Regional differences in nutrient limitation in floodplains of selected European rivers: implications for rehabilitation of characteristic floodplain vegetation

Extremely high river discharges in 1993 and 1995 along the Dutch rivers Rhine and Meuse have increased the public awareness of possible safety threats. As a result the ‘Space for Rivers’ program was implemented, aiming at restoring physical space for the rivers in combination with ecological rehabilitation. However, the development of species‐rich vegetation types in these floodplain areas is lagging behind restoration targets and biogeochemical constraints may play a vital role in this. Biogeochemical, hydrological and vegetation data were collected in 111 plots in both rehabilitated and original floodplains in regulated and more pristine river systems in The Netherlands and Poland. Soil nutrient and soil pore water data were summarized by factor analysis and the subsequent principal components were compared to vegetation and hydrological data by correlation analysis. The correlation analyses between vegetation parameters including nutrient stoichiometry and the biogeochemical soil variables resulted in a remarkable difference between pristine and impacted river systems. The results suggest a clear N‐limitation of plant growth in pristine floodplains, and apparent absence of limitation in regulated, impacted floodplains. In addition, results indicate that flooding events do not lead to one‐way transport of sediment and nutrients from the river into the floodplains; rather they indicate that highly dynamic hydrological conditions prevent soils from accumulating organic matter and nutrients. This study shows that nutrient limitation in regulated floodplains shifted from distinctly N‐limited plant growth to no nutrient limitation at all, probably due to decades of high fertilizer and manure application and nutrient input by the rivers during flooding. The consequence of our findings for rehabilitation activities is that it might be necessary to restore nitrogen limitation in floodplain systems in order to create opportunities for a species‐rich floodplain vegetation, through nutrient removal by hay‐making. Copyright © 2006 John Wiley & Sons, Ltd.     

Species composition and plant traits: Characterization of the biogeomorphological succession within contrasting river corridors

Plant communities and dynamics can be characterized according to species composition or plant traits. Here, we used species composition and plant traits to compare their effectiveness in discriminating the biogeomorphological (involving reciprocal feedbacks between physical and biological processes) and ecological (mainly biologically driven) phases of the fluvial biogeomorphological succession (FBS) model. The comparison was done between two French rivers, the largely unchannelized lower Allier and the channelized middle Garonne. One reach representative of each river section was selected for the study. Within each river reach, we chose two contrasted study sites in terms of channel and floodplain dynamics: a reference site (least altered channel and floodplain dynamics) and an altered site (laterally stabilized by riprap and constrained). In the four study sites, we sampled vegetation in 402 plots of 4 m². The 512 species identified in the plots were characterized in terms of plant traits (20) from a literature review. When comparing reaches in unconstrained ordinations and permutational multivariate analyses of variance, both species composition and plant traits led to a similar identification of the biogeomorphological and the ecological successional trajectories. Nevertheless, the trait approach was less influenced by local and regional bioclimatic, hydrogeomorphological, and anthropogenic settings and thus produced a more comprehensive and general classification of the biogeomorphological and ecological phases of the FBS model. A lower than expected contrast between the four sites was found, because neither species composition nor plant traits could entirely characterize distinct successional trajectories occurring in our reference or altered sites. Furthermore, our results contributed to a better understanding of the multiple successional trajectories that can occur in midlatitude river corridors. It also showed that relating plant traits to their effects on fluvial landform dynamics remains a core challenge in explaining succession including feedback mechanisms between hydrology, morphodynamics, and vegetation dynamics.     

