Search

At least five surface rupturing earthquakes that occurred during a

The categorization of multidimensional data into clusters is a common task in statistics. Many applications of clustering, including the majority of tasks in ecology, use data that is inherently spatial and is often also temporal. However, spatiotemporal dependence is typically ignored when clustering multivariate data. We present a finite mixture model for spatial and spatiotemporal clustering that incorporates spatial and spatiotemporal autocorrelation by including appropriate Gaussian processes (GP) into a model for the mixing proportions. We also allow for flexible and semiparametric dependence on environmental covariates, once again using GPs. We propose to use Bayesian inference through three tiers of approximate methods: a Laplace approximation that allows efficient analysis of large datasets, and both partial and full Markov chain Monte Carlo (MCMC) approaches that improve accuracy at the cost of increased computational time. Comparison of the methods shows that the Laplace approximation is a useful alternative to the MCMC methods. A decadal analysis of 253 species of teleost fish from 854 samples collected along the biodiverse northwestern continental shelf of Australia between 1986 and 1997 shows the added clarity provided by accounting for spatial autocorrelation. For these data, the temporal dependence is comparatively small, which is an important finding given the changing human pressures over this time.

Residential locations play an important role in understanding the form and function of urban systems. However, it is impossible to release this detailed information publicly, due to the issue of privacy. The rapid development of location-based services and the prevalence of global position system (GPS)-equipped devices provide an unprecedented opportunity to infer residential locations from user-generated geographic information. This article compares different approaches for predicting Twitter users' home locations at a precise point level based on temporal and spatial features extracted from geo-tagged tweets. Among the three deterministic approaches, the one that estimates the home location for each user by finding the weighted most frequently visited (WMFV) cluster of that user always provides the best performance when compared with the other two methods. The results of a fourth approach, based on the support vector machine (SVM), are severely affected by the threshold value for a cluster to be identified as the home.

Spatiotemporal disease mapping focuses on estimating the spatial pattern in disease risk across a set of nonoverlapping areal units over a fixed period of time. The key aim of such research is to identify areas that have a high average level of disease risk or where disease risk is increasing over time, thus allowing public health interventions to be focused on these areas. Such aims are well suited to the statistical approach of clustering, and while much research has been done in this area in a purely spatial setting, only a handful of approaches have focused on spatiotemporal clustering of disease risk. Therefore, this paper outlines a new modeling approach for clustering spatiotemporal disease risk data, by clustering areas based on both their mean risk levels and the behavior of their temporal trends. The efficacy of the methodology is established by a simulation study, and is illustrated by a study of respiratory disease risk in Glasgow, Scotland.

Moran's Index for spatial autocorrelation and localized index for spatial association have been widely applied in many research fields as the first step to explore and assess the spatial dependency in a set of geographic events. This article presents extensions to the equations for calculating global and localized spatial autocorrelation so to include the temporal attribute values of the geographic events being analyzed. The extended equations were successfully implemented and tested with a real world data set. In addition, simulated data sets were used to reveal how the extended equations performed. Beyond the usefulness of the extended equations, we suggest that care be taken with regard to assessing spatiotemporal patterns under the normality and randomization assumptions as different outcomes from different assumptions would require different approaches for interpretation.

The recent development in Global Positioning System (GPS) techniques has started a new era in predation studies. Estimates of kill rates based on animal movements and GPS relocation clusters have proven to be valid in several obligatory carnivores. The main focus has been to obtain accurate mean predation estimates for the management of wildlife populations. We present a model to estimate individual kill rates of moose calves by adult female brown bears in Sweden, based on spatiotemporal clustering of 30,889 bear GPS relocations and 71 moose calves verified killed during 714 field investigations in 2004-2006. In this virtually single-predator single large prey system, the omnivorous brown bear is an efficient predator on moose calves up to 4 weeks of age. The top model set only included models with cluster radii of 30 m or 50 m, indicating very high kill-site fidelity. The best model included a cluster radius of 30 m and number of periods of bear activity at the kill site as a single covariate. The mean estimated individual kill rate of 7.6 � 0.71 (n ¼ 18, xSE) moose calves per calving season is comparable to the estimate of 6.8 from a previous study of radio- tracked moose in our study area, though at a lower moose/bear ratio. The mean annual kill rates varied from 6.1 to 9.4 calves per bear. The estimated individual kill rates ranged from 2 to 15 calves per season, indicating a large individual variation in hunting skills and possibly effort. Predation and livestock depredation represent a core conflict between humans and carnivores in rural Scandinavia. Accurate predation estimates represent an important step in quantifying costs of carnivores and reducing human-carnivore conflicts. Our technique may be applied in the exploration of predation mechanisms and predator-prey interactions, and contribute to the old and global debate of problem individuals in livestock depredation. 2012 The Wildlife Society.

Recent progress in the investigation of limb malformations in free-living frogs has underlined the wide range in the types of limb malformations and the apparent spatiotemporal clustering of their occurrence. Here, we review the current understanding of normal and abnormal vertebrate limb development and regeneration and discuss some of the molecular events that may bring about limb malformation. Consideration of the differences between limb development and regeneration in amphibians has led us to the hypothesis that some of the observed limb malformations come about through misdirected regeneration. We report the results of a pilot study that supports this hypothesis. In this study, the distal aspect of the right hindlimb buds of X. laevis tadpoles was amputated at the pre-foot paddle stage. The tadpoles were raised in water from a pond in Minnesota at which 7% of surveyed newly metamorphosed feral frogs had malformations. Six percent (6 of 100) of the right limbs of the tadpoles raised in pond water developed abnormally. One truncated right limb was the only malformation in the control group, which was raised in dechlorinated municipal water. All unamputated limbs developed normally in both groups. Three major factors under consideration for effecting the limb malformations are discussed. These factors include environmental chemicals (primarily agrichemicals), encysted larvae (metacercariae) of trematode parasites, and increased levels of ultraviolet light. Emphasis is placed on the necessary intersection of environmental stressors and developmental events to bring about the specific malformations that are observed in free-living frog populations.