Development of a benthic macroinvertebrate multimetric index (MMI) for Neotropical Savanna headwater streams

Assessing the ecological impacts of anthropogenic pressures is a key task in environmental management. Multimetric indices (MMIs), based on aquatic assemblage responses to anthropogenic pressures, have been used increasingly throughout the world. The MMI approach is a low-cost, rapid field method that produces an aquatic condition index that responds precisely to anthropogenic pressures, making it useful for conservation and environmental management. We developed four candidate MMIs based on benthic macroinvertebrate assemblages sampled at 40 randomly selected sites to assess the environmental condition of streams upstream of a hydroelectric power plant in the Brazilian Neotropical Savanna biome. Those MMIs were built from landscape-adjusted and unadjusted biological metrics as well as two alternative ways of choosing metrics. The alternative MMIs performances were tested by comparing their precision to distinguish least-disturbed areas, responsiveness to discriminate least- and most-disturbed areas, and sensitivity to anthropogenic pressures at catchment and local scales. The best performing MMI had landscape-adjusted metrics and was produced through use of principal component analysis for metric selection. It included 4 metrics: Ephemeroptera richness, average tolerance score per taxon, percentage of predator individuals, and percentage of Odonata individuals adjusted by elevation. This index discriminated well the anthropogenic pressures at local- and catchment-scales, and at both scales simultaneously, as indicated by an integrated disturbance index. Our methodological development included statistical criteria for identifying least- and most-disturbed sites, calibrating for natural landscape variability, and use of non-redundant metrics. Therefore, we expect it will provide a model for environmental assessment of water resources elsewhere in Brazil and in other nations.     

Development of a multimetric plant-based index of biotic integrity for assessing the ecological state of forested,urban and agricultural natural wetlands of Jimma Highlands,Ethiopia

Wetlands are vital natural resources thereby providing ecological and socio-economic benefits to the people. However, anthropogenic activities have seriously changed the ecological conditions of wetlands worldwide, especially in developing nations like Ethiopia. Predominantly, the absence of biomonitoring tool greatly hampers the protection and management of wetlands. Therefore, the objective of this research is to develop a plant-based index of biological integrity for facilitating the management of wetlands. Accordingly, 122 plant species belonging to 37 families were collected and identified from forested, urban and agricultural wetland types and included in the analysis of the plant metrics. Initially, we reviewed and screened 35 potential metrics. Then, we selected four core metrics (% cover of native species, sensitive plant species richness, tolerant plant species richness and% of shrub species richness) using the decrease or increase responses to human disturbances, Mann-Whitney U test and redundant metric test. A trisect-quartile range system using box plots of the reference or impaired sites of wetlands was established to provide values for each core metrics. Then after, we combined the core metrics to develop the plant-based index of biological integrity. Finally, we validated the index by comparing the index response to different wetland types. Additionally, the index was validated based on the measured environmental variables that characterize the human disturbance gradient of wetlands. We found that the plant-based index is robust to discriminate the reference wetlands from impaired wetlands and can also be used as an effective tool for evaluating the long term natural wetland conditions of the Eastern African wetlands in the future.     

Biological indices applied to benthic macroinvertebrates at reference conditions of mountain streams in two ecoregions (Poland, the Slovak Republic)

The study was carried out from 2007 to 2010 in two ecoregions: the Carpathians and the Central Highlands. The objectives of our survey were to test the existing biological index metric based on benthic macroinvertebrates at reference conditions in the high- and mid-altitude mountain streams of two ecoregions according to the requirements of the EU WFD and to determine which environmental factors influence the distribution of benthic macroinvertebrates. Our results revealed statistically significant differences in the values of the physical and chemical parameters of water as well as the mean values of metrics between the types of streams at the sampling sites. RDA analysis showed that the temperature of the water, pH, conductivity, the stream gradient, values of the HQA index, and altitude were the parameters most associated with the distribution of benthic macroinvertebrate taxa and the values of the metrics. The values of biological indices should be considered according to the stream typology including altitude and geology. At the reference conditions, the suggested border values of biological indices are very harsh. The values of the biological indices of most sampling sites did not correspond to the requirements of the high status in rivers. The streams at altitudes above 1,200 m a.s.l. should be treated as another river type and new reference values should be established.     

