Identifying the ecological mechanisms that determine food web structure is critical for understanding the causes and consequences of diversity. The objective of this thesis was to identify the mechanisms structuring aquatic food webs across environmental gradients from a multi-level perspective (individual to ecosystem) using integrative methodology and field experiments to test classic ecological theory. My results demonstrate support for the dynamic constraints hypothesis, which predicts habitats with greater disturbance should have shorter food chains, but are not consistent with the ecosystem size hypothesis that predicts larger ecosystems have longer food chains. Insect and amphibian richness increased with increasing pond size and hydroperiod, indicating that insertion of new consumers into pond communities was driving variation in food-chain length. A multivariate analysis testing the influence of physicochemical variables on food-web characteristics revealed that hydroperiod and pond area had a strong influence on amphibian and invertebrate assemblages, trophic diversity and 15N range. Food-chain length did not respond strongly to any one variable, but instead responded weakly to multiple environmental variables, suggesting interacting influences on food-web structure. Conversely, the trophic niche of amphibian larvae was not influenced by pond hydroperiod, but did exhibit ontogenetic diet shifts. Populations of amphibian larvae with broader niche widths also had increased individual variation, supporting the niche variation hypothesis. In addition, I assessed whether species diversity influenced the magnitude of cross-habitat resource flow between aquatic and terrestrial habitats via emerging aquatic insects, metamorphosing amphibians, and litter deposition. Deposition into ponds far exceeded carbon exported via insect and amphibian emergences. We found a negative relationship between resource flux and the diversity of amphibians and insects, which contradicts the general pattern of positive biodiversity-ecosystem function relationships. My research strongly suggests environmental variation is a key process in shaping food-web structure and function and that multiple methodologies are needed to understand temporal and spatial dynamics of aquatic ecosystems.

Past monitoring has noted benthic macroinvertebrate community impairment downstream of both the Key Lake and Rabbit Lake uranium operations in northern Saskatchewan, Canada. The objective of this research was to try to identify the cause(s) of these impacts using a weight-of-evidence approach. Given that sediments generally accumulate contaminants that are related to metal mining activities (such as metals and radionuclides), the initial hypothesis for this research was that contaminated sediments were the primary cause of benthic community impairment at both operations.<p>In 2003 and 2004 a Sediment Quality Triad (SQT) approach confirmed the presence of an effect on benthic community structure, in addition to significant differences in surface-water, pore-water and whole-sediment chemistry at the immediate down-stream exposure sites at both uranium operations. However, no significant adverse effects were noted in 10-d whole-sediment bioassays with <i>Hyalella azteca</i>, although this lack of response could be partially due to sediment pore-water dilution resulting from the automated clean overlying water renewal process employed. Potential causes of benthic community impairment identified through the 2003 and 2004 SQTs for Key Lake include physical sediment composition, surface water pH and total ammonia, in addition to pore-water total ammonia and arsenic. Potential stressors identified at Rabbit Lake included high surface water manganese and uranium concentrations, and increases in pore-water total ammonia, manganese, iron, arsenic, and uranium levels.<p>In the summer of 2004, 4-d in-situ bioassays using <i>H. azteca</i> were conducted along with the SQTs to investigate the role both contaminated surface water and sediment played in benthic community impairment in-situ. Results from the Key Lake in-situ bioassay demonstrated that surface-water was the primary cause of acute toxicity to <i>H. azteca</i>. Results from the Rabbit Lake in-situ study also demonstrated that surface water as the primary cause of acute toxicity to <i>H. azteca</i>, although the relationship was not as strong. The cause of in-situ toxicity at Key Lake could not be correlated with any of the variables measured within the in-situ study, including trace metals, total ammonia, and pH. Of the measured constituents at Rabbit Lake, only concentrations of uranium in both surface water and pore-water were suspected of causing the observed in-situ mortality. Two data sets from two methods of surface water and pore-water collection supported these conclusions.<p>Due to time constraints and stronger cause-effect relationships, efforts were focused on the in-situ toxicity observed at Key Lake. Surface water collected in 2004 at the time of the related in-situ study was also found to be acutely toxic to <i>H. azteca</i> in separate laboratory surface water bioassays, thus verifying that contaminated surface water, not sediment, was the primary cause of the observed in-situ <i>H. azteca</i> mortality. Further information revealed that organic mill-process chemicals, which have been previously linked with sporadic effluent toxicity, were released at the Key Lake operation during the time of the in-situ experiment and associated surface water collection. Additional surface water samples collected in June and August, 2005, were not acutely toxic to <i>H. azteca</i>. Furthermore, a second bioassay with archived surface waters from the initial 2004 collection demonstrated that the water was no longer acutely toxic (i.e., acute toxicity disappeared after one-year storage). Chemistry comparisons of the toxic and non-toxic surface water samples, verified that trace metals, ammonia, pH, and major ions, including sulphate, were not the cause of toxicity, leaving only organic mill-process chemicals as a possible cause. Subsequent 4-d laboratory toxicity tests demonstrated that these process chemicals (kerosene, amine, and isodecanol) are toxic to H. azteca at the levels released in 2004, and are therefore believed to be the cause of the <i>H. azteca</i> mortality seen in the earlier in-situ experiment.<p>In short, this weight-of-evidence research provided new information on the possible causes of benthic macroinvertebrate community impairment downstream of both the Key Lake and Rabbit Lake uranium operations.

