<p>Ecologists have long sought to uncover the mechanisms behind large-scale, macroecological patterns in the distribution and abundance of species. Macroecological patterns are often attributed to the dynamics of dispersal (e.g. dispersal limitation or widespread dispersal). However, few studies actually measure dispersal to determine if dispersal rates are commensurate with the observed macroecological patterns. In this dissertation, I use population genetic analyses across many species to obtain community-level estimates of dispersal rates for two different ecological systems: birds on islands and trees in tropical rainforests. These independent estimates of dispersal then allow me to determine if macroecological patterns in these two systems can be attributed to dispersal dynamics.</p><p>In chapter two, I explore the contrasting macroecological patterns of two groups of Lesser Antillean birds. The groups' differing macroecological patterns could be due to differences in dispersal, but other authors have advocated different mechanisms. Population genetic analyses show that the two groups do differ significantly in rates of inter-island dispersal, indicating that dispersal dynamics can explain their contrasting macroecological patterns. In chapter three, I turn my attention to tropical tree communities. In contrast to studies of birds on islands, studies of trees in tropical rainforests may suffer from misidentification of individuals in the field. Using a phylogenetic approach, I determine errors rates in identification, and then assess the effect of these errors on macroecological patterns and other ecological analyses of tropical tree communities. I find that error rates are substantial, but that they have little effect on macroecological patterns. In contrast, species-level ecological analyses can be dramatically affected by these errors.</p><p>In chapter four, I return to the influence of dispersal on macroecological patterns, this time in tropical tree communities. One notable macroecological pattern in Amazonian tree communities is a high correlation in the relative abundances of species shared across communities, which could indicate high rates of dispersal between communities. However, population genetic analyses show that dispersal is severely limited between communities. Thus, some factor besides dispersal, such as differences in competitive ability or susceptibility to disease, must be driving species to achieve similar relative abundances in geographically separated communities. In contrast, I show that dispersal limitation is the likely cause of another macroecological pattern frequently observed in tropical tree communities: the decline in the compositional similarity of communities with distance. However, this is not steady-state dispersal limitation in an equilibrium framework as is conventionally thought. Instead, the dispersal limitation appears to be historical in nature, which implies a heretofore unnoticed role for historical contingency in the assembly of Amazonian tree communities.</p> / Dissertation

<p>Drylands, comprising arid and semi-arid areas and the dry subtropics, over some 40% of the world's land area and support approximately 2 billion people, including at least 1 billion who depend on dryland agriculture and grazing. 10-20% of drylands are estimated to have already undergone degradation or desertification, and lack of monitoring and assessment remains a key impediment to preventing further desertification. Change in vegetation cover, specifically in the spatial organization of vegetation may occur prior to irreversible land degradation, and can be used to assess desertification risk. Coherent spatial structures arise in the distribution of dryland vegetation where plant growth is localized in regular spatial patterns. Such "patterned vegetation" occurs across a variety of vegetation and soil types, extends over at least 18 million ha, occurs in 5 continents and is economically and environmentally valuable in its own right.</p> <p>Vegetation patterning in drylands arises due to positive feedbacks between hydrological forcing and plant growth so that the patterns change in response to trends in mean annual rainfall. Mathematical models indicate that vegetation patterns collapse to a desertified state after undergoing a characteristic set of transformations so that the condition of a pattern at any point in time can be explicitly linked to ecosystem health. This dissertation focuses on the mathematical description of vegetation patterns with a view to improving such predictions. It evaluates the validity of current mathematical descriptions of patterning for the specific case of small-scale vegetation patterns and proposes alternative hypotheses for their formation. It assesses the significance of seed dispersal in determining pattern form and dynamics for two cases: vegetation growing on flat ground with isotropic patterning, and vegetation growing on slopes and having anisotropic (i.e. directional) patterning. Thirdly, the feedbacks between local biomass density and infiltration capacity, one of the positive feedbacks believed to contribute to patterning, are quantified across a wide range of soil and climatic conditions, and new mathematical descriptions of the biomass-infiltration relationship are proposed. Finally the influence of land surface microtopography on the partitioning of rainfall into infiltration and runoff is assessed.</p> / Dissertation

