Search

A high-resolution, multi-level, primitive equation ocean model is used to examine the response of an idealized, flat-bottom, oceanic regime off northern California to steady, equatorward, local wind-forcing during the upwelling season. The model has open boundaries on all but the eastern coastal boundary on which either free-slip or zero-slip boundary conditions are imposed. Time-invariant winds, either with or without a component of wind stress curl, are used as model forcing to spin-up a classical two-dimensional, upwelling-induced coastal jet and undercurrent. Since no eddies are generated, a stability analysis of the mean flow is conducted which explores both the necessary conditions of mixed (barotropic and baroclinic) instability, through calculations of potential vorticity, and the sufficient conditions for baroclinic instability, through an application of a simple two-layer stability model. Comparisons of model results with observations of the coastal jet in the California Current System indicate that the location and the horizontal and vertical current shear associated with the model coastal jet compare favorably with observations; however the modeled jet is stronger, deeper and wider than the observed jet. Finally, the inclusion of wind stress curl and the zero-slip boundary condition are demonstrated to be important elements in model simulations of the coastal jet.

Baroclinic instability as the largest scale of motion participating in the cross frontal oceanic transport process was the theme. Classical baroclinic instability theory, leading up to recent studies in which the distinctive structure of an ocean front is included. Finite amplitude baroclinic instability in the classical model was discussed with the corresponding effect in the frontal model. Laboratory experiments on baroclinic frontal theory was surveyed. The different kinds of oceanic fronts were surveyed and additional observations were supplied. The smallest scales of motion relevant to the cross-frontal transfer problem were discussed from the oceanic standpoint and from the point of view of laboratory experiments. The main theme was supplemented, as in all previous years, by a diversity of geophysical fluid dynamical subjects, the nature of which can be seen in the abstracts included in this report. Keywords: Geophysical fluid dynamics; Baroclinic instability; and Oceanic fronts., Sponsored in part by Grant NSF-MCS82-00450

The two-layer model baroclinic stability parameter, meridional surface temperature gradients, and monthly mean meridional stationary, transient and total eddy heat transports, computed as functions of latitude and longitude for three individual Januaries, are described and discussed. Correlation analyses for all possible combinations are computed, and relationships between these quantities are discussed. The results indicate that no direct relationship exists between stationary eddy heat transports and baroclinically unstable conditions. However, a direct relationship is found between transient eddy heat transports and baroclinically unstable conditions. For example, the correlation between the transient eddy flux and the two-layer instability parameter is .52, which is statistically significant at the 99% confidence level. However, the strength of the correlation suggests that the degree of baroclinic instability only accounts for some of the variation in the transient eddy heat transport. Apparently, other factors also play an important role in the forcing of transient eddy heat transports. (Author)

Baroclinic eddies with diameters of about 10 km and maximum current speeds of about 50 cm/s have been widely observed in the central Arctic Ocean north of Alaska and Canada. The possible origin of these eddies through an instability of the mean baroclinic flow is investigated using an ocean model with exponential profiles of mean shear and Vaisala frequency. The model includes Ekman pumping at a rigid bottom and at either a free or rigid upper surface. The central Arctic Ocean where the eddies were found is baroclinically stable with no possibility of eddy production. If the eddies are spawned by this mechanism, they must be formed at a site far from where they were observed. On the periphery of the Arctic Ocean north of Alaska the combination of greater current shear, shallower depth and lack of ice cover leads to unstable conditions and the eddies apparently originate in that region. The instability theory predicts maximum velocity at the surface instead of below the surface as observed. Apparently after formation in open water the eddies are advected beneath the ice cover and dissipate the momentum of their upper layers against the ice. This is demonstrated by calculations for the diffusion of vorticity against the ice in the case of an initial exponential profile. A subsurface maximum then develops which resembles the observed profiles. (Author)

AN INITIALIZATION OF A SPECTRAL FORMULATION OF THE PRIMITIVE EQUATIONS USING A DIAGNOSTIC DIVERGENCE IS TESTED FOR A GLOBAL MODEL. The initial conditions are generated from a developing baroclinically unstable wave. A semi-implicit time scheme is developed and tested along with the usual explicit method during the course of the experiments. Results show a relatively small effect of a divergent initialization on the ensuing integrations. The semi-implicit method shows a tendency to smooth out high frequency oscillations in local tendencies. (Author)

