Designing a Computer Vision System for Underwater Robotic Interventions

This thesis deals with the challenging problem of designing systems able to perceive objects in underwater environments. In the last few decades research activities in robotics have advanced the state of art regarding intervention capabilities of autonomous systems. State of art in fields such as localization and navigation, real time perception and cognition, safe action and manipulation capabilities, applied to ground environments (both indoor and outdoor) has now reached such a readiness level that it allows high level autonomous operations. On the opposite side, the underwater environment remains a very difficult one for autonomous robots. Water influences the mechanical and electrical design of systems, interferes with sensors by limiting their capabilities, heavily impacts on data transmissions, and generally requires systems with low power consumption in order to enable reasonable mission duration. Interest in underwater applications is driven by needs of exploring and intervening in environments in which human capabilities are very limited. Nowadays, most underwater field operations are carried out by manned or remotely operated vehicles, deployed for explorations and limited intervention missions. Manned vehicles, directly on-board controlled, expose human operators to risks related to the stay in field of the mission, within a hostile environment. Remotely Operated Vehicles (ROV) currently represent the most advanced technology for underwater intervention services available on the market. These vehicles can be remotely operated for long time but they need support from an oceanographic vessel with multiple teams of highly specialized pilots. Vehicles equipped with multiple state-of-art sensors and capable to autonomously plan missions have been deployed in the last ten years and exploited as observers for underwater fauna, seabed, ship wrecks, and so on. On the other hand, underwater operations like object recovery and equipment maintenance are still challenging tasks to be conducted without human supervision since they require object perception and localization with much higher accuracy and robustness, to a degree seldom available in Autonomous Underwater Vehicles (AUV). This thesis reports the study, from design to deployment and evaluation, of a general purpose and configurable platform dedicated to stereo-vision perception in underwater environments. Several aspects related to the peculiar environment characteristics have been taken into account during all stages of system design and evaluation: depth of operation and light conditions, together with water turbidity and external weather, heavily impact on perception capabilities. The vision platform proposed in this work is a modular system comprising off-the-shelf components for both the imaging sensors and the computational unit, linked by a high performance ethernet network bus. The adopted design philosophy aims at achieving high flexibility in terms of feasible perception applications, that should not be as limited as in case of a special-purpose and dedicated hardware. Flexibility is required by the variability of underwater environments, with water conditions ranging from clear to turbid, light backscattering varying with daylight and depth, strong color distortion, and other environmental factors. Furthermore, the proposed modular design ensures an easier maintenance and update of the system over time. Performance of the proposed system, in terms of perception capabilities, has been evaluated in several underwater contexts taking advantage of the opportunity offered by the MARIS national project. Design issues like energy power consumption, heat dissipation and network capabilities have been evaluated in different scenarios. Finally, real-world experiments, conducted in multiple and variable underwater contexts, including open sea waters, have led to the collection of several datasets that have been publicly released to the scientific community. The vision system has been integrated in a state of the art AUV equipped with a robotic arm and gripper, and has been exploited in the robot control loop to successfully perform underwater grasping operations.

High performance liquid level monitoring system based on polymer fiber Bragg gratings embedded in silicone rubber diaphragm

