The Rusty Blackbird (Euphagus carolinus) has gained notoriety in recent years as one of the fastest declining North American bird species, with a global population loss of as much as 95%. Causes of the decline are not completely understood, but the high rate of forested wetland change in the southeastern United States suggests that wintering habitat degradation may be a primary driver. To better inform management on critical wintering grounds, I surveyed 68 sites in Louisiana where Rusty Blackbirds had been known to occur to address how occupancy changes with habitat type and colonization and extinction rates vary with ground cover, rainfall, and invertebrate biomass. Rusty Blackbirds use a large area while foraging on the wintering ground, therefore management may need to be targeted to even larger spatial scales. I assessed the relationship between statewide Rusty Blackbird abundance data from the Louisiana Winter Bird Atlas and landscape scale habitat within 512 unique USGS 7.5-minute quadrangles using datasets on land cover, cropland cover, and soil type. Results indicate that forested wetlands are important habitats associated with Rusty Blackbird presence, but only under certain conditions. Rusty Blackbirds prefer shallow water for foraging. At my sites, deep water cover increased with the cover of forested wetlands and may have deterred Rusty Blackbirds from using primarily forested wetland sites. The most important variables associated with transience were wet leaf litter and invertebrate biomass, which were both positively associated with colonization and negatively associated with extinction probability. For the Louisiana Winter Bird Atlas data, the top model included all explanatory variables for Rusty Blackbird abundance. Abundance increased with cover of soil hydrologic groups C, C/D, and D, which are capable of retaining surface water, suggesting that at larger scales water cover is more important than any particular habitat type. Pecans are an important food source for wintering Rusty Blackbirds and pecan orchards had the strongest positive relationship with abundance. In addition to maintaining pecan groves on the landscape, Rusty Blackbirds may benefit from management for shallowly flooded forested wetlands that can support high amounts of wet leaf litter on the grounds surface and abundant invertebrates.

The Atchafalaya River Delta system along with the Wax Lake Outlet Delta and the Mississippi River Delta are the only three areas in Louisiana where land is being gained. Beneficial use of dredge spoil from shipping channel maintenance is used on the Atchafalaya River Delta to supplement natural accretion. These dredge spoil islands have the ability to provide valuable nesting habitat for a variety of waterbirds, including Mottled Ducks. Previous studies on these islands reported mammalian predation to be a significant cause of nest failure for Mottled Ducks. I tested the hypothesis that predator reduction through trapping would increase Mottled Duck nesting success. I selected six islands based on vegetative conditions optimal for nesting vegetation and separated them into three trapped and three control islands. I found mammalian depredation of Mottled Duck nests to be rare and was not successful in detecting or trapping any predators. Instead, I found that flooding, which had been a minor issue in a previous study, to be the major cause of nest failure during the 2012 and 2013 nesting seasons. I found that Mottled Ducks strongly preferred nesting on islands that were isolated from the main delta complex. I used LIDAR elevation data as well as NOAA and pressure transducer data logger water level data to evaluate the relationship between nest elevation and nest success. I found no apparent relationship between nest elevation and nest success. Mayfield nest success for Mottled Duck nests was 20.5% in 2012 and 11.5% in 2013 with 34.5% of nests destroyed by flooding. Further research into the effects of flood duration, frequency, and incubation stage at flooding as well as considering partial loss of clutches may show a clearer relationship between nest success and the effects of flooding

