Historically, the traditional planning process has not given much attention to the road safety evaluation of development plans. To make an informed, defensible, and proactive choice between alternative plans and their safety implications, it is necessary to have a procedure for estimating and evaluating safety performance. A procedure is required for examining the influence of the urban network development on road safety, and in particular, determining the effects of the many variables that affect safety in urban planning. Safety planning models can provide a decision-support tool that facilitates the assessment of the safety implications of alternative network plans. The first objective of this research study is to develop safety planning models that are consistent with the regional models commonly used for urban transportation planning. Geographically weighted Poisson regression (GWPR), full-Bayesian semiparametric additive (FBSA), and traditional generalized linear modelling (GLM) techniques are used to develop the models. The study evaluates how well each model is able to handle spatial variations in the relationship between collision explanatory variables and the number of collisions in a zone. The evaluation uses measures of goodness of fit (GOF) and finds that the GWPR and FBSA models perform much better than the conventional GLM approach. There is little difference between the GOF values for the FBSA and GWPR models. The second objective of this research study is to examine the temporal transferability of the safety planning models and alternative updating methods. The updating procedures examine the Bayesian approach and application of calibration factors. The results show that the models are not temporally transferable in a strict statistical sense. However, relative measures of transferability indicate that the transferred models yield useful information in the application context. The results also show that the updated safety planning models using the Bayesian approach predict the number of collisions better than the calibration factor procedure.

Historically, the traditional planning process has not given much attention to the road safety evaluation of development plans. To make an informed, defensible, and proactive choice between alternative plans and their safety implications, it is necessary to have a procedure for estimating and evaluating safety performance. A procedure is required for examining the influence of the urban network development on road safety, and in particular, determining the effects of the many variables that affect safety in urban planning. Safety planning models can provide a decision-support tool that facilitates the assessment of the safety implications of alternative network plans. The first objective of this research study is to develop safety planning models that are consistent with the regional models commonly used for urban transportation planning. Geographically weighted Poisson regression (GWPR), full-Bayesian semiparametric additive (FBSA), and traditional generalized linear modelling (GLM) techniques are used to develop the models. The study evaluates how well each model is able to handle spatial variations in the relationship between collision explanatory variables and the number of collisions in a zone. The evaluation uses measures of goodness of fit (GOF) and finds that the GWPR and FBSA models perform much better than the conventional GLM approach. There is little difference between the GOF values for the FBSA and GWPR models. The second objective of this research study is to examine the temporal transferability of the safety planning models and alternative updating methods. The updating procedures examine the Bayesian approach and application of calibration factors. The results show that the models are not temporally transferable in a strict statistical sense. However, relative measures of transferability indicate that the transferred models yield useful information in the application context. The results also show that the updated safety planning models using the Bayesian approach predict the number of collisions better than the calibration factor procedure.

Aim MaxEnt, a very popular species distribution modelling technique, has been used extensively to relate species’ geographic distributions to environmental variables and to predict changes in species’ distributions in response to environmental change. Here, we test its predictive ability through time (rather than through space, as is commonly done) by modeling colonizations and extinctions. Location Continental U.S. and southern Canada. Time period 1979-2009 Major taxa studied Twenty-one species of passerine birds. Methods We used MaxEnt to relate species’ geographic distributions to the variation in environmental conditions across North America. We then modelled site-specific colonizations and extinctions between 1979 and 2009 as functions of MaxEnt-estimated previous habitat suitability and inter- annual change in habitat suitability and neighborhood occupancy. We evaluated whether the effects were in the expected direction, we partitioned model’s explained deviance, and we compared colonization and extinction model’s accuracy to MaxEnt’s AUC. Results IV Colonization and extinction probabilities both varied as functions of previous habitat suitability, change in habitat suitability, and neighborhood occupancy, in the expected direction. Change in habitat suitability explained very little deviance compared to other predictors. Neighborhood occupancy accounted for more explained deviance in colonization models than in extinction models. MaxEnt AUC correlates with extinction models’ predictive ability, but not with that of colonization models. Main conclusions MaxEnt appears to sometime capture a real effect of the environment on species’ distributions since a statistical effect of habitat suitability is detected through both time and space. However, change in habitat suitability (which is much smaller through time than through space) is a poor predictor of change in occupancy. Over short time scales, proximity of sites occupied by conspecifics predicts changes in occupancy just as well as MaxEnt. The ability of MaxEnt models to predict spatial variation in occupancy (as measured by AUC) gives little indication of transferability through time. Thus, the predictive value of species distribution models may be overestimated when evaluated through space only. Future prediction of species’ responses to climate change should make a distinction between colonization and extinction, recognizing that the two processes are not equally well predicted by SDMs.

