Exploring the role of vegetation fragmentation on aquatic conditions: Linking upland with riparian areas in Puget Sound lowland streams

https://doi.org/10.1016/j.landurbplan.2008.10.016Get rights and content

Abstract

A controversial issue in managing urbanizing watersheds is determining the scale at which conservation measures should be implemented. Current “best practices” suggest establishing riparian buffers along stream corridors and limiting impervious surfaces to prevent degradation of instream biological conditions. While there is increasing evidence that the amount of land covers (e.g., impervious surface, vegetation) has an impact on instream aquatic conditions, the effect of upland vegetation fragmentation on aquatic conditions requires further study. By using landscape metrics to quantity vegetation amount and distribution at the riparian and watershed scales, and a macroinvertebrate index to describe aquatic conditions, this study presents empirical evidence about the interactions between riparian and upland vegetation as they affect instream biological condition of 51 nested watersheds in the Puget Sound lowland. We ask if the fragmentation of vegetation within a watershed helps predict instream biological condition. In addition, we hypothesize that the fragmentation of vegetation at the riparian and watershed scales affects instream biological condition. Using parametric and non-parametric statistical analyses to test relationships, our findings suggest that the fragmentation of upland vegetation and the total amount of riparian vegetation explain the greatest amount of variation in aquatic conditions. These results help frame a management approach for conserving upland areas of vegetation through the use of land use planning techniques.

Introduction

Evidence is mounting that the way humans use the land has dramatic consequences on land cover, and that land cover change impacts ecological systems (Dale et al., 1994, Chapin et al., 2000, Foley et al., 2005). Vitousek et al. (1997) estimate that between one-third and one-half of the earth’s land cover has been transformed by human action. By altering land cover patterns and the use of land, humans have impacted the hydrologic cycle to provide freshwater for irrigation, industry, and domestic consumption (Postel et al., 1996, Vörösmarty et al., 2000). Land use and land cover changes threaten biodiversity through loss, modification, fragmentation of habitats, degradation of soil and water; and overexploitation of native species (Pimm and Raven, 2000).

The role of land cover as it affects aquatic systems is well documented in the literature, and watershed management approaches have attempted to reflect scientific understandings. Since Klien (1979) seminal work established that “stream quality impairment is first evidenced when watershed imperviousness reaches 12%, but does not become severe until imperviousness reaches 30%”, dozens of regional investigations have confirmed a relationship between the amount of a specific land cover and aquatic conditions (Osborne and Wiley, 1988, Schueler, 1994, Roth et al., 1996, Richards et al., 1996, Morley and Karr, 2002, Alberti et al., 2007). Accordingly, land use planning has applied “Best Management Practices” (BMPs) at the riparian and watershed scales to mitigate the impact of urban development. At the watershed scale, detention/retention pond ordinances and limits to the amount of allowable impervious surface are BMPs for reducing the amount and velocity of runoff entering stream systems. At the riparian scale, jurisdictions across the U.S. have adopted riparian vegetation ordinances – including critical areas, riparian buffers, and no-touch zones – to prevent reductions in streamside vegetation.

While the application of BMPs at the watershed and riparian scales helps to regulate hydrological, chemical, and trophic conditions occurring within the stream system (Gore, 1996, Stauffer et al., 2000, Konrad and Booth, 2002, Strayer et al., 2003, Stewart et al., 2001, Boyer et al., 2002, Tockner et al., 2002), several issues remain unresolved. First, in watersheds with similar types and amounts of land cover along the riparian corridor, considerable variability in aquatic conditions exists. For example, Booth et al. (2004) show that biological metrics for Little Bear creek (Washington State) almost doubled in value (Benthic Index of Biotic Integrity (B-IBI) change from 16 to 30) from one sampling point to the next downstream point. Second, there is the interaction between riparian and watershed BMPs. While it is commonplace to establish riparian buffers in urbanizing areas, models used to estimate BMP specifications do not differentiate cumulative versus incremental effects of aquatic degradation and rarely consider the presence of other BMPs in the watershed. Accordingly, BMPs are often applied incrementally with little regard to the interaction across scales (Booth and Jackson, 1997, Gergel et al., 2002). We do not know the extent to which BMPs interact to affect aquatic condition.