The objectives of my research were twofold: 1) to evaluate new technologies (camera traps and a new type of GPS collar) for studying large mammals in tropical forests, and 2) to study the ecology of the lowland tapir (Tapirus terrestris) in the Peruvian Amazon. Camera traps proved to be an efficient tool for mammal inventories in tropical forests. They recorded 24 out of 28 terrestrial medium and large sized mammal species with a survey effort of 2340 camera days. Camera traps were also able to reveal important information on habitat use, activity patterns and the use of mineral licks for five Amazonian ungulate species. There was a high spatial overlap between all the species with the grey brocket deer being the only species that was restricted to terra firme forest. White-lipped peccaries, tapirs and red brocket deer frequently used mineral licks, whereas collared peccaries and grey brocket deer were hardly ever observed at licks. A new type of GPS collar (TrackTag) tested in this study performed well under the dense canopy of a tropical forest. Position success rates of 87% for stationary collars and 48% for collars placed on tapirs were comparable to data obtained with GPS collars in temperate forests. The mean location error for stationary collars inside the forest was 28.9 m and the 95% error was 76.8 m. GPS collars placed on six tapirs for seven to 182 days showed home range sizes of 102 to 386 hectares. Tapirs were mainly nocturnal and areas used for foraging during the night differed from resting sites used during the day. Tapirs could walk up to 10 km to visit a mineral lick. Visits were irregular at intervals of a few days up to 36 days. The analysis of 135 tapir dung samples showed that tapirs ingest seeds of over 120 plant species. Seeds were found throughout the year but monthly species diversity was related to fruit availability. The size distribution of ingested seeds was related to availability. Most seeds were less then 10 mm wide, but seeds up to 25 mm were found. Both camera traps and TrackTag GPS collars greatly increased the possibilities for studying large rainforest mammals. The two technologies collect complementary information and each one is suited for a different set of questions.

Invasive alien plants have impacted southern U.S. forests. This study focuses on quantifying current distribution, rate of dispersals, and economic losses caused by four major invaders — Chinese tallow (Triadica sebifera (L.) Small), Chinese Privet (Ligustrum sinense Lour.), European Privet (Ligustrum vulgare L.), and Japanese Honeysuckle (Lonicera japonica Thunb.). Using data from USDA Forest Service’s Forest Inventory and Analysis (FIA), National Oceanic and Atmospheric Administration (NOAA), and USGS Land Use and Land Cover (LULC), I first used logistic regressions to build occupation models for these four species. Second, I used VB.NET© (Microsoft, 2003) and Environment System Research Institute (ESRI) Arc View® 9.2 Geographical Information System (GIS) to model spatially explicit dispersal dynamics of these species. Finally, I evaluated economic losses associated with these invasions. The results of occupation models showed that landscape features, climatic conditions, forest and site conditions, forest management activities and disturbances, and forest ownership have statistically significant impacts on current distributions of all four species. The results of dispersal models showed that invasions of Chinese tallow, Chinese and European privets, and Japanese honeysuckle will be likely to appear on approximately 6.65 million acres of forest land in East Texas and Louisiana, 3.81 million acres in Mississippi and Alabama, and 12.55 million acres in Mississippi and Alabama, respectively, during the next 20 years. The estimated timber losses resulting from the invasion with no control for the next 20 years at 5% discount rate is 518 million dollars for Chinese tallow, 2.72 billion dollars for Chinese and European privets, and 11.05 billion dollars for Japanese honeysuckle. From an economic point of view, it would be most cost effective to execute high intensity control for these invasive species. Based on these findings, strategies for monitoring and mitigating these invasive species are proposed.

It remains unlikely that a terrorist organization could produce or procure an actual nuclear weapon. However, the construction of a radiological dispersal device (RDD) from commercially produced radioactive sources and conventional explosives could inflict moderate human casualties and significant economic damage. The vast availability of radioactive sources and the nearly limitless methods of dispersing them demand an inclusive study of the acquisition pathways for an RDD. A complete network depicting the possible acquisition pathways for an RDD could be subjected to predictive modeling in order to determine the most likely pathway an adversary might take. In this work, a comprehensive network of RDD acquisition pathways was developed and analyzed utilizing the Bayesian network analysis software, Netica. The network includes variable inputs and motivations that can be adjusted to model different adversaries. Also, the inclusion of evidence nodes facilitates the integration of real-time intelligence with RDD plot predictions. A sensitivity analysis was first performed to determine which nodes had the greatest impact on successful completion of RDD acquisition. These results detail which portions of the acquisition pathways are most vulnerable to law enforcement intervention. Next, a series of case studies was analyzed that modeled specific adversarial organizations. The analysis demonstrates various features of the constructed Bayesian RDD acquisition network and provides examples of how this tool can be utilized by intelligence analysts and law enforcement agencies. Finally, extreme cases were studied in which the adversary was given the maximum and minimum amount of resources in order to determine the limitations of this model. The aggregated results show that successful RDD acquisition is mostly dependent on the adversary’s resources. Furthermore, the network suggests that securing radiological materials has the greatest effect on interdicting possible RDD plots. Limitations of this work include a heavy dependence on conditional probabilities that were derived from intuition, as opposed to actual historical data which does not exist. However, the model can be updated as attempted or successful RDD plots emerge in the future. This work presents the first probabilistic model of RDD acquisition pathways that integrates adversary motivations and resources with evidence of specific RDD threats.

