An Historical Perspective on the Kissimmee River Restoration Project

This paper reviews the events leading to the channelization of the Kissimmee River, the physical, hydrologic, and biological effects of channelization, and the restoration movement. Between 1962 and 1971, in order to provide flood control for central and southern Florida, the 166 km-long meandering Kissimmee River was transformed into a 90 km-long, 10 meter-deep, 100 meter-wide canal. Channelization and transformation of the Kissimmee River system into a series of impoundments resulted in the loss of 12,000–14,000 ha of wetland habitat, eliminated historic water level fluctuations, and greatly modified flow characteristics. As a result, the biological communities of the river and floodplain system (vegetation, invertebrate, fish, wading bird, and waterfowl) were severely damaged. Following completion of the canal, the U.S. Geological Survey released a report documenting the environmental concerns associated with channelization of the river. This action led to the 1971 Governor's Conference on Water Management in South Florida that produced a consensus to request that steps be taken to restore the fish and wildlife resources and habitat of the Kissimmee basin. In 1976, the Florida Legislature passed the Kissimmee River Restoration Act. As a result, three major restoration and planning studies (first federal feasibility study [1978–1985], the Pool B Demonstration Project [1984–1990], and the second federal feasibility study [1990-present] were initiated (1) to evaluate measures and provide recommendations for restoring flood-plain wetlands and improving water quality within the Kissimmee basin, (2) to assess the feasibility of the recommended dechannelization plan, and (3) to evaluate implementation of the dechannelization plan. The recommended plan calls for the backfilling of over 35 km of C-38, recarving of 14 km of river channel, and removal of two water-control structures and associated levees. Restoration of the Kissimmee River ecosystem will result in the reestablishment of 104 km² of river-floodplain ecosystem, including 70 km of river channel and 11,000 ha of wetland habitat, which is expected to benefit over 320 species of fish and wildlife.     

The Kissimmee River Restoration Project and Evaluation Program,Florida, U.S.A.

Over the past decade, restoration of the Kissimmee River in central Florida has received considerable attention from local, state, national, and international media. In terms of areal extent, project cost, and ecological evaluation it is one of the largest and most comprehensive river restoration projects in the world. The goal of reestablishing ecological integrity involves restoring the physical attributes and the hydrologic processes that were lost after channelization of the river in the 1960s. The project is expected to restore over 80 km² of floodplain wetlands and reestablish over 70 km of river channel. Restoration construction began in 1999; to date, three construction phases have been completed, with the final phase of construction slated for completion in 2019. Restoration evaluation is widely viewed as a critical component of any restoration project. Equally important is the dissemination of information gained from restoration evaluation programs. This introductory article presents a brief overview of project history and outlines the approach and logic of the Kissimmee River Restoration Evaluation Program. The following papers present the results of ecological studies conducted before and after completion of the first phase of restoration construction. This first phase reestablished flow through 23 km of reconnected river channels and seasonally inundated a large portion (approximately 2,900 ha) of the floodplain within the Phase I project area. Although these studies present interim responses prior to full hydrologic restoration, results suggest that the ecosystem is responding largely as predicted by performance measures developed prior to restoration construction.     

Interim Responses of Floodplain Wetland Vegetation to Phase I of the Kissimmee River Restoration Project: Comparisons of Vegetation Maps from Five Periods in the River's History

Phase I of the Kissimmee River Restoration Project (KRRP) reestablished intermittent inundation of the river's floodplain by backfilling 12 km of the C‐38 flood control canal in 2001. We compared floodplain vegetation maps based on 2003 and 2008 aerial imagery (2 and 7 years following completion of Phase I, respectively) to vegetation maps from 1954 (pre‐channelization), 1974 (3 years after channelization), and 1996 (25 years after channelization) to evaluate broad‐scale vegetation responses to Phase I restoration. Results indicate that the extent of wetland plant communities expanded rapidly, more than doubling in area within 2 years after completion of Phase I, and that by 2008 wetlands had nearly recovered to pre‐channelization levels. However, full reestablishment of the pre‐channelization wetland mosaic has not yet occurred. Prior to channelization, much of the floodplain was dominated by a broadleaf marsh (BLM) community associated with extended, deep annual flooding, while shorter‐hydroperiod communities dominated the floodplain in 2003 and 2008. Prior to restoration construction, the reestablishment of BLM was predicted to be slow because suitable hydrology is dependent on project components that will not be in place until all restoration components are completed (projected for 2019). Hydrologic data indicate that the duration and variability of floodplain inundation have not yet achieved restoration targets over the entire Phase I study area. Other factors affecting vegetation responses are likely involved, including the age and viability of soil seed banks, the rarity of relict propagule sources following the channelized period, and competition from an invasive wetland shrub species.     

