Exploring ecological and biogeographic questions using biological databases derived from natural history collections and surveys

Abstract

Biogeographic research has benefited from the digitizing of large databases derived from natural history collections and biological surveys. These resources made available via the Internet can be accessed by biogeographers around the world to address a multitude of ecological and geographic questions. Utilizing this data taps into hundreds of years of study and countless hours of research conducted by biologists across the globe. This dissertation could not have been completed without the availability of data collected by legions of researchers from museums, herbaria, and government agencies. By taking advantage of data collected by others, I was able to work at a geographic scale that would have been impossible had I gathered all my own data.

In chapter one, I use herbarium data to describe the temporal and spatial patterns of invasive and expansive species for the entire state of Oklahoma. Because of the inherent bias in collections of natural history specimens. I test techniques for eliminating temporal collecting bias: regression models and proportion curves. I found that patterns of species invasion and expansion in Oklahoma could be detected using these techniques which were developed for regions with longer collecting plant histories. The proportion curve analysis eliminated some biases inherent in herbarium data by reducing the effect of collecting effort. Both the regression model and proportion curve analyses illustrate the temporal invasion patterns of alien, invasive species. However, the native species did not show a clear expansion pattern. The information found in recently established herbaria may not be sensitive enough to detect the increase of abundance of native species.

Currently species distribution modelling is one of the most popular methods of utilizing large, georeferenced, biological databases. Chapter two is a brief review of the overabundant literature on species distribution modelling. Topics covered are the theoretical basis for distribution modelling, species and predictor data, modelling techniques, model evaluation, and uses for predictive maps created by modelling.

Using survey data collected for the U.S. Fish and Wildlife Service, I apply species distribution modelling techniques to predict suitable habitat for the endangered American burying beetle (Nicrophorus americanus). Using a suite of predictor variable thought to influence a burrowing insect, I built several models using a variety of modelling techniques. The Maxent modelling algorithm performed the best. However, being a generalist species, the suitable habitat for N. americanus was not well modelled. Model performance could be improved by incorporating information on the cause of N. americanus's endangered status and its population shrinkage. To improve the models and consequently the recovery effort for the species, I need to take into account interactions including congener and vertebrate competition and a reduction in optimally sized prey. Creating an accurate spatial layer of this data will be a future challenge. My hope was to produce a map of potentially suitable habitat for N. americanus that would guide conservation efforts within the state of Oklahoma. Although the model was not highly accurate, the map of suitable habitat can help to inform conservation biologists of areas that have suitable habitat for the N. americanus.

In chapter four, I return to the invasive species theme by addressing the question of whether the introduced distribution of invasive species can be predicted from its native range. I modelled the potential distribution within the United States of three alien invasive species native to Europe using the Maxent modelling technique. Using occurrence data from both the native (Europe) and introduced (US) ranges, I used reciprocal modelling to evaluate habitat discrepancies between the introduced and native ranges. This modelling approach can help to determine which environmental factors within the introduced range are different from the native range and which habitats within the native range are not represented in the introduced range. Further, reciprocal modelling can reveal potential problems with occurrence data and predictor variables in both native and introduced ranges, but it also has also been used to investigate ecological phenomena, such as niche shifts of invasive species in their introduced range. The native occurrences in Europe accurately predicted the distribution within Europe; and introduced occurrences in the US accurately predicted the US distribution. However, the reciprocal models did not perform well. The explanations for the dissociated ranges of each species in Europe and US can possibly be related to the hypotheses postulated for invasive species success. The characteristics that make a species invasive may be the cause of the species' environmental range to be different in the native and introduced regions. My aim was to see if we could use easily obtained data to model the potential areas of invasion within our state and use this information to assist conservation efforts such as early detection and rapid response. My model results indicate that the occupied niches are too inconsistent between the native and introduced ranges to make models useful at the scale we are interested in. Further modeling attempts will utilize more introduced occurrence data from areas within our region of the United States. This will entail a more concerted effort to locate available data in the areas where the species may be expanding.