The lack of cost-effective, reliable sampling methods for many wetland characteristics hinders efforts to describe the structural and functional properties of wetlands. This study evaluated techniques for sampling the subsurface hydrology and invertebrates of freshwater wetlands. The depth of rusting on mild steel rods was compared with water well measurements to determine the reliability of rust depth as a predictor of subsurface water levels. An emergence trap and a benthic coring device were compared to determine the utility of each for sampling the invertebrate fauna of a wetland. Accuracy of the rods in estimating different water table measurements (average, lowest, most recent) and comparability of rod data (within sets of five rods) were investigated for different reference points on the rods, residence times, and wetland soils. The effect of the presence of vegetation in a soil low in organic matter on rod accuracy also was evaluated. The depth of lowest formation of a rust band on the rods predicted average and most recent water table depths in peat soil (r² for regressions of rust band depth on water table depth ranged from 0.71-0.95). Estimates of average water table depths were most precise for peat soil. Accuracy and precision were considerably lower in sand and clay soils, but significant relationships (P < 0.10) between depth of rust band formation and water table depth were found for all soils (r² values for sand and clay ranged from 0.13-0.55). The presence of vegetation had no effect on rod accuracy in the sand soil. Differences in rod performance between residence times were not apparent. However, a rod residence time of 4-6 weeks is recommended to balance the time necessary for adequate rust formation on the rods and to minimize the chance of exposure to large changes in water levels. A decrease in water table depth of approximately 40 cm in one month in the clay wetland caused a month lag time in rust formation. Differences in depth of rust band formation between the five rods within replicate sets were greatest for rods from clay (mean SD = ±7.9 cm). Variability of rust band measurements within replicate sets was lower in peat (mean SD = ±2.3 cm) and sand (mean SD = ±2.6 cm). The results indicated that the rusty rod technique has serious limitations and should be applied only in situations where the use of standard methods must be restricted. Emergence traps and a benthic coring device were used to sample the invertebrates of a freshwater, emergent wetland during late spring and summer, 1989. The fauna captured by each technique, disparities between the techniques in sampling certain taxa, and factors potentially affecting abundance estimates were examined. In addition, the efficiency of each technique, expressed as the number of samples required to achieve a desired level of precision, in estimating mean abundances of the dominant invertebrate group, the Chironomidae, was evaluated. Total and monthly estimates of insect family richness were higher for continuous sampling of emergence than for monthly core samples of the benthos. Emergence traps also caught a greater variety of the insect taxa inhabiting the wetland. The precision and efficiency of each technique in estimating abundances of the dominant group, the Chironomidae, varied between months and habitats (open water; vegetation). The variation was most likely due to the natural spatial and temporal variations inherent in invertebrate populations. The number of samples required (n[subscript r]) to estimate mean Chironomidae abundances for the entire summer, June-September, to a precision of D= 0.20 (equivalent to a standard error equal to 20% of the mean), varied between techniques. Fewer sampling stations would have been required to estimate mean adult abundances using emergence traps than would have been required for estimates of larval abundances using benthic core samples. Large numbers of benthic cores (27-208 individual cores per habitat) would have generally been required for both monthly and seasonal estimates of non-insect invertebrate abundances. Labor costs for processing emergence samples were about 30% of those for benthic samples. Subsampling of dominant groups in the emergence samples would have further reduced costs. Frequent sampling throughout a season, with several different techniques, is required to completely characterize the invertebrate community of a wetland. This study compared two quantitative techniques for sampling wetland insects. Continuous sampling with emergence traps provided higher estimates of insect family richness and more precise estimates of Chironomidae abundances at a lower cost per sample than monthly core samples of the benthos. / Graduation date: 1992

