No studies have evaluated the effects of limited vehicle access on movements, survival and habitat use of Roosevelt elk (Cervus elaphus roosevelti) . We installed twenty gates, restricting motorized vehicle access by the public in seven discrete Road Management Areas (RMAs), comprising 35% of the study area. We radio-tracked 31 cow elk for 13 months in a 38,000 ha area of the Oregon Coast Range. Prior to the installation of the gates, 20 of these elk had been tracked for 14 months, allowing a paired comparison of elk movements and habitat use before and during the limited access period. The percentage of elk home ranges or core areas within the RMAs did not differ between periods, but there was a clear decrease in daily movement of elk during the limited access period. Survival rates increased during the limited access period and declined after the removal of the gates. During the limited access period, there was a significant increase in the use of open, foraging habitats and areas <150 m from roads. We conducted habitat selection analysis on vegetative cover types, distance from roads and distance from water. In general elk use of vegetative cover types was not significantly different from availability (p<0.05). Elk avoided areas <150 m from roads and selected areas >150 m from roads. Elk selected areas <150 m from streams and avoided areas >600 m from streams. Roosevelt elk should benefit from the preferred alternative of the President's forest Plan, which maintains roadless areas near streams. / Graduation date: 1996

Posting advisory speed signs at sharp horizontal curve sites is a practice well established in the United States. The purpose of these signs is to provide the driving public with a safe speed to negotiate such curves; however, the link between these signs and safety has not yet been clearly established. The first manuscript in this dissertation presents an effort to model safety as it relates to curve advisory speed signs. It proposes a statistical model relating crash frequency at 2-lane rural highways in Oregon to curve advisory speed signs and other influential factors. The Advisory Speed Crash Factor (ASCF) emerges as a sub-model that characterizes the safety effect of advisory speed signs. Results indicate that safety may be compromised if the advisory speed is either excessively prohibitive or excessively permissive. The second manuscript extends the use of the proposed ASCF to develop the OSU posting method, a new procedure that procures the "optimal" advisory speed derived from the ASCF. A field validation analysis, also presented in this manuscript, verified the meaningfulness of the proposed ASCF sub-model. The third manuscript outlines another methodology, named 'the Hybrid OSU Posting Method' in an effort to mitigate the well documented variability associated with using the Ball Bank Indicator (BBI). This method determines the advisory speed using the BBI in combination with the ASCF. Though benefits in safety performance and consistency resulted from using the Hybrid OSU method, this method is still outperformed by the computational OSU method. / Graduation date: 2012 / Access restricted to the OSU Community at author's request from May 25, 2012 - May 25, 2013