Interim Hydrologic Responses to Phase I of the Kissimmee River Restoration Project,Florida

Hydrologic conditions were evaluated during a 10‐year Interim Period following completion of Phase I of the Kissimmee River Restoration Project and initiation of environmental water releases from upstream to provide adaptive management of flow to the Phase I area. Phase I construction backfilled 13 km of flood control canal C‐38 and redirected flow into 22 km of reconnected river channel. Evaluations focused on five restoration expectations (performance measures) based on pre‐channelization hydrologic data for the Kissimmee River. Environmental releases resulted in more continuous discharge from upstream, but did not affect the magnitude of discharge. After backfilling of C‐38, water levels in the Phase I area varied with discharge and periodically inundated the floodplain. However, the long, annual recession event, characteristic of pre‐channelization, was not reestablished; instead, most Interim Period years had multiple events with shorter durations and faster recession rates. Mean channel water velocity increased during the Interim Period but was not always in the desired range. Hydrologic conditions throughout much of the Phase I area were affected by the backwater effect of the downstream water control structure. Four expectations showed improvements in terms of number of years met; however, none met the expectation targets. The inability to meet expectation targets reflects in part the incomplete or interim status of the restoration project.     

Interim Response of Dissolved Oxygen to Reestablished Flow in the Kissimmee River,Florida, U.S.A.

Restoration of the Kissimmee River should have multiple ecological benefits including improved dissolved oxygen (DO) within the river channel. Channelization of the Kissimmee River virtually eliminated flow through the natural river channel. After channelization, chronically low DO concentrations were observed in the stagnant remnant channel. Although no DO data from before channelization exist, reference estimates of pre‐channelization conditions were derived from seven relatively unimpacted streams. Stations along the Kissimmee River were sampled for 3 years before construction of the first phase of the restoration project began and for up to 8 years after the completion of construction. After Phase I construction, DO concentrations in the area of the river channel to which flow had been restored increased significantly from 2.2 to 4.9 mg/L, which is similar to DO concentrations observed in the reference streams. Mean DO concentrations for the reference streams ranged from 4.6 to 6.7 mg/L. Comparison of reference data to data from the pre‐Phase I and post‐Phase I system suggests that channelization had a negative impact on DO and that DO concentrations in the post‐Phase I Kissimmee River channel have made a significant recovery. Long‐term data trends demonstrate that DO concentrations can be negatively impacted by high flow events and that recovery from these events is generally quick, suggesting some degree of resilience in the system.     

Interim Responses of Benthic and Snag‐Dwelling Macroinvertebrates to Reestablished Flow and Habitat Structure in the Kissimmee River,Florida, U.S.A.

A critical component in the effort to restore the Kissimmee River ecosystem is the reestablishment of an aquatic invertebrate community typical of free‐flowing rivers of the southeastern United States. This article evaluates early responses of benthic and snag‐dwelling macroinvertebrates to restoration of flow and habitat structure following Phase I construction (interim period) of the Kissimmee River Restoration Project. Replicate benthic and snag samples were collected from remnant river channels in Pool A (Control site), and Pool C, the site of the first phase of restoration (Impact site). Samples were collected quarterly for 2 years prior to construction (baseline) and monthly or quarterly for 3 years following Phase I construction and restoration of flow. Baseline benthic data indicate a community dominated by taxa tolerant of organic pollution and low levels of dissolved oxygen, including the dipterans Chaoborus americanus (Chaoboridae) and the Chironomus/Goeldichironomus group (Chironomidae). Baseline snag data indicate a community dominated by gathering‐collectors, shredders, and scrapers. Passive filtering‐collector invertebrates were rare. Following restoration of flow, benthic invertebrate communities are numerically dominated by lotic taxa, including bivalves and sand‐dwelling chironomids (e.g. Polypedilum spp., Cryptochironomus spp., and Tanytarsini). Snags within the Phase I area support an invertebrate community dominated by passive filtering‐collectors including Rheotanytarsus spp. (Chironomidae) and Cheumatopsyche spp. (Hydropsychidae). Results indicate that restoration of flow has resulted in ecologically significant changes to the river habitat template not observed in Pool A. Observed shifts in benthic and snag macroinvertebrate community structure support previously developed hypotheses for macroinvertebrate responses to hydrologic restoration.     

