Scale, rank and model selection in evaluations of land cover influence on wildlife–vehicle collisions

2020 ◽  
Vol 47 (1) ◽  
pp. 44 ◽  
Author(s):  
Scott H. Markwith ◽  
Aaron H. Evans ◽  
Vanessa Pereira da Cunha ◽  
Julio Cesar de Souza
Keyword(s):  
Model Selection ◽  
Land Cover ◽  
Multiple Scales ◽  
Spatial Scales ◽  
Similar Species ◽  
Taxonomic Rank ◽  
Vehicle Collisions ◽  
Quality Models ◽  
Explanatory Variables ◽  
Appropriate Scale

Abstract ContextExamining land cover’s influences on roadkills at single predetermined scales is more common than evaluating multiple scales, but examining land cover at the appropriate scale may be necessary for efficient design of mitigation measures, and that appropriate scale may be difficult to discern a priori. In addition, the taxonomic rank at which data is analysed may influence results and subsequent conclusions concerning mitigation. AimsThe objective of the present study was to assess the influence of variation in spatial scales of land cover explanatory variables and taxonomic rank of response variables in models of wildlife–vehicle collisions (WVCs). Research questions include: (1) do the scales of land cover measurement that produce the highest quality models differ among species; (2) do the factors that influence roadkill events differ within species at different scales of measurement and among species overall; and (3) does the taxonomic rank at which data is analysed influence the selection of explanatory variables? MethodsFour frequent WVC species representing diverse taxonomic classes, i.e. two mammals (Cerdocyon thous and Hydrochaeris hydrochaeris), one reptile (Caiman yacare) and one bird (Caracara plancus), were examined. WVCs were buffered, land cover classes from classified satellite imagery at three buffer radii were extracted, and logistic regression model selection was used. Key resultsThe scale of land cover variables selected for the highest quality models (and the selected variables themselves) may vary among wild fauna. The same explanatory variables and formulae are not always included in the candidate models in all compared scales for a given species. Explanatory variables may differ among taxonomically similar species, e.g. mammals, and pooling species at higher taxonomic ranks can result in models that do not correspond with species-level models of all pooled species. ConclusionsThe most accurate analyses of WVCs will likely be found when species are analysed individually and multiple scales of predictor variable collection are evaluated. ImplicationsMitigating the effects of roadways on wildlife population declines for both common and rare species is resource intensive. Resources spent optimising models for spatially targeting management actions may reduce the amount of resources used and increase the effectiveness of these actions.

Occurrence of wildfire in the northern Great Lakes Region: Effects of land cover and land ownership assessed at multiple scales

2001 ◽  
Vol 10 (2) ◽  
pp. 145 ◽  
Author(s):  
Jeffrey A. Cardille ◽  
Stephen J. Ventura
Keyword(s):  
Land Cover ◽  
Multiple Scales ◽  
Forest Type ◽  
Spatial Scales ◽  
Fire Frequency ◽  
Usda Forest Service ◽  
Data Set ◽  
Spatially Explicit Data ◽  
In Fire ◽  
Relationship Of

Risk of wildfire has become a major concern for forest managers, particularly where humans live in close proximity to forests. To date, there has been no comprehensive analysis of contemporary wildfire patterns or the influence of landscape-level factors in the northern, largely forested parts of Minnesota, Wisconsin and Michigan, USA. Using electronic archives from the USDA Forest Service and from the Departments of Natural Resources of Minnesota, Wisconsin, and Michigan, we created and analysed a new, spatially explicit data set: the Lake States Fire Database. Most of the 18 514 fires during 1985—1995 were smaller than 4 ha, although there were 746 fires larger than 41 ha. Most fires were caused by debris burning and incendiary activity. There was considerable interannual variability in fire counts; over 80% of fires occurred in March, April, or May. We analysed the relationship of land cover and ownership to fires at two different fire size thresholds across four gridded spatial scales. Fires were more likely on non-forest than within forests; this was also true if considering only fires larger than 41 ha. An area of National or State Forest was less likely to have experienced a fire during the study period than was a forest of equal size outside National or State Forest boundaries. Large fires were less likely in State Forests, although they were neither more nor less likely to have occurred on National Forests. Fire frequency also varied significantly by forest type. All results were extremely consistent across analysis resolutions, indicating robust relationships.

