3
Effects of Roads on Ecological Conditions

INTRODUCTION

Widespread attention continues to be drawn to the ecological effects of roads, especially as the road system continues to expand. Roads are created because of changing interactions between people and their environments. They are created to facilitate access to natural resources, to connect human communities, to move goods to markets, and to move people to work. Whatever their purpose, roads, road establishment, road maintenance, and road travel have a broad variety of effects.

In this chapter, the committee addresses the following two questions as stated in its charge:

  1. What are the appropriate spatial scales for different ecological processes that might be affected by roads?

  2. What are the effects of road density on ecosystem structure and functioning and on the provision of ecosystem goods and services?

Roads have effects that can vary with a range of spatial scales. The committee’s analysis examines what is known about road effects at three scales, which are discussed later in the chapter.

For the second question, the committee used the phrase “ecological condition” for “ecosystem structure and functioning.” Because most information on road effects is given in terms of ecological structure and



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 62
Assessing and Managing the Ecological Impacts of Paved Roads 3 Effects of Roads on Ecological Conditions INTRODUCTION Widespread attention continues to be drawn to the ecological effects of roads, especially as the road system continues to expand. Roads are created because of changing interactions between people and their environments. They are created to facilitate access to natural resources, to connect human communities, to move goods to markets, and to move people to work. Whatever their purpose, roads, road establishment, road maintenance, and road travel have a broad variety of effects. In this chapter, the committee addresses the following two questions as stated in its charge: What are the appropriate spatial scales for different ecological processes that might be affected by roads? What are the effects of road density on ecosystem structure and functioning and on the provision of ecosystem goods and services? Roads have effects that can vary with a range of spatial scales. The committee’s analysis examines what is known about road effects at three scales, which are discussed later in the chapter. For the second question, the committee used the phrase “ecological condition” for “ecosystem structure and functioning.” Because most information on road effects is given in terms of ecological structure and

OCR for page 62
Assessing and Managing the Ecological Impacts of Paved Roads functioning rather than ecosystem goods and services, ecological condition is used. These terms are defined in the next section. This chapter is organized into five sections to summarize the interaction between roads and ecological conditions. After this introduction, terms and concepts are defined in the second section. A short summary of effects on ecological goods and services is provided in the third section. The fourth section summarizes published, mostly refereed literature and is organized according to two dimensions: the first is the ecological process of interest (many of which are listed in Table 3-1) and includes the effects of each process on the different levels of ecological organization (for example, abiotic, population, species, and ecosystem), and the second is the scale of the effect. Many of the effects of roads on the environment are caused by other forms of human activity and land use. Impacts of agriculture, urbanization, forest practices, and manufacturing are in many ways similar and sometimes interrelated with the impacts of roads. Information gaps are discussed in the fifth section. DEFINITIONS Ecological condition is a general term that describes the structure and functioning of ecosystems. It may refer to the status of the ecological environment at a particular time or to dynamic changes in its components and processes over time. The dynamic aspects of these condition measurements are discussed later in the chapter where both spatial and temporal dimensions are addressed. Ecosystems encompass all living organisms (biotic components) plus the nonliving environments (abiotic components) with which they interact. The abiotic components consist of hydrological and geomorphological processes, chemicals, and such disturbances as landslides, climate and weather. Levels of organization of biotic components used in this report are genetics, species and population (plants and animals), and ecosystem. Each level of biotic components has attributes of composition, structure, and functioning, and together constitute biological diversity (often called “biodiversity”). Composition refers to the identity and variety of elements in each of the biodiversity components. Structure refers to the physical organization or pattern of the elements. Ecological (or ecosystem) functioning refers to the ecological and evolutionary processes acting among the elements, or how the ecosystem works.

OCR for page 62
Assessing and Managing the Ecological Impacts of Paved Roads TABLE 3-1 Comparison of Ecosystem Goods and Servicesa and Ecosystem Structures and Processes Affected by Roads Change Due to Roads Consequence Affected Ecosystem Good Affected Ecosystem Service Chemical input from roads to water bodies Degradation of water quality, bioaccumulation Clean water Water purification, pollution abatement Chemical inputs to airshed Degradation of air quality Clean air Pollution abatement Chemical input to soils Bioaccumulation Soil fertility Pollution abatement Climate Increased temperature and rainfall Water Climate stability Hydrological processes Fluvial dynamics, sediment transport, floodplain ecology NA Flood and drought mitigation, nutrient cycling Modified habitat Plant species composition (natives and nonnatives) Biodiversity Nutrient cycling, soil fertility, seed dispersal Habitat quality, wildlife mortality Density and composition of animal species and populations Biodiversity Crop pollination, aesthetics, ecotourism aEcosystem goods and services are defined by Daily (1997). Abbreviation: NA, not available.