Restoration of the Kissimmee River: A Conceptual Model of Past and Present Fish Communities and Its Consequences for Evaluating Restoration Success

Efforts to restore fish communities of the Kissimmee River will require carefully defined criteria for assessing success. A goal of regaining communities mirroring those in the historical river may not be an appropriate target because the ecological conditions of the river before channelization are poorly known. The Kissimmee River is in a biogeographic region historically low in fish diversity, and no comparable rivers in that region remain substantially unaltered by human activity to permit their use as reference sites indicative of conditions in the Kissimmee before channelization. I propose alternative criteria for assessing restoration success emphasizing expectations for ecosystem function in similar floodplain rivers. Assessing ecosystem function will be less simple than assessing criteria such as fish condition or density of selected species. But criteria based solely on fish-population characteristics cannot be justified quantitatively. Information integrated from several levels of biotic organization (individuals, populations, communities, and systems) should be drawn upon in making conclusions about restoration success. I develop a conceptual model to outline aspects of ecosystem function that could serve as a basis for evaluation of the restoration of fish communities of the Kissimmee River. The model focuses on the dynamics of the flux of floodplain-channel nutrients and the movement of larvae, juvenile, and adult fishes and macroinvertebrates. The present community may be dominated more by species tolerant of low-oxygen conditions, such as gar and bowfin, than the restored community will be. I propose that nest sites may be the limiting recruitment success of substrate spawning species in the channelized river and that these species, including sunfish and large-mouth bass, will increase in abundance after restoration. Also, species relying on floodplain habitats, including sun-fish species, darters, and some minnows, may also increase in frequency with restoration of floodplain-channel hydro-logical conditions and habitats. The observation that no species are known to have disappeared from the Kissimmee River, and its relatively simple community structure compared to rivers of comparable size elsewhere, are encouraging for prospects of successful restoration.     

Variant restoration trajectories for wetland plant communities on a channelized floodplain

Restoration efforts are typically based on an assumption that reestablishment of altered determinants of ecological structure and function will lead to a predictable reestablishment of populations and reassembly of communities. Dechannelization and reestablishment of natural hydrologic regimes provide the basis for the ongoing restoration of the Kissimmee River in Central Florida, United States. The expected reestablishment of historically dominant broadleaf marsh (BLM) and buttonbush shrub (BB) communities was evaluated over a 10‐year period following implementation of the first phase of the restoration project. Plant species composition and cover were assessed during dry (spring) and wet (summer) season sampling periods at five sites on the restored floodplain, and four “control” sites on the channelized floodplain. Mean daily stage data from nearby gauges indicated hydroperiods and depths on the reflooded floodplain were within the range of historic hydrologic conditions that selected for BLM and BB communities on the pre‐channelization floodplain. After reflooding, pasture grass and upland shrub communities rapidly transitioned to a fluid mix of obligate and facultative wetland species. Although signature BLM and BB species, Sagittaria lancifolia (bulltongue arrowhead), Pontederia cordata (pickerel weed), and Cephalanthus occidentalis (buttonbush), recolonized all study sites, the expected reestablishment of dominant cover of these species did not occur. Results indicate that restoration of BLM and BB communities has been impeded by deep flood pulse disturbances, establishment of invasive wetland grasses, and mineralized soil characteristics of the drained floodplain.     