An improved macroinvertebrate multimetric index for the assessment of wadeable streams in the neotropical savanna

Multimetric indices (MMIs) have been successfully used to assess ecological conditions in freshwater ecosystems worldwide, and provide an important management tool especially in countries where biological indicators are fostered by environmental regulations. Nonetheless, for the neotropics, the few published papers are limited to small local scales and lack standardized sampling protocols. To fill the gaps left by previous studies, we propose a stream MMI that reflects anthropogenic impacts by using macroinvertebrate assemblage metrics from a data set of 190 sites collected from four hydrologic units in the Paraná and São Francisco River Basins, southeastern Brazil. Sites were selected through use of a probabilistic survey design allowing us to infer ecological condition to the total of 9432 kilometers of wadeable streams in the target population in the four hydrologic units. We used a filtering process to determine the least- and most-disturbed sites based on their water quality, physical habitat structure, and land use. To develop the MMI, we followed a stepwise procedure to screen our initial set of biological metrics for influence of natural variation, responsiveness and discriminance to disturbances, sampling variability, and redundancy. The final MMI is the sum of 7 scaled assemblage metrics describing different aspects of macroinvertebrate assemblage characteristics: Ephemeroptera richness, % Gastropoda individuals, Shannon-Wiener diversity index, % sensitive taxa richness, % scraper individuals, temporarily attached taxa richness, and gill respiration taxa richness. The MMI clearly distinguished the least-disturbed sites from the most-disturbed sites and showed a significant negative response to anthropogenic stressors. Of the total length of wadeable streams in the study area, 38%, 35%, and 27% were classified by the MMI as being in good, fair, and poor condition, respectively. By reducing the subjectivity of site selection, rigorously selecting the set of reference sites, and following a standardized metric screening method, we developed a robust MMI to assess and monitor ecological condition in neotropical savanna streams. This improved MMI provides an effective ecological tool to guide decision makers and managers in developing and implementing improved, cost-effective environmental policies, regulations, and monitoring of those systems.     

An advanced Index of Biotic Integrity for use in tropical shallow lowland streams in Costa Rica: Fish assemblages as indicators of stream ecosystem health

This paper presents an advanced version of the Index of Biotic Integrity (IBI), a multimetric index to indicate ecosystem health. The multimetric index has been adapted in such a way that it not only indicates overall condition but also specific causes of environmental disturbance. The newly developed index (a) uses data of tolerant as well as intolerant species in a single metric to indicate environmental disturbance, (b) does not require knowledge about species from the literature, and (c) can be applied to artificial landscapes.The metrics proposed here consist of indicator species assemblages that are selected directly for their relationship with an environmental component or specific type of environmental degradation. Thus, each metric indicates a type of environmental concern, which enables conservation practices to be targeted more effectively. Species assemblages for each single metric consist of a combination of species that can be negatively and positively related to environmental disturbances, providing a better indication of stream ecosystem health.The area studied was assumed to be too diverse for one single index. Canonical Indirect-Gradient Principal Component Analysis indicated that the optimal division of subindices based on stream typology was for streams with drainage basin sizes <10 km² and >10 km². Pearson Product-Moment Correlations were used to identify relationships between anthropogenic disturbances and the composition and abundance of fish species at impacted as well as undisturbed sites. This index proved to be useful for indicating overall stream ecosystem health as well as local onsite environmental disturbances or the environmental components of greatest concern. This index does require extensive information about measured levels of anthropogenic disturbances with the accompanying composition and abundance of fish species.     

A tool to assess the ecological condition of tropical high Andean streams in Ecuador and Peru: The IMEERA index