A lake-wide benthic survey of Lake Erie during summer 2002 indicated that Dreissena bugensis is the dominant dreissenid in Lake Erie, especially in the east basin where this species was found at every station but no Dreissena polymorpha were collected. Mean (±SD) densities of dreissenid mussels were comparable between the west (601±2,110/m2,n=49) and central (635±1,293/m2; n=41) basins, but were much greater in the east basin (9,480±11,173/m2;n=17). The greater variability in mussel density among stations and replicate samples in the central and west basins than in the east basin is attributable to the preponderance of fine-grained substrata in the nearshore, higher episodic rates of sediment deposition and periodic hypoxia in bottom waters. Although there was little change in lake-wide mean dreissenid densities between 1992 and 2002 (declining from ca. 2,636 individuals/m2 to 2,025 individuals/m2), basin-averaged shell-free dry tissue mass increased by almost four-fold from ca. 6.8±15.6 g /m2 to 24.7±71.3 g/m2 in the same interval. Up to 90% of this biomass is in the eastern basin. Other changes in 2002 include the virtual absence of mussels in the 3 to 12 mm size range, probably because of predation by round gobies, and an increase in the average size of mature mussels. The substantial changes observed between 1992 and 2002 suggest that dreissenid populations in Lake Erie were still changing rapidly in abundance and biomass, as well as species composition. The results of this survey suggest that a direct link between Dreissena spp. and hypolimnetic hypoxia in the central basin is unlikely. The dominant organisms of Lake Erie in 2002 were D. rostriformis bugensis (38%), Oligochaeta (33%), Chironomidae (18%), Sphaeriidae (2.7%), Amphipoda (2.3%) and Hydrozoa (2.2%). Mean invertebrate density was greater in the east basin, especially on hard substrates, than either the west or central basin. In the central basin, sites ≥5m supported greater numbers of organisms, than shallow (≤2m) sites in the nearshore wave zone. The greatest number of taxa were observed in the central basin, likely a resutlt of greater sampling effort there. Gammarus fasciatus comprised 80% of all amphipods, being most abundant on Dreissena-dominated hard substrates in the east basin. The introduced species, Echinogammarus ischnus occured at only 11 of 69 sites, and was the only amphipod found at 4 east basin sites, but at relatively low densities. Diporeia were not found in our survey. Hexagenia was collected at only 4 stations, all in the west basin. Chironomids were dominated by Tanytarsus and Chironomus, with Procladius, Dicrotendipes and Polypedilum also being relatively common. Oligochaete Trophic Index for 2002 indicates a similarly mesotrophic condition throughout the lake and marginal nutrient enrichment of sediments between years 1979-2002. Multivariate ordination of community data indicates clear separation of sites by year and basin as expected given the extirpation of Diporeia and the introduction of Dreissena and E. ischnus, but also reveals subtle changes in benthic structure over the last 2 decades. The benthic community of Lake Erie in 2002 does not likely represent an equilibrium condition.

Past monitoring has noted benthic macroinvertebrate community impairment downstream of both the Key Lake and Rabbit Lake uranium operations in northern Saskatchewan, Canada. The objective of this research was to try to identify the cause(s) of these impacts using a weight-of-evidence approach. Given that sediments generally accumulate contaminants that are related to metal mining activities (such as metals and radionuclides), the initial hypothesis for this research was that contaminated sediments were the primary cause of benthic community impairment at both operations.<p>In 2003 and 2004 a Sediment Quality Triad (SQT) approach confirmed the presence of an effect on benthic community structure, in addition to significant differences in surface-water, pore-water and whole-sediment chemistry at the immediate down-stream exposure sites at both uranium operations. However, no significant adverse effects were noted in 10-d whole-sediment bioassays with <i>Hyalella azteca</i>, although this lack of response could be partially due to sediment pore-water dilution resulting from the automated clean overlying water renewal process employed. Potential causes of benthic community impairment identified through the 2003 and 2004 SQTs for Key Lake include physical sediment composition, surface water pH and total ammonia, in addition to pore-water total ammonia and arsenic. Potential stressors identified at Rabbit Lake included high surface water manganese and uranium concentrations, and increases in pore-water total ammonia, manganese, iron, arsenic, and uranium levels.<p>In the summer of 2004, 4-d in-situ bioassays using <i>H. azteca</i> were conducted along with the SQTs to investigate the role both contaminated surface water and sediment played in benthic community impairment in-situ. Results from the Key Lake in-situ bioassay demonstrated that surface-water was the primary cause of acute toxicity to <i>H. azteca</i>. Results from the Rabbit Lake in-situ study also demonstrated that surface water as the primary cause of acute toxicity to <i>H. azteca</i>, although the relationship was not as strong. The cause of in-situ toxicity at Key Lake could not be correlated with any of the variables measured within the in-situ study, including trace metals, total ammonia, and pH. Of the measured constituents at Rabbit Lake, only concentrations of uranium in both surface water and pore-water were suspected of causing the observed in-situ mortality. Two data sets from two methods of surface water and pore-water collection supported these conclusions.<p>Due to time constraints and stronger cause-effect relationships, efforts were focused on the in-situ toxicity observed at Key Lake. Surface water collected in 2004 at the time of the related in-situ study was also found to be acutely toxic to <i>H. azteca</i> in separate laboratory surface water bioassays, thus verifying that contaminated surface water, not sediment, was the primary cause of the observed in-situ <i>H. azteca</i> mortality. Further information revealed that organic mill-process chemicals, which have been previously linked with sporadic effluent toxicity, were released at the Key Lake operation during the time of the in-situ experiment and associated surface water collection. Additional surface water samples collected in June and August, 2005, were not acutely toxic to <i>H. azteca</i>. Furthermore, a second bioassay with archived surface waters from the initial 2004 collection demonstrated that the water was no longer acutely toxic (i.e., acute toxicity disappeared after one-year storage). Chemistry comparisons of the toxic and non-toxic surface water samples, verified that trace metals, ammonia, pH, and major ions, including sulphate, were not the cause of toxicity, leaving only organic mill-process chemicals as a possible cause. Subsequent 4-d laboratory toxicity tests demonstrated that these process chemicals (kerosene, amine, and isodecanol) are toxic to H. azteca at the levels released in 2004, and are therefore believed to be the cause of the <i>H. azteca</i> mortality seen in the earlier in-situ experiment.<p>In short, this weight-of-evidence research provided new information on the possible causes of benthic macroinvertebrate community impairment downstream of both the Key Lake and Rabbit Lake uranium operations.