Piscivore Responses to Enhancement of the Channelized Kissimmee River,Florida, U.S.A.

Evaluation of the success of ecosystem restoration projects requires identification of appropriate ecological metrics. Comparison of reconstructed food webs (or subsets thereof) from restored and non‐restored habitats may be a valuable tool to evaluate restoration success because food webs help identify critical predator–prey relationships, keystone species, relative importance of direct and indirect trophic interactions, and other aspects of ecological function. We compared the diets of apex predatory fishes collected from enhanced and non‐enhanced portions of the channelized Kissimmee River, Florida, USA to determine whether food web structure responded to experimental hydrologic manipulations. Diets were reconstructed for black crappie (Pomoxis nigromaculatus), bowfin (Amia calva), chain pickerel (Esox niger), Florida gar (Lepisosteus platyrhincus), largemouth bass (Micropterus salmoides), and warmouth (Lepomis gulosus) collected from enhanced and non‐enhanced portions of the Kissimmee River. Prey eaten by apex predatory fishes in the enhanced portion of the Kissimmee River were quantitatively and qualitatively different from prey eaten in non‐enhanced portions of the river. Predators in the enhanced portion of the river had fewer empty stomachs, more prey items per individual, more prey types per individual, more fish prey per individual, greater overall richness of prey, and a multivariate suite of prey distinct from predators in non‐enhanced portions of the river. Results from hydrologic manipulations suggest that large‐scale restoration of hydrologic linkages between the main channel and floodplain habitats will positively affect food web structure and ecosystem function in the Kissimmee River.     

Geomorphic Responses to Interim Hydrology Following Phase I of the Kissimmee River Restoration Project,Florida

Channelization of the Kissimmee River eliminated flow through the river channel, which allowed the formation of a largely organic deposition layer (ODL) on the river channel bed and stopped active sand transport needed to maintain point bars on meander bends. In 2001, completion of the first phase of dechannelization for the Kissimmee River Restoration Project (KRRP) reestablished flow to the river channel in the Phase I area. This study evaluated changes in the ODL and the number of meander bends with active point bar development (MBPB) following Phase I of dechannelization. Evaluations involved comparing interim measurements made after flow was reestablished to the river channel in the Phase I area but before full completion of KRRP with (a) baseline measurements made before dechannelization and (b) predicted changes based on reference measurements representing the pre‐channelization system. ODL thickness was measured in core samples on fixed transects perpendicular to the river channel. The ODL was thinner during the Interim Period than the Baseline Period and this decrease exceeded the expected change predicted from the reference condition. MBPB was assessed with aerial photography. MBPB increased from the baseline measurement of 0 bends to interim measurements of 27 bends in 2002 and 72 bends in 2009, approximating the increase predicted from the reference condition. The decrease in ODL thickness and the increase in MBPB to levels that meet or approximate the reference condition indicate that these aspects of the river channel are recovering following reestablishment of flow.     

Multiscale Patterns of Riparian Plant Diversity and Implications for Restoration

Planning riparian restoration to resemble historic reference conditions requires an understanding of both local and regional patterns of plant species diversity. Thus, understanding species distributions at multiple spatial scales is essential to improve restoration planting success, to enhance long‐term ecosystem functioning, and to match restoration planting designs with historic biogeographic distributions. To inform restoration planning, we examined the biogeographic patterns of riparian plant diversity at local and regional scales within a major western U.S.A. drainage, California's Sacramento—San Joaquin Valley. We analyzed patterns of species richness and complementarity (β‐diversity) across two scales: the watershed scale and the floodplain scale. At the watershed scale, spatial patterns of native riparian richness were driven by herbaceous species, whereas woody species were largely cosmopolitan across the nearly 38,000 km² study area. At the floodplain scale, riparian floras reflected species richness and dissimilarity patterns related to hydrological and disturbance‐driven successional sequences. These findings reinforce the importance of concurrently evaluating both local and regional processes that promote species diversity and distribution of native riparian flora. Furthermore, as restoration activities become more prevalent across the landscape, strategies for restoration outcomes should emulate the patterns of species diversity and biogeographic distributions found at regional scales.     