Liquid-level sensing technologies have attracted great prominence, because such measurements are essential to industrial applications, such as fuel storage, flood warning and in the biochemical industry. Traditional liquid level sensors are based on electromechanical techniques; however they suffer from intrinsic safety concerns in explosive environments. In recent years, given that optical fiber sensors have lots of well-established advantages such as high accuracy, costeffectiveness, compact size, and ease of multiplexing, several optical fiber liquid level sensors have been investigated which are based on different operating principles such as side-polishing the cladding and a portion of core, using a spiral side-emitting optical fiber or using silica fiber gratings. The present work proposes a novel and highly sensitive liquid level sensor making use of polymer optical fiber Bragg gratings (POFBGs). The key elements of the system are a set of POFBGs embedded in silicone rubber diaphragms. This is a new development building on the idea of determining liquid level by measuring the pressure at the bottom of a liquid container, however it has a number of critical advantages. The system features several FBG-based pressure sensors as described above placed at different depths. Any sensor above the surface of the liquid will read the same ambient pressure. Sensors below the surface of the liquid will read pressures that increase linearly with depth. The position of the liquid surface can therefore be approximately identified as lying between the first sensor to read an above-ambient pressure and the next higher sensor. This level of precision would not in general be sufficient for most liquid level monitoring applications; however a much more precise determination of liquid level can be made by linear regression to the pressure readings from the sub-surface sensors. There are numerous advantages to this multi-sensor approach. First, the use of linear regression using multiple sensors is inherently more accurate than using a single pressure reading to estimate depth. Second, common mode temperature induced wavelength shifts in the individual sensors are automatically compensated. Thirdly, temperature induced changes in the sensor pressure sensitivity are also compensated. Fourthly, the approach provides the possibility to detect and compensate for malfunctioning sensors. Finally, the system is immune to changes in the density of the monitored fluid and even to changes in the effective force of gravity, as might be obtained in an aerospace application. The performance of an individual sensor was characterized and displays a sensitivity (54 pm/cm), enhanced by more than a factor of 2 when compared to a sensor head configuration based on a silica FBG published in the literature, resulting from the much lower elastic modulus of POF. Furthermore, the temperature/humidity behavior and measurement resolution were also studied in detail. The proposed configuration also displays a highly linear response, high resolution and good repeatability. The results suggest the new configuration can be a useful tool in many different applications, such as aircraft fuel monitoring, and biochemical and environmental sensing, where accuracy and stability are fundamental. © (2015) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.

Medical emergency alarm dissemination in urban environments

During medical emergencies, the ability to communicate the state and position of injured individuals is essential. In critical situations or crowd aggregations, this may result difficult or even impossible due to the inaccuracy of verbal communication, the lack of precise localization for the medical events, and/or the failure/congestion of infrastructure-based communication networks. In such a scenario, a temporary (ad hoc) wireless network for disseminating medical alarms to the closest hospital, or medical field personnel, can be usefully employed to overcome the mentioned limitations. This is particularly true if the ad hoc network relies on the mobile phones that people normally carry, since they are automatically distributed where the communication needs are. Nevertheless, the feasibility and possible implications of such a network for medical alarm dissemination need to be analysed. To this aim, this paper presents a study on the feasibility of medical alarm dissemination through mobile phones in an urban environment, based on realistic people mobility. The results showed the dependence between the medical alarm delivery rates and both people and hospitals density. With reference to the considered urban scenario, the time needed to delivery medical alarms to the neighbour hospital with high reliability is in the order of minutes, thus revealing the practicability of the reported network for medical alarm dissemination. © 2013 Elsevier Ltd. All rights reserved.

The effects of material variations in high value manufacturing industries

It has never been easy for manufacturing companies to understand their confidence level in terms of how accurate and to what degree of flexibility parts can be made. This brings uncertainty in finding the most suitable manufacturing method as well as in controlling their product and process verification systems. The aim of this research is to develop a system for capturing the company’s knowledge and expertise and then reflect it into an MRP (Manufacturing Resource Planning) system. A key activity here is measuring manufacturing and machining capabilities to a reasonable confidence level. For this purpose an in-line control measurement system is introduced to the company. Using SPC (Statistical Process Control) not only helps to predict the trend in manufacturing of parts but also minimises the human error in measurement. Gauge R&R (Repeatability and Reproducibility) study identifies problems in measurement systems. Measurement is like any other process in terms of variability. Reducing this variation via an automated machine probing system helps to avoid defects in future products.Developments in aerospace, nuclear, oil and gas industries demand materials with high performance and high temperature resistance under corrosive and oxidising environments. Superalloys were developed in the latter half of the 20th century as high strength materials for such purposes. For the same characteristics superalloys are considered as difficult-to-cut alloys when it comes to formation and machining. Furthermore due to the sensitivity of superalloy applications, in many cases they should be manufactured with tight tolerances. In addition superalloys, specifically Nickel based, have unique features such as low thermal conductivity due to having a high amount of Nickel in their material composition. This causes a high surface temperature on the work-piece at the machining stage which leads to deformation in the final product.Like every process, the material variations have a significant impact on machining quality. The main cause of variations can originate from chemical composition and mechanical hardness. The non-uniform distribution of metal elements is a major source of variation in metallurgical structures. Different heat treatment standards are designed for processing the material to the desired hardness levels based on application. In order to take corrective actions, a study on the material aspects of superalloys has been conducted. In this study samples from different batches of material have been analysed. This involved material preparation for microscopy analysis, and the effect of chemical compositions on hardness (before and after heat treatment). Some of the results are discussed and presented in this paper.