Complex fluvial processes influence floodplains. River modifications in the 1930s have affected hydrogeomorphic processes influencing the lower White River in southeastern Arkansas. The overall objective of this study was to better understand the hydrologic and geomorphic influence on the floodplain forest. We used the HEC-RAS model to quantify hydrologic relationships within the floodplain before and after 1930s river modifications. The model can replicate flooding within 3-5 m. Despite river modifications, HEC-RAS modeling showed headwater floods influenced the upper reach of the floodplain while backwater floods from the Mississippi River influenced the lower reach of the floodplain. Post-1930s incision that occurred from the confluence to the middle reach of the lower floodplain reduced the flooding extent primarily in frequent (< 5-year return interval) headwater floods. In contrast, incision only reduced flooding extent in the smallest (1-year return interval) backwater floods, and larger backwater events were largely unaffected. Modeled flooding regimes for PNV classes were more distinguishable among floodplain reaches than among PNV classes. The upper reach in the floodplain flooded more often from headwater floods, but the lower reach in the floodplain flooded deeper by backwater floods. Post-1930s incision reduced flooding depth and flooding extent the most in the riverine backwater upper zone , riverine backwater lower zone, and riverine overbank natural levee classes. The largest reductions in flooding depth and extent within these classes occurred during the more frequent floods, which are most important for ecological processes

No description available.

Avian diversity in degraded fragmented Amazonian landscapes depends on the persistence of species in cleared and disturbed areas. Regenerating forest facilitates bird dispersal within degraded Amazonian landscapes and may tip the balance in favor of persistence in previously depauperate habitat patches. Despite the potential value of Amazonian second growth, we lack comparisons of demography among second growth, continuous forest, forest fragments in regenerating landscapes, and truly isolated fragments. Here, we used point-count and capture data to compare Amazonian bird communities among continuous forest plots, 100 ha forest fragments with adjacent second growth, 100 ha forested islands bounded by water, young and older second growth plots. We also compared differences in survival, population growth and age ratios between primary and mature secondary forest. Among foraging guilds, understory insectivores and obligate-flocking species were nearly absent in islands and young second growth. Fragments surrounded by a regenerating matrix were surprisingly species rich, suggesting that a developing matrix may mitigate extinction associated with fragmentation. Survival and population growth was lower in mature second growth relative to primary forest for all foraging guilds except frugivores, gap specialists and ant-following birds. Similarly, age ratios were skewed towards more adult insectivorous birds in older forest; these findings suggest that dominant individuals may preferentially use older forest. Our findings reinforce that true islands are extinction-driven systems with distinct, depauperate communities. Islands are not appropriate comparisons to forest fragments in some landscapes. In contrast, succession of bird communities in second growth facilitates recolonization of forest fragments, permitting fragments as small as 100 ha to support bird communities similar to continuous forest.

Each year, floodplain habitats are inundated and dewatered throughout the lower Mississippi River during the flood pulse. Many organisms, particularly fishes, are facultative or obligate users of inundated floodplain habitats for foraging and reproduction. The abundance and distribution of these fishes are influenced by annual floods, ranging from weeks to months depending on intensity and duration of the pulse. One major consequence of anthropogenic alterations to the river has been the sequential loss of connectivity of floodplain habitats. In light of these changes, a concentrated effort has been made to restore the integrity of the LMR floodplain. An integral part of restoration efforts depends on understanding characteristics of aquatic habitats that promote reproduction, growth, and survival of floodplain species. In collaboration with USFWS and GCPOLCC, this project investigated fish-habitat relationships and identified aquatic habitat conditions that promote healthy alligator gar populations. My objectives for this study were to: 1) identify floodplain habitat features associated with desired fish assemblage characteristics based on the role of Alligator Gar as a surrogate species; and 2) develop a sampling protocol for surveying alligator gar with side-scan sonar. The sampling program involved collecting fishes with gill nets throughout the flood pulse and documenting species and sizes of fishes collected in relation to habitat characteristics. A total of 373 fishes representing 14 species were sampled across 62 sites during two pulse periods (2013-2014). The most frequently caught species was Smallmouth Buffalo, which made up approximately half of the total catch followed second by Bigmouth buffalo. Considering two species of conservation concern that were present, Paddlefish (n=8) were found strictly in OPWA and FLVG. Conversely, Alligator Gar (n=31), which were caught at 22 sites with 6 sites producing multiple fish. These sites also yielded significant abundances of other fish species, including Gizzard Shad (53.8%; n=13), Common Carp (45.5%; n=22), and Longnose Gar (60.0%; n=5). Analyses revealed that fish assemblage structure was strongly related to habitats in REWA and OPWA, distance to river, sample year, and river stage. Alligator gar were reliably detected with side-scan sonar, yielding 788 images collected and a total estimate of 515 gar with approximate total lengths > 1m. These data will assist in developing sound conservation strategies throughout the LMRV to identify areas that fit the USFWSs needs in prioritization of conservation and floodplain restoration projects and the Services initiative for strategic habitat conservation. A better understanding of this floodplain system and the characteristics that contribute to its habitat value will hopefully provide the basis for development of management programs to enhance floodplain fish diversity and accessibility of floodplain habitats to riverine species.