Bird abundance and species richness on Florida lakes: influence of trophic status,lake morphology,and aquatic macrophytes

Data from 46 Florida lakes were used to examine relationships between bird abundance (numbers and biomass) and species richness, and lake trophic status, lake morphology and aquatic macrophyte abundance. Average annual bird numbers ranged from 7 to 800 birds km^–2 and bird biomass ranged from 1 to 465 kg km^–2. Total species richness ranged from 1 to 30 species per lake. Annual average bird numbers and biomass were positively correlated to lake trophic status as assessed by total phosphorus (r = 0.61), total nitrogen (r = 0.60) and chlorophyll a (r = 0.56) concentrations. Species richness was positively correlated to lake area (r = 0.86) and trophic status (r = 0.64 for total phosphorus concentrations). The percentage of the total annual phosphorus load contributed to 14 Florida lakes by bird populations was low averaging 2.4%. Bird populations using Florida lakes, therefore, do not significantly impact the trophic status of the lakes under natural situations, but lake trophic status is a major factor influencing bird abundance and species richness on lakes. Bird abundance and species richness were not significantly correlated to other lake morphology or aquatic macrophyte parameters after the effects of lake area and trophic status were accounted for using stepwise multiple regression. The lack of significant relations between annual average bird abundance and species richness and macrophyte abundance seems to be related to changes in bird species composition. Bird abundance and species richness remain relatively stable as macrophyte abundance increases, but birds that use open-water habitats (e.g., double-crested cormorant, Phalacrocorax auritus) are replaced by species that use macrophyte communities (e.g., ring-necked duck, Aythya collaris).     

Conceptual Evaluation of Factors Potentially Affecting Restoration of Habitat Structure within the Channelized Kissimmee River Ecosystem

A critical element of the ongoing effort to restore the ecological integrity of Florida's Kissimmee River ecosystem is the reestablishment of pre-channelization habitat structure and function. Restoration of habitat will form the basis for responses by most biological components of the ecosystem and will provide a key indicator of the success of the restoration effort. This paper evaluates the relative importance of a range of abiotic and biotic habitat parameters in the existing and historic Kissimmee River ecosystem and provides a conceptual framework for predicting expected spatial and temporal responses of river and floodplain habitats to the restoration project. Among the ecological factors and process that influenced the development, dynamics, and maintenance of river and floodplain habitat structure, hydrology is expected to be of central importance in eliciting restoration responses in the Kissimmee River Ecosystem. Based on the assumption that the restoration plan will reestablish historic hydrologic characteristics, predictions are made of expected responses by geomorphic and vegetative components of the Kissimmee River's habitat structure. Recommendations are made regarding key habitat parameters requiring long term tracking and analysis and utilization of a geographic information system(GIS). A hierarchical habitat classification scheme is provided as a foundation for all components of the restoration evaluation program.     

The Influence of Habitat,Season, and Detectability on Abundance Estimates across an Amazonian Parrot Assemblage

The Amazon basin holds very high parrot species richness but almost nothing is known of parrot population densities in the region or how these vary between species, habitats, sites, and seasons. Such data are becoming important as humans impact on increasing areas of the region. Seventeen parrot species were surveyed using a line transect distance sampling method over 3 yr in floodplain and terra firme forests at two sites in the Tambopata region of southeast Peru. Density estimates for most species were in the range of 3.3–7.8/km², with Brotogeris cyanoptera and Amazona farinosa reaching densities of 22 and 23/km² in floodplain forest during the dry season. Parrot densities were higher in floodplain forest than in terra firme forest at both sites. The parrot communities of terra firme forests were similar across sites and seasons, but those in floodplain forests differed widely across sites and across seasons. Upper canopy birds are notoriously difficult to survey. We introduce a procedure to correct for the likely violation of the assumption that all birds on the transect line are detected (distance sampling assumption g(0) = 1). We correct g(0) based on calling rates of birds using a cue‐counting technique. Multipliers for g(0) differ across species and site, but not season. This method yielded density estimates on average 22 percent higher (6–40% higher in individual species) than those from the standard method.     

