FISH ASSEMBLAGES AT ENGINEERED AND NATURAL CHANNEL STRUCTURES IN THE LOWER MISSOURI RIVER: IMPLICATIONS FOR MODIFIED DIKE STRUCTURES

Large rivers throughout the world have been modified by using dike structures to divert water flows to deepwater habitats to maintain navigation channels. These modifications have been implicated in the decline in habitat diversity and native fishes. However, dike structures have been modified in the Missouri River USA to increase habitat diversity to aid in the recovery of native fishes. We compared species occupancy and fish community composition at natural sandbars and at notched and un‐notched rock dikes along the lower Missouri River to determine if notching dikes increases species diversity or occupancy of native fishes. Fish were collected using gill nets, trammel nets, otter trawls, and mini fyke nets throughout the lower 1212 river km of the Missouri River USA from 2003 to 2006. Few differences in species richness and diversity were evident among engineered dike structures and natural sandbars. Notching a dike structure had no effect on proportional abundance of fluvial dependents, fluvial specialists, and macrohabitat generalists. Occupancy at notched dikes increased for two species but did not differ for 17 other species (81%). Our results suggest that dike structures may provide suitable habitats for fluvial species compared with channel sand bars, but dike notching did not increase abundance or occupancy of most Missouri River fishes. Published in 2011 by John Wiley & Sons, Ltd.     

Effects of grade control structures on the macroinvertebrate assemblage of an agriculturally impacted stream

Nearly 400 rock rip‐rap grade control structures (hereafter GCS) were recently placed in streams of western Iowa, USA to reduce streambank erosion and protect bridge infrastructure and farmland. In this region, streams are characterized by channelized reaches, highly incised banks and silt and sand substrates that normally support low macroinvertebrate abundance and diversity. Therefore, GCS composed of rip‐rap provide the majority of coarse substrate habitat for benthic macroinvertebrates in these streams. We sampled 20 sites on Walnut Creek, Montgomery County, Iowa to quantify macroinvertebrate assemblage characteristics (1) on GCS rip‐rap and at sites located (2) 5–50 m upstream of GCS, (3) 5–50 m downstream of GCS and (4) at least 1 km from any GCS (five sites each). Macroinvertebrate biomass, numerical densities and diversity were greatest at sites with coarse substrates, including GCS sites and one natural riffle site and relatively low at remaining sites with soft substrates. Densities of macroinvertebrates in the orders Ephemeroptera, Trichoptera, Diptera, Coleoptera and Acariformes were abundant on GCS rip‐rap. Increases in macroinvertebrate biomass, density and diversity at GCS may improve local efficiency of breakdown of organic matter and nutrient and energy flow, and provide enhanced food resources for aquatic vertebrates. However, lack of positive macroinvertebrate responses immediately upstream and downstream of GCS suggest that positive effects might be restricted to the small areas of streambed covered by GCS. Improved understanding of GCS effects at both local and ecosystem scales is essential for stream management when these structures are present. Copyright © 2007 John Wiley & Sons, Ltd.     

Is it Worth the Money? The Functionality of Engineered Shallow Stream Banks as Habitat for Juvenile Fishes in Heavily Modified Water Bodies

Heavily modified water bodies (HMWB) are characterized by monotonous and straightened channel morphologies with high degrees of bank enforcement. They often lack shallow bank habitats, which are considered important for critical life stages of fishes. In this study, three principle options to engineer shallow stream zones were assessed concerning the value of the created habitats for larval, juvenile and adult stages of fishes in 30 sites from three HMWB. The construction scheme of the juvenile habitats comprised different degrees of embankment ranging from rip‐rap structures with steep bank angles to almost nature‐like construction schemes with very flat river‐banks and sparing usage of structural enrichment such as boulders and dead wood. In general, the differences between the three habitat types were more pronounced in density of different life stages than in the presence or absence of species or certain life stages. A steep bank angle and a high degree of engineering such as placement of rip‐rap embankment, boulders or dead wood structure in the habitats were hardly accepted by early larval and juvenile stages of rheophilic fishes. In contrast, the construction scheme of a nature‐like habitat with a flat bank angle (<10%), low water depth (mean = 24 cm) and a sparing usage of coarse woody debris (CWD) and boulders had the highest success. Other investigated habitat types did not provide additional benefit, neither in terms of supporting additional species and life stages, nor in high individual numbers and should thus only be implemented when land for restoration is scarce and nature‐like habitats cannot be realized. Copyright © 2016 John Wiley & Sons, Ltd.     