Modeling habitat suitability is beneficial for management and conservation of a species. Although data-rich models are commonly used, opinion-based models may be a beneficial alternative to estimate suitable habitat locations. Despite the increasing use of habitat models, few studies have linked habitat model covariates (i.e., land cover, weather, and normalized difference vegetation indexes (NDVI)) to demographic parameters. This study evaluates model performance and transferability of maximum entropy (MaxEnt) and expert opinion models for predicting American beaver (Castor canadensis) distribution in the southeastern US. I also investigated the relationship of environmental and habitat model covariates to beaver survival. The model’s predictive performance and transferability were evaluated using the area under the curve (AUC) index. Both model approaches performed well at predicting beaver presence. While MaxEnt had better performance, the expert models predicted greater areas as suitable for beaver. Beaver survival was estimated for northern Alabama and was found to be influenced by NDVI and weather covariates in this study.

The Roanoke logperch (Percina rex) is a federally endangered large darter that occurs only within the Roanoke and Chowan drainages of Virginia. This dissertation examines multi-scale habitat use patterns by logperch in three river systems in Virginia, including comparisons among rivers and life stages. The first study in this dissertation compares microhabitat use patterns of logperch among the Roanoke, Pigg, and Nottoway rivers. My objectives are to: 1) compare available microhabitat and microhabitat use by logperch among these rivers; and 2) examine the transfer of habitat models among rivers. Habitat availability in the three rivers indicates that the Nottoway River is least impacted by human activity, while the Pigg River is most impacted. The Roanoke and Pigg rivers are found within the same region of Virginia and share many habitat characteristics. Logperch consistently use silt free, loosely embedded gravel in all rivers and can occupy a variety of depths and velocities to accommodate substrate requirements. Microhabitat models transfer better between the similar Pigg and Roanoke rivers. The second study in this dissertation compares micro- and meso-habitat use patterns by Roanoke logperch in the Roanoke and Nottoway rivers. My objectives are to: 1) compare micro- and meso-habitat use patterns of logperch in the Roanoke and Nottoway rivers; and 2) examine transfer of habitat models at both scales. An increase in scale from micro- to meso- habitat did not improve model transfer. Habitat selectivity and transfer was strongest at the microhabitat scale. Logperch appear to be microhabitat substrate specialists and mesohabitat generalists. The final study in this dissertation examines ontogenetic patterns of habitat use by Roanoke logperch in the Roanoke and Nottoway rivers. My goals are to: 1) examine habitat use by three age classes of logperch and 2) compare ontogenetic patterns of habitat use between the Roanoke and Nottoway rivers. In the Roanoke River, adult and subadult logperch primarily used run and riffle habitat, often over gravel substrate. Subadults were found in lower water velocities and more embedded microhabitats than adults. Young-of-year logperch were found in shallow, stagnant backwaters and secondary channels. In the Nottoway River, both adult and subadult logperch were found over sand and gravel in deep, low velocity pools and runs. Subadults were observed in slightly more silted, lower velocity habitat. Younger age classes of logperch appear to be more vulnerable to sedimentation caused by human activity. Evidence in this dissertation strongly indicates that logperch have strict substrate requirements and the distribution of habitat types and pathways of dispersal will be critical for completion of the logperch life cycle. A watershed-level conservation approach that addresses sediment loading and preserves ecological processes that provide ephemeral, seasonal, and persistent types of habitat required over logperch ontogeny will be most effective for management geared towards the recovery of this endangered species. / Ph. D.

Aim: The abundances and distributions of some species are more closely matched to variations in climate than others. Species traits that might influence how well the distribution and abundance of a species are matched to climatic variation include life history (e.g., body size and dispersal ability), ecology (e.g., habitat specialization and territoriality) and demography (e.g., population size). Here, we used a survey of bird abundances across the USA to assess the extent to which species abundances and distributions are predicted by climate (i.e., climate matching) and how species traits relate to interspecific variation in climate matching. Location: USA. Time period: 1983–2018. Major taxa studied: Birds. Methods: Species abundances were obtained from the North American Breeding Bird Survey. Climate matching was estimated as the predictive performance of species–climate models fitted using boosted regression trees and generalized additive models and modelled as a function of species traits. Results: Species traits explained 56% of the variation in climate matching among species. Intermediate-sized species were more well matched to climate than smaller or larger species, as were species that lived primarily in forested compared with open habitats, species that were locally more abundant and species that were more territorial. Alternatively, species that were more specialized or had high variability in abundance among sites were less well matched to climate. We also found that species classified as “near threatened” were more well matched to climate, suggesting that these species might be more vulnerable to climate change. However, species classified as “vulnerable” were more decoupled from climate than those of “least concern”, possibly owing to ecological drift associated with progressive population declines. Main conclusions: Our findings provide an ecological basis for understanding the extent to which species abundances and distributions match broad climatic gradients, which can provide the groundwork to improve our ability to predict distributions under global change.