The density and stability of birds in shrubland and drainage line woodland in the southern Karoo,South Africa

Comparative surveys of bird species richness and abundance showed that a total of 69 bird species was recorded on 1 km transects (n = 996) through shrubland, with a total estimated density of birds of 32.7 ± 32.8 (S.D.) birds/km², of which resident birds made up 24.4 ± 21.0 birds/km², nomads made up 1.9 ± 11.7 birds/km² and local nomads made up 6.3 ± 20.1 birds/km². There was no correlation in general between total bird numbers or total bird biomass and rain, but numbers and biomass of nomads showed the strongest correlations with rainfall two months previously. A total of 86 species was recorded on 276 surveys along a 1 km stretch of drainage line woodland. Species richness varied from an average of about nine species to 24 species each month, and only showed a marked change during the 'dry' period of 1990 to early 1991. Density of birds in this woodland was 59.2 ± 20.8 birds/km. Biomass of birds was relatively constant throughout the study period, with marked increases in certain months when large species were present.     

Fine‐scale genetic diversity and landscape‐scale genetic structuring in three foundational bulrush species: implications for wetland revegetation

The appropriate sourcing of seeds for restoration is critical for establishing foundational plant species that support ecosystem functions and services. Genetic analyses of such species can yield insights into patterns of genetic diversity and structuring to inform seed collections. Here we document, for three foundational bulrush species, distinct genetic patterns to guide restoration of wetlands along the iconic Great Salt Lake, the largest lake in western North America. Specifically, Schoenoplectus acutus and Schoenoplectus americanus had moderate levels of site‐scale genet richness and relatively low genet richness levels within 1‐m² plots. These patterns contrast with Bolboschoenus maritimus, which had higher levels of site‐ and plot‐level genet richness, and has therefore likely experienced more recent seedling establishment. At the landscape scale, we found some evidence for genetic isolation of individuals at more remote sites (namely Fish Springs National Wildlife Refuge in the West Desert of Utah), but all species are relatively well dispersed over hundreds of kilometers, a pattern most likely to occur via avian dispersal. In our mechanistic dispersal assessment, we found abundant bulrush seeds present in waterfowl gizzards and those seeds germinated readily despite (or because of) partial digestion. Migratory waterfowl likely facilitate the broad dispersal of all species and may aid in bulrush establishment by breaking seed dormancy. These findings suggest that seeds for restoration should be collected within and among seed source sites to ensure a diverse restoration seed lot that does not disrupt gene flow patterns.     

Diversity of waterbirds wintering in Douz wetlands (south Tunisia): factors affecting wetland occupancy and species richness

Understanding the environmental factors shaping wetland attractiveness for waterbirds is an important question in wetland ecology and for conservation purposes. However, detailed data from numerous North African wetlands, notably those situated in inland areas, are still lacking. Thus, the aim of this study was to identify the factors influencing wetland use by waterbirds wintering in one of such poorly known habitat systems, namely the Saharan wetland complex of Douz, in south-western Tunisia. Thirty-four waterbird species (20 wading birds and 14 waterfowl species) were found to winter in this area. Using discriminant function analyses, we found that wetland size was the unique variable discriminating between occupied and unoccupied sites for total waterbirds and wading birds, while waterfowl distribution was related to both wetland size and water depth. Wetland size also provided the strongest predictor of species richness of wading birds, waterfowl and total waterbirds. Overall, our findings highlight the importance of wetland size as a key factor determining the attractiveness of wetlands for waterbirds wintering in the Saharan wetland complex of Douz. The possible explanations of this wetland size effect are discussed.     

