Rapid assessment of rivers using macroinvertebrates: A procedure based on habitat-specific sampling,family level identification and a biotic index

Abstract This paper describes a simple and inexpensive procedure for the rapid biological assessment of water quality in rivers and streams in eastern Australia. The procedure involves the standardized collection of samples of 100 macroinvertebrates from defined habitat types within a water body. Specimens are identified to family level only and a biotic index is calculated. Proposed future testing and evaluation are described, and the limitations of the rapid approach are discussed.     

Primary nutrients and the biotic integrity of rivers and streams

1. Controls to reduce loadings of primary nutrients to maintain biotic integrity in rivers and streams have not been widely implemented because the relation between nutrients and chlorophyll, and its consequences for higher trophic levels, is confounded in lotic ecosystems by their openness, the variable degree of nutrient limitation and by the effect of physical factors.
2. The relationship between primary nutrients and biotic integrity in rivers and streams was tested using biological, physical and chemical information collected since 1982 from similar locations in streams throughout Ohio using standard procedures.
3. There was a negative correlation between nutrients, especially total phosphorus, and biotic integrity. The deleterious effect of increasing nutrient concentration on fish communities in low order streams was detectable when nutrient concentrations exceeded background conditions (total inorganic nitrogen and phosphorus > 0.61 mg L^–1 and 0.06 mg L^–1, respectively).
4. These results suggest that the control of release of toxins and oxygen‐demanding wastes to rivers is insufficient to protect aquatic life, and confirm the importance of non‐point sources of pollution in catchment planning as well as the combined effect of habitat and riparian quality on nutrient assimilation.     

Development and application of benthic macroinvertebrate-based multimetric indices for the assessment of streams and rivers in the Taihu Basin,China

The development of biological indicators for assessing ecological conditions in streams and rivers is urgently needed in China, particularly in heavily impacted regions. The aim of this study was to develop and apply benthic macroinvertebrate-based multimetric indices (MMIs) for the assessment of streams and rivers in the western hill and eastern plain aquatic ecoregions of the Taihu Basin. MMIs were based on samples collected from October 16 to November 8 2012 at 120 sites in streams and rivers. Least disturbed sites defined the reference conditions. Chemical water quality, physical habitat, and land use were used as criteria to identify reference sites in the basin. Metrics related to benthic macroinvertebrate richness, composition, diversity and evenness, pollution tolerance, and functional feeding groups were screened by range, sensitivity, responsiveness, and redundancy tests. Total number of taxa; percentage of Ephemeroptera, Trichoptera and Odonata (% ETO); Berger–Parker’s index (BP); Biotic index (BI); and percentage of filterers–collectors (% FC) were used to construct the MMI for the western hill aquatic ecoregion (MMI_W). Total number of taxa, percentage of Crustacea and Odonata (% CO), BP, BI, and % FC were used to construct the MMI for the eastern plain aquatic ecoregion (MMI_E). The MMI scores were obtained by combining the rating categories (excellent, good, fair, poor, and very poor). The MMIs were tested using a separate subset of the data, and the results indicated that the newly developed MMIs were robust in terms of percentage of sites correctly classified, coefficient of variation, box-separation ratios, and separation powers. The ecological status was then evaluated based on the MMI scores. The results indicated that the general ecological status of streams and rivers in the Taihu Basin was rated lower than “good”, the western hill aquatic ecoregion was rated “fair” and the eastern plain aquatic ecoregion was rated “poor”. Moreover, the MMIs showed a significant negative response to an increasing gradient of disturbance. Therefore, these preliminary MMIs can be used as assessment tools in ecological biomonitoring and management of the Taihu Basin.     