Long‐term fish assemblages of inner bends in a large river

We sampled fishes at 17 inner bend sites on the Wabash River in 2008 to compare with collections from 1977 to 1997. We used the same seine collection methods as previous years and collected a total of 37 species. Mean site Shannon‐Wiener diversity, species richness, evenness and abundance for all years were similar. We used multivariate analyses to test for patterns in fish assemblage structure among all sites and all years. The multivariate analyses resulted in distinct assemblages for each collection‐year, suggesting shifts in assemblage composition among years. We used separate multivariate analyses to examine fish assemblage variation within individual years. Variation that corresponded to an upstream–downstream pattern was present in 1977 and 2008, but not in 1997. A hydrologic analysis based on daily discharge revealed that eight large flood events occurred from 1928 to 2007, with four of these events during the recent 20 years. We quantified substrate variation at the 17 sites in 2008 and identified a longitudinal gradient in dominant substrate categories with gravel upstream and sand downstream that was correlated with the first axis of the 2008 fish assemblage ordination. We suggest that observed changes in fish assemblages in decadal periods were from hydrologic impacts of large floods on local habitats. Copyright © 2010 John Wiley & Sons, Ltd.     

INTRA‐ANNUAL VARIATION IN RIVER–RESERVOIR INTERFACE FISH ASSEMBLAGES: IMPLICATIONS FOR FISH CONSERVATION AND MANAGEMENT IN REGULATED RIVERS

While much is known about the fish assemblages, habitats, and ecology of rivers and reservoirs, there has been limited study of the fish assemblages in transitional habitats between these lotic and lentic habitats. Data about these river–reservoir interface (RRI) fish assemblages are needed to guide integrated management efforts of river–reservoir ecosystems. The aim of these efforts is to recommend flows for natural river function, conserve native riverine fish assemblages, and maintain reservoir sport fisheries. We used a multigear approach to assess the fish assemblages of four RRIs in the Colorado River Basin, Texas. In addition to characterizing RRI fish assemblages using species richness and evenness metrics, and habitat‐use guilds, we used a multivariate approach to evaluate intra‐annual shifts in species composition and abundance. All RRIs had high species richness and evenness values and included both macrohabitat generalist and fluvial species. RRIs also contained high proportions of the fish species available within each river–reservoir ecosystem, ranging from 55% to 80%. Observed intra‐annual shifts in RRI fish assemblages resulted from changes in abundance of dominant species rather than changes in species composition, with abundance of most species increasing from early spring to summer. Fish species responsible for intra‐annual shifts included mostly floodplain and migratory species, suggesting that species both used littoral habitats within RRIs and migrated through RRIs to river and reservoir habitats. The diversity of fishes found within RRIs highlights the importance of including these areas in future conservation and management efforts of river–reservoir ecosystems. Copyright © 2013 John Wiley & Sons, Ltd.     