Increased abundance of snails and trematode parasites of <Emphasis Type="Italic">Fundulus heteroclitus</Emphasis> (L.) in restored New Jersey wetlands

The Hackensack Meadowlands District is a large heavily degraded, brackish marsh system in the urbanized northeastern region of New Jersey, USA. Six study sites were used, three of which were restored (Mill Creek, Skeetkill Creek and Vince Lombardi), and three others were unrestored (Richard DeKorte Park, Cedar Creek and Kingsland Creek). Highly significant differences were found with respect to snail abundance and gill parasite abundance. In the three restored sites, significantly more Littoridinops tenuipes were found, and Fundulus heteroclitus had significantly more digenean trematode metacercariae gill infections than at unrestored sites. As habitat quality improves following restoration, the number of suitable digenean trematode parasite hosts multiplies as substrate for benthic invertebrates (first intermediate host) increases and usage by other species, such as Fundulus spp. (second intermediate host), is encouraged, which then attracts more wading birds (definitive host). Though the restoration process enhances trophic complexity, including primary consumers (gastropods), secondary consumers (fish) and tertiary consumers (wading birds), and ultimately parasite diversity, restoration also helps facilitate parasite life cycles.     

Determining habitat quality for species that demonstrate dynamic habitat selection

Determining habitat quality for wildlife populations requires relating a species' habitat to its survival and reproduction. Within a season, species occurrence and density can be disconnected from measures of habitat quality when resources are highly seasonal, unpredictable over time, and patchy. Here we establish an explicit link among dynamic selection of changing resources, spatio‐temporal species distributions, and fitness for predictive abundance and occurrence models that are used for short‐term water management and long‐term restoration planning. We used the wading bird distribution and evaluation models (WADEM) that estimate (1) daily changes in selection across resource gradients, (2) landscape abundance of flocks and individuals, (3) conspecific foraging aggregation, and (4) resource unit occurrence (at fixed 400 m cells) to quantify habitat quality and its consequences on reproduction for wetland indicator species. We linked maximum annual numbers of nests detected across the study area and nesting success of Great Egrets (Ardea alba), White Ibises (Eudocimus albus), and Wood Storks (Mycteria americana) over a 20‐year period to estimated daily dynamics of food resources produced by WADEM over a 7490 km² area. For all species, increases in predicted species abundance in March and high abundance in April were strongly linked to breeding responses. Great Egret nesting effort and success were higher when birds also showed greater conspecific foraging aggregation. Synthesis and applications: This study provides the first empirical evidence that dynamic habitat selection processes and distributions of wading birds over environmental gradients are linked with reproductive measures over periods of decades. Further, predictor variables at a variety of temporal (daily‐multiannual) resolutions and spatial (400 m to regional) scales effectively explained variation in ecological processes that change habitat quality. The process used here allows managers to develop short‐ and long‐term conservation strategies that (1) consider flexible behavioral patterns and (2) are robust to environmental variation over time.     

Predictability of flood pulse driven assembly rules for restoration of a floodplain plant community

Community assembly rules were formulated to evaluate the restoration of wet prairie along the periphery of the floodplain of the Kissimmee River in central Florida. Restoration of this plant community is expected to be driven by the reestablishment of flood pulse hydrology following the ongoing dechannelization of the river. Assembly rules were assessed with plant species composition and cover data from 15 permanent plots on the restored floodplain and 6 control plots on the channelized floodplain. These sites were sampled biannually from 1998 to 2010. Mean annual hydroperiods and depths confirmed that the frequency, duration and amplitude of post-restoration flood pulses at study sites were similar to historic reference locations. Elimination of pasture grasses (primarily Paspalum notatum Flüggé) following restoration of the flood pulse validated the hypothesized deletion rule for initial transformation of the wet prairie zone. Predicted increased dominance of obligate and facultative wetland species, a “community addition rule”, also was confirmed. An index of weighted averages of wetland indicator taxa showed significant short-term responses to antecedent hydroperiods and depths, and a restoration trajectory for wetland plant species. As predicted, recruitment of wet prairie indicator species from the extant seed bank correlated with reestablishment of the flood pulse, but was greatest when inundation extended from the wet season into the dry season. Restoration of a wetland plant community did not result in the predicted increase in species richness and diversity. Colonization and expansion of the exotic grass, Hemarthria altissima (Poir.) Stapf & C.E. Hubb., disrupted community reassembly processes. By summer 2007, mean cover of this species and several other exotic grasses increased to 24%, and necessitated herbicide treatments. Assembly rules provided useful predictions for the initial restoration of wet prairie vegetation, but were eventually confounded by the spread of an exotic species that was new to the regional flora.