Multi-scale Spatiotemporal Analyses of Moose-vehicle Collisions: a Case Study in Northern Vermont
Moose-vehicle collisions are a major environmental and safety concern in regions throughout the United States and Europe, and collisions have increased substantially in Europe and North America mainly due to increasing traffic volume. This case study takes a closer look at Moose-vehicle collisions in the state of Vermont, where one third of all reported moose-vehicle collisions result in motorist injury or fatality. An adapted kernal density estimator was implemented to detect and analyze high density collisions hotspots on four major roads in Northeastern Highlands of Vermont and uses a dataset from 1983 to 1999. The analyses produced a couple interesting findings. The first finding is there are seven major density peaks. The second is the kernal estimaor in time for all roads showed seasonally cyclic components with the majority of collisions happening between May and October, and annual increased number of collisions. The third finding is an observation that density hotspots shift in time, suggesting changes in moose movement patterns.
Methods, Tools, and Data
Methods: Ripley's K-function for cluster testing, Univariate Ripley's K-function applied in space or time, and Bivariate K-function for spatiotenporal analysis.
Tools: Kernel Density Estimator, and exploratory analysis of Kernal Density Estimation.
Data: Moose-vehicle collision dataset from 1983-1999.
Books and Publications
Giorgos Mountrakis & Kari Gunson (2009) Multi‐scale spatiotemporal analyses of moose–vehicle collisions: a case study in northern Vermont, International Journal of Geographical Information Science, 23:11, 1389-1412, DOI: 10.1080/13658810802406132