Status assessment in acid-sensitive and non-acid-sensitive Maryland coastal planin streams using an integrated biological,chemical, physical,and land-use approach

This study was designed to: (1) evaluate the ecological status of acid-sensitive and non acid-sensitive Maryland coastal plain streams using biological (Index of biotic Integrity [IBI] for fish), chemical and physical habitat conditions; (2) determine if a low IBI for coastal plain stream fish can be related to stream sensitivity from acidic inputs and (3) correlate land use activities and watershed size in the coastal plain streams with biological, chemical and physical conditions. IBI values obtained using 12 community metrics for Maryland coastal plain stream fish demonstrated that there were no significant differences in these values when acid-sensitive and non-acid-sensitive streams were compared. However, other complementary data in acid-sensitive streams such as absence of the acid-sensitive species, blacknose dace and higher numbers and biomass of tolerant species suggested that these streams may be impacted. IBI values were also found to be negatively correlated with various trace metals in acid-sensitive streams but not in non-acid-sensitive areas. Chemical conditions such as trace metals and nutrients were associated with land use activities. Highest concentrations of trace metals (chromium, nickel, and cadmium) were found in streams with the highest percentage of low residential housing. Nitrate concentrations were significantly higher in streams found in agricultural areas than in forested areas. Agriculturally dominated streams with highest nitrate concentrations (> 10 mg l^-1) also contained the highest percentage of livestock feeding operations. The mean IBI score for streams draining agricultural land was higher than the mean value for forested streams when all streams were compared. However, when several streams that were only marginally forested (< 50%) were removed from the analysis, the IBI scores did not differ significantly by land use. Two physical habitat indices exhibited a strong associated with each other. Each habitat index also correlated with IBI values.     

Distribution of epipelic diatoms in artificial fishponds along environmental and spatial gradients

Although epipelic diatoms play a key role in primary production of many ecosystems, many aspects of their biodiversity, ecology and geographical distribution are poorly understood. The present study is based on sampling of 45 man-made fishponds in the Czech Republic covering an environmental gradient from oligo/dystrophic highland ponds within protected areas to the eutrophic/hypertrophic lowland ponds used for intensive fish production. Diatom distribution patterns assessed using biomass and species composition variables were assessed along environmental and geographical gradients. In total, 185 epipelic diatom taxa were found in the investigated samples. The differences in species composition between sites were correlated with environmental parameters, but not with the geographic distance of the localities. This pattern might suggest that niche-based control, rather than the effect of dispersal limitation, is the main driving force in the species composition of epipelic diatoms in fishponds. The alpha-diversity of sites correlated with altitude, nitrogen and chlorophyll a concentrations but did not correlate with the area of the ponds. The significant relationships between local abundance of species and their regional occupancy were very similar to previous studies of diatoms in boreal streams. In addition, these data concur with patterns known for multicellular organisms suggesting that in this respect diatoms may not differ from groups of organisms with larger body sizes. Handling editor: J. Padisak     

Assessing effects of water abstraction on fish assemblages in Mediterranean streams

1. Water abstraction strongly affects streams in arid and semiarid ecosystems, particularly where there is a Mediterranean climate. Excessive abstraction reduces the availability of water for human uses downstream and impairs the capacity of streams to support native biota. 2. We investigated the flow regime and related variables in six river basins of the Iberian Peninsula and show that they have been strongly altered, with declining flows (autoregressive models) and groundwater levels during the 20th century. These streams had lower flows and more frequent droughts than predicted by the official hydrological model used in this region. Three of these rivers were sometimes dry, whereas there were predicted by the model to be permanently flowing. Meanwhile, there has been no decrease in annual precipitation. 3. We also investigated the fish assemblage of a stream in one of these river basins (Tordera) for 6 years and show that sites more affected by water abstraction display significant differences in four fish metrics (catch per unit effort, number of benthic species, number of intolerant species and proportional abundance of intolerant individuals) commonly used to assess the biotic condition of streams. 4. We discuss the utility of these metrics in assessing impacts of water abstraction and point out the need for detailed characterisation of the natural flow regime (and hence drought events) prior to the application of biotic indices in streams severely affected by water abstraction. In particular, in cases of artificially dry streams, it is more appropriate for regulatory agencies to assign index scores that reflect biotic degradation than to assign ‘missing’ scores, as is presently customary in assessments of Iberian streams.     