scholarly journals Urbanisation and wing asymmetry in the western honey bee (Apis mellifera, Linnaeus 1758) at multiple scales

PeerJ ◽  
2018 ◽  
Vol 6 ◽  
pp. e5940 ◽  
Author(s):  
Ryan J. Leonard ◽  
Katie K.Y. Wat ◽  
Clare McArthur ◽  
Dieter F. Hochuli
Keyword(s):  
Apis Mellifera ◽  
Land Cover ◽  
Honey Bee ◽  
Multiple Scales ◽  
Spatial Scales ◽  
Wing Shape ◽  
Phenotypic Traits ◽  
Wing Size ◽  
Geometric Morphometric ◽  
The Impact

Changes in the mean and variance of phenotypic traits like wing and head morphology are frequently used as indicators of environmental stress experienced during development and may serve as a convenient index of urbanization exposure. To test this claim, we collected adult western honey bee (Apis mellifera Linnaeus 1758, Hymenoptera, Apidae) workers from colonies located across an urbanization gradient, and quantified associations between the symmetries of both wing size and wing shape, and several landscape traits associated with urbanization. Landscape traits were assessed at two spatial scales (three km and 500 m) and included vegetation and anthropogenic land cover, total road length, road proximity and, population and dwelling density. We then used geometric morphometric techniques to determine two wing asymmetry scores—centroid size, a measure of wing size asymmetry and Procrustes distance, a measure of wing shape asymmetry. We found colony dependent differences in both wing size and shape asymmetry. Additionally, we found a negative association between wing shape asymmetry and road proximity at the three km buffer, and associations between wing shape asymmetry and road proximity, anthropogenic land cover and vegetation cover at the 500 m buffer. Whilst we were unable to account for additional variables that may influence asymmetry including temperature, pesticide presence, and parasitism our results demonstrate the potential usefulness of wing shape asymmetry for assessing the impact of certain landscape traits associated with urbanization. Furthermore, they highlight important spatial scale considerations that warrant investigation in future phenotypic studies assessing urbanization impact.

scholarly journals Generalized Linear Models outperform commonly used canonical analysis in estimating spatial structure of presence/absence data

PeerJ ◽  
2020 ◽  
Vol 8 ◽  
pp. e9777
Author(s):  
Lélis A. Carlos-Júnior ◽  
Joel C. Creed ◽  
Rob Marrs ◽  
Rob J. Lewis ◽  
Timothy P. Moulton ◽  
...  
Keyword(s):  
Model Selection ◽  
Spatial Structure ◽  
Generalized Linear Models ◽  
Species Interactions ◽  
Linear Models ◽  
Spatial Scales ◽  
Error Rates ◽  
Ecological Communities ◽  
Forward Selection ◽  
Explanatory Variables