I surveyed random (mail and internet panel surveys) and convenience samples (web surveys) of Louisiana waterfowl hunters following the 2009-2010, 2011-2012, and 2012-2013 seasons. I hypothesized significant differences in effort, success, satisfaction and demographics, but no difference in attitudes, in responses between random and convenience survey methods. I mailed to stratified random samples of 2,500 in 2010, 6,400 in 2012, and 2,500 each for mail and on-line internet panel response in 2013. An identical web survey, hosted each year during the same time period, facilitated open-access response. I received 1,676 usable responses in 2010 (727 mail and 949 web), 2,382 usable responses in 2012 (1,096 mail and 1,286 web), and 2,121 usable responses in 2013 (590 mail, 145 internet panel, and 1,407 web). Cost per usable response averaged $39.10 for mail surveys, $41.70 for the internet panel survey, and $3.20 for web surveys. Compared with the mail surveys and internet panel, respondents to the web surveys hunted more frequently, harvested more waterfowl, and placed higher levels of self-identity or importance on waterfowl hunting. However, I noted similarities in attitudes toward regulatory alternatives across survey methods in all 3 studies. I tested the random and convenience samples using binary logistic regression of variables measuring effort, success, satisfaction, and demographics, exceeding statistical standards for classification accuracy in 2010 and 2012, but not 2013, resulting in a mixed conclusion for these characteristics. Identical tests of attitudinal variables failed to meet statistical standards for classification accuracy in all 3 studies, confirming the inability to distinguish respondents by survey method using attitudinal variables. Polar reclassification of attitudinal responses into bichotomous categories led to identical managerial conclusions in 12 of 13 Likert-scaled questions, irrespective of survey method. Results support my hypothesis of no significant differences in hunter attitudes between random and convenience samples. These findings identify an opportunity to increase stakeholder feedback at reduced costs using web-based surveys. I suggest that survey methodology be carefully linked to survey objectives, and that open web surveys may be used to supplement random surveys in investigations of stakeholder attitudes to inform development of natural resource policy.

With the ongoing advance of the agricultural frontier in the Amazon basin, it is inevitable that heterogeneous landscapes will play a key role in conservation. These landscapes are mostly composed of patchworks of small forest fragments, secondary forests and roads. Conservation, however must take species interactions into consideration as they play a pivotal part the maintenance of several biological processes in the tropics. One of the most conspicuous interspecific interactions are seen in mixed-species flocks of birds, which in the Amazon, represent one of the best organized systems of bird aggregations. In this research, I assess how flock spatial behavior and species compositions are affected by changes in habitat structure. I followed 29 mixed-species flocks in different landscapes types such as secondary forests, forest fragments of 10 and 100 ha, and mixes of primary and secondary forest patches. As flocks foraged through their territories, I recorded their species composition every 30 minutes and georeferenced their movements every 30 seconds. Flocks spatial behavior was severely affected by anthropogenic features such as forest edges and secondary forests as flocks respond strongly to vegetation height. Using step-selection models, it was possible to reproduce flock movements and show that they prefer taller vegetation and lower areas of topography such as stream valleys. Due to this behavior, flocks avoided areas where canopy height was below 15 meters, and extensive areas of secondary below this height hold unstable flocks that do not persist for long periods. The ones that persisted showed home ranges that were much larger than what was observed in primary forest. Time spent in secondary forest was dependent on vegetation height, but not area, which seems to be shaped by intraspecific interactions. Flock social structure is also severely affected by habitat structure. Flock species richness did not show a predictable pattern, but participation was negatively affected. In fact, our data indicates that flock social structure may take longer to recover than spatial behavior. Assessing a 30-year mist-net capture dataset, we were able to determine that indeed, decreased species participation seems to be a more important driver in flock dissolution than local extinction.

