The importance of land-use legacies for modeling present-day species distributions

Context

Land-use legacies play an important role in shaping contemporary species distributions. However, land-use legacies are rarely considered in species distribution models (SDMs) that aim to model present-day species distributions across the landscape, even though they can lead to a species absence in suitable areas. SDMs that do not account for land-use legacies will likely result in biased predictions of species distributions.

Objective

We examine the importance of land-use legacies for modeling present-day distributions of tree species at a regional scale, assessing how the addition of land-use legacy variables improves predictive power of SDMs.

Methods

We generated land-use legacy variables using raster layers of reconstructed historical agricultural land use and 3310 inventory plots. SDMs were developed for six forest tree species based on climatic, edaphic, and topographic variables, and with (SDM_LU) and without (SDM_Base) land-use legacy variables. We compared the predictive power between SDM_LU and SDM_Base models and then quantified the local importance of land-use legacy variables relative to other abiotic variables.

Results

Our results show that the importance of land-use legacy variables for present-day species distributions and the improvement on the predictive power of SDMs is species-specific. The inclusion of land-use legacy variables improved SDMs primarily by lowering errors of commission and increasing the overall accuracy of prediction.

Conclusion

The influence of land-use legacies on SDMs suggests that, for some tree species, incorporating land-use legacies can accurately identify suitable areas that are not occupied by the species due to land-use legacies, and advance our understanding of their present-day distributions.

     

Effect of landscape context on anuran communities in breeding ponds in the National Capital Region,Canada

Land cover change, predominantly habitat conversion to agricultural use and urbanization, has recently been recognized as the primary cause of biodiversity loss in terrestrial ecosystems. We evaluated the relative effects of urban and agricultural landscapes on anuran species richness and the abundance of six anuran species found at breeding ponds in and around the cities of Ottawa, Ontario and Gatineau, Quebec. We performed six call surveys at 29 permanent focal ponds surrounded by one of three landscape contexts: primarily urban, primarily agricultural, and primarily forested. We also measured three local pond variables to control for the effects of local habitat quality in our analyses. We found that anuran species richness was significantly lower in breeding ponds in urban landscapes compared to forested and agricultural landscapes, which exhibited no significant difference in species richness. The abundances of individual anuran species were also consistently lower in urban landscapes for all species except one, which exhibited no response to landscape type. Three species had their highest abundances in ponds in forested landscapes, whereas two species had their highest abundances in ponds in agricultural landscapes. We conclude that ponds embedded in urban landscapes support lower biodiversity than ponds in agricultural settings. We suggest that landscapes composed of a mosaic of forest and open habitats surrounding wetlands would hold the highest biodiversity of these species.     

Pond density as a determinant of aquatic species richness in an urban landscape

Green spaces within urban areas provide services and benefits to human populations and habitat for a variety of species. Freshwater, in the form of rivers, canals, lakes, reservoirs and ponds, is an important component of urban greenspaces. This paper focuses on ponds; and specifically ponds within urban areas. This work is timely as during 2008 ponds were designated, in the UK, as habitats of national conservation importance. Yet, while farmland ponds have received considerable attention, there has been little work on the ecology and landscape ecology of urban ponds. Ecological data was collected from 37 ponds in the Borough of Halton (northwest England) over a period of 2 years (2005–2006). The median species richness in these ponds was 28 invertebrate species and 10 macrophyte species. A highly significant correlation was observed between pond density and species richness. The relationship between the richness of different taxa varied according to scale; becoming more significant within pond clusters than within a single pond. These findings have significance for those involved in planning and managing urban environments, further strengthening the need for functional ecological connectivity in urban areas. With pressure to increase infill development, and thus raise housing density, a greater understanding of the affect of urban design on pond ecology will be of importance to urban planners and ecologists alike.     