Performance of different biotic indices assessing the ecological status of rivers in the Central Himalaya

Biotic indices based on benthic macroinvertebrates are increasingly being used to assess the ecological quality of running water worldwide. In Nepal in the central Himalaya, five biotic indices have been developed in succession over the last decade to assess the ecological status of rivers, namely: the NEPbios, BRSbios, GRSbios, HKHbios, and NEPbios-Extended. The implementation of control and protection policies should be based on indices with proven reliability. Such reliability refers to the ability of the index to detect the correct status about the health of the assessed environment. So in the present study, we compare the performance of these five indices to determine the most robust, reliable biotic index for Nepal. In total, 248 biological samples belonging to first- to sixth-order streams were analyzed. Each sampling site was preclassified in situ using a screening protocol and later validated with physicochemical parameters to reflect the actual field condition. Overall, the GRSbios proved to be the most reliable, followed by the NEPbios-Extended, HKHbios, NEPbios, and BRSbios. The results were further validated by the ROC (Receiver Operating Characteristic) curve method. The study recommends applying the GRSbios to ecologically assess rivers in Nepal. In addition, we discuss modifying the GRSbios method to consider the future impacts of global change.     

An approach for the identification of exemplar sites for scaling up targeted field observations of benthic biogeochemistry in heterogeneous environments

Headwater streams influence the biogeochemical characteristics of large rivers and play important roles in regional and global carbon budgets. The combined effects of seasonality and land use change on the biogeochemistry of headwater streams, however, are not well understood. In this study we assessed the influence of catchment land use and seasonality on the composition of dissolved organic matter (DOM) and ecosystem metabolism in headwater streams of a Kenyan river. Fifty sites in 34 streams draining a gradient of catchment land use from 100% natural forest to 100% agriculture were sampled to determine temporal and spatial variation in DOM composition. Gross primary production (GPP) and ecosystem respiration (ER) were determined in 10 streams draining primarily forest or agricultural catchments. Absorbance and fluorescence spectrophotometry of DOM reflected notable shifts in composition along the land use gradient and with season. During the dry season, forest streams contained higher molecular weight and terrestrially derived DOM, whereas agricultural streams were dominated by autochthonous production and low molecular weight DOM. During the rainy season, aromaticity and high molecular weight DOM increased in agricultural streams, coinciding with seasonal erosion of soils and inputs of organic matter from farmlands. Most of the streams were heterotrophic. However, GPP and ER were generally greater in agricultural streams, driven by higher dissolved nutrient (mainly TDN) concentrations, light availability (open canopy) and temperature compared with forest streams. There were correlations between freshly and autochthonously produced DOM, GPP and ER during both the dry and wet seasons. This is one of the few studies to link land-use with organic carbon dynamics and DOM composition. Measures of ecosystem metabolism in these streams help to affirm the role of tropical streams and rivers as important components of the global carbon cycle and demonstrate that even semi-intensive, smallholder agriculture can have measurable effects on riverine ecosystem functioning.     

Assessment of organic pollution effect considering differences between lotic and lentic stream habitats

Based on the requirements of the Water Framework Directive, a macroinvertebrate-based assessment system to evaluate the ecological quality of streams has been developed by AQEM project consortium. In the Czech Republic the impact of organic pollution was principal pressure studied, but some morphological degradation of some sampling sites could not be avoided. A multimetric assessment system for three stream types was developed. Detrended Correspondence Analysis was used for the detection of the response of macroinvertebrate communities to the gradient of organic degradation. Significant relationships between abiotic (BOD, TOC, nutrients) and biotic (saprobic index, ASPT) indicators of organic enrichment/eutrophication were identified. Separate storage of the riffle and pool components of each multi-habitat sample allowed differences between these habitats to be compared in context of the metrics applied in the assessment system. Lotic and lentic habitats differed in taxonomic composition, ecological traits and biotic indices. The separate assessment of the riffle and pool parts of samples provides additional useful information when combined effects of organic pollution and morphological degradation are to be considered.     