An initialization of a spectral formulation of the primitive equations using a diagnostic divergence is tested for a global model. The initial conditions are generated from a developing baroclinically unstable wave. A semi-implicit time scheme is developed and tested along with the usual explicit method during the course of the experiments. Results show a relatively small effect of a divergent initialization on the ensuring integrations. The semi-implicit method shows a tendency to smooth out high frequency oscillations in local tendencies. (Author)

The California Current System, CCS, consists of four separate currents that are spatially and seasonally distributed of the west coast of the United States. They are: 1) the California Current, a southward flowing surface jet; 2) the California Undercurrent, a northward flowing sub-surface jet; 3) the Davidson Current, a poleward flowing surface jet north of Point Conception; and 4) the Southern California Countercurrent, a northward flowing surface current south of Point Conception in the California Bight. Offshore jets or filaments are a well observed phenomenon in the CCS. Their surface and subsurface structures as well as their spatial and temporal variability are well documented. The methods by which these filaments are formed is not as well understood as their structural form. The most common generation theories are: 1) variations in wind stress coupled with topographic irregularities, 2) dynamic instability, and 3) geostrophic turbulence. This thesis attempts to identify the factor(s) responsible for filament formation south of Cape Mendocino by numerically simulating the CCS using a two-layer, non-linear, primitive equation model. It is shown that baroclinic instability is the primary method by which filaments are formed in this region and that other factors such as barotropic instability, friction and nonlinearities can alter the characteristics of the fastest growing baroclinically unstable wave. Keywords: PE ocean model; Coastal jet flow; Topography; Eddies; Friction.

Cloud overlap parameter α was estimated on remote sensing data. This parameter is a measure of the relative weight of maximum (α = 1) and random (α = 0) overlap, and may be used to diagnose relative contribution of convective and stratiform cloudiness to total cloud fraction. Cloud overlap parameter α was calculated on passive satellite data MODIS and CERES for total cloud fraction; for both cases the vertical structure of cloud was assessed by using active satellite data CALIPSO. The global annual mean α is 0.36 (for CERES) and 0.26 (for MODIS), which points to the dominance of random overlap. Values of α are higher over land than over the ocean. The most prominent annual cycle of α is noted in the monsoon regions where α is close to 1 in winter and almost 0 in summer. The maximum cloud overlap is associated with small values of cloud fraction and occurs in subtropical highs over the ocean and in subtropical and polar deserts over land. The random cloud overlap occurs in regions with large values of cloud fraction (e.g., ITCZ and midlatitudinal storm tracks). Midlatitude oceanic lows are characterized by negative values of α, mostly in summer. Presumably, the assumption of the minimum overlap of cloud layers should be used in these regions due to strong baroclinic instability and horizontal shift of cloud layers.

Cloud overlap parameter α was estimated on remote sensing data. This parameter is a measure of the relative weight of maximum (α = 1) and random (α = 0) overlap, and may be used to diagnose relative contribution of convective and stratiform cloudiness to total cloud fraction. Cloud overlap parameter α was calculated on passive satellite data MODIS and CERES for total cloud fraction; for both cases the vertical structure of cloud was assessed by using active satellite data CALIPSO. The global annual mean α is 0.36 (for CERES) and 0.26 (for MODIS), which points to the dominance of random overlap. Values of α are higher over land than over the ocean. The most prominent annual cycle of α is noted in the monsoon regions where α is close to 1 in winter and almost 0 in summer. The maximum cloud overlap is associated with small values of cloud fraction and occurs in subtropical highs over the ocean and in subtropical and polar deserts over land. The random cloud overlap occurs in regions with large values of cloud fraction (e.g., ITCZ and midlatitudinal storm tracks). Midlatitude oceanic lows are characterized by negative values of α, mostly in summer. Presumably, the assumption of the minimum overlap of cloud layers should be used in these regions due to strong baroclinic instability and horizontal shift of cloud layers.