Responses of anuran communities to rapid urban growth in Shanghai,China

Urbanization affects amphibian communities through habitat loss, fragmentation, and degradation of habitat quality. The effects of these changes in habitat at different scales vary depending on the sensitivity of individual species. We assessed the breeding distribution of anurans along an urban–rural gradient in Shanghai, China, a region experiencing intensive urbanization. Our results showed that urban density had a significantly negative influence on the overall anuran abundance and diversity and that the responses of individual species to urbanization varied. Pond age was an overall predictor in models describing the responses of Pelophylax nigromaculatus, Fejervarya multistriata, and M. fissipes and total anuran abundance. The quality of habitat at a pond was also important, and the high abundance of Bufo gargarizans and Pelophylax plancyi was associated with ponds with aquatic vegetation coverage. Urban density showed strong negative effects on B. gargarizans, total anuran abundance, and species richness. The broad-scale landscape variables associated with forests, agricultural fields, and wetlands surrounding breeding ponds have been shown to affect anuran abundance and species richness. The response of individual species, total abundance, and species richness to urbanization reflected differences in their ecological requirements. We quantified the effects of urbanization on frogs in a rapidly urbanizing region, and our results demonstrated that both multi-spatial and temporal variables affect anurans in Shanghai. Our results emphasized the importance of anuran conservation planning in urbanized areas to preserve and/or restore terrestrial habitat and to improve connectivity between ponds and other wetlands.     

Predicting mammal species richness and distributions: testing the effectiveness of satellite-derived land cover data

Mapping species richness and distributions is an important aspect of conservation and land use planning, but the time and cost of producing maps from field surveys is prohibitive. It is useful, therefore, if mappable environmental variables, from a readily accessible source, can be used as surrogates for species attributes. We evaluated the power of satellite-derived land cover information, from the Land Cover Map of Great Britain, to predict species richness and occurrences of terrestrial mammals in one hundred 10×10 km quadrats, from four regions of Britain. The predictive power of the land cover data was relatively poor – with a few exceptions, land cover explained less than half of the variation in mammal species richness and occurrence in regression models. Predictive power was considerably stronger when regions were analyzed separately than when analyzed together, and best fitting models varied between regions and between mammal taxa. Predictive power was also affected (positively or negatively depending on taxon) when PCA-ordinated land cover variables were used as predictors. The predictive strength of the land cover data was probably limited mostly by the high proportion of British mammal species with geographic distributions changing rapidly and independently of land cover (and hence the non-saturation of preferred habitats), and to a lesser extent by shortcomings in the mammal and land cover data, and the influence of landscape factors other than land cover on mammal distributions. The results suggest that regional stratification is essential when attempting to predict species richness and distributions, even across relatively limited areas such as Great Britain. We conclude that caution is necessary in using results from environmental information systems such as this as a basis for conservation and land use planning decisions.     

Models of upland species’ distributions are improved by accounting for geodiversity

Context

Recent research suggests that novel geodiversity data on landforms, hydrology and surface materials can improve biodiversity models at the landscape scale by quantifying abiotic variability more effectively than commonly used measures of spatial heterogeneity. However, few studies consider whether these variables can account for, and improve our understanding of, species’ distributions.

Objectives

Assess the role of geodiversity components as macro-scale controls of plant species’ distributions in a montane landscape.

Methods

We used an innovative approach to quantifying a landscape, creating an ecologically meaningful geodiversity dataset that accounted for hydrology, morphometry (landforms derived from geomorphometric techniques), and soil parent material (data from expert sources). We compared models with geodiversity to those just using topographic metrics (e.g. slope and elevation) and climate data. Species distribution models (SDMs) were produced for ‘rare’ (N?=?76) and ‘common’ (N?=?505) plant species at 1 km² resolution for the Cairngorms National Park, Scotland.

Results

The addition of automatically produced landform geodiversity data and hydrological features to a basic SDM (climate, elevation, and slope) resulted in a significant improvement in model fit across all common species’ distribution models. Adding further geodiversity data on surface materials resulted in a less consistent statistical improvement, but added considerable conceptual value to many individual rare and common SDMs.

Conclusions

The geodiversity data used here helped us capture the abiotic environment’s heterogeneity and allowed for explicit links between the geophysical landscape and species’ ecology. It is encouraging that relatively simple and easily produced geodiversity data have the potential to improve SDMs. Our findings have important implications for applied conservation and support the need to consider geodiversity in management.

     

Integrated models that unite local and regional data reveal larger-scale environmental relationships and improve predictions of species distributions

Context

The scale of environmental relationships is often inferred through the use of species distribution models. Yet such models are frequently developed at two distinct scales. Coarse-scale models typically use information-poor (e.g., presence-only) data to predict relative distributions across geographic ranges, whereas fine-scale models often use richer information (e.g., presence–absence data) to predict distributions at local to landscape scales.

Objectives

We unite presence–absence and presence-only data to predict occurrence of species, what we refer to as integrated distribution models. We determine if integrated models improve predictions of species distributions and identification of characteristic spatial scales of environmental relationships relative to presence–absence modeling and ensemble modeling that averages predictions from separate presence-only and presence–absence models.