The effect of organic pollution on the abundance and distribution of aquatic oligochaetes in an urban water basin, Taiwan

Aquatic oligochaetes are abundant in polluted areas and are, therefore, commonly used as bioindicators to study organic pollution in rivers and streams. In order to develop a species-level oligochaete biotic index to reflect the River Pollution Index (RPI) in the Taichung Water Basin in Taiwan, we conducted a systematic sampling scheme to collect aquatic oligochaetes from the sediment samples of watercourses in the Taichung Water Basin, Taiwan. We evaluated the relationships between aquatic oligochaetes and the sewage pollution using statistical methods. The distribution of aquatic oligochaetes in relation to environmental variables, such as water quality and sediment characteristics of the regional urban contaminated streams was expressed by Canonical Correspondence Analysis (CCA). We identified 17 species of aquatic oligochaetes (Annelida, Clitellata) including 3 species of Tubificidae, 13 species of Naididae, and 1 species of Enchytraeidae from the watercourses of an urban region in the Taichung Water Basin in Taiwan, during the summer and winter of 2005 and 2006. A positive correlation was found between the total abundance of aquatic oligochaetes and the RPI (r = 0.58, P < 0.05). However, only population density of the most abundant tubificid, Limnodrilus hoffmeisteri, increased with increasing RPI values and a significantly negative correlation was found between the population density of the naidid, Nais communis, and RPI values. The results of CCA indicated that certain naidids, such as Aulophorus furcatus and Allonais gwaliorensis also tolerated extremely polluted environments in upper stream or stony habitats, implying that tubificids should not be the sole representation of simple biotic indices but should also include pollution-tolerant naidids. We found that the community structure of aquatic oligochaetes was influenced by short-term variations in microhabitat rather than according to seasonal factors in our study region. The results proved that aquatic oligochaetes were sensitive enough to provide a supplement for the regional urban pollution assessment applications for biotic indicators at the species-level.     

Primary producers in a Pampean stream: temporal variation and structuring role

Low current velocities, high nutrient levels, the lack of riparian forest vegetation, and the development of dense and rich macrophyte communities characterize Pampean streams. The objective of this study was to describe the main physical, chemical, and biological characteristics of a headwater Pampean stream as well as to analyze the role of macrophytes and phytobenthos. The study was conducted in a stream considered to be not much disturbed by human activities. Samples of water and organisms (macrophytes, benthic algae and invertebrates) were taken monthly for 14 months in two sampling stations, in fast flow and slow flow sites. Macrophyte biomass and diversity increased in spring and summer, and they decreased in autumn, when the plant community was greatly affected by an important flood. Phytobenthos biomass was lower in late summer, possibly due to the establishment of a dense cover of the floating macrophyte Lemna gibba L. Density of amphipods and gastropods greatly increases in spring and summer, jointly with the macrophyte development. Analysis of correlation showed that current velocity is the most important factor influencing macrophyte biomass and phytobenthos structure, while depth, nutrients, and herbivores are linked factors. Pampean streams could be considered systems dynamically fragile, because habitat heterogeneity is generated by aquatic vegetation, a substratum that varies along time.     

Benthic macroinvertebrate assemblages of coastal and continental streams and large rivers of southwestern British Columbia, Canada

In southwestern British Columbia (BC, Canada) and within a relatively small geographic area, lotic environments range from streams in coastal rainforests, to streams in arid continental grasslands, to very large rivers. Little is known about the invertebrate communities in large rivers in general, or in the streams of continental BC. The purpose of this study was to determine whether the benthic invertebrate community structure changes spatially between small coastal and small interior streams; between small streams versus large rivers; and whether changes in the benthic community are related to the environmental conditions. Kicknet samples and environmental data were collected from three coastal streams, three continental streams and two large rivers (discharge of 781 and 3620 m³/s, respectively). The large river sites had low invertebrate abundance, species richness and diversity, relative to the small streams. The coastal streams had the highest species richness and the continental streams had the highest invertebrate abundance. A number of taxa were specific to each class of stream. Invertebrate abundance decreased with river size, and increased with elevation, pH, conductivity, alkalinity, NO₂NO₃-N, total Kejldahl nitrogen and percent carbon in suspended solids.     

