Nitric oxide is a critical signaling molecule in the microcirculatory control of vascular resistance. It enables the maintenance of blood pressure and cardiac output at optimal levels, via flow mediated endothelial nitric oxide production which serves to modulate vascular tone. Understanding nitric oxide production and bioavailability in the vasculature facilitates a number of theoretical models and clinical applications. Hematocrit influences the vaso-active role of nitric oxide, since both blood viscosity and shear stress increase with red blood cell concentrations. Moreover, increases in hematocrit cause cell free layer thickness to decrease and hemoglobin scavenging of nitric oxide in red blood cells to increase. To capture this behavior at the scale of an individual blood vessel, we construct a model of the autoregulatory response of blood vessels. The model incorporates coupled models of blood flow and nitric oxide transport, thereby explaining how changes in hematocrit influence vascular response. We also develop a model that captures the effects of changing hematocrit on blood rheology, velocity profiles and wall shear stress measurements. This model accounts for the non-Newtonian, shear-thinning properties of blood. A clinical application of these ideas is demonstrated by showing how changes in the blood rheology via plasma expanders can result in the restoration of cardiac performance. These individual- vessel models inform our analysis of the structure of microvascular networks. Specifically, we examine how hematocrit, pressure and shear stress are distributed in optimally configured model networks. Our approach allows for simulation of vascular networks which exhibit the broad characteristics of networks observed in-vivo. At the cellular level, we examine the biochemistry of nitric oxide production inside endothelial cells. We construct a model which simulates the endothelial nitric oxide production cycle, following application of shear stress. Our model predictions for both steady and transient shear induced nitric oxide production are shown to be in broad agreement with experimental data. Collectively, this dissertation significantly enhances our understanding of the dual and often competing roles of nitric oxide and hematocrit in the microcirculation

Environmental movements have been hindered in utilizing disparate impact as an effective legal mechanism for change. Since the 2001 Sandoval ruling limited the private right of action for Title VI disparate impact claims, environmental justice advocates have adopted the disparate impact framework as a persuasive tool to analyze, investigate, and challenge inequitable development. Concurrently, ecosystem services have blossomed as a growing field. The ecosystem services framework asserts that ecosystems providee conomic and health benefits for communities. However, this framework faces challenges with value recognition and visibility, lack of implementation within existing institutional frameworks, and inequitable access. This article explicitly combines the disparate impact and ecosystem services frameworks together to strengthen each other. Specifically, this article argues that incorporating ecosystem services within disparate impact analyses can provide new persuasive data and evidence for environmental justice movements. Additionally, environmental justice advocates utilizing ecosystem services frameworks can increase the field’s visibility, data, and provide more information regarding inequitable ecosystem access.

Rapid advances in semiconductor technology have led to an era where entire systems, consisting of complex, heterogeneous components, can be integrated on a single chip, referred to as System-on-Chip (SoC). Due to the escalating cost in designing customized application- specific SoCs, recent years have witnessed the emergence of 'platform-based' SoCs. These systems consist of largely pre-designed, general-purpose components that can be re- targeted towards numerous applications, thereby amortizing design costs. A key determinant to the success of such platforms is the extent to which they can be customized to meet the diverse requirements imposed by different applications. Modern SoC platforms are mostly limited to providing a one-time (static) customization of the platform hardware. However, with the convergence of multiple diverse applications on the same platform, each imposing time-varying requirements, there is a growing need for SoC platforms that can be dynamically configured. Provisioning for such configurability and exploiting it at run-time is the focus of this dissertation. This dissertation proposes SoC platforms featuring multiple, dynamic configurability options, and illustrates their advantages over existing design styles. It introduces the concept of Dynamic Platform Management, a methodology for the run-time customization of such platforms in a coordinated manner, to satisfy the time-varying requirements imposed by the executing applications. The dissertation addresses the problem of provisioning for dynamic configurability by proposing a novel, on-chip communication architecture that features a dynamically configurable topology. Dynamic management techniques are then presented for platforms featuring multiple, run-time configurable components. In particular, it considers platforms consisting of configurable processors, flexible memory architectures, and configurable communication architectures. Finally, it investigates the benefits of synergistically combining techniques for configuring platform hardware with techniques for adapting application behavior. Experiments conducted on a large number of SoC designs, and an implementation of dynamic platform management on the Altera Excalibur development board demonstrate that configurable platforms with dynamic platform management result in significantly superior application performance, more efficient utilization of platform resources, and improved energy efficiency compared to conventional static approaches. Hence, the techniques described in this dissertation will facilitate more wide-spread adoption of the platform-based approach, leading to low-cost, yet function-rich and energy- efficient devices

This paper provides a detailed description of how deictic motion events are encoded in a Tangsa variety called Hakhun, spoken in Arunachal Pradesh and Assam in India, and in Sagaing Region in Myanmar. Deictic motion events in Hakhun are encoded by a set of two motion verbs, their serial or versatile verb counterparts, and a set of two ventive particles. Impersonal deictic motion events are encoded by the motion verbs alone, which orient the motion with reference to a center of interest. Motion events with an SAP figure or ground are simultaneously encoded by the motion verbs and ventive particles. These motion events evoke two frames of reference: a home base and the speech-act location. The motion verbs anchor the motion with reference to the home base of the figure, and the ventives (or their absence) anchor the motion with reference to the location of the speaker, the addressee, or the speech-act. When the motion verbs are concatenated with other verbs, they specify motion associated with the action denoted by the other verb(s).