Methods

We apply recent advances in integrated distribution models to predict Sherman’s fox squirrel (Sciurus niger shermani) distribution in north-central Florida. Presence-only data were collected through a citizen-science program across its geographic range, while presence–absence data were collected using camera trapping surveys across 40 landscapes.

Results

Integrated models estimated environmental relationships with greater precision and identified larger characteristic scales for environmental relationships than using presence–absence data alone. In addition, integrated models tended to have greater predictive performance, which was more robust to the amount of presence–absence and presence-only data used in modeling, than presence–absence and ensemble models.

Conclusions

Integrated distribution models hold much potential for improving our understanding of environmental relationships, the scales at which environmental relationships operate, and providing more accurate predictions of species distributions. Many avenues exist for further advancement of these modeling approaches.

     

An invasive species’ relationship with environmental variables changes across multiple spatial scales

To further our understanding of invasive species?? novel distributions, knowledge of invasive species?? relationships with environmental variables at multiple spatial scales is paramount. Here, we investigate which environmental variables and which spatial scales best explain the invasive mute swan??s (Cygnus olor) distribution in southern Ontario (Canada). Specifically we model mute swan distribution changes according to ecologically-relevant spatial scales: average territory size radius, 140?m; median dispersal distance of cygnets, 3,000?m; and average activity distance of males, 8,000?m. For individual spatial scales, global models using variables measured at each particular scale result in the highest Akaike weights, AUC, and Cohen??s Kappa values. Yet composite models (models combining variables measured at different scales) elicit the best models, as determined by higher Akaike weights and high AUC and Cohen??s Kappa values. Overall, percent water, waterbody perimeter density, temperature, precipitation, and road density are positively correlated with mute swan distribution, while percent forest and elevation are negatively correlated at all scales of analysis. Only percent water and annual precipitation are more influential in determining mute swan distribution at the 3,000 and 8,000?m zone scales than the territory scale. While most species distribution models are performed at a single scale, the results of our study suggest that composite models reflecting a species?? ecological needs provide models of better fit with similar, if not better, predictive accuracy. When analyzing species distributions, we also recommend that ecologists consider the scale of the underlying landscape processes and the effect that this may have on their modelling outcomes.     

The scaling of geographic ranges: implications for species distribution models

Context

The geographic ranges of many species are responding to ongoing environmental change. Processes operating at different levels of biological organization, with corresponding spatial extents and grains and temporal rates, interact with the evolving configuration of environmental conditions to determine range dynamics.

Objectives

To synthesize understanding of scales and scaling, including relevant biological levels of organization, focusing on the processes that mediate species-environment relationships and the models used to make inferences about species distributions.

Methods

We review concepts related to the scaling of geographic ranges and implications for the most commonly used analytic methods, using simple simulations to illustrate important issues.

Results

Many processes lead to species distributions being dependent on environmental conditions within sites and within a neighborhood. Studies with large extents and fine grains can cut across several levels of biological organization (individual, within-population, and metapopulation processes) complicating interpretation. Many geographic ranges are not in dynamic equilibrium, but common models used for inference assume equilibrium. Interspecific interactions shape species distributions at multiple scales, and arguments for ignoring species interactions also assume equilibrium.

Conclusions

There is a need for timely science to inform policy and management decisions; however, we must also strive to provide predictions that best reflect our understanding of ecological systems. Species distributions evolve through time and reflect responses to environmental conditions that are mediated through individual and population processes. Species distribution models that reflect this understanding, and explicitly model dynamics, are likely to give more accurate predictions.

     

Pond area and distance from continuous forests affect amphibian egg distributions in urban green spaces: A case study in Sapporo,Japan

We examined factors that affect egg distributions of amphibians, which are currently declining globally, in urban areas. We counted the number of egg masses of two amphibian species (Hynobius retardatus and Rana pirica) in ponds in urban green spaces during two years (2011 and 2012) and investigated the effects of local and landscape factors on them. Forest area in green spaces and pond area had positive effects, and distance from continuous forests in the suburbs and the rate of pond shore protection had negative effects on the number of eggs. Among these factors, distance from continuous forests and pond area had strong effects on egg distributions. These results suggest the importance of water habitat size and the need for colonization by amphibians from surrounding continuous forests. To conserve urban amphibian assemblages, the preservation of ponds with sufficient area and continuous forests close to urban habitats would be important.     