Finally, and perhaps connected to the inter-relationship among BMPs in the watershed, is that previous studies have produced mixed conclusions about the importance of riparian-scale versus coarse-scale physical factors in affecting aquatic condition (Richards and Host, 1994, Allan et al., 1997, Lambert and Allan, 1999, Wang et al., 2001, Wang et al., 2002, Dovciak and Perry, 2002, Fausch et al., 2002, Roy et al., 2003, Roy et al., 2007, Snyder et al., 2003). For example, Roth et al. (1996) examined the relationship among land use, land cover and instream biological conditions at various spatial scales in southeastern Michigan. While they found that watershed-scale land use is the strongest predictor of instream biological condition, a similar study in the same region by Lambert and Allan (1999) found that riparian-scale land cover characteristics explained more of the variation of instream biology. Although differences in study designs and changes in watershed conditions from one year to the next (i.e., development, removal of riparian land cover) may have influenced the results, both studies examined identical streams, used similar measures of land cover characteristics, as well as metrics for instream biological conditions (IBI). In another study, Roy et al. (2003) report that values of a macroinvertebrate IBI were strongly correlated with both catchment and riparian land cover over a range of 5–61% total urban area and 34–95% forest area in 100-m buffers. However, macroinvertebrate indices were more strongly correlated with environmental factors quantified at the reach-scale, including variation in substrate size and ion concentrations (Roy et al., 2003).

Several possible explanations may be offered to address the aforementioned variability of results when linking riparian and watershed conditions to instream biological measures. First, “landscape legacies” (Allan, 2004) may have a lasting impact on the condition of stream systems. Wang et al. (2001) found that values of IBI varied strongly along an urbanization gradient. This was interpreted as the legacy of similar habitat degradation at all sites under the common, prior influence of agriculture. Second, minor changes in land use may have an impact on the geomorphic features within a watershed and affect localized hydrological regimes across watersheds (Booth and Jackson, 1997, Trimble, 1997). Modifications in watershed hydrology impact instream biota through changes in the quantity and quality of water in the stream channel (Paul and Meyer, 2001). In addition, land use and land cover conditions are generally recognized as instrumental in affecting energy sources, habitat structures, chemical constituents, and biotic interactions within the stream system (Karr and Rossano, 2001), and have lasting consequences on instream conditions. Due to the mechanistic pathways through which these environmental characteristics influence aquatic biota, understanding the structure and dynamics of macroinvertebrate communities in streams, including their responses to human disturbance, can be enhanced by examining environmental and anthropogenic effects at multiple spatial and temporal scales (Peterjohn and Correll, 1984, Vannote et al., 1980, Allan and Johnson, 1997, Malmqvist, 2002, Weigel et al., 2003, Alberti, 2005).

Another possible explanation, and the topic explored in this study, is that previous research has largely relied on the composition of land cover in watersheds, such as percent of agricultural land and total amount of impervious or vegetated surface, disregarding the role of configuration or fragmentation of the land cover within the watershed (previous studies have examined separate scales – the riparian zone and whole watershed – but these studies do not explicitly consider the role of fragmentation of vegetation as it affects instream aquatic conditions). Watershed metrics that quantify spatial fragmentation of the land cover add important information to those that simply quantify landscape composition to help explain the variability in stream conditions (Alberti et al., 2007). Because configuration metrics are spatially explicit (Herzog and Lausch, 2001, Turner et al., 2001, Gergel et al., 2002), they account for the distributional effects of land uses or land covers on stream conditions. This is an important element since the distribution of land cover can be affected by land use planning activities, and spatial configuration may link more explicitly to management activities aimed at reducing the impact of urban development on aquatic systems.

This study explores the role of watershed vegetation patterns in explaining variations in aquatic conditions in the Puget Sound lowland. Vegetation patterns are defined as both the amount and the spatial distribution of vegetation cover in a watershed. By building upon previous studies examining the relationship between composition and configuration of vegetation and instream biological conditions, we examine this relationship across scales (riparian corridor and whole watershed). Specifically, we ask how vegetation patterns at the riparian and watershed scales help to explain instream biological conditions. To address this question we test two null hypotheses: (1) no significant relationship exists between riparian vegetation patterns and instream biological conditions that are not already explained by watershed vegetation patterns; and (2) fragmentation (a measure of configuration) of upland vegetation is not related to instream biological conditions. By explicitly describing the linkages between vegetation amount and distribution, and instream biological conditions, this research aims to expand scientific understanding of the role of vegetation in explaining instream biotic conditions. This study will also attempt to provide urban and regional watershed management agencies with tools for systematically describing and monitoring watershed conditions from the riparian to the watershed scales.