Many developing countries have witnessed fast and rapid growth in the last two decades due to the high development rate of economic activity in these countries. Many transportation projects have been constructed. In the same time both population growth and vehicle ownership rate increased; resulting in increasing levels of road crashes. Road traffic crashes in Gulf Cooperation Council (GCC) is considered a serious problem that has deep effects on GCC's population as well as on the national productivity through the loss of lives, injuries, property damage and the loss of valuable resources. From a recent statistical study of traffic crashes in Oman, it was found that in 2013 there were 7,829 crashes occurred for a total of 1,082,996 registered vehicles. These crashes have resulted in 913, 5591, and 1481 fatal, injury and property damage only crashes, respectively (Directorate General of Traffic, 2014), which is considered high rates of fatalities and injuries compared to other more developed countries. This illustrates the seriousness and dangerousness of the safety situation in GCC countries and Oman particularly. Thus, there is an urgent need to alleviate the Severity of the traffic safety problem in GCC which in turn will set a prime example for other developing countries that face similar problems. Two main data sources from Riyadh, the capital city of Kingdom of Saudi Arabia (KSA) and Muscat, the capital city of Sultanate of Oman have been obtained, processed, and utilized in this study. The Riyadh collision and traffic data for this study were obtained in the form of crash database and GIS maps from two main sources: the Higher Commission for the Development of Riyadh (HCDR) and Riyadh Traffic Department (RTD). The Muscat collision and traffic data were obtained from two main sources: the Muscat Municipality (MM) and Royal Oman Police, Directorate General of Traffic (DGC). Since the ARC GIS is still not used for traffic crash geocoding in Oman, the crash data used in the analysis were extracted manually from the filing system in the DGC. Due to the fact that not all developing countries highway agencies possess sufficient crash data that enable the development of robust models, this problem gives rise to the interest of transferability of many of the models and tools developed in the US and other developed nations. The Highway Safety Manual (HSM) is a prime and comprehensive resource recently developed in the US that would have substantial impact if researchers are able to transfer its models to other similar environment in GCC. It would save time, effort, and money. The first edition of the HSM provides a number of safety performance functions (SPFs), which can be used to predict collisions on a roadway network. This dissertation examined the Transferability of HSM SPFs and developing new local models for Riyadh and Muscat. In this study, first, calibration of the HSM SPFs for Urban Four-lane divided roadway segments (U4D) with angle parking in Riyadh and the development of new SPFs were examined. The study calibrates the HSM SPFs using HSM default Crash Modification Factors (CMFs), then new local CMFs is proposed using cross-sectional method, which treats the estimation of calibration factors using fatal and injury data. In addition, new forms for specific SPFs are further evaluated to identify the best model using the Poisson-Gamma regression technique. To investigate how well the safety performance model fits the data set, several performance measures were examined. The performance measures summarize the differences between the observed and predicted values from related SPFs. Results indicate that the jurisdiction-specific SPFs provided the best fit of the data used in this study, and would be the best SPFs for predicting severe collisions in the City of Riyadh. The study finds that the HSM calibration using Riyadh local CMFs outperforms the calibration method using the HSM default values. The HSM calibration application for Riyadh crash conditions highlights the importance to address variability in reporting thresholds. One of the findings of this research is that, while the medians in this study have oversize widths ranging from 16ft-70ft, median width has insignificant effect on fatal and injury crashes. At the same time the frequent angle parking in Riyadh urban road networks seems to increase the fatal and injury collisions by 52 percent. On the other hand, this dissertation examined the calibration of the HSM SPFs for Urban intersections in Riyadh, Kingdom of Saudi Arabia (KSA) and the development of new set of models using three year of collision data (2004-2006) from the city of Riyadh. Three intersection categories were investigated: 3-leg signalized, 4-leg signalized, and 3-leg unsignalized. In addition, new forms for specific SPFs are further evaluated to identify the best model using the Poisson-Gamma regression technique. Results indicate that the new local developed SPFs provided the best fit of the data used in this study, and would be the best SPFs for predicting severe crashes at urban intersections in the City of Riyadh Moreover, this study examined the calibration of the HSM SPFs for Fatal and Injury (FI), Property Damage Only (PDO) and total crashes for Urban Four-lane divided roadway segments (U4D) in Muscat, Sultanate of Oman and the development of new SPFs. This study first calibrates the HSM SPFs using the HSM methodology, and then new forms for specific SPFs are further evaluated for Muscat's urban roads to identify the best model. Finally, Riyadh fatal and injury model were validated using Muscat FI dataset. Comparisons across the models indicate that HSM calibrated models are superior with a better model fit and would be the best SPFs for predicting collisions in the City of Muscat. The best developed collision model describes the mean crash frequency as a function of natural logarithm of the annual average daily traffic, segment length, and speed limit. The study finds that the differences in road geometric design features and FI collision characteristics between Riyadh and Muscat resulted in an un-transferable Riyadh crash prediction model. Overall, this study lays an important foundation towards the implementation of HSM methods in multiple cities (Riyadh and Muscat), and could help their transportation officials to make informed decisions regarding road safety programs. The implications of the results are extendible to other cities and countries and the region, and perhaps other developing countries as well.