Temporal distribution and taxonomic composition differences of larval fish in a large navigable river: A comparison of artificial dike structures and natural habitat

Distributions of larval fish in a regulated river were determined by sampling 15 sites in 2003 using larval fish light traps. Typical of large river systems in the Mississippi River drainage, collections in the Kanawha River were dominated (>93%) by Cyprinidae and Percidae species. Catch per unit effort (CPUE) along artificial dike structures were higher than CPUE along other shoreline sites, indicating the importance of these structures as nurseries. Taxonomic richness within artificial dike structures was greater than or equal to normal shoreline reference sites. Cyprinidae (Pimphales type and Notropis type) were collected significantly more on artificial structures than reference sites (p < 0.01). We collected higher numbers and observed greater CPUE of some sport fishes (Percidae) at the artificial dike structures. Percidae CPUE was significantly higher (p < 0.01) on artificial dike structures than reference sites. Percidae distribution appears to be positively influenced by the introduction of artificial dike structures in comparison to other abundant fish groups. Artificial dike structures encouraged taxonomic and larval fish diversity, and positively influenced some game fish species. The results of this study suggest that artificial dike structures are habitats that increase larval fish diversity in large, navigable rivers. Copyright © 2010 John Wiley & Sons, Ltd.     

JUVENILE SALMON RESPONSE TO THE PLACEMENT OF ENGINEERED LOG JAMS (ELJS) IN THE ELWHA RIVER,WASHINGTON STATE,USA

Engineered log jams (ELJs) are increasingly being used in large rivers to create fish habitat and as an alternative to riprap for bank stabilization. However, there have been few studies that have systematically examined how juvenile salmonids utilized these structures relative to other available habitat. We examined Chinook salmon (Oncorhynchus tshawytscha), coho salmon (O. kisutch) and trout (O. mykiss and O. clarki) response to the placement of engineered log jams (ELJs) in the Elwha River, Washington State, USA. We used summer snorkel surveys and a paired control‐treatment design to determine how engineered log jams in a large river system affect the density of juvenile salmon. We hypothesized that densities of juvenile salmonids would be greater in habitats with ELJs than in habitats without ELJs in the Elwha River and that this ELJ effect would vary by species and size class. Juvenile salmonid density was higher in ELJ units for all control‐treatment pairs except for one pair in 2002 and one pair in 2003. Positive mean differences in juvenile salmon densities between ELJ and non‐ELJ units were observed in two of four years for all juvenile salmon, trout greater than 100 mm and juvenile Chinook salmon. Positive mean differences occurred in one of 4 years for juvenile coho salmon and trout less than 100 mm. The results suggest that ELJs are potentially useful for restoring juvenile salmon habitat in the Elwha River, Washington State, USA. Copyright © 2011 John Wiley & Sons, Ltd.     

Response of fish assemblages to restoration of rapids habitat in a Great Lakes connecting channel

Rapids habitats are critical spawning and nursery grounds for multiple Laurentian Great Lakes fishes of ecological importance such as lake sturgeon, walleye, and salmonids. However, river modifications have destroyed important rapids habitat in connecting channels by modifying flow profiles and removing large quantities of cobble and gravel that are preferred spawning substrates of several fish species. The conversion of rapids habitat to slow moving waters has altered fish assemblages and decreased the spawning success of lithophilic species. The St. Marys River is a Great Lakes connecting channel in which the majority of rapids habitat has been lost. However, rapids habitat was restored at the Little Rapids in 2016 to recover important spawning habitat in this river. During the restoration, flow and substrate were recovered to rapids habitat. We sampled the fish community (pre- and post-restoration), focusing on age-0 fishes in order to characterize the response of the fish assemblage to the restoration, particularly for species of importance (e.g. lake whitefish, walleye, Atlantic salmon). Following restoration, we observed a 40% increase in age-0 fish catch per unit effort, increased presence of rare species, and a shift in assemblage structure of age-0 fishes (higher relative abundance of Salmonidae, Cottidae, and Gasterosteidae). We also observed a “transition” period in 2017, in which the assemblage was markedly different from the pre- and post-restoration assemblages and was dominated by Catostomidae. Responses from target species were mixed, with increased Atlantic salmon abundance, first documented presence of walleye and no presence of lake sturgeon or Coregoninae.     