WETLAND HABITAT PATCHES AS ECOLOGICAL COMPONENTS OF LANDSCAPE MEMORY IN A HIGHLY MODIFIED FLOODPLAIN

Landscape memory and ecological memory are relatively new concepts developed independently in geomorphology and ecology. Eco‐geomorphology strives to enrich the concept of landscape memory with that of ecological memory, as ecosystems build inherent parts of landscapes and are controlled by geomorphological factors. This integration is illustrated by a small‐scale eco‐geomorphological case study of links between ecosystems and their hydrogeomorphological template, in the floodplain of a medium size Hungarian River. Despite that the river is channelized by 1835 and most of the study area is under intensive agricultural cultivation, the seminatural vegetation along the various reaches shows significant variations according to geomorphological background variables: detrended correspondence analysis yields results of p = 0.002 significance. Our results prove that geomorphological patterns in floodplains, defined by landscape memory, and the related vegetation patterns, as manifestations of ecological memory, jointly determine the face of floodplain landscapes. The integration of ecological memory into the definition of landscape memory may help meet the challenges of eco‐geomorphology and facilitate a holistic understanding of landscapes, and thus be instrumental for reconstructing past environments—both in theory (scientific research) and practice (rehabilitation projects). Copyright © 2013 John Wiley & Sons, Ltd.     

Differential macrohabitat use by birds on the unregulated Ovens River floodplain of southeastern Australia

In many lowland floodplains around the world, upriver interferences to flows (weirs, dams, off‐takes) have led to much reduced frequency and duration of flooding. As a result, many floodplain wetlands are now inundated relatively rarely if at all. Given regulation of most lowland rivers in southeastern Australia, we assessed use of wetlands by birds in the essentially unregulated Ovens River in northeastern Victoria. Twelve sites (0.4–1.2 ha) were studied after flooding. Four sites were ‘permanent billabongs’, four were temporary wetlands and the other four were randomly selected woodland sites >60 m from the nearest water body (including the river) acting as ‘control’ or ‘reference’ sites. Aquatic birds were not recorded using woodland sites, but many species were differentially associated with either billabongs or temporary wetlands. A surprising number of non‐aquatic birds either exclusively or differentially were associated with wetland sites compared with woodland sites. We concluded that heterogeneous macrohabitat will increase local avian biodiversity on lowland floodplains. Moreover, densities and diversity of non‐aquatic, woodland species also increased with the presence of wetlands. Temporary wetlands were used differently from permanent billabongs by birds, especially in foraging methods. This suggests that the reinstatement of major flooding on heavily regulated floodplains would be ecologically advantageous for birds by providing foraging and breeding opportunities. Copyright © 2002 John Wiley & Sons, Ltd.     

Mesoscale river restoration enhances the diversity of floodplain vegetation

Effective river restoration aims for the recovery of ecosystem functions by restoring processes and connectivity to the floodplain. At the straightened lowland river Stör in northern Germany, a sequence of 15 new meanders was created in 2008, with wavelengths up to 70 m. The newly created areas within the meander bends range in size from 215 to 1,115 m² and function as a series of 15 restored floodplain sites, which are subject to succession. After 7 years of restoration measures, we investigated the vegetation dynamics on the (a) restored floodplains and compared them with adjacent floodplain sites that were used as (b) low‐intensity grazed grassland or as (c) abandoned grassland. We analysed the species diversity, functional vegetation parameters, and plant communities of 200 plots within the floodplain area of the three floodplain types and of 246 plots at their river banks. Plant species diversity and composition differed with respect to restoration measure and site management. Restored floodplains revealed a higher coverage in species of wet grasslands and softwood forests and higher species diversity than abandoned grasslands. Grazed grasslands showed the highest species number and coverages of pioneer vegetation. The banks indicated fewer differences in species composition between floodplain types. The construction of restored floodplains revealed greater overall plant diversity due to promoting the development of typical floodplain vegetation. Shallow meanders with increased flooding intensity and the creation of a varying microreliefs are recommended as combined river/floodplain measures in order to foster processes and connectivity between valley components.     