Section snippets

Study site selection

The Puget Sound region covers an area of more than 41,440 km2 and comprises about 80% land and 20% water (Lasmanis, 1991). Puget Sound is an estuary—a semi-enclosed, glacial fjord where saltwater from the ocean is mixed with fresh water draining from the surrounding watersheds. Fresh water inflow from rivers amounts to a yearly mean of 41,000 cubic feet per second (ft3/s), ranging between a peak of about 367,000 ft3/s to a minimum of about 14,000 ft3/s. Dense coniferous forests dominate the Puget

Results

Results of the Pearson’s correlation analysis provided the basis for examining the relationship between the amount and fragmentation of watershed and the amount and fragmentation of riparian vegetation conditions. The amount of watershed vegetation and the fragmentation of that vegetation was highly correlated (r = 0.88, P < 0.01), and the relationship between the amount of watershed and riparian vegetation, while weak, was significant (r = 0.28, P < 0.01). Because of the high correlation between these

Discussion

Our study emphasized the measurement of spatial vegetation patterns – as defined by amount and fragmentation of vegetation – across riparian and watershed scales. We can reject our first null hypothesis because the percent of riparian vegetation was significantly correlated to instream biological conditions, while controlling for amount of watershed vegetation. Alternatively, the distribution of riparian vegetation, as characterized here by riparian AI, did not remain significant when

Conclusions

A result of the urbanization process is the fragmentation of previously contiguous patches of vegetation (Paul and Meyer, 2001). Much of our understanding of the impact of vegetation fragmentation on ecosystems has focused on the demographic effects on bird and mammal populations (Ambuel and Temple, 1983, Freemark and Collins, 1992, Robinson et al., 1995, Tewksbury et al., 1998). By building on such research, this study examines the role of forest fragmentation on aquatic conditions.

Acknowledgements

The authors wish to thank several people that were instrumental in the design and implementation of this study, including Derek Booth, Jim Karr, Gordon Bradley, Stephan Coe, Daniele Spirindelli, Jeff Hepinstall, and Bekkah Coburn. This study was funded by the National Science Foundation’s Integrated Graduate Education and Research Traineeship (IGERT) program.

Vivek Shandas is an assistant professor in the Nohad A. Toulan School of Urban Studies and Planning and a research associate in the Center for Urban Studies at Portland State University.

References (103)

  • M. Alberti et al.

    The impact of urban patterns on aquatic ecosystems: an empirical analysis in Puget lowland sub-basins

    Landscape and Urban Planning

    (2007)
  • J.R. Dorney et al.

    Composition and structure of an urban woody plant community

    Urban Ecology

    (1984)
  • J. Linehan et al.

    Greenway planning: developing a landscape ecological network approach

    Landscape and Urban Planning

    (1995)
  • M. Alberti

    The effects of urban patterns on ecosystem function

    International Regional Science Review

    (2005)
  • M. Alberti et al.

    Urban land-cover change analysis in central Puget Sound

    Photogrammetric Engineering & Remote Sensing

    (2004)
  • J.D. Allan

    Landscapes and riverscapes: the influence of land use on river ecosystems

    Annual Reviews of Ecology, Evolution and Systematics

    (2004)
  • J.D. Allan et al.

    Catchment-scale analysis of aquatic ecosystems

    Freshwater Biology

    (1997)
  • J.D. Allan et al.

    The influence of catchment land use on stream integrity across multiple spatial scales

    Freshwater Biology

    (1997)
  • B. Ambuel et al.

    Area-dependent changes in bird communities and vegetation of southern Wisconsin forests

    Ecology

    (1983)
  • D.B. Booth et al.

    Urbanization of aquatic systems-degradation thresholds, stormwater detention, and limits of mitigation

    Journal of American Water Resources Association

    (1997)
  • D.B. Booth et al.

    Reviving urban streams: land use, hydrology, biology, and human behavior

    Journal of the American Water Resources Association

    (2004)
  • E.W. Boyer et al.

    Anthropogenic nitrogen sources and relationships to riverine nitrogen export in the northeastern U.S.A.

    Biogeochemistry

    (2002)
  • C.J.F. Braak et al.

    Data Analysis in Community and Landscape Ecology

    (1995)
  • F.S. Chapin et al.

    Consequences of changing biodiversity

    Nature

    (2000)
  • B. Clausen et al.

    Relationships between benthic biota and hydrological indices in New Zealand Streams

    Freshwater Biology

    (1997)
  • V.H. Dale et al.

    Effects of forest fragmentation on neotropical fauna

    Conservation Biology

    (1994)
  • P. Dutilleul et al.

