Classification Tree Models for Predicting Distributions of Michigan Stream Fish from Landscape Variables

Steen, Paul J.M. Van; Zorn, Troy G.; Seelbach, Paul W.; Schaeffer, Jeffrey S.

doi:10.1577/t07-119.1

Cited by 51 publications

(86 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our study is one of the few studies that have used an independent data set to validate fish species occurrence models and to quantify confidence in model predictions (e.g., Leftwich et al 1997;Steen et al 2008). Most often, species occurrence models fail to accurately predict species distributions outside of the region for which the models were developed.…”

Section: Discussionmentioning

confidence: 99%

“…Furthermore, an understanding of species-habitat relationships can provide insight into the effects of land use practices, habitat alterations, and climate change on species distributions (Wang et al 2003;Wall et al 2004;Lyons et al 2010). Modeling of species distributions is an important tool for addressing many issues in conservation (Guisan and Thuiller 2005), and the use of predictive occurrence models to further the understanding of fish species' relationships with habitat features in freshwater systems is common (e.g., Olden and Jackson 2001;Rich et al 2003;Steen et al 2008). As habitat loss and degradation continue to threaten fish biodiversity in North America (Miller et al 1989;Richter et al 1997;Jelks et al 2008), species distribution models are playing an increasingly important role in conservation.…”

mentioning

confidence: 99%

“…Habitat variables measured at large scales are often able to explain substantial variation in fish assemblages, particularly in highly variable landscapes (e.g., Rocky Mountain streams; Rahel and Hubert 1991) or across large spatial scales (e.g., entire midwestern USA; MarshMatthews and Matthews 2000). Additionally, the modeling of species occurrences with large-scale habitat variables is economical because such variables are easily measured with a GIS, thereby reducing the need for costly field sampling (Oakes et al 2005;Steen et al 2008). However, in other regions, large-scale habitat variables such as land use may not effectively explain variation in fish assemblage structure (Rich et al 2003;Heitke et al 2006) or may only do so indirectly (Rowe et al 2009a).…”

mentioning

confidence: 99%

See 2 more Smart Citations

Habitat Associations of Fish Species of Greatest Conservation Need at Multiple Spatial Scales in Wadeable Iowa Streams

Sindt

Quist

Pierce

2012

N American J Fish Manag

View full text Add to dashboard Cite

Fish and habitat data were collected from 84 wadeable stream reaches in the Mississippi River drainage of Iowa to predict the occurrences of seven fish species of greatest conservation need and to identify the relative importance of habitat variables measured at small (e.g., depth, velocity, and substrate) and large (e.g., stream order, elevation, and gradient) scales in terms of their influence on species occurrences. Multiple logistic regression analysis was used to predict fish species occurrences, starting with all possible combinations of variables (5 large-scale variables, 13 small-scale variables, and all 18 variables) but limiting the final models to a maximum of five variables. Akaike's information criterion was used to rank candidate models, weight model parameters, and calculate model-averaged predictions. On average, the correct classification rate (CCR = 80%) and Cohen's kappa (κ = 0.59) were greatest for multiple-scale models (i.e., those including both largescale and small-scale variables), intermediate for small-scale models (CCR = 75%; κ = 0.49), and lowest for large-scale models (CCR = 73%; κ = 0.44). The occurrence of each species was associated with a unique combination of large-scale and small-scale variables. Our results support the necessity of understanding factors that constrain the distribution of fishes across spatial scales to ensure that management decisions and actions occur at the appropriate scale. RightsWorks produced by employees of the U.S. Government as part of their official duties are not copyrighted within the U.S. The content of this document is not copyrighted. AbstractFish and habitat data were collected from 84 wadeable stream reaches in the Mississippi River drainage of Iowa to predict the occurrences of seven fish species of greatest conservation need and to identify the relative importance of habitat variables measured at small (e.g., depth, velocity, and substrate) and large (e.g., stream order, elevation, and gradient) scales in terms of their influence on species occurrences. Multiple logistic regression analysis was used to predict fish species occurrences, starting with all possible combinations of variables (5 large-scale variables, 13 small-scale variables, and all 18 variables) but limiting the final models to a maximum of five variables. Akaike's information criterion was used to rank candidate models, weight model parameters, and calculate model-averaged predictions. On average, the correct classification rate (CCR = 80%) and Cohen's kappa (κ = 0.59) were greatest for multiple-scale models (i.e., those including both large-scale and small-scale variables), intermediate for small-scale models (CCR = 75%; κ = 0.49), and lowest for large-scale models (CCR = 73%; κ = 0.44). The occurrence of each species was associated with a unique combination of large-scale and small-scale variables. Our results support the necessity of understanding factors that constrain the distribution of fishes across spatial scales to ensure that management decisions and actio...