Baroclinic instability in the ocean is a primary cause of mesoscale eddies, which are pockets of water in the scale of 100km that have different density, thermal, and rotational characteristics than their surroundings. First observed in the early 1900s, eddies are thought to be a predominant reason for the heat flux between the equator and the poles in both the ocean and the atmosphere. In attempt to understand this process better, this study uses a series of numerical simulations performed on high performance computing systems. The calculations are based on the Massachusetts Institute of Technology general circulation model, which is used to compare lateral heat transport between different simulations. The specific objectives of this project include i) Comparison the direct and remote interactions of shear with topographic slope. The direct scenario is one in which the shear extends throughout the entire ocean depth and is therefore in direct contact with the sea floor, whereas in the remote scenario there is a spatial separation between the shear in the upper half of the basin and the bottom topography, ii) Analysis of the system response to changes in the zonal and meridional seafloor slope and iii), Investigation of the effect of orientation changes in the mean large-scale current on cross-flow fluxes. The lateral heat transport and diffusivity of these simulations are then compared to our analytic model, known as Growth Rate Balance, which is based on the balance between growth rate (primary) instabilities deduced from linear theory and numerically generated secondary instabilities., http://archive.org/details/directndremoteef1094556130, Lieutenant Commander, United States Navy, Approved for public release; distribution is unlimited.

The effect of topography on the evolution of a disturbance in a baroclinically unstable mean flow is studied using analytical and numerical simulations. In particular, dynamical mechanisms involved in cyclogenesis in the lee of long, meridional barriers similar to the Rocky Mountains are explored. The rapid growth observed in lee cyclogenesis is highly dependent on the superposition of a growing baroclinic wave with a steady, orographically forced wave of the same scale. As the baroclinic wave moves over the mountain, development is masked by the orographically- forced, high-pressure ridge. As it moves down the lee side of the mountain, the baroclinic wave appears to grow rapidly due to superposition with the forced, lee-side trough. Indications of enhanced, lee-side baroclinic instability are present in the numerical simulations, but the effect on wave development is minor compared to the effect of superposition. The rapid continuous-mode growth which has recently been demonstrated by Farrell (1982) is not observed in any of the simulations., http://archive.org/details/anumericalnalyti1094521469, Major, United States Air Force, Approved for public release; distribution is unlimited.

A global, six-level primitive equation model was numerically integrated to examine the interaction of a planetary wave with a growing mid-latitude baroclinic wave. The purpose of this study was to determine the contribution of the baroclinic wave to the initiation of a monsoon surge. Three experiments were run. Experiment I depicted the effect of the planetary wave on the southward migration of cold air. The cyclogenesis of a mid-latitude baroclinic wave was shown in Experiment II. Experiment III combined the growing baroclinic wave with the stable planetary wave in an attempt to simulate a cold surge. Characteristics of a monsoon surge were present in the numerical simulation but not to the extent or detail of observed surges . The results indicated the important role of planetary wave advections in the formation of cold surges., http://archive.org/details/numericalsimulat1094521471, Captain, United States Air Force, Approved for public release; distribution is unlimited.

A global, six-level, primitive equation model was numerically integrated to examine the interaction of a mid-latitude baroclinic wave with several different mountain-valley configurations. The purpose was to determine the contribution of a well-developed baroclinic wave to the initiation of a monsoon surge. It was found that these waves can initiate a surge. Such surges are weak and limited in extent but they do exhibit the characteristics of observed monsoon surges. The latitudinal location of the east-west mountain range has a large impact on the strength of the disturbance and thus on the initiation of the surge. However, the size of the valley is not an important consideration. The surge is strongest at the lowest levels and dissipates rapidly with height. Overall results indicate surge forcing by the baroclinic wave is possible but other forcing mechanisms are necessary in order to simulate the stronger, more extensive surges observed during the winter monsoon., http://archive.org/details/numericalsimulat1094519922, Captain, United States Air Force, Approved for public release; distribution is unlimited.

Six vertical discretization schemes are compared for a linear, baroclinic, vorticity-divergence equation model. Variables are defined for two staggered and one unstaggered grids. -A finite difference and a Galerkin finite element approximation are formulated for each of the three grids. The models are tested in three experiments. The largest difference between the grids occurs in the mid-atmosphere diabatic heat source experiment and the unstaggered grid provides the best results. For the staggered grids, the results of the finite element models are not more accurate than the corresponding finite difference results. Oscillations occur in the temperature profiles near the surface for both staggered finite element models., http://archive.org/details/acomparisonofsix1094521203, Captain, United States Air Force, Approved for public release; distribution is unlimited.