The study intended to investigate the phylogeography of mid-altitude species, Japalura brevipes, which is the endemic to Taiwan. I used mitochondrial cytochrome b sequences and morphological characters to investigate the phylogeography. The molecular results showed that there are 16 haplotypes. The length of partial cytochrome b sequences is 617 bp including 84 informative sites, and the mean genetic distance is 5.4%. In phylogenetic analyses, three major clades were found in the phylogenetic trees. Moreover, these clades (north, central-north and south) corresponded to the geographic distribution. Analysis of population genetic structure revealed significant differentiation among populations, and most of the haplotypes were restricted to the locally. The morphological analyses showed that there is no difference between sexes at 18 characters, with the exception of the axilla-groin length and dorsal-crest scales. The Principal Component Analysis and Canonical Discriminate Analysis using 16 morphological characters showed the different results. Principal Component Analysis failed to separate populations. However, Canonical Discriminate Analysis could discriminate the populations of Guanwu, Nanheng, Sihyuantkou and Wuling Farm from others. Phylogeography of J. brevipes belonged to category 1 defined by Avise (2000). I infer refuge effect and dispersal ability hypotheses to explain the genetic structure of J. brevipes.

In the event that a radiological dispersal device (RDD) is detonated in the U.S. or near U.S. interests overseas, it will be crucial that the actors involved in the event can be identified quickly. If irradiated nuclear fuel is used as the dispersion material for the RDD, it will be beneficial for law enforcement officials to quickly identify where the irradiated nuclear fuel originated. One signature which may lead to the identification of the spent fuel origin is the isotopic composition of the RDD debris. The objective of this research was to benchmark a forward model methodology for predicting isotopic composition of spent nuclear fuel used in an RDD while at the same time optimizing the fidelity of the model to reduce computational time. The code used in this study was Monteburns-2.0. Monteburns is a Monte Carlo based neutronic code utilizing both MCNP and ORIGEN. The size of the burnup step used in Monteburns was tested and found to converge at a value of 3,000 MWd/MTU per step. To ensure a conservative answer, 2,500 MWd/MTU per step was used for the benchmarking process. The model fidelity ranged from the following: 2-dimensional pin cell, multiple radial-region pin cell, modified pin cell, 2D assembly, and 3D assembly. The results showed that while the multi-region pin cell gave the highest level of accuracy, the difference in uncertainty between it and the 2D pin cell (0.07% for 235U) did not warrant the additional computational time required. The computational time for the multiple radial-region pin cell was 7 times that of the 2D pin cell. For this reason, the 2D pin cell was used to benchmark the isotopics with data from other reactors. The reactors from which the methodology was benchmarked were Calvert Cliffs Unit #1, Takahama Unit #3, and Trino Vercelles. Calvert Cliffs is a pressurized water reactor (PWR) using Combustion Engineering 14??14 assemblies. Takahama is a PWR using Mitsubishi Heavy Industries 17??17 assemblies. Trino Vercelles is a PWR using non-standard lattice assemblies. The measured isotopic concentrations from all three of the reactors showed good agreement with the calculated values.

To determine possible causes of declining Rio Grande wild turkey (RGWT; Meleagris gallopavo intermedia) abundance in the southern Edwards Plateau, research was conducted on 4 sites, 2 with stable (S [SA and SB]; Kerr and Real counties) and 2 with declining (D [DA and DB]; Bandera County) RGWT populations. RGWTs were trapped, radio-tagged, and tracked. Ranges were constructed with 95% kernels. Data on brood survival and invertebrate and predator abundances were combined with range characteristics to assess habitat at a landscape scale. Annual range sizes did not differ in year 1, but were larger in S than in D in year 2. Range sizes in S increased from year 1 to year 2 while there was no change in range sizes in D. Range overlap was higher in D than S in both years. Movement distances remained consistent in S for both years, but were larger in D during year 1. During year 1 and year 2, RGWT females exhibited larger reproductive ranges and less range overlap in S. Invertebrate abundance for 4 insect orders was 2.5–15.9 times greater in S than in D while coyote abundance was 2–3 times greater in D than in S. Results were similar in year 3, with the exception of SB, where reproductive ranges and spatial arrangement were smaller than all other sites. My results refute the conventional assumption that larger ranges are indicative of poorer habitat quality. Range overlap suggests that useable space may have been limiting in D in the less productive year 2. In D, multiple broods used the same reproductive range, presumably depleting resources faster than in S. Greater predator abundance in D increased the risk of brood predation. The smaller reproductive spatial arrangement of SB females in year 3 correlates there being >3 times the percentage of females missing in other sites. If SB females moved further in year 3 than the detection distance of the radio telemetry equipment, the results would fit the pattern of greater dispersion distance in SA. RGWT females may attempt to separate themselves from other breeding females, possibly to avoid nest or brood predation and/or potential competition for brood resources.