Background Ecological communities tend to be spatially structured due to environmental gradients and/or spatially contagious processes such as growth, dispersion and species interactions. Data transformation followed by usage of algorithms such as Redundancy Analysis (RDA) is a fairly common approach in studies searching for spatial structure in ecological communities, despite recent suggestions advocating the use of Generalized Linear Models (GLMs). Here, we compared the performance of GLMs and RDA in describing spatial structure in ecological community composition data. We simulated realistic presence/absence data typical of many β-diversity studies. For model selection we used standard methods commonly used in most studies involving RDA and GLMs. Methods We simulated communities with known spatial structure, based on three real spatial community presence/absence datasets (one terrestrial, one marine and one freshwater). We used spatial eigenvectors as explanatory variables. We varied the number of non-zero coefficients of the spatial variables, and the spatial scales with which these coefficients were associated and then compared the performance of GLMs and RDA frameworks to correctly retrieve the spatial patterns contained in the simulated communities. We used two different methods for model selection, Forward Selection (FW) for RDA and the Akaike Information Criterion (AIC) for GLMs. The performance of each method was assessed by scoring overall accuracy as the proportion of variables whose inclusion/exclusion status was correct, and by distinguishing which kind of error was observed for each method. We also assessed whether errors in variable selection could affect the interpretation of spatial structure. Results Overall GLM with AIC-based model selection (GLM/AIC) performed better than RDA/FW in selecting spatial explanatory variables, although under some simulations the methods performed similarly. In general, RDA/FW performed unpredictably, often retaining too many explanatory variables and selecting variables associated with incorrect spatial scales. The spatial scale of the pattern had a negligible effect on GLM/AIC performance but consistently affected RDA’s error rates under almost all scenarios. Conclusion We encourage the use of GLM/AIC for studies searching for spatial drivers of species presence/absence patterns, since this framework outperformed RDA/FW in situations most likely to be found in natural communities. It is likely that such recommendations might extend to other types of explanatory variables.

scholarly journals Dispersal and Land Cover Contribute to Pseudorabies Virus Exposure in Invasive Wild Pigs

EcoHealth ◽  
2021 ◽  
Author(s):  
Felipe A. Hernández ◽  
Amanda N. Carr ◽  
Michael P. Milleson ◽  
Hunter R. Merrill ◽  
Michael L. Avery ◽  
...  
Keyword(s):  
Land Cover ◽  
Disease Transmission ◽  
Wildlife Conservation ◽  
Pseudorabies Virus ◽  
Spatial Scales ◽  
Information Criterion ◽  
Small Scale ◽  
Wild Pigs ◽  
Bacterial Agent ◽  
Open Canopy

AbstractWe investigated the landscape epidemiology of a globally distributed mammal, the wild pig (Sus scrofa), in Florida (U.S.), where it is considered an invasive species and reservoir to pathogens that impact the health of people, domestic animals, and wildlife. Specifically, we tested the hypothesis that two commonly cited factors in disease transmission, connectivity among populations and abundant resources, would increase the likelihood of exposure to both pseudorabies virus (PrV) and Brucella spp. (bacterial agent of brucellosis) in wild pigs across the Kissimmee Valley of Florida. Using DNA from 348 wild pigs and sera from 320 individuals at 24 sites, we employed population genetic techniques to infer individual dispersal, and an Akaike information criterion framework to compare candidate logistic regression models that incorporated both dispersal and land cover composition. Our findings suggested that recent dispersal conferred higher odds of exposure to PrV, but not Brucella spp., among wild pigs throughout the Kissimmee Valley region. Odds of exposure also increased in association with agriculture and open canopy pine, prairie, and scrub habitats, likely because of highly localized resources within those land cover types. Because the effect of open canopy on PrV exposure reversed when agricultural cover was available, we suggest that small-scale resource distribution may be more important than overall resource abundance. Our results underscore the importance of studying and managing disease dynamics through multiple processes and spatial scales, particularly for non-native pathogens that threaten wildlife conservation, economy, and public health.

scholarly journals How far from the road should land cover be assessed? A case study on mesopredator mortality on roads

2021 ◽  
Vol 67 (2) ◽  
Author(s):  
Angelika Nieszała ◽  
Daniel Klich
Keyword(s):  
Land Cover ◽  
Red Fox ◽  
Home Range Size ◽  
Buffer Size ◽  
Local Conditions ◽  
Explanatory Variables ◽  
The Road ◽  
Spatially Correlated ◽  
European Badger ◽  
The Impact