    The Mantel test versus Pearson’s correlation analysis: assessment of the differences for biological and environmental studies

    Journal of Agricultural, Biological and Environmental Statistics

    (2000)
  • T.A. Dillaha et al.

    Vegetative filter strips for agricultural nonpoint source pollution control

    Transactions of the American Society of Agricultural Engineers

    (1989)
  • C.P. Doberstein et al.

    The effect of fixed-count sub-sampling on macroinvertebrate biomonitoring in small streams

    Freshwater Biology

    (2000)
  • M.J. Dole-Olivier et al.

    Response of invertebrates to lotic disturbance: is the hyporheic zone a patchy refugium?

    Freshwater Biology

    (1997)
  • A. Dovciak et al.

    In search of effective scales for stream management: does agroecoregion, watershed, or their intersection best explain the variance in stream macroinvertebrate communities?

    Environmental Management

    (2002)
  • K.D. Fausch et al.

    Landscapes to riverscapes: bridging the gap between research and conservation of stream fishes

    BioScience

    (2002)
  • J.A. Foley et al.

    Global consequences of land use

    Science

    (2005)
  • L.S. Fore et al.

    Assessing invertebrate responses to human activities: evaluating alternative approaches

    Journal of North American Benthological Society

    (1996)
  • R.T.T. Forman et al.

    Road Ecology: Science and Solutions

    (2002)
  • M.J. Fortin et al.

    How to test the significance of the relation between spatially autocorrelated data at the landscape scale: a case study using fire and forest maps

    Ecoscience

    (2002)
  • T.B. Francis et al.

    Incorporating science into the environmental policy process: a case study from Washington State

    Ecology and Society

    (2004)
  • K. Freemark et al.

    Landscape ecology of birds breeding in temperate forest fragments

  • S.E. Gergel et al.

    Landscape indicators of human impacts to riverine systems

    Aquatic Sciences

    (2002)
  • J.A. Gore

    Responses of aquatic biota to hydrological change

  • H.S. He et al.

    An aggregation index (AI) to quantify spatial patterns on landscapes

    Landscape Ecology

    (2000)
  • F. Herzog et al.

    Supplementing land-use statistics with landscape metrics: some methodological considerations

    Environmental Monitoring and Assessment

    (2001)
  • W.C. Hession et al.

    Influence of bank vegetation on channel morphology in rural and urban watersheds

    Geology

    (2003)
  • J. Jaeger

    Landscape division, splitting index, and effective mesh size: new measures of landscape fragmentation

    Landscape Ecology

    (2000)
  • E.J. Kaiser et al.

    Urban Land Use Planning

    (1995)
  • J.R. Karr

    Biological integrity: a long-neglected aspect of water resource management

    Ecological Applications

    (1991)
  • J.R. Karr

    Defining and measuring river health

    Freshwater Biology

    (1999)
  • J.R. Karr et al.

    Restoring Life in Running Waters: Better Biological Monitoring

    (1999)
  • Karr, J.R., Fausch, K.D., Angermeirer, P.L., Yant, P.R., Schlosser, I.J., 1986. Assessing Biological Integrity in...
  • J.R. Karr et al.

    Applying public health lessons to protect river health

    Ecology and Civil Engineering

    (2001)
  • B.L. Kearns et al.

    A Benthic Index of Biotic Integrity (B-IBI) for rivers of the Tennessee Valley

    Ecological Applications

    (1994)
  • King County, 2000. King County Spatial Data Warehouse (KCSDW), high resolution satellite image, EMERGE dataset....
  • R.S. King et al.

    Spatial considerations for linking watershed land cover to ecological indicators in streams

    Ecological Applications

    (2005)
  • Kleindl, W.J., 1995. A benthic index of biotic integrity for Puget Sound lowland streams, Washington, USA. MS Thesis....
  • R.D. Klien

    Urbanization and stream quality impairment

    Journal of the American Water Resources Association

    (1979)
  • Konrad, C.P., Booth, D.B., 2002. Hydrologic Trends Resulting from Urban Development in Western Washington Streams. U.S....
  • M. Lambert et al.

    Assessing biotic integrity of streams: effects of scale in measuring the influence of land use/cover and habitat structure on fish and macroinvertebrates

    Environmental Management

    (1999)
  • R. Lasmanis

    The geology of Washington

    Rocks & Minerals

    (1991)
  • P. Legendre et al.

    Numerical Ecology

    (1998)
  • Cited by (34)

    • What physical habitat factors determine the distribution of gastropods in neotropical headwater streams?