DISCONNECTING THE FLOODPLAIN: EARTHWORKS AND THEIR ECOLOGICAL EFFECT ON A DRYLAND FLOODPLAIN IN THE MURRAY–DARLING BASIN,AUSTRALIA

Globally, dams and water extractions are well‐recognised disruptors of flow regimes in floodplain wetlands, but little is known of the hydrological and ecological impacts of floodplain earthworks constructed for irrigation, flood mitigation and erosion control. We mapped the distribution of earthworks with high‐resolution SPOT (Système Probatoire d'Observation de la Terre) imagery in an internationally recognised Ramsar wetland, the Macquarie Marshes of the Murray–Darling Basin, Australia. There were 339 km levees, 1648 km channels, 54 off‐river storages and 664 tanks (0.5–5 m high), detected within the 4793 km² floodplain study area. Earthworks reduced localised flooding compared with undeveloped sites. The most pronounced disconnection of the original floodplain (73.0%) occurred where earthworks were most concentrated compared with areas with few earthworks (53.2%). We investigated relationships between hydrological connectivity and mortality of the perennial flood‐dependent river red gum Eucalyptus camaldulensis at 55 floodplain sites (225 × 150 m). Over half of the river red gums were dead at 21.8% of the sites. Earthworks blocked surface flows to flood‐dependent vegetation and drowned vegetation in artificially inundated off‐river storages. Mortality was due to impacts of earthworks and potentially exacerbated by effects of river regulation, water extraction and climate. River red gums were healthiest in narrow river corridors where earthworks confined flows and flows could recede freely. Rehabilitation of flood‐dependent ecosystems should focus on reinstating lateral connectivity and protecting environmental flows. Copyright © 2011 John Wiley & Sons, Ltd.     

PAST AND PREDICTED FUTURE CHANGES IN THE LAND COVER OF THE UPPER MISSISSIPPI RIVER FLOODPLAIN,USA

This study provides one historical and two alternative future contexts for evaluating land cover modifications within the Upper Mississippi River (UMR) floodplain. Given previously documented changes in land use, river engineering, restoration efforts and hydro‐climatic changes within the UMR basin and floodplain, we wanted to know which of these changes are the most important determinants of current and projected future floodplain land cover. We used Geographic Information System data covering approximately 37% of the UMR floodplain (3232 km²) for ca 1890 (pre‐lock and dam) and three contemporary periods (1975, 1989 and 2000) across which river restoration actions have increased and hydro‐climatic changes have occurred. We further developed two 50‐year future scenarios from the spatially dependent land cover transitions that occurred from 1975 to 1989 (scenario A) and from 1989 to 2000 (scenario B) using Markov models. Land cover composition of the UMR did not change significantly from 1975 to 2000, indicating that current land cover continues to reflect historical modifications that support agricultural production and commercial navigation despite some floodplain restoration efforts and variation in river discharge. Projected future land cover composition based on scenario A was not significantly different from the land cover for 1975, 1989 or 2000 but was different from the land cover of scenario B, which was also different from all other periods. Scenario B forecasts transition of some forest and marsh habitat to open water by the year 2050 for some portions of the northern river and projects that some agricultural lands will transition to open water in the southern portion of the river. Future floodplain management and restoration planning efforts in the UMR should consider the potential consequences of continued shifts in hydro‐climatic conditions that may occur as a result of climate change and the potential effects on floodplain land cover. Published 2011. This article is a U.S. Government work and is in the public domain in the USA.     

Predicting Long‐Term Changes in Riparian Bird Communities in Floodplain Landscapes

As anthropogenic impacts on riverine ecosystems expand, both aquatic and terrestrial ecosystems are influenced over large spatiotemporal scales. We predicted how riparian bird communities changed in response to long‐term changes in floodplain landscapes such as woodland expansion (i.e. rapid increases in vegetation cover on gravel bars and the progress of vegetation succession due to a decrease in the frequency and magnitude of flood disturbance). To test the hypothesis that woodland expansion after dam construction reduces the abundance of gravel bar‐nesting birds and increases the abundance of forest‐nesting birds, we estimated historical changes between past and present bird abundances using species distribution models across multiple rivers that were either unregulated or regulated by dams. We created past and present vegetation maps from remote sensing images and used habitat quantities as explanatory variables in the species distribution models. As we hypothesized, the estimated abundance of gravel bar‐nesting birds decreased and that of forest‐nesting birds increased because of woodland expansion in some regulated rivers. This suggests that anthropogenic alterations of riverine conditions (e.g. dam construction) can affect terrestrial ecosystems (e.g. riparian bird communities) through changes in floodplains (e.g. woodland expansion). In addition, our findings highlight the efficacy of combining spatial and temporal analyses when examining long‐term ecological dynamics. Copyright © 2013 John Wiley & Sons, Ltd.     

