters (2) sampled during multiple 24 h periods to relate spatiotemporal water column distributions of postlarval and early juveniles blue crabs with circulation patterns, and used a hydrodynamic model to recreate dispersal trajectories from eastern (inlet) to western sound nursery habitats and (3) examined the environmental (wind, diel cycle, tidal phase) and biological (ontogenetic, density-dependent) factors that contribute to early juvenile blue crab secondary dispersal from near-inlet nursery habitats. During our study, surface currents responded synchronously to wind-forcing by generally flowing in the same direction as the wind. Particle-tracking simulations suggested that dispersal from Oregon and Hatteras Inlets to across-sound nursery habitats resulted from the combined use of tidal and wind-driven currents. Simulation results and observed crab distributions further indicated that Oregon Inlet was the primary supplier of postlarval blue crabs (dispersing in surface waters at night) throughout Pamlico Sound, as postlarvae ingressing through Hatteras Inlet were not retained within our study area. Furthermore, Oregon Inlet supplied early juvenile blue crabs (dispersing in bottom waters at night) to northwestern sound habitats, while crabs from Hatteras Inlet dispersed to mid- and eastern-sound regions. Results from our study in near-inlet settlement habitats confirmed the importance of tides to mediating dispersal partway into Pamlico Sound, as early juvenile blue crabs responded to increasing conspecific density in settlement habitats by using flood-tide transport near the inlets to rapidly leave these habitats. Based on our findings, we make recommendations regarding the prioritization of nursery habitats for conservation and fisheries management.

Since the 1960?s several successful missions have been sent to Mars to gain a better understanding of the planet. In 2009, the Mars Science Laboratory (MSL) mission is scheduled to launch as part of the National Aeronautics and Space Administration (NASA) Mars Exploration Program. To assure the safety of this mission, an understanding of the Martian atmosphere is necessary. This is the first mission that may determine the landing site based on weather conditions. As such, potential landing sites at Terby Crater, Melas Chasma, Gale Crater, and Nili Fossae Trough were studied. Due to limited observations of Mars, the Planetary Weather Research and Forecast (WRF) Mars general circulation model was used to represent the Martian atmosphere. Model validation was conducted against Viking Lander 1, Viking Lander 2, and Mars Pathfinder data and showed that the Planetary WRF model was able to reasonably represent the Martian atmosphere. This research is divided into two parts. The first part focuses on density, temperature, and wind profiles at each potential landing site. These profiles are used to determine the amount of variability engineers must account for in the spacecraft design specifications. All profile deviations were within the MSL design specifications. The largest deviations occurred at Terby Crater while the smallest deviations occurred at Nili Fossae Trough. It appears that the large topographic features of Hellas Basin and Valles Marineris affect the local airflow patterns around Terby Crater and Melas Chasma. The second part focuses on these topographically-forced atmospheric perturbations using two of the largest features on Mars, Hellas Basin and Olympus Mons. Vertically propagating waves were generated over Olympus Mons during the night, while a strong daytime convective boundary layer and diabatic heating plume occurred during the day. Hellas Basin was dominated by cyclonic motion throughout the night and vertically propagating waves along the western edge of the basin during the day. In general, the Planetary WRF model compared to conventional mountain wave theory and was able to model topographic disturbances with coarse resolution. Limitations of the model are discussed.

Estuarine habitats serve as important nurseries for many species of juvenile fishes and invertebrates. Due to concerns about declining fishery stocks and habitat degradation, it is critical that the value of different habitats be evaluated, so that areas that are most important can be prioritized for habitat conservation and restoration. This dissertation used Atlantic croaker (Micropogonias undulatus) as a model species to quantitatively examine patterns of early juvenile nursery habitat use, factors influencing growth and mortality, and to test the widely held assumption that ?growth is a valid indicator of habitat quality?. A unique aspect of this study was consideration of how growth and survival of juvenile Atlantic croaker were influenced by the joint effects of individual (larval and juvenile growth) and cohort (density-dependent) characteristics, as well as biotic and abiotic characteristics of the nursery. Late-larval and early juvenile Atlantic croaker were collected weekly, using a 1-m beam trawl in river, upper-estuary, and creek habitats of two estuaries (White Oak River estuary and Newport River estuary, North Carolina, USA), sampled during two seasons (fall and spring) of two recruitment years (2001-2002 and 2002-2003). Subsequent analysis of otolith microstructure was used to provide estimates of growth rates, population age structure, and mortality. Juvenile Atlantic croaker were most abundant in oligohaline river nursery areas (primary nursery area), with the upper-estuary serving as an important ?refuge? habitat following periods of high freshwater outflow that forced fish down-estuary. In addition to management implications (i.e., time and area closures of shrimp trawling), such flood events have important implications for growth-based assessment of habitat quality: A positive covariance between larval and juvenile growth was found only after periods in which croaker were displaced from river nursery areas, when food was apparently in short supply. Atlantic croaker cohorts from both estuaries also exhibited evidence for selective mortality favoring survival of individuals that were faster growing as larvae and early juveniles, as well as evidence for density-dependent growth (i.e., slower average cohort growth rates at higher conspecific density). The results of this study indicate that growth is not a robust indicator of habitat quality. Unless factors such as previous environmental history, density-dependence, and selective mortality are accounted for, growth-based assessment of habitat quality may reach incorrect conclusions regarding the value of different habitats.