The main objective of this study was to develop a highland Andean streams ecological assessment tool for managers to determine the biological quality in this broad area of South America. Sampling was conducted during the dry season at 123 sites in eight watersheds of high Andean streams from south of Peru to North of Ecuador. The sites were at elevations above 2000 m a.s.l., and ranged in anthropogenic disturbance from none or little (reference) to highly disturbed. Using the physicochemical, hydromorphological and aquatic macroinvertebrate assemblage attributes of the reference sites, two different elevation bioregions were identified (from 2000 to 3500 m a.s.l. and those sites at altitudes higher than 3500 m a.s.l.). Differences between these two bioregions were related to the change in altitude of the most relevant environmental factors, i.e., temperature, oxygen content of the water and the extent of forested vegetation in the basin and in the riparian zone. These features were paramount to having different macroinvertebrate assemblages as demonstrated by an MDS analysis of our data. Two versions of the multi-metric index IMEERA were developed (the acronym comes from the Spanish name ‘Índice Multimétrico del Estado Ecológico para Ríos Altoandinos’) that corresponded to the two bioregions. In the lower altitude bioregion (Bosque river type, IMEERA-B index), the pressure gradient was driven by the organic pollution and the hydromorphological degradation. While in the higher elevation bioregion (Páramo and Puna river types; IMEERA-P river type), the gradient was driven by the organic pollution and the habitat heterogeneity. The IMEERA B index includes six macroinvertebrate metrics using its richness, habit characteristics and tolerance/intolerance to disturbances (EPT taxa, % clingers, % climbers, intolerant taxa, ABI and % tolerant taxa). The IMEERA P index was calculated using four metrics corresponding to macroinvertebrate richness and its tolerance/intolerance to disturbances (total taxa, intolerant taxa, ABI and % tolerant taxa). The index was validated with a set of independent data from the headwaters of Guayllabamba River in Ecuador.     

Incorporating natural variability in the bioassessment of stream condition in the Atlantic Forest biome,Brazil

Most bioassessment programs in Brazil face difficulties when scaling up from small spatial scales because larger scales usually encompass great environmental variability. Covariance of anthropogenic pressures with natural environmental gradients can be a confounding factor in the evaluation of biologic responses to anthropogenic pressures. The objective of this study was to develop a multimetric index (MMI) with macroinvertebrates for two stream types and two ecoregions in the Atlantic Forest biome in Rio de Janeiro state, Brazil. We hypothesized that by using two approaches – (1) testing and adjusting metrics to landscape parameters, and (2) selecting metrics using a cluster analysis to avoid metrics redundancy – the final MMI would perform better than the traditional approach (unadjusted metrics, one metric representing each category). Four MMIs were thus developed: MMI-1 – adjusted MMI with metrics selected after cluster analysis); MMI-2 – adjusted MMI with one metric from each category; MMI-3 – unadjusted MMI with metrics selected after cluster analysis; MMI-4 – unadjusted MMI with one metric from each category. We used three decision criteria to assess MMI’s performance: precision, responsiveness and sensitivity. In addition, we tested the MMI’s by using an independent set of sites to validate the results. Although all MMIs performed well in the three criteria, adjusting metrics to natural variation increased MMI response and sensitivity to impairment. In addition, the selected MMI-2 was able to classify sites of two stream types and two ecoregions. The use of cluster analysis, however, did not avoid high redundancy between metrics of different branches. The MMI-4 had the poorest performance among all tested MMIs and it was not able to distinguish adequately reference and impaired sites from both ecoregions. We present some considerations on the use of metrics and on the development of MMI’s in Brazil and elsewhere.     

Headwater streams and forest management: Does ecoregional context influence logging effects on benthic communities?

Effects of forest management on stream communities have been widely documented, but the role that climate plays in the disturbance outcomes is not understood. In order to determine whether the effect of disturbance from forest management on headwater stream communities varies by climate, we evaluated benthic macroinvertebrate communities in 24 headwater streams that differed in forest management (logged-roaded vs. unlogged-unroaded, hereafter logged and unlogged) within two ecological sub-regions (wet versus dry) within the eastern Cascade Range, Washington, USA. In both ecoregions, total macroinvertebrate density was highest at logged sites (P = 0.001) with gathering-collectors and shredders dominating. Total taxonomic richness and diversity did not differ between ecoregions or forest management types. Shredder densities were positively correlated with total deciduous and Sitka alder (Alnus sinuata) riparian cover. Further, differences in shredder density between logged and unlogged sites were greater in the wet ecoregion (logging × ecoregion interaction; P = 0.006) suggesting that differences in post-logging forest succession between ecoregions were responsible for differences in shredder abundance. Headwater stream benthic community structure was influenced by logging and regional differences in climate. Future development of ecoregional classification models at the subbasin scale, and use of functional metrics in addition to structural metrics, may allow for more accurate assessments of anthropogenic disturbances in mountainous regions where mosaics of localized differences in climate are common.     

