Remains of a mammoth, other Pleistocene fauna, and a wooden bowl were recovered from the Vernor site located in Clute, Brazoria County on the Texas Gulf Coast. Stratigraphy, sedimentology, and geochronology were used to establish the depositional history of the site. The geologic evidence suggests that these sediments were deposited during a period of fluvial activity by an ancient meander belt of the Brazos River, known today as Oyster Creek, which characterized this region during the Late Pleistocene and Early Holocene. Organics associated with the wooden bowl were radiocarbon dated to 4205 + 30 yr B.P. (UCIAMS-12039), while sand grains associated with the remains of the mammoth were dated using the luminescence technique to 66,000 + 7000 yr B.P. (UIC1383). According to these dates and their positions in the stratigraphic record, it was established that the mammoth and other Pleistocene age fauna preceded human occupation, and are not contemporaneous with the wooden bowl.

We collected 6,823 cloacal swabs over four years (2005–2006: 1,460; 2006– 2007: 2,171; 2007–2008: 2,424; and 2008–2009: 768) from 30 potential avian host species. Most samples (88.3 percent) were from dabbling ducks (genus Anas), while diving ducks (genus Aythya) accounted for 5.0 percent, and geese (genera Anser, Chen, and Branta) 3.0 percent of the samples tested. Waterfowl (Anatidae) comprised 98.7 percent of samples, with 1.8 percent from non-migratory dabbling ducks (genus Anas). All samples were screened for avian influenza virus (AIV) by AIV-matrix real-time RT-PCR (rRT-PCR); all rRT-PCR positive samples (541) were processed for virus isolation as well as 4,473 rRT-PCR negative samples. Differences were observed in apparent prevalence estimates over the four years between virus isolation (0.5, 1.3, 3.9, and 0.7 percent) and rRT-PCR (5.9, 6.5, 11.2, and 5.5 percent). We isolated 138 AIVs, of which two were obtained from rRT-PCR negative samples. Unlike previous reports of seasonal variation in AIV prevalence, we documented differences in prevalence estimates among months using rRT-PCR only during 2008–2009 and by virus isolation only during 2006–2007 and 2007–2008. Several of the AIV subtypes we identified are common in North America (e.g., H3, H4, and H6); H3N8 and H4N6 were the most common subtype combinations isolated. Similar to most surveillance studies, we found no significant difference in AIV infection based on host sex, but did find that juveniles were more likely to be positive for AIV than adults. We also documented that dabbling ducks were more likely to be positive for AIV than diving ducks, although not all dabbling ducks are equally likely to be positive. Molecular sequence analysis revealed no insertions of multiple basic amino acids at the cleavage site, which supported the identification of low pathogenic AIV. Phylogenetic anlyses performed on H5, H6, H7, N1, N2, N3, and N4 subtypes sequenced indicated similarity to other North American isolates with the exception of seven H6 which were more similar in amino acid translation to an isolate from Japan. In sum, this is the first multiyear study of avian influenza viruses on waterfowl wintering grounds of the Central Flyway, a historically understudied area of North America.

Coastal wetlands protect coastlines through efficient storage of organic carbon (OC) that decreases wetland vulnerability to sea level rise (SLR). Accelerated SLR is driving saltwater intrusion and altering vegetation communities and biogeochemical conditions in coastal wetlands with uncertain implications. We quantified changes in OC stocks and fluxes driven by 1) saltwater and phosphorous intrusion on freshwater and brackish marshes, 2) vegetation along an experimental saltmarsh to mangrove gradient, 3) saltwater intrusion and vegetation change across a marsh to mangrove ecotone, and 4) vegetation change and mangrove forest development along a marsh to mangrove ecotone. Increasing salinity in freshwater marshes decreased root biomass and soil elevation within one year. In brackish marshes, increased salinity decreased root productivity and biomass and increased root breakdown rate (k), while added salinity did not increase elevation loss. In our experimental marsh-mangrove ecotone, mangrove vegetation promoted higher organic carbon (OC) storage by increasing above and belowground biomass and reducing organic matter k. However, mangroves also increased belowground k, and decreased allochthonous marine subsidies, indicating the potential for OC storage trade-offs. In the Southeast Everglades, we identified strong interior-coastal gradients in soil stoichiometry and mangrove cover. Interior freshwater soil conditions increased k, while total soil OC stocks decreased toward the coast indicating that saltwater intrusion is driving large scale soil OC loss. In the southeast Everglades, mangrove expansion increased root biomass and root productivity, but did not mitigate the overall loss of OC stocks toward the coast. Similarly, in the southwest Everglades, saltwater intrusion drove a decrease in soil OC. However, mangrove encroachment drove a rapid recovery and increased OC stocks. Mangrove encroachment doubled aboveground biomass within the last ten years, increased it 30 times in the last 30 years, and doubled belowground biomass after 20 years. Our research shows that 1) moderate saltwater intrusion without mangrove encroachment will lead to a loss in OC stocks and potentially lead to wetland elevation loss and submergence, 2) in the absence of a change in saltwater intrusion, mangrove expansion can enhance OC storage 3) mangrove expansion can mitigate OC loss during saltwater intrusion, but this pattern depends on mangrove recruitment and ecosystem productivity.

Commonly referred to as “red tide”, harmful algal blooms (HABs) formed by Karenia brevis occur frequently in the Gulf of Mexico (GOM). A bloom is defined as cell abundances >105 cells L-1. This thesis will focus primarily on Karenia brevis, formerly known as Gymnodinium breve, in the Gulf of Mexico. K. brevis is harmful because it produces brevetoxin, a ladder-frame polyether that acts as a potent neurotoxin in vertebrates. K. brevis commonly causes fish kills, respiratory irritation in humans, and Neurotoxic Shellfish Poisoning (NSP) if ingested. Blooms of K. brevis occur almost annually along the West Florida Shelf (WFS) in the late summer and early fall, when the coastal current is favorable for bloom initiation. Along the Texas-Louisiana shelf (TLS) however, blooms of K. brevis are infrequent and sporadic. While much is known of the blooms along the WFS due to their frequent presence, little is known of the mechanisms driving the blooms along the TLS due to their inconsistent presence. To understand the stochastic nature of HABs along the TLS, historical data of bloom occurrences from 1996 to present were compared with NOAA station PTAT2 wind, sea-level pressure, air and water temperature data and NCEP NARR-A sea-level pressure data. The difference in the monthly-mean along-shore component of the wind was statistically significant between bloom and non-bloom years in September (p<<0.001) and April (p=0.0015), with bloom years having a strong downcoast current. Monthly mean water temperature values yielded similar results between bloom and non-bloom years. Both March and September monthly-mean water temperature values were lower during non-bloom years with p-values of 0.01 and 0.048, respectively. These results suggest the possibly of forecasting for HABs along the TLS with currently measured, publicly available data.