AbstractThe methods used to assess the significance of land cover in the vicinity of a road for the mortality of mesopredators are diverse. In assessing the effect of land cover along the road on road causalities, scientists use various buffer sizes, or even no buffer along the road. The aim of this study was to verify how results of land cover effects on the mortality of mesopredators on roads may differ when analyzing various buffer sizes from the road. We assessed road causalities in the Warmian-Masurian voivodeship (Poland) from 3 consecutive years: 2015, 2016, and 2017. The roads were divided into equal sections of 2000 m each with buffer size of radius: 10, 250, 500, and 1000 m. We analyzed the number of road kills of red fox and European badger separately in a generalized linear model, whereas explanatory variables we used land cover types (based on the Corine Land Cover inventory) and traffic volume. Mean annual mortality from road collisions amounts to 2.36% of the red fox population and 3.82% of the European badger population. We found that the buffer size determines the results of the impact of land cover on mesocarnivore mortality on roads. The red fox differed from the European badger in response to land cover depending on the buffer size. The differences we have shown relate in particular to built-up areas. Our results indicate a 500-m buffer as best reflecting the land cover effects in road kills of both species. This was confirmed by model evaluation and a tendency to use or avoid the vicinity of human settlements of the analyzed species. We concluded that buffer size will probably affect mostly the significance of cover types that are spatially correlated with roads, positively or negatively. We suggest that the home range size of given species in local conditions should be assessed before determining the size of the buffer for analysis.

Implications of temporal and spatial scale for Atlantic salmon (Salmo salar) research

1998 ◽  
Vol 55 (S1) ◽  
pp. 9-21 ◽  
Author(s):  
Carol L Folt ◽  
Keith H Nislow ◽  
Mary E Power
Keyword(s):  
Atlantic Salmon ◽  
Salmo Salar ◽  
Multiple Scales ◽  
Spatial Scales ◽  
Effective Means ◽  
Research Strategies ◽  
Atlantic Salmon Salmo Salar ◽  
Temporal And Spatial ◽  
And Migration ◽  
Dependent Processes

The Atlantic salmon (Salmo salar) is a model species for studying scale issues (i.e., the extent, duration, and resolution of a study or natural process) in ecology. Major shifts in behavior and habitat use over ontogeny, along with a relatively long life span and large dispersal and migration distances, make scale issues critical for effective conservation, management, and restoration of this species. The scale over which a process occurs must be linked to the research design and we illustrate this with a discussion of resource tracking by Atlantic salmon. Identifying scale inconsistencies (e.g., when a process is evident at one scale but not another) is shown to be an effective means by which some scale-dependent processes are understood. We review the literature to assess the temporal and spatial scales used in Atlantic salmon research and find most current studies appear to sacrifice spatial and temporal extent for increased resolution. Finally, we discuss research strategies for expanding the temporal and spatial scales in salmon research, such as conducting multiple scales studies to elucidate scale inconsistencies, identifying mechanisms, and using techniques and approaches to generalize across studies and over time and space.

scholarly journals Improving and parameterising nitrogen and phosphorus modelling for application of LUCI in New Zealand

2021 ◽  
Author(s):  
◽  
Martha Trodahl
Keyword(s):  
Water Quality ◽  
New Zealand ◽  
Land Cover ◽  
Land Management ◽  
Stream Water ◽  
Nutrient Loss ◽  
Catchment Scale ◽  
Quality Models ◽  
Export Coefficient ◽  
Water Quality Models