      2022, Water Biology and Security
      Citation Excerpt :

      Lowland streams are easily accessible to humans and more often located near urban centers or agricultural areas. This results in greater occurrence of anthropogenic stressors, including altered basin land use and riparian vegetation, eutrophication, damming, channelization, and introduced invasive species (Graf et al., 2016; Horsák et al., 2009; Shandas and Alberti, 2009). The positive correlation of Melanoides tuberculata with stable hard substrate corroborates previous studies in South American reservoirs (Karatayev et al., 2007; Linares et al., 2020), where high densities of individuals of this species were found in sites with hard substrates.

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

      2016, Ecological Indicators
      Citation Excerpt :

      Furthermore, presence of natural cover is evidence of greater habitat complexity and, consequently, of higher biological condition scores (Sponseller et al., 2001; Hrodey et al., 2009). Others have reported high correlations between MMI scores and catchment land use and cover for forested areas (Sponseller et al., 2001; Shandas and Alberti, 2009), agricultural areas (Wang et al., 1997; Hrodey et al., 2009), and impervious (urban) areas (Morley and Karr, 2002; Wang et al., 2003). At the local scale, MMI-1 responded largely to physical habitat variables (Table 4).

    • Understanding the dynamic of greenspace in the urbanized area of Beijing based on high resolution satellite images

      2015, Urban Forestry and Urban Greening
      Citation Excerpt :

      Consequently, local residents suffer less from diseases caused by high temperature (Kenney et al., 2014; Sun et al., 2014), and may also consume less energy to maintain a comfortable indoor temperature. The increase in greenspace and change in spatial configuration may also have significant impacts on urban biodiversity (Kong et al., 2010), stream biotic conditions (Shandas and Alberti, 2009), and air quality (Nowak et al., 2006; Nowak et al., 2014). The change analysis that included the gain and loss of greenspace revealed that the actual changes in percent cover of greenspace were much larger than the net increase, since the lost greenspace was compensated by gained greenspace.

    • Predicting biological condition in southern California streams

      2012, Landscape and Urban Planning
      Citation Excerpt :

      Identification and understanding of stressor–response relations have been of long-term interest to aquatic ecologists studying stream ecosystems (e.g., Allan, 2004); however, consideration of large scale land use and land cover was difficult until the results of remote sensing became widely available. Remote sensing and computer-based geographic information systems (GISs) have greatly increased our ability to include considerations of land use in assessments of biological condition (Alberti et al., 2007; Falcone et al., 2010; Lee, Hwang, Lee, Hwang, & Sung, 2009; Shandas & Alberti, 2009; Turak et al., 2011). However, the use of GIS data involves a major assumption regarding the nature of the stressor–response relation.

    • Assessing riparian vegetation structure and the influence of land use using landscape metrics and geostatistical tools

      2011, Landscape and Urban Planning
      Citation Excerpt :

      In addition, we compared the expected and the observed responses of the landscape metrics to land use. Bibliographic sources, such as Aguiar et al. (2000), Aguiar and Ferreira (2005), Guirado et al. (2007), Schuft et al. (1999), Shandas and Alberti (2009), Timm et al. (2004), Wu et al. (2000), and expert judgment were used to suggest the behaviour, positive or negative relationships, of the landscape metrics influenced by land use. Stretches of the Margem and the Chouto and the upstream section of the Sôr are of medium valley width.

    • Ecological engineering in a new town development: Drainage design in The Woodlands, Texas

      2010, Ecological Engineering
      Citation Excerpt :

      This indicates the watershed is experiencing a wet period before this rainfall event. In land use planning, three methods are generally used to capture the impervious surface area of development: (1) use parcel data to quantify the impervious area (Alley and Veenhuis, 1983; Rogers and DeFee, 2005), (2) classify Landsat remote sensing imagery to extract the impervious area (Alberti et al., 2007; Shandas and Alberti, 2009), and (3) digitize high-resolution aerial photographs to delineate the impervious area (Light, 1993; Jennings and Jarnagin, 2002). This study used the first method to calculate the impervious area from 1972 to 2002 using the Geographic Information System (GIS).

    View all citing articles on Scopus

    Vivek Shandas is an assistant professor in the Nohad A. Toulan School of Urban Studies and Planning and a research associate in the Center for Urban Studies at Portland State University.

    Marina Alberti is an associate professor of Urban Design and Planning and the director of the Urban Ecology Research Lab at the University of Washington.

    View full text