1998-2001年长江口近海鱼类群聚结构及其与环境因子的关系

根据1998-2001年长江口近海4个航次鱼类拖网调查数据,运用群落生态学分类和排序方法,分析了长江口近海鱼类群聚结构特征及其与环境因子的关系。筛选后的64个站次的48种长江口近海鱼类分属10目28科,鲈形目种类最多,灯笼鱼目,鲈形目和鲱形目鱼类丰度共占总丰度的 99.81% 。龙头鱼、七星底灯鱼和黄鲫为秋季优势种,银鲳和皮氏叫姑鱼为春季优势种。相同季节年度间的鱼类群聚差异不显著,春季(5月)和秋季(11月)鱼类群聚有明显的季节分化。CCA分析显示：1998-2001年长江口近海鱼类群聚的主要影响因素是水深、底层温度、底层溶解氧、表层总磷和pH值。     

Effects of habitat enhancement on 0‐group fishes in a lowland river

Loss of habitat complexity through river channelization can have adverse affects on riverine fauna and flora through reductions in abundance and diversity of species. Habitat enhancement schemes are used to improve the physical and biological heterogeneity of riverine habitats. Between 1996 and 1997 the Environment Agency undertook a habitat enhancement scheme on the Huntspill River, Somerset, England to improve conditions for coarse (non‐salmonid) fishes. The scheme involved reducing bank gradients and the construction of off‐channel bays in parts of the channel, all of which were planted with willow (Salix sp.) and common reed (Phragmites australis). The effectiveness of the enhancement scheme was investigated by comparing 0‐group fish assemblages in manipulated and unmanipulated sites. Abundance and diversity of 0‐group fishes was significantly higher in manipulated habitats. There was no significant difference detected in the effects of the different types of enhancement measure used. The significance of microhabitats produced by habitat enhancement schemes is discussed with respect to spawning, nursery and refuge sites for 0‐group coarse fish assemblages. Copyright © 2001 John Wiley & Sons, Ltd.     

Spatial effects of reservoirs on fish assemblages in Great Plains streams in Kansas,USA

Reservoirs are important components of modern aquatic ecosystems that have negative impacts on native aquatic biota both up‐ and downstream. We used a landscape‐scale geographic information system (GIS) approach to quantify the spatial effects of 19 large reservoirs on upstream prairie fish assemblages at 219 sites in Kansas, USA. We hypothesized that fish assemblage structure would vary with increasing distance from a reservoir and that the abundance of reservoir fishes in upstream reaches would decline with distance from a reservoir. Ordination of sample sites showed variation in fish assemblage structure occurred primarily across river basins and with stream size. Variance partitioning of a canonical ordination revealed that the pure effect of reservoir distance explained a small but significant (6%; F = 4.90, P = 0.002) amount of variability in fish assemblage structure in upstream reaches. Moreover, reservoir species catch per unit of effort (CPUE) significantly declined with distance from a reservoir, but only in fourth‐ and fifth‐ order streams (r² = 0.32, P < 0.001 and r² = 0.49, P < 0.001, respectively). Finally, a multivariate regression model including measures of stream size, catchment area, river basin, and reservoir distance successfully predicted CPUE of reservoir species at sites upstream of Kansas reservoirs (R² = 0.45, P < 0.001). Overall, we found significant upstream effects of reservoirs on Kansas stream fish assemblages, which over time has led to a general homogenization of fish assemblages because of species introductions and extirpations. However, characteristic reservoir species are present throughout these systems and the importance of spatial proximity to reservoirs is probably dependent on the availability of suitable habitat (e.g. deep pools) in these tributary streams. Copyright © 2005 John Wiley & Sons, Ltd.     