Site modelling methods for detecting hydrologic alteration of flood frequency and flood duration in the floodplain below the carlyle dam,lower Kaskaskia River,Illinois, USA

Reports concerning the influence of dams on river hydrology vary among researchers, interest groups and government agencies. These often contradicting statements may occur because changes in hydrology caused by dams are distinct for each dam and river watershed. The objective of this research was to use site specific techniques to determine if the 1967 installation of the Carlyle Dam, lower Kaskaskia River, Illinois, altered flood frequency and duration within the forested floodplain located below the dam. Results indicated a decrease in flood duration and frequency, and a decrease in annual flood frequency variation at a site 6.4 km below the dam. Pre‐dam versus post‐dam differences in flood frequency and duration at the site 32.2 km below the dam were related to climate rather than dam effects. Although dam impacts are a concern, this research shows that distance downstream from the dam and downstream tributary and watershed characteristics should be considered before assuming that the dam has changed hydrologic parameters for portions of rivers. This research also indicates that areas of the lower Kaskaskia River may still maintain hydrologic ecological integrity, and could be targeted for restoration and adaptive management purposes. Hydrologic modelling combined with river gage and on‐site well measurement techniques presented in this study could provide detailed flood frequency and duration information for land use, sociological and geomorphological questions in focus areas within river floodplains. Copyright © 2008 John Wiley & Sons, Ltd.     

The effect of altered flow regime on the frequency and duration of bankfull discharge: Murrumbidgee River,Australia

On meandering rivers with well‐developed floodplains, bankfull stage has geomorphological and ecological significance because it approximates the level of connection between the channel and the floodplain. As a river rises to bankfull stage, sediment begins to be deposited on the floodplain, wetlands are progressively inundated and organisms migrate between the channel and floodplain habitats. On many rivers large headwater dams have reduced the frequency and duration of floodplain inundation downstream. However, the lack of reliable pre‐regulation flow data has made it difficult to quantify the effects of river regulation. This study used historical regulated and modelled natural flow data to determine the effects of regulation on the frequency and duration of bankfull flows on the Murrumbidgee River, one of Australia's largest and most heavily regulated rivers. In combination with floodplain surveys the flow data show that regulation has halved the frequency and duration of bankfull flows. This reduction in channel–floodplain connection has implications for the ecological health of the Murrumbidgee River. Copyright © 2005 John Wiley & Sons, Ltd.     

On the beautiful diverse Danube? Danubian floodplain forest vegetation and flora under the influence of river eutrophication

The aim of this study was to investigate and compare the flora and vegetation of three well‐preserved hardwood floodplain forest areas along the river Danube and to discuss whether possible differences between the floodplain forests can be linked to river eutrophication. Flora and vegetation data from three study areas located on the Upper, Middle and Lower Danube in Central and Eastern Europe were compared using univariate and multivariate statistical methods. Special attention was paid to floristic composition, plant functional types, and ecological indicator values. We found that the three studied hardwood floodplain forests appeared to be rather different regarding floristic composition and herb‐layer vegetation. Despite the high beta diversity, the distribution of the plant functional types indicated generally equal habitat conditions, which were quite stable. The diversity of herb‐layer vegetation decreased downstream, while the indicators of nutrient availability pointed to increasing nutrient supply. The factor light apparently played a minor role for herb‐layer diversity. There is a remarkable congruence between the results of our floodplain vegetation analysis and the longitudinal river eutrophication patterns as described in the literature. We conclude that the nutrient input into Danubian hardwood floodplain forests increased downstream, resulting in higher nutrient availability for plants. This promoted especially the growth of tall and competitive forbs, which outcompeted other plant species. Even if the importance of the various eutrophication patterns is difficult to quantify, our study provides evidence that anthropogenic eutrophication has a distinct effect on the flora and vegetation of Danubian hardwood floodplain forests. Copyright © 2010 John Wiley & Sons, Ltd.     

