Hierarchical Modeling
Hierarchical models, also known as mutl-level and mixed effects models, have advanced the field of population ecology with inferences about population dynamics at broad spatial and temporal scales. The analysis of hierarchical models has been facilitated by recent advances in Bayesian analysis, and computationally intensive techniques such as Markov Chain Monte Carlo (see Mathematical modeling).
Hierarchical Models for Estimation of Population Parameters
Much of wildlife research consists of the description of variation in data. Some of the variation results from spatial and temporal change in populations, while some results from biologically irrelevant sampling variation induced by the process of data collection. Distinguishing relevant from irrelevant variation is the first task of statistical analysis, but the job does not end there. Even if the true values of population parameters were known, without the contamination of sampling variation, the investigation of population processes would require an evaluation of pattern among parameters.
Many critical wildlife surveys, such as the North American Breeding Bird Survey (BBS), are analyzed using complex hierarchical models. These models are generally multi-scale and contain random effects; the standard approaches to model selection and assessment of model fit are often inappropriate and no simple way exists to compare alternative models. However, a clear need exists for these assessments. Many alternative models new exist for analysis of BBS data, and simply presenting multiple results without clear guidance on which model is most appropriate will lead to confusion among users of BBS data and limit use of this important survey.
Design and Analysis of Surveys for Estimation of Temporal and Spatial Change in Animal Populations
Population status information is required for management of migratory bird populations, and structured decision making and adaptive anagement place additional emphasis on the need for rigorous survey designs and robust estimation methods. The North American Breeding Bird Survey (BBS) and Christmas Bird Count (CBC) provide continent-scale information on breeding and wintering populations of >450 species of North American birds, and for many species these two surveys are our only data source for population status and trend information. Appropriate analyses of these important surveys require sophisticated methods to accommodate variation in survey efficiency over the large areas covered by the surveys and to control for factors that influence detection of birds. Factors such as observer quality and effort, if not appropriately controlled for in the analysis, can lead to biased estimates of population change.
Hierarchical Models of Animal Abundance and Occurrence
Research goals of this project are to develop models, statistical methods, sampling strategies and tools for inference about animal population status from survey data. Survey data are always subject to a number of observation processes that induce bias and error. In particular, inferences are based on spatial sampling – we can only ever sample a subset of locations where species occur --and imperfect detection – species or individuals might go undetected in the sample. Principles of hierarchical modeling can be applied directly to accommodate both features of ecological data. Prior to the development of hierarchical models at PWRC, studies of unmarked populations focused on simplistic descriptions of distribution patterns and temporal trends. Hierarchical models have advanced the field of population ecology by enabling the estimation of demographic and movement parameters that previously could only be obtained using costly field methods. Ecologists can now make inferences about population dynamics at broad spatial and temporal scales using models designed specifically for this task.
Development of Computer Software for the Analysis of Animal Population Parameters
Biologists at USGS Patuxent, as well as cooperating agencies are constantly looking for new ways of answering questions about the status of animal populations and how animal populations change over time. To address these questions, data are collected on captures and or sightings of animals which can be used to estimate parameters which affect the population using legacy software. Over time, new questions and methods for addressing these questions arise which require new computer software.
Hierarchical models, also known as mutl-level and mixed effects models, have advanced the field of population ecology with inferences about population dynamics at broad spatial and temporal scales. The analysis of hierarchical models has been facilitated by recent advances in Bayesian analysis, and computationally intensive techniques such as Markov Chain Monte Carlo (see Mathematical modeling).
Hierarchical Models for Estimation of Population Parameters
Much of wildlife research consists of the description of variation in data. Some of the variation results from spatial and temporal change in populations, while some results from biologically irrelevant sampling variation induced by the process of data collection. Distinguishing relevant from irrelevant variation is the first task of statistical analysis, but the job does not end there. Even if the true values of population parameters were known, without the contamination of sampling variation, the investigation of population processes would require an evaluation of pattern among parameters.
Many critical wildlife surveys, such as the North American Breeding Bird Survey (BBS), are analyzed using complex hierarchical models. These models are generally multi-scale and contain random effects; the standard approaches to model selection and assessment of model fit are often inappropriate and no simple way exists to compare alternative models. However, a clear need exists for these assessments. Many alternative models new exist for analysis of BBS data, and simply presenting multiple results without clear guidance on which model is most appropriate will lead to confusion among users of BBS data and limit use of this important survey.
Design and Analysis of Surveys for Estimation of Temporal and Spatial Change in Animal Populations
Population status information is required for management of migratory bird populations, and structured decision making and adaptive anagement place additional emphasis on the need for rigorous survey designs and robust estimation methods. The North American Breeding Bird Survey (BBS) and Christmas Bird Count (CBC) provide continent-scale information on breeding and wintering populations of >450 species of North American birds, and for many species these two surveys are our only data source for population status and trend information. Appropriate analyses of these important surveys require sophisticated methods to accommodate variation in survey efficiency over the large areas covered by the surveys and to control for factors that influence detection of birds. Factors such as observer quality and effort, if not appropriately controlled for in the analysis, can lead to biased estimates of population change.
Hierarchical Models of Animal Abundance and Occurrence
Research goals of this project are to develop models, statistical methods, sampling strategies and tools for inference about animal population status from survey data. Survey data are always subject to a number of observation processes that induce bias and error. In particular, inferences are based on spatial sampling – we can only ever sample a subset of locations where species occur --and imperfect detection – species or individuals might go undetected in the sample. Principles of hierarchical modeling can be applied directly to accommodate both features of ecological data. Prior to the development of hierarchical models at PWRC, studies of unmarked populations focused on simplistic descriptions of distribution patterns and temporal trends. Hierarchical models have advanced the field of population ecology by enabling the estimation of demographic and movement parameters that previously could only be obtained using costly field methods. Ecologists can now make inferences about population dynamics at broad spatial and temporal scales using models designed specifically for this task.
Development of Computer Software for the Analysis of Animal Population Parameters
Biologists at USGS Patuxent, as well as cooperating agencies are constantly looking for new ways of answering questions about the status of animal populations and how animal populations change over time. To address these questions, data are collected on captures and or sightings of animals which can be used to estimate parameters which affect the population using legacy software. Over time, new questions and methods for addressing these questions arise which require new computer software.