Liquid desiccant dehumidification and regeneration process to meet cooling and freshwater needs of desert greenhouses

Agriculture accounts for ~70% of freshwater usage worldwide. Seawater desalination alone cannot meet the growing needs for irrigation and food production, particularly in hot, desert environments. Greenhouse cultivation of high-value crops uses just a fraction of freshwater per unit of food produced when compared with open field cultivation. However, desert greenhouse producers face three main challenges: freshwater supply, plant nutrient supply, and cooling of the greenhouse. The common practice of evaporative cooling for greenhouses consumes large amounts of fresh water. In Saudi Arabia, the most common greenhouse cooling schemes are fresh water-based evaporative cooling, often using fossil groundwater or energy-intensive desalinated water, and traditional refrigeration-based direct expansion cooling, largely powered by the burning of fossil fuels. The coastal deserts have ambient conditions that are seasonally too humid to support adequate evaporative cooling, necessitating additional energy consumption in the dehumidification process of refrigeration-based cooling. This project evaluates the use of a combined-system liquid desiccant dehumidifier and membrane distillation unit that can meet the dual needs of cooling and freshwater supply for a greenhouse in a hot and humid environment.

The impact of airport shopping environments and dwell time on consumer spending

This article provides new insight into how the ambience and design of shopping environments impact onspending behaviour. Environmental cues in a retail area influence emotional states of by-passers, which in turn influence spending levels. Past research suggested that this effect only applies to shops with moderate arousal level. Also, several studies failed to confirm a relationship between emotions and spending levels. This is surprising, since high arousal environments (e.g., amusement parks, sports stadiums and airports) often feature a wide range of retail outlets. Based on survey data collected in a live airport shopping area, this study finds a relationship between pleasure emotions associated with the retail area and recalled consumer spending, but also the time available for shopping (which in an airport is constrained). Also, visitors’ emotional state was influenced by the ambience (e.g., cleanliness, noise levels, lighting) as well as the design (e.g., easy wayfinding, seating areas) of the retail area. Shopper’s arousal levels did not explain variations in spending level. Implications for researchers and managers are discussed as well as suggestions for future research.

Organic matter dynamics in the Florida coastal Everglades: A molecular marker and isotopic approach

Everglades National Park (ENP) is about to undergo the world's largest wetland restoration with the aim of improving the quality, timing and distribution of water flow. The changes in water flow are hypothesized to alter the nutrient fluxes and organic matter (OM) dynamics within ENP, especially in the estuarine areas. This study used a multi-proxy approach of molecular markers and stable δ 13C isotope measurements, to determine the present day OM dynamics in ENP. ^ OM dynamics in wetland soils/sediments have proved to be difficult to understand using traditional geochemical approaches. These are often inadequate to describe the multitude of OM sources (e.g. higher land plant, emergent vegetation, submerged vegetation) to the soils/sediments and the complex diagenetic processes that can alter the OM characteristics. A multi-proxy approach, however, that incorporates both molecular level and bulk parameter information is ideal to comprehend complex OM dynamics in aquatic environments. Therefore, biomass-specific molecular markers or proxies can be useful in tracing the sources and processing of OM. This approach was used to examine the OM dynamics in the two major drainage basins, Shark River Slough and Taylor River Slough, of ENP. Freshwater to marine transects were sampled in both systems for soils/sediments and suspended particulate organic matter (SPOM) to be characterized through bulk OM analyses, lipid biomarker determinations (e.g. sterols, fatty acids, hydrocarbons and triterpenoids) and compound-specific stable carbon isotope (δ 13C) determinations. ^ One key accomplishment of the research was the assessment of a molecular marker proxy (Paq) to distinguish between emergent/higher plant vegetation from submerged vegetation within ENP. This proxy proved to be quite useful at tracing OM inputs to the soils/sediments of ENP. A second key accomplishment was the development of a 3-way model using vegetation specific molecular markers. This novel, descriptive model was successfully applied to the estuarine areas of Taylor and Shark River sloughs, providing clear evidence of mixing of freshwater, estuarine and marine derived OM in these areas. In addition, diagenetic transformations of OM in these estuaries were found to be quite different between Taylor and Shark Rivers, and are likely a result of OM quality and hydrological differences. ^