River regulation influences the composition and distribution of inland frog communities

Frogs are widespread through inland rivers and floodplains and are an important component of floodplain food chains. Despite this, studies of frog communities in inland river systems are limited and the impacts of river regulation on frog communities have received very little attention. Surveys for frogs, tadpoles and egg masses along with assessment of vegetation, hydrology and water chemistry were conducted along 10 km reaches of three creek systems in the Lachlan River catchment, a major regulated river in Australia's Murray–Darling Basin. A total of 23 sample sites were surveyed at locations above and below in‐stream weirs as well as adjacent floodplain depressions. The hydrological regimes of sample sites were classified according to the length of time that they were known to hold water (water permanence). The sample sites fell into two distinct categories, 14 were classified as permanent and occurred upstream of weirs while nine were classified as temporary and were located downstream of weirs and in depressions adjacent to the weir pool. Permanent sample sites had a significantly higher percentage of dead standing timber and were deeper with less aquatic vegetation cover than temporary sample sites. Seven frog species were identified; there were no significant differences in species richness between permanent and the temporary waterbodies but the composition of frog assemblages differed significantly between them. This suggests that alteration to the hydrology of inland creek systems can lead to changes in the distribution frog species, with some becoming more common due to increases in the availably of permanent waterbodies while others decline due to reductions in the availability of seasonally flooded waterbodies. Copyright © 2010 John Wiley & Sons, Ltd.     

Trends in ecological river engineering in Korea

Ecological river engineering can be defined as the design and implementation of river works and river restoration works for the benefit of human society. It also guarantees the sustainable ecological functions of a river, such as its habitats and self-purification of its water. It is currently in the beginning stages in Korea, utilizing scientific knowledge on the processes of aquatic ecosystem degeneration and a methodology for solving the ecological problems in artificially altered rivers currently under development. The changes in river management and work practices in Korea may be best explained with a chronologically progressing sequence of ‘Natural’, ‘Disaster-prevention’, ‘Occupied’, ‘Park’, and ‘Close-to-nature’ rivers. Since the 1960s, the focus on river management and work has shifted from flood control only, to both flood control and riverine habitat conservation and restoration. Five research topics have been selected for this article, and the progress of each research area is briefly described with a representative picture in each topic. They are as follows: (1) flow resistance due to vegetation, (2) environmental flow, (3) floodplain vegetation modeling, (4) small dam removal, and (5) river restoration. For the future prospects of research on ecological river engineering in Korea, a necessity of further research on floodplain vegetation recruitment and succession, which can explain the so-called ‘white river’ and ‘green river’, is underlined, among others. Finally, two ongoing large research programs on river ecosystem restoration, of which are sponsored by the Government of Korea, are briefly introduced, followed by the introduction of a near-prototype experiment facility recently completed mainly for research on ecological river engineering.     

Habitat diversity and fish assemblage structure in local river widenings: A case study on a swiss river

We investigated habitat availability and fish assemblage structure in three local river widenings, completed 3–14 years ago, and five adjacent canalized reaches on the river Thur, a seventh‐order river in Switzerland. To account for seasonal variability, surveys were repeated in winter and summer 2005. Results were compared with historical pre‐disturbance data to evaluate whether the current abiotic and biotic conditions in the study reaches have attained historic near‐natural levels. Hydro‐physical habitat diversity (depth, flow velocity, cover availability) was considerably greater in the two longer widenings (>900 m length) than in the canalized reaches and in the shortest widening (300 m length), with higher proportions of shallow or deep areas of different flow velocities. However, the comparison of current and historical near‐natural shoreline lengths indicated that the current geomorphological complexity is still considerably impaired in all reaches. No overall significant relationship was found between the reach type (canalized or rehabilitated) and the number of species or the total fish abundance which were strongly correlated with the availability of suitable cover and moderate flow velocity. However, highest winter abundances were observed in deep, well‐structured backwaters of the rehabilitated reaches, documenting their significance as wintering habitats. Assemblage structure and composition were similar in canalized and rehabilitated reaches. Compared to the historical data, however, fewer and different dominant species were found, and guild composition changed towards a higher representation of generalists and tolerant species. Copyright © 2008 John Wiley & Sons, Ltd.     