Mayfly bioindicator thresholds for several anthropogenic disturbances in neotropical savanna streams

Anthropogenic disturbances are widely recognized as major threats to terrestrial and aquatic biodiversity worldwide, including areas located in non-forest ecosystems. Headwater streams in the neotropical savanna are severely threatened by large-scale landscape changes that degrade local habitat characteristics and lead to biodiversity loss. The objective of our study was to evaluate Ephemeroptera assemblages as bioindicators of catchment land use and cover, local streambed and riparian vegetation conditions, and instream water quality. To do so, we sampled mayfly nymphs in 184 stream sites across a broad disturbance gradient in four hydrologic units of the Brazilian neotropical savanna. We selected seven metrics without significant co-variation with natural variability: % catchment urban, riparian vegetation condition index (RCOND), human disturbances of the stream channel and riparian zone (W1_HALL), substrate mean embeddedness (XEMBED), dissolved oxygen (mg L⁻¹), pH, and total phosphorus (mg L⁻¹). We ran threshold indicator taxa analysis (TITAN) for each disturbance metric to detect change points in mayfly genera responses (whether sensitive or tolerant) and assemblage turnover pattern. TITAN showed that 20 of the 39 genera found were robust bioindicators (based on purity and reliability values >0.95), sixteen of them being sensitive to increased disturbance. The most sensitive genera were Tricorythopsis (Leptohyphidae) and Camelobaetidius (Baetidae), showing decreased abundance to most disturbance metrics. We found a turnover pattern of mayfly genera in response to W1_HALL in a narrow variation range. For total phosphorus, the benchmark value defined in Brazilian Federal Legislation is higher than the turnover threshold of several mayfly genera. This indicates that we will lose many sensitive genera even within the limits imposed by national environmental legislation. The indicator taxa approach, based on multiple taxa rather than univariate metrics or single indicator species, demonstrates the value of quantitative ecological information for conserving and managing freshwater ecosystems globally.     

From compositional to functional biodiversity metrics in bioassessment: A case study using stream macroinvertebrate communities

While compositional diversity is a common metric for assessing human impacts on aquatic communities, functional diversity is scarcely employed, though highly desirable from the perspective of the European Water Framework Directive. Using abundance data from 99 minimally disturbed sites (i.e., no or very weak anthropogenic impact) from a national survey, we studied the spatial variability of compositional and functional biodiversity metrics across a predefined ecoregional classification. Metrics of compositional diversity comprised taxonomic and EPT richness and Simpson diversity. Functional diversity metrics were based on Rao's Quadratic Entropy (RQE), which described the differences among benthic invertebrate genera in eleven biological traits (e.g., size, life cycle, reproduction types, feeding habits). Using generalized linear models we show that taxonomic richness may vary greatly across ecoregions, contrasting with Simpson diversity and functional metrics that varied weakly in response to natural environmental variability. Functional diversity metrics, because of their stability in response to natural environmental variability, may be useful tools for assessing human impairment to ecosystem function. We further tested the response of functional diversity metrics to a specific human impact (sewage) and demonstrated significant modifications of functional diversity downstream of sewage pollution. Further investigations are required to test the ability of functional diversity metrics to precisely and accurately indicate different types of human impacts.     

A fish index of biotic integrity for South Dakota's Northern Glaciated Plains Ecoregion

Ecosystem goods and services in streams are impaired when their biotic communities are degraded by anthropogenic stressors. An index of biotic integrity (IBI) translates community structure into a standardized ecoregion-specific stream health score. Documenting stream health is especially important in the Northern Glaciated Plains (NGP) Ecoregion, which is undergoing rapid landscape alterations through increased agriculture production. Our objectives were to develop a fish IBI and validate candidate reference sites for NGP wadeable perennial streams. Fish were sampled from 54 sites (consisting of reference sites, known-condition least and most disturbed sites, and random sites) during summers 2006–2011. Candidate metrics were sorted into nine metric classes based on attributes of fish assemblage form and function. Metric values were screened using metric range, signal-to-noise ratios, responsiveness to disturbance, and redundancy tests until each metric class contained only those metrics most responsive to anthropogenic stressors. The final IBI consisted of six metrics that were reflective of prairie stream fish assemblages, and differentiated between known-condition least and most disturbed sites. The mean reference sampling site IBI scores were found to be similar to both least and most disturbed sites (Mann–Whitney U-test; P < 0.05). Twelve reference site scores were below the NGP's median (69), whereas the other 11 sites were above the median and were representative of least disturbed conditions. We now have developed a standardized bioassessment tool for evaluating stream health, as well as a baseline for long-term monitoring in a dynamic ecoregion.     

