In recent years neural network have been shown to be quite effective in solving difficult combinatorial optimization problems. In this work a Hopfield neural network is used to schedule operations in a dataflow graph. This is an important step in behavioral synthesis systems. These operations must be assigned to a limited number of control steps, functional units, and busses. Also, there is an objective to minimize the lengths of data paths. Current methods which do this type of scheduling typically rely on heuristic algorithms. The neural network devised to solve this problem is one of the most complex to date. A special mechanism, "flag" neurons, was developed to enable the neural network to encode a bussing constraint. The neural network has been tested with problems from literature and problems randomly generated. The results have been consistently superior to those produced by a heuristic algorithm called ALAP.

An important wetland type in the southwestern United States requiring study is the cienega. Cienega is a Spanish term used for wetlands found in semi-arid grasslands, associated with perennial springs and headwater streams. A field study was conducted at Canelo Hills Cienega in Southeastern Arizona for one year in which changes in water levels, soil water content, and stream flow were monitored. Plant composition, soil classification, and basic geological characteristics were determined as well. Water level fluctuations and flow gradients indicate that this cienega is mostly groundwater dependent and is an effluent system maintaining the perennial nature of the adjacent stream. Mean water levels ranged from.9 cm above the surface to 0.85 m below. Water level fluctuation within the cienega ranged from 1.12 m/yr to 0.18 m/yr. Vegetation occurred in distinct patterns which varied across the cienega in relation to water levels and availability. Dominant genera included Eleocharis, Carex, Juncus, Poa, and Bidens.

This is a landscape study of the Spanish-American village of Chimayo, New Mexico where the vestige of a fascinating pre-modern folk landscape persists into the present. The landscape in Chimayo has undergone significant restructuring in the present century--from colonial agricultural landscape to modern suburban townscape--yet an image of the traditional rural landscape remains. The complex modern economic and social forces (e.g., automobiles, telecommunications, tourism) which, early in this century, supplanted Chimayo's colonial subsistence agricultural economy has also restructured its landscape. This project has revealed that artifacts, spaces, and place images can persist in a changing rural landscape. Moreover, in the evolution of the vernacular landscape of Chimayo, New Mexico, the pre-modern folk landscape has been partially conserved, both physically and in myth, and persists as an integral part of the contemporary landscape.

This thesis presents a discrete time implementation and analysis of the generalized impulse response model proposed by Altes. This model is a generalization of the weighted sum of time delayed delta functions typically used to describe the impulse response of scattering targets. The target chosen for the analysis is the rigid acoustic sphere. The coefficients of the generalized model are used to calculate estimates of the radius of the sphere, which is known in advance for testing purposes. The accuracy of the radius estimates indicates the accuracy of the model coefficients. The generalized model is shown to be superior from the standpoint of radius estimation. An estimator for the time of arrival of a signal of unknown but deterministic form is derived. It is based upon a generalized likelihood ratio test whose structure accommodates the generalized model. This estimator performs well in high levels of noise.