Effects of low‐head dams on fish assemblages in subtropical streams: Context dependence on species category and data type

Anthropogenic disturbances may cause cosmopolitan species to replace endemic species, which will alter both the within‐community diversities and between‐community similarities of stream fish assemblages. In this study, we used data collected from headwater streams within the Xin'an basin, China, to evaluate the effects of low‐head dams on the alpha diversity and community similarity of fish assemblages. Our aims were to determine whether the changes in fish diversities and similarities related to dam‐associated disturbance are dependent on the species category (i.e., indigenous vs. native‐invasive species) or data type (i.e., occurrence‐based vs. abundance‐based indices). We found that low‐head dams significantly decreased the alpha diversity of the indigenous species in the impoundments but increased that of the native‐invasive species. However, the magnitude of this change was weakened if the two categories of fishes were not distinguished. Additionally, low‐head dams significantly decreased the occurrence‐based similarities of the indigenous fishes but increased those of the native‐invasive fishes. Despite the positive correlation between the occurrence‐based and abundance‐based indices, the former significantly overestimated the community similarities. Although most pairs of communities showed the same direction of changes for the two indices, some cases presented contrasting outcomes, including “perceived homogenization” (i.e., occurrence‐based differentiation but abundance‐based homogenization) and “perceived differentiation.” Our results suggest the importance of distinguishing indigenous and native‐invasive species and considering both occurrence‐based and abundance‐based indices in future research on how anthropogenic activities affect fish diversities and similarities in streams.     

DISTRIBUTION AND ABUNDANCE OF STREAM FISHES IN RELATION TO BARRIERS: IMPLICATIONS FOR MONITORING STREAM RECOVERY AFTER BARRIER REMOVAL

Dams are ubiquitous in coastal regions and have altered stream habitats and the distribution and abundance of stream fishes in those habitats by disrupting hydrology, temperature regime and habitat connectivity. Dam removal is a common restoration tool, but often the response of the fish assemblage is not monitored rigorously. Sedgeunkedunk Stream, a small tributary to the Penobscot River (Maine, USA), has been the focus of a restoration effort that includes the removal of two low‐head dams. In this study, we quantified fish assemblage metrics along a longitudinal gradient in Sedgeunkedunk Stream and also in a nearby reference stream. By establishing pre‐removal baseline conditions and associated variability and the conditions and variability immediately following removal, we can characterize future changes in the system associated with dam removal. Over 2 years prior to dam removal, species richness and abundance in Sedgeunkedunk Stream were highest downstream of the lowest dam, lowest immediately upstream of that dam and intermediate farther upstream; patterns were similar in the reference stream. Although seasonal and annual variation in metrics within each site was substantial, the overall upstream‐to‐downstream pattern along the stream gradient was remarkably consistent prior to dam removal. Immediately after dam removal, we saw significant decreases in richness and abundance downstream of the former dam site and a corresponding increase in fish abundance upstream of the former dam site. No such changes occurred in reference sites. Our results show that by quantifying baseline conditions in a small stream before restoration, the effects of stream restoration efforts on fish assemblages can be monitored successfully. These data set the stage for the long‐term assessment of Sedgeunkedunk Stream and provide a simple methodology for assessment in other restoration projects. Copyright © 2011 John Wiley & Sons, Ltd.     

PATTERNS OF ICHTHYOFAUNAL DISTRIBUTION AND CONNECTIVITY IN NAVIGABLE AND FREE‐FLOWING REACHES OF A MAJOR RIVER SYSTEM: THE ALLEGHENY RIVER IN PENNSYLVANIA