The effect of exposure on landscape scale soil surface temperatures and species distribution models

Species distribution models (SDMs) often use elevation as a surrogate for temperature or utilise elevation sensitive interpolations from weather stations. These methods may be unsuitable at the landscape scale, especially where there are sparse weather stations, dramatic variations in exposure or low elevational ranges. The goal of this study was to determine whether radiation, moisture or a novel estimate of exposure could improve temperature estimates and SDMs for vegetation on the Illawarra Escarpment, near Sydney, Australia. Forty temperature sensors were placed on the soil surface of an approximately 12,000 ha study site between November 2004 and August 2006. Linear regression was used to determine the relationship with environmental factors. Elevation was correlated more with moderate temperatures (winter maximums, summer minimums, spring and autumn averages) than extreme temperatures (summer maximums, winter minimums). The correlation (r ²) between temperature and environmental factors was improved by up to 0.38 by incorporating exposure, moisture and radiation in the regressions. Summer maximums and winter minimums were predominately determined by exposure to the NW and coastal influences respectively, while exposure to the NE and SW was important during other seasons. These directions correspond with the winds that are most influential in the study area. The improved temperature estimates were used in Generalised Additive Models for 37 plant species. The deviance explained by most models was increased relative to elevation, especially for moist rainforest species. It was concluded that improving the accuracy of seasonal temperature estimates could improve our ability to explain the patchy distribution of many species. Electronic supplementary material The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

Spatial correlates of amphibian use of constructed wetlands in an urban landscape

Many amphibian species rely on both aquatic and terrestrial habitats to complete their life cycles. Therefore, processes operating both within the aquatic breeding habitat, and in the surrounding uplands may influence species distributions and community composition. Moreover, changes in land use adjacent to breeding site may degrade aquatic habitats. To assess land use effects on pond-breeding amphibian assemblages, we investigated relationships between land use, breeding habitat conditions, and breeding amphibian use of constructed wetlands in urban environments of the Baltimore metropolitan area, USA. Forest and impervious surface associations with species richness and occurrence occurred at spatial scales ranging from 50 to 1,000 m, with strongest relationships at 500 m. Forest and impervious surface cover within 1,000 m of ponds were also related to water and sediment quality, which in turn were capable of explaining a proportion of the observed variation in species richness and occurrence. Taken together, our results suggest that forest and other land covers within relatively proximal distances to ponds (i.e., within 50–1,000 m) may be influencing species richness directly via the provisioning of upland habitat, and indirectly via influences on within pond habitat quality. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

The pond network: can structural connectivity reflect on (amphibian) biodiversity patterns?

Landscape connectivity is a very recurrent theme in landscape ecology as it is considered pivotal for the long term conservation of any organism’s populations. Nevertheless, this complex concept is still surrounded by uncertainty and confusion, largely due to the separation between structural and functional connectivity. Amphibians are the most threatened vertebrates around the globe, in Europe mostly due to habitat alteration, and to their particular life cycle. Pond breeding amphibians are considered to be organised in metapopulations, enhancing the importance of landscape connectivity in this group of animals. We sampled the amphibian species present in two pond groups in Central Western Spain. We applied the graph theory framework to these two pond networks in order to determine the importance of each pond for the entire network connectivity. We related the pond importance for connectivity with the species richness present in each pond. We tested if connectivity (partially) determined the presence of the amphibian species sampled using logistic regression. The results show that the structural connectivity of the pond network impacts on the amphibian species richness pattern and that the importance of the pond for the connectivity of the network is an important factor for the presence of some species. Our results, hence, attest the importance of (structural) landscape connectivity determining the pattern of amphibian (functional) colonization in discrete ponds.     

Species absence in developed landscapes: an experimental evaluation

Context

Conversion of landscapes is widely associated with loss of biodiversity. While there are several competing hypotheses for the local extinction of species in developed landscapes, experimental approaches are seldom applied to elucidating mechanisms.

Objectives

In this study, we focus on the habitat degradation hypothesis and predict that poor quality of relictual wetlands in developed landscapes contributes to the absence of wood frogs (Rana sylvatica = Lithobates sylvaticus) by decreasing their performance.

Methods

In a translocation experiment, we reared wood frog larvae within enclosures in seven ponds where they naturally occur and in five ponds in developed landscapes where they are absent. Premature pond drying precluded assessing performance in one present pond and one absent pond.

Results

Absent ponds were surrounded by upland buffers dominated by developed land covers while ponds with wood frog breeding populations were surrounded primarily by intact forest. Ponds were largely similar in their attributes. Survival and growth rate did not differ between pond types. Development tended to be slightly more rapid in some absent ponds perhaps related to higher water temperatures.