show abstract

Section: Discussionmentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Habitat Associations of Fish Species of Greatest Conservation Need at Multiple Spatial Scales in Wadeable Iowa Streams

Sindt

Quist

Pierce

2012

N American J Fish Manag

View full text Add to dashboard Cite

show abstract

“…Classification trees were then used to predict the occurrence of the stream fish assemblages identified with cluster analyses using the land use, landscape and in-stream variables described above. Classification trees are a powerful method to analyse complex ecological data (De'ath and Fabricius 2000) and have recently been used to describe the distribution of individual fish species in arctic lakes (Hershey et al 2006) and in Michigan streams (Steen et al 2008). The library rpart in S-Plus (Insightful Corp.) using untransformed response and predictor variables was used to generate the classification trees, and separate models were developed for each of the four spatial scales (50 m riparian corridor, 500 m riparian corridor, the entire catchment and in-stream variables) and for the two stream types.…”

Section: Discussionmentioning

confidence: 99%

Prediction of stream fish assemblages from land use characteristics: implications for cost-effective design of monitoring programmes

Kristensen

Baattrup‐Pedersen

Andersen

2011

Environ Monit Assess

View full text Add to dashboard Cite

Increasing human impact on stream ecosystems has resulted in a growing need for tools helping managers to develop conservations strategies, and environmental monitoring is crucial for this development. This paper describes the development of models predicting the presence of fish assemblages in lowland streams using solely cost-effective GIS-derived land use variables. Three hundred thirty-five stream sites were separated into two groups based on size. Within each group, fish abundance data and cluster analysis were used to determine the composition of fish assemblages. The occurrence of assemblages was predicted using a dataset containing land use variables at three spatial scales (50 m riparian corridor, 500 m riparian corridor and the entire catchment) supplemented by a dataset on in-stream variables. The overall classification success varied between 66.1-81.1% and was only marginally better when using in-stream variables than when applying only GIS variables. Also, the prediction power of a model combining GIS and in-stream variables was only slightly better than prediction based solely on GIS variables. The possibility of obtaining precise predictions without using costly in-stream variables offers great potential in the design of monitoring programmes as the distribution of monitoring sites along a gradient in ecological quality can be done at a low cost.

show abstract

“…For the US, a comparable dataset is the NHD plus system (1:24K scale), which provides climate, hydrology, and land-use information summarized within the entire upstream network above each stream reach. Many freshwater species distribution modeling efforts have utilized the NHDplus data (1:24k) and architecture because of topological connectivity and habitat predictors offered by the resource [102][103][104][105][106][107] ( Figure 11). Although NHDplus is a convenient database to support freshwater species distribution modeling, it does not adequately represent 1st order streams, the majority of which provide habitat for freshwater taxa (Figure 11).…”

Section: A Synopsis Of Globalmentioning

confidence: 99%

A Synopsis of Global Mapping of Freshwater Habitats and Biodiversity: Implications for Conservation

McManamay¹,

Griffiths²,

DeRolph³

et al. 2018

Pure and Applied Biogeography

View full text Add to dashboard Cite

Accurately mapping freshwater habitats and biodiversity at high-resolutions across the globe is essential for assessing the vulnerability and threats to freshwater organisms and prioritizing conservation efforts. Since the 2000s, extensive efforts have been devoted to mapping global freshwater habitats (rivers, lakes, and wetlands), the spatial representation of which has changed dramatically over time with new geospatial data products and improved remote sensing technologies. Some of these mapping efforts, however, are still coarse representations of actual conditions. Likewise, the resolution and scope of global freshwater biodiversity compilation efforts have also increased, but are yet to mirror the spatial resolution and fidelity of mapped freshwater environments. In our synopsis, we find that efforts to map freshwater habitats have been conducted independently of those for freshwater biodiversity; subsequently, there is little congruence in the spatial representation and resolution of the two efforts. We suggest that global species distribution models are needed to fill this information gap; however, limiting data on habitat characteristics at scales that complement freshwater habitats has prohibited global high-resolution biogeography efforts. Emerging research trends, such as mapping habitat alteration in freshwater ecosystems and trait biogeography, show great promise in mechanistically linking global anthropogenic stressors to freshwater biodiversity decline and extinction risk.

show abstract

Classification Tree Models for Predicting Distributions of Michigan Stream Fish from Landscape Variables

Cited by 51 publications

References 50 publications

Habitat Associations of Fish Species of Greatest Conservation Need at Multiple Spatial Scales in Wadeable Iowa Streams

Habitat Associations of Fish Species of Greatest Conservation Need at Multiple Spatial Scales in Wadeable Iowa Streams

Prediction of stream fish assemblages from land use characteristics: implications for cost-effective design of monitoring programmes

A Synopsis of Global Mapping of Freshwater Habitats and Biodiversity: Implications for Conservation

Contact Info

Product

Resources

About