Development of dual fish multi-metric indices of biological condition for streams with characteristic thermal gradients and low species richness

Biological indicators based on fish assemblage characteristics are used to assess stream condition worldwide. Fish-based bioassessment poses challenges in Southern New England, the USA, due to the effects of within-watershed thermal gradients on fish assemblage types, low regional species richness, and lack of minimally disturbed sites. Dual multi-metric indices (MMI) of biological condition were developed for wadeable streams based on fish assemblage characteristics sampled across watershed landscapes with varying levels of human disturbance. A coldwater MMI was developed using streams with drainage area of ≤15 km², and a mixed-water MMI for streams with drainage areas of >15 km². For each MMI development, candidate metrics represented by ecological classes were sequentially tested by metric range, within-year precision, correlation with stream size, responsiveness to landscape-level human disturbances, and redundancy. Resultant coldwater and mixed-water MMI were composed of 5 and 7 metrics, respectively. Stream sites tended to score similarly when the two MMI were applied to transitional sites, i.e., drainage areas of 5–40 km². However, some sites received high scores from the mixed-water MMI and intermediate scores from the coldwater MMI. It was thus difficult to ascertain high-quality mixed-water streams from potential coldwater streams which currently support mixed-water assemblages due to ecological degradation. High-quality coldwater streams were restricted to stream sites with drainage areas ≤15 km². The newly developed fish-based MMI will serve as a useful management tool and the dual-MMI development approach may be applicable to other regions with thermal gradients that transition from coldwater to warmwater within watersheds.     

Response of three lotic assemblages to riparian and catchment‐scale land use: implications for designing catchment monitoring programmes

1. Although many studies have focussed on the effects of catchment land use on lotic systems, the importance of broad (catchment) and fine (segment/reach) scale effects on stream assemblages remain poorly understood. 2. Nine biological metrics for macrophytes (498 sites), benthic macroinvertebrates (491) and fish (478) of lowland and mountain streams in four ecoregions of France and Germany were related to catchment and riparian buffer land use using partial Redundancy Analysis and Boosted Regression Trees (BRTs). 3. Lotic fauna was better correlated (mean max., r = 0.450) than flora (r = 0.277) to both scales of land use: the strongest correlations were noted for mountain streams. BRTs revealed strong non‐linear relationships between mountain assemblage metrics and land use. Correlations increased with increasing buffer lengths, suggesting the importance of near‐stream land use on biotic assemblages. 4. Several metrics changed markedly between 10–20% (mountain ecoregions) and 40–45% (lowland) of arable land use, irrespective of the buffer size. At mountain sites with >10% catchment arable land use, metric values differed between sites with <30% and sites with >30% forest in the near‐stream riparian area. 5. These findings support the role of riparian land use in catchment management; however, differences between mountain and lowland ecoregions support the need for ecoregion‐specific management.     

Assessment of biotic condition of Atlantic Rain Forest streams: A fish-based multimetric approach

We developed a preliminary fish-based multimetric index (MMI) to assess biotic condition of Atlantic Rain Forest streams in Southeastern Brazil. We used least-disturbed sites as proxies of reference conditions for metric development. To determine the disturbance gradient we used an Integrated Disturbance Index (IDI) that summarized the multiple disturbances measured at local/regional catchment scales in a single index, describing the totality of exposure of the streams to human pressures. For our 48 sites, nine were least-disturbed (IDI < 0.25), five were most-disturbed (IDI > 1.35) and 34 were intermediate. Initially, we considered 41 candidate metrics selected primarily from previous studies. We screened this pool of candidate metrics using a series of tests: range test, signal-to-noise test, correlation with natural gradients, responsiveness test, and redundancy test. After screening, we selected six metrics for the MMI: % Characiform individuals, % water column native individuals, % benthic invertivorous individuals, % tolerant species, % intolerant species, and % detritivorous individuals. Metrics such as diversity, dominance, species richness and biomass that have been historically used for assessing ecosystem condition failed one or more screening tests. We conclude that an IDI and rigorous metric screening are critical to the MMI development process and for meaningful assessments of stream condition.     