<p>Over the last 50 years freshwater and marine environments have become severely impaired due to contamination from pathogens, heavy metals, sediment, industrial chemicals and nutrients (MEA 2005b). In many countries, including New Zealand, increased nitrogen (N) and phosphorus (P) loading to terrestrial and freshwater environments from diffuse nutrient sources are of particular concern (MEA 2005a; PCE 2015b; Steffen et al. 2015) and many governments now mandate control of diffuse nutrient loss to water. Water quality models are invaluable tools that can assist with decision making around this widespread issue through exploration of the current situation and future scenarios.  Many water quality models exist, functioning at a variety of temporal and spatial scales and varying in detail and complexity. However, few, if any, simultaneously represent sub-field to catchment scale processes and outcomes, both of which are required to fully address water quality issues associated with diffuse nutrient sources. Those that do, likely require extensive time and expertise to operate. Water quality models embedded in the Land Utilisation and Capability Indicator (LUCI), an ecosystem service decision support framework, offer the opportunity to overcome these limitations. Being highly spatially explicit, yet straightforward to use, they can inform and assist individual land owners, catchment managers and other stakeholders with planning, decision making and management of water quality at sub-field to landscape scale.  To model diffuse nutrient losses LUCI, like many catchment scale water quality models, requires some form of estimated nutrient loss, or export coefficient, from land units within the catchment of interest. To be representative export coefficients must consider climate, soil, topography, and land cover and management variables. A number of methods of export coefficient derivation exist, although generally they consider only very limited geo-climatic, land cover and land management variables.  The principal aim of this study is development of algorithms capable of calculating New Zealand site specific N and P export coefficients from detailed geo-climatic, land cover and land management variables, for application in LUCI water quality models. Algorithms for pastoral land cover are developed from a large dataset comprising real pastoral farm input and output data from nutrient budgeting model OVERSEER. Algorithms are extended to land covers other than pasture, albeit in a limited manner. This is achieved through rescaling of the pastoral algorithms to account for relative differences in literature reported N and P losses from pasture and a variety of other New Zealand land covers. Application of the developed algorithms in LUCI water quality models results in positioning of export coefficients at the DEM grid square scale (≤ 15 m x 15 m for New Zealand). In addition, intra-basin configuration is considered in LUCI, at the same grid square scale, as water and nutrient flows are cascaded through the catchment. Application of the export coefficient calculating algorithms are applied to two contrasting New Zealand catchments. Tuapaka catchment, an 85ha agricultural foothill catchment in Manawatu, North Island, and Lake Rotorua catchment, a 502 km2 volcanic, mixed land cover catchment in Bay of Plenty, North Island.  This research is supported by Ravensdown, a farmer owned co-operative, which plans to use LUCI extensively to advise and assist farmers with water quality issues. The ability to model mitigation strategies in LUCI is an important capability. Therefore, this research also includes a review of five particularly important on-farm mitigation strategies, which will later be used by the wider LUCI development team to assist with better parameterisation and improved performance of mitigation options in LUCI.  Application of the developed algorithms at farm to catchment scale in LUCI results in considerably more nuanced, detailed maps and data showing N and P sources and pathways, compared to LUCI’s previously used ‘one export coefficient per land cover’ approach. Although results indicate absolute nutrient loss values are not always ‘correct’ compared to either OVERSEER predictions or in-stream water quality measurements, these differences appear comparable to those seen with similar water quality models. In addition, the issue of representativeness of OVERSEER predictions and in-stream water quality measurements exists.  Nevertheless improvement to absolute predictions is always an aim. This research indicates further improvements to LUCI water quality predictions could result from refinement of both pastoral and other land cover algorithms, and from improved representation of attenuation processes in LUCI, including groundwater representation. However, lack of measured on-land and in-stream N and P loss data is a major challenge to both algorithm refinement and to evaluation of results. In addition, more detailed spatial data would provide more nuanced results from algorithm application.  Although the algorithm application context in this research is LUCI water quality models applied in New Zealand, this does not preclude application of the developed algorithms in other export coefficient based, catchment scale water quality models. Using spatial data pertaining to climate, soil, topographic and land management variables, land units of combined variables can be identified and the algorithms applied, resulting in explicitly positioned export coefficients that can be fed into the catchment scale water quality model of interest. Therefore, developments made here potentially represent a wider contribution to catchment scale modelling using export coefficients.</p>

scholarly journals Are Distribution Patterns Correlated with Plant Traits?