Effects of stream size on taxa richness and other commonly used benthic bioassessment metrics

Benthic macroinvertebrate samples were collected from natural substrates in disturbed and undisturbed South Carolina upper coastal plain streams to determine if taxa richness and other bioassessment metrics were significantly related to stream size as predicted by the River Continuum Concept (RCC). Linear, quadratic, and lognormal regression models indicated that stream width was positively related to total number of taxa; number of Ephemeroptera, Plecoptera, and Trichoptera (EPT) taxa; and total number of organisms. Linear regression showed that the expected number of taxa at undisturbed sites ranged from 35 in 2.0 m wide streams to 64 in 16.0 m streams. Comparable values were 8–20 for EPT and 109–261 for number of organisms. Stream width was inversely related to biotic index values indicating a decrease in average organism tolerance with increasing stream size. ANCOVA showed that the effects of stream size were similar for disturbed and undisturbed sites. Rank correlations and multidimensional scaling (MDS) showed that Lepidoptera and Trichoptera were more abundant in larger streams and Annelida in smaller streams. Stream size related changes in benthic macroinvertebrate community composition are often ignored in bioassessment protocols; however, failure to adjust metrics for stream size can lead to erroneous conclusions. Adjustments are possible by analyzing regression residuals stripped of stream size related variance, dividing the area beneath the maximum taxa richness line into equal size units for metric scoring, or scaling metrics based on predicted reference values. Electronic supplementary material Electronic supplementary material is available for this article at and accessible for authorised users.     

Occupancy dynamics of escaped farmed Atlantic salmon in Canadian Pacific coastal salmon streams: implications for sustained invasions

Farmed non-native Atlantic salmon (Salmo salar) is the largest agriculture export product of British Columbia, Canada. Chronic low-volume escapes of salmon from farms into Pacific waters (“leakage”) are typically undetectable (Britton et al. 2011). Analysis of escape-reporting from farmers indicates that reports greatly underestimate the true number of Atlantic salmon inadvertently released from open-net pen rearing sites (Morton and Volpe 2002). To quantify the spatial extent of escaped Atlantic salmon in Canadian Pacific rivers, we systematically snorkel-surveyed 41 known Pacific salmon (Oncorhynchus spp.)-supporting rivers and creeks on Vancouver Island over a span of 3 years. We estimated and accounted for imperfect detections using multi-season occupancy models. We detected Atlantic salmon in 36.6 % of surveyed rivers. After accounting for imperfect detection, occupancy models estimated that over half of surveyed streams across the study area contained Atlantic salmon, and that 97 % of streams with high native salmon diversity were occupied by Atlantic salmon. Even in intensive snorkel surveys, Atlantic salmon are detected in occupied streams only 2/3 the time, suggesting abundance and distribution of non-native salmon is greater than indicated by the only existing data. Further, Atlantic salmon are more likely to occupy streams with high native Pacific salmon diversity—and more likely to maintain occupancy across years—potentially increasing competitive pressure on native salmonids. Understanding local biotic and abiotic predictors of Atlantic salmon occupancy, stream colonization, and local extinction requires more data; the same is true for the effects of escaped Atlantic salmon on local salmon diversity and sustainability. These data for the first time show that Atlantic salmon occupy Pacific coastal rivers for multiple years. The impact of Atlantic salmon occupancy in British Columbia rivers must be factored into policy decisions regarding the future of salmon farming in the provincial waters.     