Effects of natural hydrological variability on fish assemblages in small Mediterranean streams: Implications for ecological assessment

Small Mediterranean streams are shaped by predictable seasonal events of flooding and drying over an annual cycle, and present a strong inter and intra-annual variation in flow regime. Native fish assemblages in these streams are adapted to this natural environmental variability. The distinction of human-induced disturbances from the natural ones is thus a crucial step before assessing the ecological status of these streams. In this aim, the present study evaluates the effects of natural hydrological variability on fish assemblages from disturbed and least disturbed sites in small intermittent streams of south Portugal. Data were collected over the last two decades (1996–2011) in 14 sites located in the Guadiana and Sado river basins. High variability of fish assemblages was strongly dependent on human-induced disturbances, particularly nutrient/organic load and sediment load, and on natural hydrological variability. Natural hydrological variability can act jointly with anthropogenic disturbances, producing changes on fish assemblages structure of small intermittent streams. In least disturbed sites, despite the natural disturbances caused by inter-annual rainfall variations (including drought and flood events), fish assemblages maintained a long-term stability and revealed a high resilience. On the contrary, disturbed sites presented significantly higher variability on fish assemblages and a short and long-term instability, reflecting a decrease on the resistance and resilience of fish assemblages. Under these conditions, fish fauna integrity is particularly vulnerable and the ecological assessment may be influenced by natural hydrological variations. High hydrological variability (especially if it entails high frequency of dryer years and meaningful cumulative water deficit) may affect the impact of the human pressures with significant and consistent consequences on fish assemblage composition and integrity. In this study, fish metrics that maximize the detection of human degradation and minimize the response to natural variability were based on the relative abundance of native species (insectivorous species, eurytopic species, water column species, native lithophilic species), relative abundance of species with intermediate tolerance and relative number of exotic species. Results highlight the importance of assessing temporal variability on stream biomonitoring programs and emphasize the need to improve the assessment tools, accounting for long-term changes in fish assemblages, namely by selecting the most appropriate fish metrics that respond to anthropogenic disturbances but exhibit low natural temporal variability, essential both in the characterization of the biological reference conditions and in the development of fish indexes in intermittent streams.     

Regional and local drivers of macroinvertebrate assemblages in boreal springs

Aim To identify the most important environmental drivers of benthic macroinvertebrate assemblages in boreal springs at different spatial scales, and to assess how well benthic assemblages correspond to terrestrially derived ecoregions. Location Finland. Methods Benthic invertebrates were sampled from 153 springs across four boreal ecoregions of Finland, and these data were used to analyse patterns in assemblage variation in relation to environmental factors. Species data were classified using hierarchical divisive clustering (twinspan ) and ordinated using non‐metric multidimensional scaling. The prediction success of the species and environmental data into a priori (ecoregions) and a posteriori (twinspan ) groups was compared using discriminant function analysis. Indicator species analysis was used to identify indicator taxa for both a priori and a posteriori assemblage types. Results The main patterns in assemblage clusters were related to large‐scale geographical variation in temperature. A secondary gradient in species data reflected variation in local habitat structure, particularly abundance of minerogenic spring brooks. Water chemistry variables were only weakly related to assemblage variation. Several indicator species representing southern faunistic elements in boreal springs were identified. Discriminant function analysis showed poorer success in classifying sites into ecoregions based on environmental than on species data. Similarly, when classifying springs into the twinspan groups, classification based on species data vastly outperformed that based on environmental data. Main conclusions A latitudinal zonation pattern of spring assemblages driven by regional thermal conditions is documented, closely paralleling corresponding latitudinal patterns in both terrestrial and freshwater assemblages in Fennoscandia. The importance of local‐scale environmental variables increased with decreasing spatial extent. Ecoregions provide an initial stratification scheme for the bioassessment of benthic macroinvertebrates of North European springs. Our results imply that climate warming, landscape disturbance and degradation of spring habitat pose serious threats to spring biodiversity in northern Europe, especially to its already threatened southern faunistic elements.     