基于河道治理的河流生态修复

阐述河流整治带来的负面影响 ,论证对受损的河流进行生态修复的必要性 .基于河道治理的河流生态修复是指河道形态和河床断面的修复及恢复丧失的河岸带植被和湿地群落 ,可采用的具体方法有 :恢复河道的连续性 ;重现水体流动多样性 ;给河流更多的空间 ;慎重选择河道整治方案 .在介绍国内外相关研究进展情况的基础上对河流生态修复的研究和应用提出合理建议 .     

Small hydropower dam alters the taxonomic composition of benthic macroinvertebrate assemblages in a neotropical river

Hydropower dams substantially modify lotic ecosystems. Most studies regarding their ecological impacts are based on large dams and provide little information about the far more abundant effects of small hydropower dams. Our aim was to characterize the ecological effects of a small hydropower dam and run‐of‐the‐river reservoir on the structure of benthic macroinvertebrate assemblages in the Pandeiros River located in the neotropical savanna of Brazil. We tested the hypothesis that benthic macroinvertebrate assemblages in sites directly affected by the dam and reservoir would show a different taxonomic structure compared with those in free‐flowing sites. We expected to find sensitive native species associated with the free‐flowing sites, whereas resistant and non‐native invasive taxa were expected in impounded sites. We also explored associations between the presence of native and non‐native invasive taxa to each habitat type. We found that the structure of benthic macroinvertebrate assemblages was significantly different between free‐flowing and impounded sites. Also, we found that the dam and reservoir facilitated colonization of non‐native invasive species (Corbicula fluminea and Melanoides tuberculata) because only in those sites they were found in high abundance, in contrast to the free‐flowing sites. Although the environmental conditions imposed by the impoundment altered the structure of benthic macroinvertebrate assemblages, the effects were limited to sites closest to the dam. Our results highlight the necessity of understanding physical habitat changes caused by the presence and management of run‐of‐the‐river dams and reservoirs.     

Impacts of water resource development on hydrological connectivity of different floodplain habitats in a highly variable system

Floodplains and their associated wetlands are important features of semiarid and arid landscapes, providing habitat and refugia for native species as well as contributing to human needs for freshwater. Globally, floodplain habitats are some of the most modified ecological communities because of water resource development and land‐use changes. However, the hydrological changes that have occurred in highly variable semiarid and arid systems are rarely quantified in a way that helps us understand the consequences for different floodplain habitat types. This study investigated changes in floodplain‐river connectivity that have occurred because of water resource development on four floodplain habitat types in the Lachlan River Catchment, Australia: (a) temporary floodplain lakes, (b) intermittent river red gum (Eucalyptus camaldulensis) swamps, (c) intermittent black box (Eucalyptus largiflorens) swamps, and (d) terminal wetlands (wetlands along distributary creeks). Changes to floodplain‐river connectivity characteristics were calculated using their commence to fill thresholds and flow scenarios derived from a river hydrology model, enabling comparison of long‐term data sets (120 years) encompassing a range of climate conditions. Connection regime metrics have changed significantly in all floodplain habitats except intermittent black box swamps. Temporary floodplain lakes have experienced the greatest reduction in number of connection events (60% reduction), followed by intermittent river red gum swamps (55% reduction). Intermittent black box swamps and terminal wetlands have experienced the least change in number of connection events (35% reduction). The nature of the change in connection suggests a change in vegetation communities will occur in response to long‐term hydrological change.     