Diatom-based paleolimnological reconstruction of Quaternary environments in a Florida sinkhole lake

Despite lake sensitivity to climate change, few Florida paleolimnological studies have focused on changes in hydrology. Evidence from Florida vegetation histories raise questions about long-term hydrologic history of Florida lakes, and a 25-year limnological dataset revealed recent climate-driven effects on Lake Annie. The objectives of this research are (1) to use modern diatom assemblages to develop methods for reconstruction of climatic and anthropogenic change (2) to reconstruct both long-term and recent histories of Lake Annie using diatom microfossils. Paleoenvironmental reconstruction models were developed from diatom assemblages of various habitat types from modern lakes. Plankton and sediment assemblages were similar, but epiphytes were distinct, suggesting differences in sediment delivery from different parts of the lakes. Relationships between a variety of physical and chemical data and the diatoms from each habitat type were explored. Total phosphorus (TP), pH, and color were found to be the most relevant variables for reconstruction, with sediment and epiphyte assemblages having the strongest relationships to those variables, six calibration models were constructed from the combination of these habitat types and environmental variables. Reconstructions utilizing the weighted averaging models in this study may be used to directly reveal TP, color, and pH changes from a sediment record, which might be suggestive of hydrologic change as well. These variables were reconstructed from the diatom record from both a long-term (11,000 year) and short-term (100 year) record and showed an interaction between climate-driven and local land-use impacts on Lake Annie. The long-term record begins with Lake Annie as a wetland, then the lake filled to a high stand around 4000 years ago. A period of relative stability after that point was interrupted near the turn of the last century by subtle changes in diatom communities that indicate acidification. Abrupt changes in the diatom communities around 1970 AD suggest recovery from acidification, but concurrent hydrologic change intensified anthropogenic effects on the lake. Diatom evidence for alkalization and phosphorus loading correspond to changes seen in the limnological record.

Carbon and nitrogen stable isotopic patterns in South Florida coastal ecosystems: Modern and paleoceanographic perspectives

Long term management plans for restoration of natural flow conditions through the Everglades increase the importance of understanding potential nutrient impacts of increased freshwater delivery on coastal biogeochemistry. The present study sought to increase understanding of the coastal marine system of South Florida under modern conditions and through the anthropogenic changes in the last century, on scales ranging from individual nutrient cycle processes to seasonal patterns in organic material (OM) under varying hydrodynamic regime, to century scale analysis of sedimentary records. In all applications, carbon and nitrogen stable isotopic compositions of OM were examined as natural recorders of change and nutrient cycling in the coastal system. ^ High spatial and temporal variability in stable isotopic compositions were observed on all time scales. During a transient phytoplankton bloom, δ 15N values suggested nitrogen fixation as a nutrient source supporting enhanced productivity. Seasonally, particulate organic material (POM) from ten sites along the Florida Reef Tract and in Florida Bay demonstrated variable fluctuations dependent on hydrodynamic setting. Three separate intra-annual patterns were observed, yet statistical differences were observed between groupings of Florida Bay and Atlantic Ocean sites. The POM δ 15N values ranged on a quarterly basis by 7‰, while δ 13C varied by 22‰. From a sediment history perspective, four cores collected from Florida Bay further demonstrated the spatial and temporal variability of the system in isotopic composition of bulk OM over time. Source inputs of OM varied with location, with terrestrial inputs dominating proximal to Everglades freshwater discharge, seagrasses dominating in open estuary cores, and a marine mixture of phytoplankton and seagrass in a core from the boundary zone between Florida Bay and the Gulf of Mexico. Significant shifts in OM geochemistry were observed coincident with anthropogenic events of the 20th century, including railroad and road construction in the Florida Keys and Everglades, and also the extensive drainage changes in Everglades hydrology. The sediment record also preserved evidence of the major hurricanes of the last century, with excursions in geochemical composition coincident with Category 4-5 storms. ^

