This thesis is comprised of a collection of short fiction depicting the childhoods and family lives of characters living in small northeastern towns. These seven stories explore the impact that the adult world has on the children who observe and seek to understand it, as well as how the events of childhood and the past can permeate and shape the lives of teenagers and adults trying to make important human connections in their present lives.

This is a study about disasters, vulnerability and power. With regards to social justice organizing a particular research problem guides the work, specifically that emancipatory projects are often initiated and steered by privileged actors who do not belong to the marginalized communities they wish to strengthen, yet the work is based on the belief that empowerment requires self-organizing from within. Through an ethnographic field study of social justice organizing in the wake of Hurricane Sandy in Rockaway, New York City, the thesis explores whether and how vulnerable groups were empowered within the Occupy Sandy network. It is a process study that traces outside activists attempts at empowering storm-affected residents over time, from the immediate relief phase to long-term organizing in the recovery phase. The activists aimed to put to practice three organizing ideals: inclusion, flexibility and horizontality, based on a belief that doing so would enhance empowerment. The analysis demonstrates that collaboration functioned better in the relief phase than in the long-term recovery phase. The same organizing ideals that seem to have created an empowering milieu for storm-affected residents in the relief phase became troublesome when relief turned to long-term recovery. The relief phase saw storm-affected people step up and take on leadership roles, whereas empowerment in the recovery phase was conditional on alignment with outside activists’ agendas. Internal tensions, conflicts and resistance from residents toward the outside organizers marked the recovery phase. It seems that length of collaborative projects is not the only factor for developing trust but so is complexity. The more complex the activities over which partners are to collaborate the less easy it is. Based on this we could further theorize that the more complex the work is the more challenging it is for privileged groups to give away control. The internal struggles of the organization partially explain the failures to influence an urban planning process that the organization attempted to impact, which connects the micro-processes with broader change processes toward transformation of vulnerability.

The 2004 Indian Ocean and 2011 Tohoku tsunami events have shown the destructive power of tsunami inundation to the constructed environment in addition to the tragic loss of life. A comparable event is expected for the Cascadia Subduction Zone (CSZ) which will impact the west coast of North America. Research efforts have focused on understanding and predicting the hazard to mitigate potential impacts. This thesis presents two manuscripts which pertain to estimating infrastructure damage and determining design loads of tsunami inundation. The first manuscript estimates damage to buildings and economic loss for Seaside, Oregon, for CSZ events ranging from 3 to 25 m of slip along the entire fault. The analysis provides a community scale estimate of the hazard with calculations performed at the parcel level. Hydrodynamic results are obtained from the numerical model MOST and damage estimates are based on fragility curves from the recent literature. Seaside is located on low lying coastal land which makes it particularly sensitive to the magnitude of the events. For the range of events modeled, the percentage of building within the inundation zone ranges from 9 to 88%, with average economic losses ranging from $2 million to $1.2 billion. The second manuscript introduces a new tsunami inundation model based on the concept of an energy grade line to estimate the hydrodynamic quantities of maximum flow depth, velocity, and momentum flux between the shoreline and extent of inundation along a 1D transect. Using the numerical model FUNWAVE empirical relations were derived to tune the model. For simple bi-linear beaches the average error for the tuned model in flow depth, velocity, and momentum flux were 10, 23, and 10%, respectively; and for complex bathymetry at Rockaway Beach, Oregon, without recalibration, the errors were 14, 44, and 14% for flow depth, velocity, and momentum flux, respectively. / Graduation date: 2013