Stream epilithic, epipelic and epiphytic diatoms: habitat fidelity and use in biomonitoring

We compared the use of epilithic, epiphytic, and epipelic diatom communities in stream biomonitoring by investigating species composition and relationships with measured water quality variables in two tributaries of the Grand River, Ontario, Canada. Although ordination showed some separation of the epilithon from other habitats, clear habitat preferences and seasonality were not identified. Canonical correspondence analysis showed that measured water quality variables explained the major gradients in the diatom data for all three habitats. The diatoms from each habitat were strongly related to alkalinity, suspended solids, biological oxygen demand and conductivity, and in addition, epipelic and epiphytic diatoms showed significant relationships with total phosphorus. It was concluded that although some species were more abundant in one habitat than the others, the community structure of the epilithic, epiphytic and epipelic habitats from these streams was not consistently different. The results indicate no apparent benefit to sampling discrete habitats for water quality monitoring using diatoms, however, the best relationship are obtained using the sum of the diatom data from all three habitats.     

The invertebrate species index (ISI) for streams in southeast Queensland,Australia

The invertebrate species index (ISI) is a new biotic index to assess stream health in southeast Queensland, Australia, using benthic macroinvertebrates. The index aims to refine stream monitoring, in particular for eutrophication, as nutrient input is a major stressor of streams in the region. Biotic indices previously used for the region were based on sensitivity scores for macroinvertebrate families and orders, and were valid for all streams across the continent. The ISI is based on species level and tailored to the specific traits of southeast Queensland, thus yielding an increased level of detection of biological change. This will improve monitoring of environmental impact on a regional and local scale. The ISI is a site-specific index calculated as the weighted average (WA) of species’ sensitivity scores (S₁₀), with a species-specific indicator weight (W) and the abundance (A) of each species used as weights. S₁₀ scores for 203 species of benthic macroinvertebrates ranging from 10 (species most sensitive to pollution) to 1 (tolerant of excessive pollution) were derived by means of WA regression and calibration using site scores representing an environmental impact gradient. W measures the indicator strength of the species, and was derived from the weighted standard deviation of the S₁₀. The initial site scores for the WA modeling were derived using canonical correspondence analysis (CCA) to ordinate the sites along a gradient associated with 12 abiotic variables. The data on benthic macroinvertebrates are based on 212 quantitative samples collected in wadeable freshwater streams in southeast Queensland. Two major stream types, (1) small creeks, mainly of uplands and (2) larger streams and rivers of lowlands, were recognised for the region, and for both types the ISI range representing reference condition was established. These reference conditions can be used to establish ecological quality ratios by comparing observed to expected indices and to define ecological quality classes. The ISI is the first biotic index for streams in Australia that uses sensitivity scores and indicator weights for macroinvertebrate species. There is a growing trend in Australia to identify stream macroinvertebrates to species level and to study their specific traits and ecological requirements. The reasons for this are manifold; assessing and monitoring stream health is only one of them. For most regions of Australia, no common ground exists, so far, on how to use species data for stream health assessment. The new biotic index fills this gap for southeast Queensland in providing a standard for the use of species level data in stream health assessment.     

The bioassessment of fish and macroinvertebrates in a Mediterranean–Atlantic climate: Habitat assessment and concordance between contrasting ecological samples