Connectivity throughout large riverine networks is often compromised by lock and dam (L/D) structures designed to facilitate year‐round navigation. The resultant interruption of flow inhibits free passage of aquatic biota potentially isolating mainstem and tributary communities. Our objectives were (i) to evaluate the impact of a series of navigational L/D structures on two targeted fish assemblages (TFAs): large‐bodied (>250 mm total length) pelagic and benthic (darter) communities and (ii) to examine patterns of tributary and mainstem connectivity. We systematically captured fishes utilizing gillnets, benthic trawls and backpack electrofishers from an impounded and a free‐flowing reach extending over 203 km of the Allegheny River in Pennsylvania. Species richness among the large‐bodied targeted fish assemblage was distributed somewhat evenly throughout each pool, peaking near dams and in an undredged pool. Depauperate darter assemblages of low species richness characterized most of the navigable reach with diversity and abundance concentrated in L/D tailrace areas. By contrast, darter communities inhabiting the free‐flowing reach were more diverse, abundant and evenly distributed, indicating the influence of uninterrupted connectivity. Community similarity at mainstem/tributary junctions increased with increasing tributary size with 50% of the mainstem species complement also inhabiting the largest tributary. This study underscores the importance of dams as barriers to ichthyofaunal connectivity, particularly to those benthic fishes which are relatively immobile and habitat specific, and the importance of mainstem/tributary junctions as avenues of riverine connectivity. Copyright © 2013 John Wiley & Sons, Ltd.     

Fish Assemblage Structure and Habitat Associations in a Large Western River System

Longitudinal gradients of fish assemblage and habitat structure were investigated in the Kootenai River of northern Idaho. A total of 43 500‐m river reaches was sampled repeatedly with several techniques (boat‐mounted electrofishing, hoop nets and benthic trawls) in the summers of 2012 and 2013. Differences in habitat and fish assemblage structure were apparent along the longitudinal gradient of the Kootenai River. Habitat characteristics (e.g. depth, substrate composition and water velocity) were related to fish assemblage structure in three different geomorphic river sections. Upper river sections were characterized by native salmonids (e.g. mountain whitefish Prosopium williamsoni), whereas native cyprinids (peamouth Mylocheilus caurinus, northern pikeminnow Ptychocheilus oregonensis) and non‐native fishes (pumpkinseed Lepomis gibbosus, yellow perch Perca flavescens) were common in the downstream section. Overall, a general pattern of species addition from upstream to downstream sections was discovered and is likely related to increased habitat complexity and additions of non‐native species in downstream sections. Assemblage structure of the upper sections were similar, but were both dissimilar to the lower section of the Kootenai River. Species‐specific hurdle regressions indicated the relationships among habitat characteristics and the predicted probability of occurrence and relative abundance varied by species. Understanding fish assemblage structure in relation to habitat could improve conservation efforts of rare fishes and improve management of coldwater river systems. Copyright © 2015 John Wiley & Sons, Ltd.     

Establishment patterns of non-native fishes: Lessons from the Duluth-Superior harbor and lower St. Louis River, an invasion-prone Great Lakes coastal ecosystem

We conducted a 2-year, multi-gear survey in the lower St. Louis River, which includes the Duluth-Superior harbor, an international shipping port and non-native species invasion “hotspot.” Our objectives were to quantify the contribution of non-native species to the overall fish assemblage and assess their spatial distribution and abundance. We captured 10 non-native fishes; none, however, were first detections. Non-native fishes composed roughly one quarter of the total species richness, were found in 84% of samples, and composed 15% of the total abundance. The spatial distribution and abundance of non-native fishes was dependent on gear selection (and thereby habitat selection), sampling location, and abundance measure (individuals or biomass). We used a recently published non-native species establishment framework to integrate catch data from the different gears. Viewed in the context of this framework, we identified only two non-native fishes, Eurasian ruffe (Gymnocephalus cernuus) and round goby (Neogobius melanostomus), as both widespread and abundant, whereas we identified three as localized and rare. Moreover, the time since first detection of non-native fishes was not predictive of their frequency of occurrence across the study area, underscoring the importance of environmental and biological factors in controlling fish establishment success. Although non-native fishes constitute a considerable portion of the trawl catch in the river channel, from a multi-gear system-wide perspective, native fishes collectively make up the majority of the fish biomass and abundance in the lower St. Louis River.     