Bioelectrical impedance analysis (BIA) has been proposed as an easy and accurate technique for a non-lethal index of condition in fish; however, factors affecting its accuracy have yet to be properly identified. The technique uses the resistance (R) and reactance (Xc) of water and lipids in contiguous tissues to estimate lipid content. Studies have investigated the use of BIA on fish, but with mixed results, possibly caused by unrecognized sources of influences. Our study explored possible sources of variance for BIA measurements in gulf killifish, Fundulus grandis, through laboratory feeding and fasting experiments, and experimental exposures to acute changes in temperature and salinity. In our feeding and fasting experiments, wild caught F. grandis were held in recirculating tanks and either fed ad libitum or withheld food for 28 days and sampled every 7 days. In the fed and fasted experiments, BIA measurements, plasma osmolarity, Lecrens Kn, and % organ lipid were significantly different by day with most variables showing a gradual change over time (ANOVA, p <.05). However, in the fasted experiment BIA measurements did not follow the trends of declining condition indicated by Kn and % organ lipid In our experiments with environmental factors, fish acclimated to 24⁰C and 12 ppt salinity where exposed to tanks randomly assigned treatments of salinities 2, 12, 24 and temperatures 2, 12, 24⁰C. Reactance was consistently significantly higher at salinity 2 ppt, and R significantly decreased as temperature increased (ANOVA,Tukey-Kramer p <.05). Our results indicate additional possible sources of variance including sex, size, temperature and changes in body chemistry. Some of the mixed results from other studies may be explained by these sources of variance. If these sources of variance are shown to be easily correctable, BIA may prove to be an effective index of condition.

Anthropogenic regulation of the Middle Rio Grande (MRG) of New Mexico has altered the hydrogeomorphology of the system and led to a rapid decline in breeding riparian-obligate birds as native riparian forests are replaced by exotic tamarisk stands. In this study, I evaluated the relationships among hydrogeomorphology, vegetation structure and composition, and avian communities of the San Acacia Reach (SAR) of the MRG. I conducted avian point count surveys and collected habitat data to determine avian community structure and abundance, geomorphic feature, surface flooding, and vegetation structure and composition along 44 transects throughout the SAR. A total of 999 point count surveys and 1,801 vegetation plots were completed throughout the 2012 and 2013 breeding season. Sixty-nine land bird species were detected over the two breeding seasons. Avian guilds responded differently to various hydrogeomorphic conditions. Ground-nesting birds and low shrub-nesting birds were insensitive to hydrogeomorphic changes as they do not rely on native understory, but can use exotic understory or woody debris. In contrast, canopy-nesting birds required native overstory; therefore, they were sensitive to hydrogeomorphic changes as native overstory species, such as cottonwood (Populus deltoides) and Gooddings willow (Salix gooddingii), require surface floods to germinate and establish. Similarly, water-obligate species were also sensitive to hydrogeomorphic changes as they require close proximity to surface water. I also evaluated hydrogeomorphic relationships with the habitat of the Western Yellow-billed Cuckoo (Coccyzus americanus occidentalis), a candidate species under the Endangered Species Act, using cuckoo presence/absence survey data, spatial statistics in ArcGIS 10, and vegetation and hydrology data. Results indicated that management of riparian forests that promotes overbank floods and regeneration and survival of cottonwood and Gooddings willow overstory, with a mixed understory of Gooddings willow, coyote willow (Salix exigua), and New Mexico olive (Forestiera pubescens) would provide long-term habitat for Western Yellow-billed Cuckoos. I also quantified the amount of variability within Hink and Ohmart vegetation structure classes, a vegetation classification scheme widely used throughout the MRG. Results indicated that including woody stem density classes in conjunction with the current Hink and Ohmart classification methodology will better inform songbird habitat management prescriptions.