Patterns in the species composition of fish assemblages among Wisconsin streams

Synopsis To better understand patterns of fish assemblage composition in Wisconsin streams in relation to major environmental gradients, I carried out multivariate direct gradient analysis (canonical correspondence analysis) of two large independent datasets on fish species abundance in Wisconsin streams. Analysis of the two datasets yielded similar results, suggesting that observed patterns and relationships were real. Stream sites were distributed along fish species-environment gradients, but segregation into distinct stream temperature and geographic groups was also evident. The strongest gradient in both datasets was related to summer water temperature patterns, and encompassed a transition from small, coldwater streams dominated by salmonids, cottids, certain cyprinids, and few other species, to both small and large, warmwater streams dominated by a high diversity of different cyprinids, catostomids, ictalurids, centrarchids, and percids. A second gradient in both datasets was complex but largely geographic. Within it, sites from each of the four ecoregions that occupy Wisconsin formed fairly discrete groups. The strongest differences were between sites in the two southern Wisconsin ecoregions, the Driftless Area and the Southeastern Wisconsin Till Plains, that were dominated by certain cyprinids, ictalurids, and centrarchids, and sites in the two northern Wisconsin ecoregions, the North Central Hardwood Forests and the Northern Lakes and Forests, that were dominated by a different set of cyprinids and ictalurids, plus some petromyzontids, salmonids, catostomids, and percids. Sites from the Driftless Area that were mostly higher-gradient (steep stream slope) and had many riffle-dwelling species could also be distinguished from sites in the Southeastern Wisconsin Till Plains that were mostly lower-gradient and had many pool-dwelling species. The patterns of fish assemblage composition among sites and the associated fish species-environment relationships that were revealed by the analyses provided a framework for developing an ecologically meaningful hierarchical classification of Wisconsin stream sites based on stream thermal regime, ecoregion, stream size, and stream gradient.     

A causal examination of the effects of confounding factors on multimetric indices

The development of multimetric indices (MMIs) as a means of providing integrative measures of ecosystem condition is becoming widespread. An increasingly recognized problem for the interpretability of MMIs is controlling for the potentially confounding influences of environmental covariates. Most common approaches to handling covariates are based on simple notions of statistical control, leaving the causal implications of covariates and their adjustment unstated. In this paper, we use graphical models to examine some of the potential impacts of environmental covariates on the observed signals between human disturbance and potential response metrics. Using simulations based on various causal networks, we show how environmental covariates can both obscure and exaggerate the effects of human disturbance on individual metrics. We then examine from a causal interpretation standpoint the common practice of adjusting ecological metrics for environmental influences using only the set of sites deemed to be in reference condition. We present and examine the performance of an alternative approach to metric adjustment that uses the whole set of sites and models both environmental and human disturbance effects simultaneously. The findings from our analyses indicate that failing to model and adjust metrics can result in a systematic bias towards those metrics in which environmental covariates function to artificially strengthen the metric–disturbance relationship resulting in MMIs that do not accurately measure impacts of human disturbance. We also find that a “whole-set modeling approach” requires fewer assumptions and is more efficient with the given information than the more commonly applied “reference-set” approach.     

A multi-assemblage,multi-metric biological condition index for eastern Amazonia streams

Multimetric indices (MMIs) are widely used for assessing ecosystem condition and they have been developed for a variety of biological assemblages. However, when multiple assemblages are assessed at sites, the assessment results may differ because of differing physiological sensitivities to particular stressor gradients, different organism size and guilds, and the effects of different scales of disturbances on the assemblages. Those differences create problems for managers seeking to avoid type-1 and type-2 statistical errors. To alleviate those problems, we used an anthropogenic disturbance index for selecting and weighting metrics, modeled metrics against natural variability to reduce the natural variability in metrics, and developed an MMI based on both fish and aquatic insect metrics. We evaluated eight different ways of calibrating and combining candidate metrics and found that MMIs with unweighted and modeled aquatic insect and fish metrics were the preferred MMI options.