2021 ◽  
Author(s):  
◽  
Benjamin Magana-Rodriguez
Keyword(s):  
Life History ◽  
Leaf Area ◽  
Spatial Patterns ◽  
Life History Traits ◽  
Multiple Scales ◽  
Spatial Scales ◽  
Regional Scale ◽  
Distribution Patterns ◽  
Spatial Arrangement ◽  
Scaling Relationship

<p>The current crisis in loss of biodiversity requires rapid action. Knowledge of species' distribution patterns across scales is of high importance in determining their current status. However, species display many different distribution patterns on multiple scales. A positive relationship between regional (broad-scale) distribution and local abundance (fine-scale) of species is almost a constant pattern in macroecology. Nevertheless interspecific relationships typically contain much scatter. For example, species that possess high local abundance and narrow ranges, or species that are widespread, but locally rare. One way to describe these spatial features of distribution patterns is by analysing the scaling properties of occupancy (e.g., aggregation) in combination with knowledge of the processes that are generating the specific spatial pattern (e.g., reproduction, dispersal, and colonisation). The main goal of my research was to investigate if distribution patterns correlate with plant life-history traits across multiple scales. First, I compared the performance of five empirical models for their ability to describe the scaling relationship of occupancy in two datasets from Molesworth Station, New Zealand. Secondly, I analysed the association between spatial patterns and life history traits at two spatial scales in an assemblage of 46 grassland species in Molesworth Station. The spatial arrangement was quantified using the parameter k from the Negative Binomial Distribution (NBD). Finally, I investigated the same association between spatial patterns and life-history traits across local, regional and national scales, focusing in one of the most diverse families of plant species in New Zealand, the Veronica sect. Hebe (Plantaginaceae). The spatial arrangement was investigated using the mass fractal dimension. Cross-species correlations and phylogenetically independent contrasts were used to investigate the relationships between plant life-history traits and spatial patterns on both data bases. There was no superior occupancy-area model overall for describing the scaling relationship, however the results showed that a variety of occupancy-area models can be fit to different data sets at diverse spatial scales using nonlinear regression. Additionally, here I showed that it is possible to deduce and extrapolate information on occupancy at fine scales from coarse-scale data. For the 46 plantassemblage in Molesworth Station, Specific leaf area (SLA) exhibits a positive association with aggregation in cross-species analysis, while leaf area showed a negative association, and dispersule mass a positive correlation with degree of aggregation in phylogenetic contrast analysis at a local-scale (20 × 20 m resolution). Plant height was the only life-history trait that was associated with degree of aggregation at a regional-scale (100 × 60 mresolution). For the Veronica sect. Hebe dataset, leaf area showed a positive correlation with aggregation while specific leaf area showed a negative correlation with aggregation at a fine local-scale (2.5-60 m resolution). Inflorescence length, breeding system and leaf area showed a negative correlation with degree of aggregation at a regional-scale (2.5-20 km resolution). Height was positively associated with aggregation at national-scale (20-100 km resolution). Although life-history traits showed low predictive ability in explaining aggregation throughout this thesis, there was a general pattern about which processes and traits were important at different scales. At local scales traits related to dispersal and completion such as SLA , leaf area, dispersule mass and the presence of structures in seeds for dispersal, were important; while at regional scales traits related to reproduction such as breeding system, inflorescence length and traits related to dispersal (seed mass) were significant. At national scales only plant height was important in predicting aggregation. Here, it was illustrated how the parameters of these scaling models capture an important aspect of spatial pattern that can be related to other macroecological relationships and the life-history traits of species. This study shows that when several scales of analysis are considered, we can improve our understanding about the factors that are related to species' distribution patterns.</p>

scholarly journals RANDOM AND SYSTEMATIC LAND-COVER TRANSITIONS IN NORTH-EASTERN WOLLEGA, ETHIOPIA

2017 ◽  
Vol 332 ◽  
pp. 3-15 ◽  
Author(s):  
Alemayehu Adugna ◽  
Assefa Abegaz ◽  
Asmamaw Legass ◽  
Diogenes L. Antille
Keyword(s):  
Land Cover ◽  
Agricultural Land ◽  
Spatial Scales ◽  
Random Processes ◽  
Processes Of Change ◽  
Landsat Data ◽  
Loss Ratio ◽  
Agricultural Abandonment ◽  
North Eastern ◽  
Transi Tions