Benchmarking Fluvial Dynamics for Process‐Based River Restoration: the Upper Rhine River (1816–2014)

Multi‐temporal analysis of river‐floodplain processes is a key tool for the identification of reference conditions or benchmarks and for the evaluation of deviations or deficits as a basis for process‐based river restoration in large modified rivers. This study developed a methodology for benchmarking fluvial processes at river segment level, focusing on those interrelations between morphodynamics (aggradation, erosion, channel shift) and vegetation succession (initial, colonization, transition) that condition habitat structure. Habitat maps of the free‐flowing Upper Rhine River downstream from Iffezheim dam (France–Germany border) were intersected with a geographic information system‐based approach. Patches showing trajectories of anthropization, changeless, progression and regression allowed for the identification of natural and human‐induced processes over almost 200 years. Before channelization, the riverine system was characterized by a shifting habitat mosaic with natural heterogeneity, high degree of surface water connectivity and equilibrium between progression and regression processes. On the other hand, the following 175 years of human interventions led to severe biogeomorphologic deficits evidenced by loss of natural processes and habitat heterogeneity, hydrological disconnection between the river and its floodplain and imbalance of progression versus regression dynamics. The main driving forces of change are found in hydromorphological impacts (channelization, regulation and hydropower plant construction). Regression processes are now almost absent and have to be the objective of process‐based river restoration measures for the studied river‐floodplain system. A sustainable view on water management and river restoration should aim at a more resilient riverine system by balancing the recovery of natural processes with societal needs. Copyright © 2016 John Wiley & Sons, Ltd.     

Rodent use of anthropogenic and ‘natural’ desert riparian habitat,lower Colorado River,Arizona

The role of native trees, Fremont cottonwood (Populus fremontii) and Goodding willow (Salix gooddingii), in structuring the riparian small mammal assemblage on rivers in the American desert southwest is unclear. It is unknown, for example, whether these trees directly or indirectly provide the food or shelter necessary for the presence of any species. Because of the rapid and widespread decline of gallery forest, due in part to river regulation, the retention of remnant stands and replacement of lost stands are major regional conservation issues. To elucidate small mammal–forest linkages, we compared patterns of macrohabitat use among terrestrial small mammals at two rehabilitated and one unmanipulated alluvial floodplain site along the highly regulated lower Colorado River. We also compared current patterns to the Colorado River faunal associations Joseph Grinnell documented in 1910, prior to significant flow regulation. We used grid‐based, capture‐mark‐recapture techniques at two revegetation sites, each a mosaic of six distinct macrohabitats, including planted cottonwood/willow, to associate species with specific macrohabitats. We also trapped a ‘reference’ grid containing naturally regenerating cottonwood and willow at a site on the lower Bill Williams River floodplain. Despite very poor development of cottonwood plantings at one of the revegetated sites, each supported at least nine species and harbored all seven species that Grinnell associated with areas flooded nearly every year. The set of species Grinnell associated with cottonwood/willow stands (Peromyscus maniculatus, Reithrodontomys megalotis, and Sigmodon arizonae) was trapped at both revegetation sites but entirely absent at the reference site. The Bill Williams site may be inaccessible to Sigmodon, but the absence of the other two species is probably a consequence of differences in floodplain structure and functioning among the study sites as well as between the Bill Williams site and historic Colorado River riparian areas. Our data suggest the richness of the native lower Colorado River riparian small mammal assemblage is unrelated to the presence or absence of cottonwood/willow trees, but does depend in part upon the presence or absence of dense herbaceous vegetation. Resource managers attempting to rehabilitate degraded desert riverine ecosystems need to consider understory as well as overstory plant species in revegetation efforts. Copyright © 1999 John Wiley & Sons, Ltd.     

河流健康诊断与生态修复

河流生态修复的总体目标是恢复河流健康，实现人与河流的和谐共存。从河流系统诸功能与河流健康之间的关系出发。分析了河流系统平衡的特征和影响因素，初步建立了河流健康诊断指标体系。根据河流受损功能、受损程度及其可修复性。提出了不同阶段河流治理和生态修复的对策。目前采用的主要技术包括：缓冲区修复、植被恢复、河道补水、生物-生态修复、生境修复、水生生物群落修复技术等。河流生态修复是一个多目标、多变量、多层次、多约束条件的动态优化问题，需要因地制宜、全面考虑、科学决策。