Local biomass control on the composition and reactivity of particulate organic matter in aquatic environments

Freshwater ecosystems have been recognized as important components of the global carbon cycle, and the flux of organic matter (OM) from freshwater to marine environments can significantly affect estuarine and coastal productivity. The focus of this study was the assessment of carbon dynamics in two aquatic environments, namely the Florida Everglades and small prairie streams in Kansas, with the aim of characterizing the biogeochemistry of OM. In the Everglades, particulate OM (POM) is mostly found as a layer of flocculent material (floc). While floc is believed to be the main energy source driving trophic dynamics in this oligotrophic wetland, not much is known about its biogeochemistry. The objective of this study was to determine the origin/sources of OM in floc using biomarkers and pigment-based chemotaxonomy to assess specific biomass contributions to this material, on a spatial (freshwater marshes vs. mangrove fringe) and seasonal (wet vs. dry) scales. It was found that floc OM is derived from the local vegetation (mainly algal components and macrophyte litter) and its composition is controlled by seasonal drivers of hydrology and local biomass productivity. Photo-reactivity experiments showed that light exposure on floc resulted in photo-dissolution of POC with the generation of significant amounts of both dissolved OM (DOM) and nutrients (N & P), potentially influencing nutrient dynamics in this ecosystem. The bio-reactivity experiments determined as the amount and rate of CO2 evolution during incubation were found to vary on seasonal and spatial scales and were highly influenced by phosphorus limitation. Not much is known on OM dynamics in small headwater streams. The objective of this study was to determine carbon dynamics in sediments from intermittent prairie streams, characterized by different vegetation cover for their watershed (C4 grasses) vs. riparian zone (C3 plants). In this study sedimentary OM was characterized using a biomarker and compound specific carbon stable isotope approach. It was found that the biomarker composition of these sediments is dominated by higher plant inputs from the riparian zone, although inputs from adjacent prairie grasses were also apparent. Conflicting to some extent with the River Continuum Concept, sediments of the upper reaches contained more degraded OM, while the lower reaches were enriched in fresh material deriving from higher plants and plankton sources as a result of hydrological regimes and particle sorting.

Tracking rates of ecotone migration due to salt-water encroachment using fossil mollusks in coastal South Florida

We determined the rate of migration of coastal vegetation zones in response to salt-water encroachment through paleoecological analysis of mollusks in 36 sediment cores taken along transects perpendicular to the coast in a 5.5 km2 band of coastal wetlands in southeast Florida. Five vegetation zones, separated by distinct ecotones, included freshwater swamp forest, freshwater marsh, and dwarf, transitional and fringing mangrove forest. Vegetation composition, soil depth and organic matter content, porewater salinity and the contemporary mollusk community were determined at 226 sites to establish the salinity preferences of the mollusk fauna. Calibration models allowed accurate inference of salinity and vegetation type from fossil mollusk assemblages in chronologically calibrated sediments. Most sediments were shallow (20–130 cm) permitting coarse-scale temporal inferences for three zones: an upper peat layer (zone 1) representing the last 30–70 years, a mixed peat-marl layer (zone 2) representing the previous ca. 150–250 years and a basal section (zone 3) of ranging from 310 to 2990 YBP. Modern peat accretion rates averaged 3.1 mm yr)1 while subsurface marl accreted more slowly at 0.8 mm yr)1. Salinity and vegetation type for zone 1 show a steep gradient with freshwater communities being confined west of a north–south drainage canal constructed in 1960. Inferences for zone 2 (pre-drainage) suggest that freshwater marshes and associated forest units covered 90% of the area, with mangrove forests only present along the peripheral coastline. During the entire pre-drainage history, salinity in the entire area was maintained below a mean of 2 ppt and only small pockets of mangroves were present; currently, salinity averages 13.2 ppt and mangroves occupy 95% of the wetland. Over 3 km2 of freshwater wetland vegetation type have been lost from this basin due to salt-water encroachment, estimated from the mollusk-inferred migration rate of freshwater vegetation of 3.1 m yr)1 for the last 70 years (compared to 0.14 m yr)1 for the pre-drainage period). This rapid rate of encroachment is driven by sea-level rise and freshwater diversion. Plans for rehydrating these basins with freshwater will require high-magnitude re-diversion to counteract locally high rates of sea-level rise.