Conclusions

Despite the highly altered landscapes surrounding them, we find no evidence that absent wetlands provide inferior habitat for wood frog larval recruitment. Performance in absent ponds matched or exceeded that observed in present ponds implying that absence of this species may stem from influences mediated by the upland landscape. These results provide a caution to the typically unexamined presumption that relictual habitats in developed landscapes are degraded in their utility for wildlife.

     

Biological soil crust distribution is related to patterns of fragmentation and landuse in a dryland agricultural landscape of southern Australia

The dryland agricultural landscape of north-west Victoria, Australia, includes isolated remnants of eucalypt woodland that are exposed to ongoing disturbance from sheep grazing and cropping activity. Biological soil crusts are a functionally important feature of these woodland communities. We used a modern form of regression (boosted regression tree (BRT) models) to investigate relationships between crust abundance and environmental and landscape variables. We also investigated whether the use of broad morphological groups of crust organisms is more informative than simply measuring total crust cover. Remnant size was the single most influential variable for crust abundance, with negligible crust cover in small patches (<5 ha). The BRT model also identified relationships between crust abundance and available P, soil C and perennial grass. We argue that disturbance from stock grazing and camping is the mechanism driving these relationships. Other variables related to crust abundance were proximity to the windward edge, litter cover and tree cover. Morphological groups showed a differential response to some variables, suggesting assessment of total cover may mask important patterns in community structure. Crust disturbance represents a serious issue for maintenance of ecosystem function in the study region, particularly loss of crusts from small remnants because the majority of remnants are small.     

Using climate variables to predict small mammal occurrence in hot, dry environments

Many species are considered to be threatened by the direct and synergistic effects of increased temperatures and less reliable rainfall caused by anthropogenic climate change. The climate of arid Australia is naturally defined by high temperatures and highly variable rainfall. In these environments the survival of small mammals is facilitated by a range of physiological, morphological and behavioural adaptive traits (e.g. insectivorous diets, torpor, nocturnal activity and sheltering in burrows). Features within the landscape associated with these traits may have a proximal influence on species occurrence and be critical for predicting the distribution of arid species. Here we explored the contribution of coarse- and fine-scale environmental variables, sampled from 150 to 10,000 m from known species records, to the development of more robust species distribution models. The inclusion of environmental variables representing topography, soil and vegetation better modelled occurrence for all five small mammal species (up to 55.2 % DE increase over climate-only models) and increased discrimination of species occurrence in an out-of-sample cross-validation for Antechinomys laniger, Dasyuroides byrnei and Notomys cervinus (e.g. increase in area-under-the curve by 0.4). For these species, model performance (% DE) increased with the resolution of environmental data used, reflecting a coarse-scale relationship with distance to watercourse (e.g. at 1,000 m), but masking a fine-scale relationship with rainfall during winter (at 150 m). The inclusion of a fine-scale variable shrub cover under 0.5 m (SD) improved the representation of S. macroura occurrence. These results show improved representation of habitats and resources using course- and fine-scale variables, which may be vital to model the occurrence of some species and their refuge habitats, but as static variables may limit accurate predictions of the effect of climate change on species range dynamics.     

Continuous predictors of species distributions support categorically stronger inference than ordinal and nominal classes: an example with urban bats

Understanding of how species distributions are driven by landscape-level processes has been obscured by null or inconsistent findings from poorly designed studies. We explore how differences in the way potential drivers of species distributions are defined can influence their perceived effects. Specifically, we evaluate how much statistical power is lost when continuous variables are discretised, and how the use of qualitatively defined nominal variables impacts biological interpretation of results. We fitted generalized linear models to dependent variables relating to bat distribution (species richness, diversity, relative abundance of functional groups and individual species) obtained from 36 sites across Melbourne, Australia, and independent variables that were continuous (percentage tree cover, dwelling density), ordinal (dichotomised continuous variables) or nominal (land-use, urban context). We found that models fitted with continuous predictors had better fit and explanatory power than those fitted with ordinal predictors for all response variables. Ordinal models failed to detect statistically significant effects for 4 of the 11 response variables that were successfully modelled with continuous data, suggesting Type II errors had occurred. Models fitted with nominal data explained a comparable amount of variation in some dependent variables as continuous models. However, interpretation of the mechanisms behind responses to nominal categorical levels was obscured because environmental conditions within them were confounded and not homogenous. To gain better understanding from nominal predictors would therefore require further investigation. Our findings show that careful consideration must be given to the choice of environmental variables used for species distribution modelling and how those variables are defined.     