Human intervention and drought conditions in Mediterranean rivers may necessitate the collection of organisms from a different habitat type to that specified by the original bioassessment protocol; it is therefore critical to understand the consequences of adaptive sampling strategies in these lotic systems. In a survey of 31 streams (2nd–4th order) in northern Portugal we collected separate macroinvertebrate samples from running-water and standing-water habitats and fish samples from a delineated reach, inclusive of all habitat types. In addition to collecting habitat data specific to macroinvertebrates and fish we conducted a generalized River Habitat Survey (RHS) and derived scores designed to report habitat quality and habitat degradation. Ecological data were described in terms of taxonomic structure, functional organization and by biotic indices and interpreted in relation to habitat data, comparing samples within and between sites for concordance. Macroinvertebrates from respective habitats differed in their relationship with habitat variable with running-water samples more strongly related to substrate and water quality and standing-water samples more strongly reflecting habitat characteristics at the reach scale. Running- and standing-water samples from the same-site varied widely, indicating that substituting standing-water for running-water samples in macroinvertebrate-based bioassessment carries a high risk of misclassification. However, a consistent bias in index scores suggests that a habitat-adjusted interpretation of biotic indices could be employed to improve measurement precision. Variation in key parameters of the fish community corresponded to changes in water quality and habitat structure and to reach-scale features recorded by the RHS. Trends between macroinvertebrate and fish measurements indicated common patterns for oxygen/flow sensitive taxa and for trophic organization that were related to similar habitat preferences. Whilst macroinvertebrates demonstrated no relationship with derived scores of habitat quality, the fish-based Index of well-being (Iwb) was strongly related to the Habitat Modification Score. Overall these data indicate how different ecological samples can be used to focus on different aspects of habitat quality and are suggestive of strategies for both the collection and interpretation of ecological data that would improve assessment performance. This study emphasizes the need to adapt assessment methods not only to the species pool but also to the differences in the managed and natural habitat characteristics of these types of Mediterranean streams.     

Detritivores in Kenyan highland streams: more evidence for the paucity of shredders in the tropics?

1. The relationship between coarse particulate organic matter (CPOM) standing stock and benthic macroinvertebrate assemblages in Kenyan highland streams was determined by sampling seven sites on three rivers (2000–2700 m a.s.l.). Taxa recorded were allocated to functional feeding groups using published literature, mouthpart analysis and examination of gut contents. Patterns were compared with five structurally similar streams in three areas of Europe (south-west France, south-east England, north-east England).
2. Number of individuals and proportion of detritivores in Kenyan streams were equivalent to, or greater than, those in European sites. Shredders were, however, almost completely absent from Kenyan sites, despite high standing stocks of CPOM. Shredders were abundant in all European sites.
3. The phenomenon of low shredder abundance has been observed in other tropical streams in south-east Asia and Central and South America but, in contrast to these regions, the African rivers studied were devoid of shrimps or fish which may occupy the shredding niche elsewhere.
4. These preliminary data suggest that shredder-mediated detritus processing, which is a key functional component of streams in the North Temperate Zone, does not operate in East African streams. There are three possible reasons for this. The first is that tropical African rivers are functionally different to those in temperate regions. This could be because of enhanced microbial activity replacing shredder activity at high temperatures. Alternatively, it could be a result of low palatability of detrital inputs from dominant riparian trees in the region. The second and third are methodological: that our allocation to functional feeding groups is incorrect, and that our sampling methods missed a potentially key shredding taxon – the freshwater crab Potamonautes sp.     

Headwaters are critical reservoirs of microbial diversity for fluvial networks

Streams and rivers form conspicuous networks on the Earth and are among nature''s most effective integrators. Their dendritic structure reaches into the terrestrial landscape and accumulates water and sediment en route from abundant headwater streams to a single river mouth. The prevailing view over the last decades has been that biological diversity also accumulates downstream. Here, we show that this pattern does not hold for fluvial biofilms, which are the dominant mode of microbial life in streams and rivers and which fulfil critical ecosystem functions therein. Using 454 pyrosequencing on benthic biofilms from 114 streams, we found that microbial diversity decreased from headwaters downstream and especially at confluences. We suggest that the local environment and biotic interactions may modify the influence of metacommunity connectivity on local biofilm biodiversity throughout the network. In addition, there was a high degree of variability in species composition among headwater streams that could not be explained by geographical distance between catchments. This suggests that the dendritic nature of fluvial networks constrains the distributional patterns of microbial diversity similar to that of animals. Our observations highlight the contributions that headwaters make in the maintenance of microbial biodiversity in fluvial networks.