Long-Term Effects of Adding Nutrients to an Oligotrophic Coastal Environment

Management of ecological disturbances requires an understanding of the time scale and dynamics of community responses to disturbance events. To characterize long-term seagrass bed responses to nutrient enrichment, we established six study sites in Florida Bay, USA. In 24 plots (0.25 m2) at each site, we regularly added nitrogen (N) and phosphorus (P) in a factorial design for 7 years. Five of the six sites exhibited strong P limitation. Over the first 2 years, P enrichment increased Thalassia testudinum cover in the three most P-limited sites. After 3 years, Halodule wrightii began to colonize many of the P-addition plots, but the degree of colonization was variable among sites, possibly due to differences in the supply of viable propagules. Thalassia increased its allocation to aboveground tissue in response to P enrichment; Halodule increased in total biomass but did not appear to change its aboveground: belowground tissue allocation. Nutrient enrichment did not cause macroalgal or epiphytic overgrowth of the seagrass. Nitrogen retention in the study plots was variable but relatively low, whereas phosphorus retention was very high, often exceeding 100% of the P added as fertilizer over the course of our experiments. Phosphorus retentions exceeding 100% may have been facilitated by increases in Thalassia aboveground biomass, which promoted the settlement of suspended particulate matter containing phosphorus. Our study demonstrated that lowintensity press disturbance events such as phosphorus enrichment can initiate a slow, ramped successional process that may alter community structure over many years.

A synthesis of long-term research by the Florida Coastal Everglades LTER Program

This paper synthesizes research conducted during the first 5–6 years of the Florida Coastal Everglades Long-Term Ecological Research Program (FCE LTER). My objectives are to review our research to date, and to present a new central theme and conceptual approach for future research. Our research has focused on understanding how dissolved organic matter (DOM) from upstream oligotrophic marshes interacted with a marine source of the limiting nutrient, phosphorus (P), to control productivity in the oligohaline estuarine ecotone. We have been working along freshwater to marine transects in two drainage basins located in Everglades National Park (ENP). The Shark River Slough transect (SRS) has a direct connection to the Gulf of Mexico, providing this estuarine ecotone with a source of marine P. The oligohaline ecotone along our southern Everglades transect (TS/Ph), however, is separated from this marine P source by the Florida Bay estuary. We originally hypothesized an ecosystem productivity peak in the SRS ecotone, driven by the interaction of marine P and Everglades DOM, but no such productivity peak in the TS/Ph ecotone because of this lack of marine P. Our research to date has tended to show the opposite pattern, however, with many ecosystem components showing enhanced productivity in the TS/Ph ecotone, but not in the SRS ecotone. Water column P concentrations followed a similar pattern, with unexpectedly high P in the TS/Ph ecotone during the dry season. Our organic geochemical research has shown that Everglades DOM is more refractory than originally hypothesized. We have also begun to understand the importance of detrital organic matter production and transport to ecotone dynamics and as the base of aquatic food webs. Our future research will build on this substantial body of knowledge about these oligotrophic estuaries. We will direct our efforts more strongly on biophysical dynamics in the oligohaline ecotone regions. Specifically, we will be focusing on inputs to these regions from four primary water sources: freshwater Everglades runoff, net precipitation, marine inputs, and groundwater. We are hypothesizing that dry season groundwater inputs of P will be particularly important to TS/Ph ecotone dynamics because of longer water residence times in this area. Our organic geochemical, biogeochemical, and ecosystem energetics work will focus more strongly on the importance of detrital organics and will take advantage of a key Everglades Restoration project, scheduled for 2008 or 2009, that will increase freshwater inputs to our SRS transect only. Finally, we will also begin to investigate the human dimensions of restoration, and of a growing population in south Florida that will become increasingly dependent on the Everglades for critical ecosystem services (including fresh water) even as its growth presents challenges to Everglades sustainability.