All life is organized hierarchically. Lower levels, such as cells and zooids, are nested within higher levels, such as multicellular organisms and colonial animals. The process by which a higher-level unit forms from the coalescence of lower-level units is known as “individuation”. Individuation is defined by the strength of functional interdependencies among constituent lower-level units. Interdependency results from division of labor, which is evidenced in colonial metazoans as zooid polymorphism. As lower-level units specialize for certain tasks, they become increasing dependant on the rest of the collective to perform other tasks. In this way, the evolution of division of labor drives the process of individuation. This study explores several ways in which polymorphism evolves in colonial marine invertebrates such as cnidarians, bryozoans, and urochordates. A previous study on the effect of environmental stability on polymorphism is revisted and reinterpreted. A method for quantifying colonial-level individuation by measuring the spatial arrangement of polymorphic zooids is proposed and demonstrated. Most significantly, a comparison across all colonial marine invertebrate taxa reveals that polymorphism only appears in those colonial taxa with moderately to strongly compartmentalized zooids. Weakly compartmentalized and fully compartmentalized taxa are universally monomorphic. This pattern is seen across all colonial marine invertebrate taxa and is interpreted as a “rule” governing the evolution of higher-level individuation in the major taxa of colonial marine invertebrates. The existence of one rule suggests that there may be others, including rules that transcend levels of biological hierarchy. The identification of such rules would strongly suggest that new levels in the hierarchy of life evolve by a universal pattern that is independent of the type of organism involved. / Dissertation

Intermittent headwater streams serve important functions in semi-arid rangelands, both for humans and wildlife. However, few studies have assessed species-environment relationships for fish and benthic macroinvertebrate assemblages in these systems. Additionally, no published studies could be found that addressed the influence of juniper coverage in watersheds on assemblage structure of these taxa. Increased juniper coverage in recent decades is believed to be associated with decreased water yields in central Texas streams. During summer 2003 and spring 2004, I examined potential effects of juniper cover on aquatic ecology. Fishes, benthic macroinvertebrates, and the physicochemical habitat were investigated in spring-fed headwater tributaries of the Pedernales River. My objectives were to: 1) describe the typical fish and benthic macroinvertebrate assemblages in headwater creeks of the Pedernales River basin; 2) compare seasonal variability of fish and benthic macroinvertebrate assemblages; 3) identify species-environment relationships in this river basin; and 4) evaluate the influence of juniper coverage in the watershed, relative tolocal and landscape-level environmental factors, on the structure of fish and benthic macroinvertebrate assemblages. A total of 4,181 individual fish were collected in summer 2003 samples, 551 fish were collected in spring 2004, 59,555 macroinvertebrates were captured in summer 2003 samples, and 51,500 macroinvertebrates were collected in spring 2004. Assemblages were typical for the area and habitat conditions. Faunal richness was lower in spring than in summer, possibly due to a combination of sampling after a relatively dry period in the spring, and lack of winter refugia in the form of deep pools. Fish assemblages may structure based principally on abiotic factors in spring, the harsher season (less available water), whereas predation pressure may influence structure in summer. Another important environmental gradient for both fish and invertebrate assemblages contrasts pool and run mesohabitats. In general, juniper cover was weakly associated with fish and invertebrate assemblages, although it tended to be associated with relatively high quality habitat for sensitive taxa (flowing runs with coarse substrate; deep, connected pools). In these intermittent streams, local-scale environmental factors probably are the dominant influences on fish and benthic macroinvertebrate assemblages. Implications for future studies are discussed.

Thesis (Ph. D.)--University of Washington, 2003. / Vita. Includes bibliographical references (leaves 82-95).

No description available.

No description available.