Error assessment in decision-tree models applied to vegetation analysis

Methods were developed to evaluate the performance of a decision-tree model used to predict landscape-level patterns of potential forest vegetation in central New York State. The model integrated environmental databases and knowledge on distribution of vegetation. Soil and terrain decision-tree variables were derived by processing state-wide soil geographic databases and digital terrain data. Variables used as model inputs were soil parent material, soil drainage, soil acidity, slope position, slope gradient, and slope azimuth. Landscapescale maps of potential vegetation were derived through sequential map overlay operations using a geographic information system (GIS). A verification sample of 276 field plots was analyzed to determine: (1) agreement between GIS-derived estimates of decision-tree variables and direct field measurements, (2) agreement between vegetation distributions predicted using GIS-derived estimates and using field observations, (3) effect of misclassification costs on prediction agreement, (4) influence of particular environmental variables on model predictions, and (5) misclassification rates of the decision-tree model. Results indicate that the prediction model was most sensitive to drainage and slope gradient, and that the imprecision of the input data led to a high frequency of incorrect predictions of vegetation. However, in many cases of misclassification the predicted vegetation was similar to that of the field plots so that the cost of errors was less than expected from the misclassification rate alone. Moreover, since common vegetation types were more accurately predicted than rare types, the model appears to be reasonably good at predicting vegetation for a randomly selected plot in the landscape. The error assessment methodology developed for this study provides a useful approach for determining the accuracy and sensitivity of landscape-scale environmental models, and indicates the need to develop appropriate field sampling procedures for verifying the predictions of such models.     

Land use-based stakeholders mapping for natural heritage preservation – Case study: Hampstead Heath Ponds,London

Hampstead Heath as a natural heritage site in the middle of London plays a significant role in the English ancient landscape and retains some essential elements of historical features such as geology, vegetation and animal species, and the early water reservoir and source to fulfil the needs of people since hundreds of years ago. This study examines the spatial distribution of buildings as a tangible aspect of stakeholders’ settling places by applying spatial analysis, constituting proximity analysis and nearest neighbour analysis. The spatial data were collected from some sources such as OpenStreetMap, london.gov.uk, English Heritage, ArcGIS open data, and observation on the site. The results indicate that Hampstead No. 1 Pond was the pond with the densest buildings around, other than unnamed small ponds. The spatial distribution of buildings around the Hampstead Heath ponds was clustered with a Nearest Neighbour Ratio of 0.471575 and z-score of − 27.611095. The observation results of seven ponds supported the evidence of Hampstead No. 1 as the most visited pond and Highgate No. 1 in second place. In contrast, Bird Sanctuary was the most sustainable pond with nihil buildings and the fewest visitors identified. The urban heritage-led framework highlights that each actor plays a role and key impact on the entire system, divided into three sub-systems: economic, urban heritage environment, and socio-political. It is suggested that related stakeholders anticipate potential conflict caused by the commercial pressure from commercial actors such as Property Developers and Business Owners on this heritage site in the future. Involvement of stakeholders, communication, and regulatory systems can be possible actions to suppress this pressure as part of the heritage management system.     

Use of landscape pattern metrics and multiscale data in aquatic species distribution models: a case study of a freshwater mussel

The distributions of freshwater mussels are controlled by landscape factors operating at multiple spatial scales. Changes in land use/land cover (LULC) have been implicated in severe population declines and range contractions of freshwater mussels across North America. Despite widespread recognition of multiscale influences few studies have addressed these issues when developing distribution models. Furthermore, most studies have disregarded the role of landscape pattern in regulating aquatic species distributions, focusing only on landscape composition. In this study, the distribution of Rabbitsfoot (Quadrula cylindrica) in the upper Green River system (Ohio River drainage) is modeled with environmental variables from multiple scales: subcatchment, riparian buffer, and reach buffer. Four types of landscape environment metrics are used, including: LULC pattern, LULC composition, soil composition, and geology composition. The study shows that LULC pattern metrics are very useful in modeling the distribution of Rabbitsfoot. Together with LULC compositional metrics, pattern metrics permit a more detailed analysis of functional linkages between aquatic species distributions and landscape structure. Moreover, the inclusion of multiple spatial scales is necessary to accurately model the hierarchical processes in stream systems. Geomorphic features play important roles in regulating species distributions at intermediate and large scales while LULC variables appear more influential at proximal scales.