The Impact of Changes in Water Level and Human Development on Forage Fish Assemblages in Great Lakes Coastal Marshes

Changes in water levels and development of shorelines are expected to negatively affect coastal marshes. The small-bodied fish assemblage was sampled in the inner marsh vegetation zone in five Les Cheneaux bays with differing levels of development. Observations were made from 1996 to 2004 during which time summer water levels varied from 177.2 m to 176.0 m (chart datum = 176.0 m). Each marsh was sampled for 10 consecutive days in July and August using gangs of five baited commercial minnow traps. Assemblage composition was assayed by species richness, the number of native minnow species, the percentage of selected tolerant fishes (bowfin, Amia calva, mudminnow, Umbra limi, common carp, Cyprinus carpio, and brown bullhead, Ameiurus nebulosus), and catch-per-unit-effort (CPUE). There were no consistent relationships between fish assemblage measures and year, water level, annual change in water level, exposure, and water temperature. Fish assemblage measures except CPUE were impacted by the density of building along the shoreline, a measure of development. Impervious surface area was < 4.5% and was not consistently related to fish assemblage measures.     

Fish community response to habitat variables in two restored side channels of the lower Platte River,Nebraska

Anthropogenic alterations to large rivers ranging from impoundments to channelization and levees have caused many rivers to no longer access the floodplain in a meaningful capacity. Floodplain habitats are important to many riverine fishes to complete their life‐history strategies. The fish community and species of fish that inhabit floodplain habitats are often dictated by the type of habitat and the conditions within that habitat (e.g., temperature, water velocity, depth, and discharge). As mitigation and restoration projects are undertaken, it is imperative that managers understand how various habitat components will affect the fish community in floodplain habitats. We collected fish and habitat data from two restored side channels with different structural designs on the lower Platte River, Nebraska, to determine how habitat variables predicted species diversity and individual species presence. We found a decrease in discharge in the main‐stem river resulted in increased diversity in one of the side channels, with the greatest diversity values occurring during summer. No habitat variables performed well for predicting fish species diversity for an adjacent side channel with more uniform depth and velocity and no groundwater inputs. However, several native riverine fish species in this side channel were shown to be associated with high temperature, dissolved oxygen, main‐stem discharge, and discharge variability. These results highlight the importance of considering the physical design of restored floodplain habitats when attempting to enhance fish communities.     

REMOVAL OF AQUATIC WEEDS GREATLY ENHANCES FISH COMMUNITY RICHNESS AND DIVERSITY: AN EXAMPLE FROM THE BURDEKIN RIVER FLOODPLAIN,TROPICAL AUSTRALIA

Hydrological changes associated with irrigation, in conjunction with increased nutrient concentrations and aquatic plant densities, have greatly impacted fish habitat values on the Burdekin River floodplain. The two most significant weeds in the Burdekin floodplain are water hyacinth (Eichhornia crassipes) and para grass (Urochloa mutica). Water hyacinth creates a base for the para grass (and then other weeds) to grow out into deeper water, creating weed mats that can then only be removed by mechanical means or floods of the largest magnitude. We attempted to rehabilitate floodplain lagoons by the mechanical removal of floating weed mats and monitored the result by measuring the subsequent effects on fish habitats and fish communities. Prior to weed removal, fish habitats were generally of poor quality, and fish community structure was skewed away from rich native assemblages to depauperate communities dominated in some instances by alien species. Poor water quality arising from floating weed mats is considered to be the main determinant of reduced fish abundance and diversity. After mechanical weed removal, recovery of water quality and physical habitat led to the re‐establishment of many native fish species. Key refuge habitats within the distribution channels were a critical source of recruits for fishes dispersing during times of elevated seasonal flows into the newly rehabilitated reaches. This study demonstrates that floating alien weed mats have significant negative effects upon aquatic communities and that mechanical removal of these weed mats (as opposed to chemical removal) results in dramatic improvements in native fish species richness and abundance. Copyright © 2011 John Wiley & Sons, Ltd.