I studied amphibians and their associated wetlands in the Missouri River floodplain in Iowa from 2010-2013. I had the opportunity to study the effect of a catastrophic flood on an anuran community in the Missouri River floodplain. Three species (plains leopard frog, Woodhouse's toad, and Blanchard's cricket frog) had only minor changes in adult male occupancy rate in the two years after the flood. Colonization rates for these species were positively associated with wetlands that were shallower near the shore and they did not appear to be affected by reduced vegetation. Three other species or species complexes (northern leopard frog, the gray treefrog complex, and boreal chorus frog) had greatly reduced occupancy rates in the two years after the flood. Colonization rates for these species were relatively low, and they had high extinction rates. Colonization rates for these species were not associated with any measured habitat characteristic. Future flood events will likely continue to make northern leopard frogs, gray treefrogs, and boreal chorus frogs a less important part of the ecological community. Although some species may fare well under extreme climate events forecast under climate change scenarios, I hypothesize that many species will be challenged.

Amphibian conservation has often relied on auditory call surveys to determine habitat associations of anuran species. These surveys are restricted to only a single important life stage, however, and management recommendations from call surveys alone risk creating ecological traps or population sinks. Calling adult male surveys had established that slope at the inner edge of the wetland, wetland area, and percentage of bare ground within 1 m of the wetland were important habitat associations of calling adults of various species. I surveyed tadpoles and metamorphs of 5 anuran species and compared habitat associations for extinction rates of tadpoles and metamorphs with habitat associations for calling adults and found little overlap. I also estimated occupancy rates for calling adults, tadpoles, and metamorphs. Occupancy estimates indicate that reproductive success is variable, but the reasons are unknown. Occupancy rates of tadpoles and metamorphs were often lower than occupancy rates of adult calling males. Too often amphibian conservation and management proceeds with little information on critical aquatic life stages and I recommend greater emphasis on the entire life cycle to avoid potential misinformed conservation actions.

The importance of isolated wetlands disconnected from the river in large river floodplains has largely been ignored and consequently there is little management supporting this important component of productivity and biodiversity. Isolated and fishless wetlands in the upper elevations of floodplains may support a community largely absent from the rest of the floodplain. Flow regulation has rendered the upper elevations of many floodplains nearly devoid of wetlands but restoration of large river floodplains places no emphasis on isolated wetlands, despite concern over their status elsewhere. Primary drivers of community composition in isolated wetlands are predatory fish and hydrology. Wetlands are likely colonized by fish, and I used a GIS model of the floodplain topography to estimate the connection stage (the river stage when a wetland floods), and fill stage (river stage corresponding to the minimum elevation of the wetland, which is an indication of hydroperiod) of isolated wetlands in my study site. I compared the characteristics of current isolated wetlands to historical hydrology to determine how often fishless isolated wetlands are present. Productivity and biodiversity of the floodplain could be increased greatly by increasing the number of wetlands with a connection stage corresponding to river stage of 8-9 m and a fill stage of 5-7 m. I recommend restoration of large river floodplains give due consideration to the entire gradient of wetland types that were present before flow regulation so that the full complement of biodiversity and productivity can be restored.

I used a the multi-season occupancy model to estimate occupancy rates of successive life stages in Chapter 3, herein referred to as multiple life stage design. The constraints in a multi-stage analysis, namely, colonization rates of 0 and the successively decreasing occupancy rates for each life stage, require special considerations during the study design phase. I used simulations to explore the robustness of the parameter estimates when incorrectly assuming that colonization rate is 0. I explored the effect of design considerations such as the number of sites, detection probability, and number of surveys on confidence intervals of occupancy rates. Unacceptable bias in nearly all parameters is induced by true values of the colonization rate as low as 0.05 when the colonization rate is fixed to 0. Confidence intervals for occupancy rates are most improved by increasing the number of sites, but also by increasing detection probability and increasing the number of surveys. Under a simulated scenario with 50 sites, confidence intervals overlapped if the difference between occupancy rates was approximately 0.3 or less. Under excellent simulated study conditions with 200 sites, confidence intervals overlapped when the difference between occupancy rates was approximately 0.15 or less. These results will aid researchers in making appropriate study design decisions to avoid bias and meet their objectives in researching life stages.