The influence of body size on the ecology of coastal fish predators in the Bahamas

Body size is a fundamental structural characteristic of organisms, determining critical life history and physiological traits, and influencing population dynamics, community structure, and ecosystem function. For my dissertation, I focused on effects of body size on habitat use and diet of important coastal fish predators, as well as their influence on faunal communities in Bahamian wetlands. First, using acoustic telemetry and stable isotope analysis, I identified high variability in movement patterns and habitat use among individuals within a gray snapper (Lutjanus griseus) and schoolmaster snapper (L. apodus) population. This intrapopulation variation was not explained by body size, but by individual behavior in habitat use. Isotope values differed between individuals that moved further distances and individuals that stayed close to their home sites, suggesting movement differences were related to specific patterns of foraging behavior. Subsequently, while investigating diet of schoolmaster snapper over a two-year period using stomach content and stable isotope analyses, I also found intrapopulation diet variation, mostly explained by differences in size class, individual behavior and temporal variability. I then developed a hypothesis-testing framework examining intrapopulation niche variation between size classes using stable isotopes. This framework can serve as baseline to categorize taxonomic or functional groupings into specific niche shift scenarios, as well as to help elucidate underlying mechanisms causing niche shifts in certain size classes. Finally, I examined the effect of different-sized fish predators on epifaunal community structure in shallow seagrass beds using exclusion experiments at two spatial scales. Overall, I found that predator effects were rather weak, with predator size and spatial scale having no impact on the community. Yet, I also found some evidence of strong interactions on particular common snapper prey. As Bahamian wetlands are increasingly threatened by human activities (e.g., overexploitation, habitat degradation), an enhanced knowledge of the ecology of organisms inhabiting these systems is crucial for developing appropriate conservation and management strategies. My dissertation research contributed to this effort by providing critical information about the resource use of important Bahamian fish predators, as well as their effect on faunal seagrass communities.

Water Uptake Patterns of an Invasive Exotic Plant in Coastal Saline Habitats

Schinus terebinthifolius Raddi (Schinus) is one of the most widely found woody exotic species in South Florida. This exotic is distributed across environments with different hydrologic regimes, from upland pine forests to the edges of sawgrass marshes and into saline mangrove forests. To determine if this invasive exotic had different physiological attributes compared to native species in a coastal habitat, we measured predawn xylem water potentials (Ψ), oxygen stable isotope signatures (δ18O), and sodium (Na+) and potassium (K+) contents of sap water from plants within: (1) a transition zone (between a mangrove forest and upland pineland) and (2) an upland pineland in Southwest Florida. Under dynamic salinity and hydrologic conditions, Ψ of Schinus appeared less subject to fluctuations caused by seasonality when compared with native species. Although stem water δ18O values could not be used to distinguish the depth of Schinus and native species' water uptake in the transition zone, Ψ and sap Na+/K+ patterns showed that Schinus was less of a salt excluder relative to the native upland species during the dry season. This exotic also exhibited Na+/K+ ratios similar to the mangrove species, indicating some salinity tolerance. In the upland pineland, Schinus water uptake patterns were not significantly different from those of native species. Differences between Schinus and native upland species, however, may provide this exotic an advantage over native species within mangrove transition zones.