Population structure can be defined as a difference in the frequencies of alleles among populations, and it is often caused by barriers to dispersal. I used mitochondrial DNA sequences to examine the population structure of two large, aquatic salamanders, Amphiuma means (the Two-Toed Amphiuma) and Siren lacertina (the Greater Siren), at two spatial scales in peninsular Florida. Both species appear to have similar life histories, including poor dry land dispersal capabilities, and I hypothesized that they would exhibit significant population structure at both broad and fine scales. Wetland loss and landscape fragmentation are certain to have caused decreases in abundance and even extirpation of local populations of both species. Understanding current levels of genetic diversity is important in conservation efforts, particularly if individuals must be translocated in the future. A portion of the cytochrome b gene was analyzed for 34 Amphiuma means and 21 Siren lacertina sequences. A neighbor-joining tree and haplotype network showed that A. means exhibited statistically significant population structure at the broad, but not the fine scale. The lack of structure at the fine scale was attributed to periodic flooding among drainage basins in the low-lying study area. The same analyses showed that S. lacertina did not exhibit statistically significant population structure at the broad scale (analyses were not conducted at the fine scale because of low sample size). The difference in broad-scale population structure between the two species is the result of a difference in gene flow, which in turn, can be the result of a difference in effective population size (Ne), female dispersal capability and/or behavior, chance of bait-bucket transfer, or biogeographic history. The best of these possible explanations seems to be a difference in biogeographic history. For the two sites from which both species were sampled, estimated population divergence date was 3.3-6.4 million years ago (mya) for A. means and 0.04-0.5 mya for S. lacertina. The large difference in estimated divergence dates suggests that S. lacertina colonized peninsular Florida from somewhere else following a glacial sea level rise or an interglacial sea level fall, at a time when A. means was already present in refugia within peninsular Florida. This is the first study to explicitly examine the population genetics of A. means and S. lacertina, and future studies should examine their population genetics range wide, to elucidate how gene flow and/or historical biogeography may have contributed to population structure elsewhere.

Studies of secondary seed dispersal by small mammals have largely been focused on the interaction between nut-bearing tree species and sciurid rodents such as squirrels, and on heteromyid rodents in the southwestern United States. However, there is now evidence that wind-dispersed tree species such as pines also undergo a process of secondary seed dispersal, where animals redistribute (cache) seeds that have already fallen to the ground, often in microhabitats more suitable for successful seed germination. In Florida sandhill, where fire suppression has threatened wind-dispersed longleaf pine ( Pinus palustris) by encouraging the encroachment of hardwoods such as sand live oak ( Quercus geminata), secondary seed dispersal may be an important factor in determining community composition and persistence of longleaf pine systems. Using a combination of seed depots and seed predator exclosures, I looked at both longleaf pine and sand live oak in terms of whether small animals such as squirrels ( Sciurus carolinensis) and cotton mice ( Peromyscus gossypinus) cache the seeds, and where the seeds of these two tree species best germinate. Since sand live oak acorns are prone to infestation by weevils ( Curculio spp.), I also examined whether nut condition affects acorn germination potential. I found that longleaf pine seeds are cached by small mammals to a small degree. While these seeds are not moved great distances from where they originate, they are often redistributed into microhabitats that promote successful seed germination. Caging experiments indicated that seeds were most likely to germinate when buried in open areas between adult trees, and to some degree, under shrub cover. On the other hand, sand live oak acorns appear to face heavy predation by large seed predators such as raccoons ( Procyon lotor) and wild pigs (Sus scrofa). Those acorns that do escape predation, including weevil-infested acorns, may provide an opportunity for seedling establishment. However, it appears that sand live oak depends heavily on vegetative sprouting for regeneration. This suggests that even in the absence of fire, longleaf pines in Florida sandhill are able to persist through secondary seed dispersal by small animals coupled with heavy seed predation on competing sand live oak.