Africa has seen significant changes in land cover at different spatial scales. Changes in Land Use and Land Cover (LULC) include deforestation and subse- quent use of the land for arable cropping, conversion to grassland or urbanization. The work reported in this article was conducted to examine land cover transi- tions in north-eastern Wollega (Ethiopia) between 2005 and 2015. The analysis focused on land cover transitions that occurred systematically or randomly, and identified the main drivers for these changes. Landsat data from 2005 and 2015 were examined to better unders- tand the various dimensions of land cover transitions, namely: swaps, losses, gains, persistency and vulnerability. Results showed that shrubland exhibited the largest gain (22%), with a 63% gain- to-loss ratio, a 47% gain-to-persistence ratio and a positive net change-to-persis- tence ratio of 46%. Cropland showed the largest loss (19%) while grassland was the most stable type of land cover des- pite some fluctuation (»10%) observed during the 10-year period. The land cover transition was dominated by systematic processes, with few random processes of change. Systematic land cover transitions such as agricultural abandonment and vegetation re-growth were attributed to regular or common processes of change. This study suggests that the implementa- tion of practices conducive to sustainable intensification of existing agricultural land, supported by policies that promote increased diversification of Ethiopian agriculture, would mitigate pressure on forests by avoiding their future conver- sion to cropland.

Landscape Influences on Stream Habitats and Biological Assemblages

2006 ◽  
Keyword(s):  
Land Cover ◽  
Ictalurus Punctatus ◽  
Environmental Variables ◽  
Agricultural Land ◽  
Fish Assemblages ◽  
Spatial Scales ◽  
Species Abundance ◽  
Ecological Condition ◽  
Basin Scale ◽  
Assemblage Metrics

<em>Abstract.</em>—We analyzed data from 38 sites on 31 large rivers in Wisconsin to characterize the influence of environmental variables at the basin, reach, and site scales on fish assemblages. Electrofishing and site habitat data were collected for a distance of 1.6 km per site. Environmental variables included conductivity, substrate, and fish cover at the site scale; distance to impoundments, dams, and length of riverine habitat at the reach scale; and land cover, climate, and geology at the basin scale. Of the 77 fish species found, 39 occurred in more than 10% of the sites and were retained for analyses of fish abundance and biomass. Redundancy analysis (RDA) was used to relate species abundance, biomass, and 16 assemblage metrics to environmental variables at the three spatial scales. The site and basin scales defined fishes along a gradient from high conductivity, fine substrate, and agricultural land cover to low conductivity, rocky substrate, and forested land cover. For abundance and biomass, the strongest assemblage pattern contrasted northern hog sucker <em>Hypentelium nigricans</em>, blackside darter <em>Percina maculata</em>, and logperch <em>P. caprodes </em>with common carp <em>Cyprinus carpio</em>, channel catfish <em>Ictalurus punctatus</em>, and sauger <em>Sander canadensis</em>. The <em>H. nigricans </em>group, along with high values of index of biotic integrity and some assemblage metrics (percent lithophilic spawners, percent round-bodied suckers), corresponded with the forested end of the ecological gradient, whereas the <em>C. carpio </em>group and percent anomalies corresponded with the agricultural end. Natural environmental conditions, including bedrock geology type, bedrock depth, surficial geology texture, basin area, and precipitation, also influenced the fish assemblage. Partial RDA procedures partitioned the explained variation among spatial scales and their interactions. We found that widespread land cover alterations at the basin scale were most strongly related to fish assemblages across our study area. Understanding the influence of environmental variables among multiple spatial scales on fish assemblages can improve our ability to assess the ecological condition of large river systems and subsequently target the appropriate scale for management or restoration efforts.

Sign in / Sign up

Export Citation Format

Share Document