OCR for page 62
Assessing and Managing the Ecological Impacts of Paved Roads Diversity of the genetic component refers to the variation in genes within a particular species, subspecies, or population. A relevant measure of the genetic component is allelic diversity, a measure of its structure is heterozygosity, and a measure of its functioning is gene flow. Diversity of the population and species component refers to the variety of living species and their populations at the local, regional, or global scale. A relevant measure of this component is species abundance, a measure of its structure is population age, and a measure of its functioning is demographic processes, such as births and deaths. Diversity of the ecosystem component refers to the number of species, biotic communities, and ecosystem types and to the genetic variation in the organisms present. Relevant measures of this component include a variety of measures of species diversity (see NRC 2001), including the ratio of native to nonnative species; various measures of genetic diversity (see Hedrick 2004); and variations in trophic activities and structure, such as food-chain length and feeding adaptations. ECOLOGICAL CONDITION—ECOSYSTEM GOODS AND SERVICES Ecological condition incorporates the concepts of ecosystem goods and services through the ecosystem component. Ecosystem goods are the materials and elements (for example, water, food, fiber, and fuel) that are products of ecosystems and used for a variety of human needs. Ecosystem services are the benefits that people obtain from ecosystems. Those include goods, such as food and water; services, such as regulation of floods, droughts, land degradation, and disease; supporting services, such as soil formation and nutrient cycling; and cultural goods, such as recreational, spiritual, religious, and other nonmaterial benefits (Millennium Ecosystem Assessment 2003). An enumeration of road effects on ecosystem goods and services is marginally addressed in this report. The effects of roads on those ecological structures and processes that are of direct and indirect use to humans are, however, discussed in detail. Ecosystem structure and functioning can be translated into ecosystem goods and services, as described in subsequent sections. Roads affect ecosystem goods and services in many ways. The documented road-associated changes to be discussed can be translated into equivalent alterations in ecosystem goods and services (Table 3-1).

OCR for page 62
Assessing and Managing the Ecological Impacts of Paved Roads Among the changes are altered local climatic conditions, altered nutrient cycling, loss of flood and drought mitigation, loss of soil fertility, and changes in biodiversity. Ecological Conditions and Scale Roads interact with ecosystems across a wide range of scales. For example, at small scales, heavy metal molecules accumulate in soils adjacent to roads. At intermediate scales, roads disrupt soil structures and hydrological pathways and alter plant and animal communities. At large scales (regions to nation), roads alter migration patterns and increase spread of exotic organisms. Many effects can occur at more than one spatial scale (for example, effects on migration patterns). The literature review in the next section documents effects of roads on ecological conditions at three scales. The smallest scale assessed is the road segment. This scale generally extends to a hundred meters (see Figures 1-1, 1-2, and 1-3). The intermediate scale is identified as a system of a geographic region—defined either politically (for example, a state or province) or ecologically (for example, a watershed or eco-region). This scale extends from one to tens of kilometers (see Figures 1-4 and 1-5). The largest scale is the macroscale and is defined by regional ecological units (for example, eco-regions) (Bailey et al. 1994) to national political boundaries. This scale extends from hundreds to thousands of kilometers (see Figures 1-6 and 1-7). LITERATURE REVIEW The committee developed an annotated bibliography (Appendix B) of road effects on ecological conditions, with an emphasis on spatial scale. The review included only studies that directly measured the effects of roads on the surrounding environment. Effects were categorized as either abiotic or biotic. Abiotic effects included the effects on hydrogeomorphic process, the effects of road-related chemicals on water and air quality, and the effects of other disturbances, such as landslides, local climate, and lighting. The three subcategories of biotic effects were genetic, species and population, and ecosystem. Within each subcategory, the effects of roads on structure, functioning, and composition were documented.

OCR for page 62
Assessing and Managing the Ecological Impacts of Paved Roads Every aspect of roads (as with many human activities) has some interaction with the surrounding environment, including road construction, operation, and maintenance. However, the committee’s review focuses on operation—that is, the effects of roads and their structures (for example, culverts) and vehicles that use them. The literature review and synthesis provides an overview of the current understanding, trends, and information gaps relating to the effects of roads and traffic on ecological conditions and the spatial scales at which roads affect ecological conditions. The available information at different scales of ecological effects is also examined. Approach The committee collected and reviewed over 500 journal articles and conference proceedings. The literature was obtained primarily from scientific journals, although some reports were obtained from the grey literature. The list of studies was not meant to be exhaustive but nonetheless captured the majority of accessible literature. The focal area of the review was North America, but research findings from Europe and Australia were also included. Table 3-2 summarizes the available bibliographic information. For each of the ecological conditions and spatial scales, the committee qualitatively categorized the number of studies as none, few, several, or many. For some subcategory and scale assessments, little information was obtained; however, there was a general consensus among committee members that more information was available but not in formats readily accessible. The findings of the studies are summarized in the following sections. Ecological Significance of Road Attributes The presence or absence of roads is not the only factor that governs impacts on the surrounding environment. A major impact of roads is related to their use—that is, traffic. The density of the road network, the volume of traffic on a roadway or road segment, the road surface, and other engineered features also affect the extent of ecological effects of a road. This section briefly discusses the direct individual ecological impacts of various road attributes.

OCR for page 62
Assessing and Managing the Ecological Impacts of Paved Roads TABLE 3-2 Summary of Number of Studies Addressing Different Types of Road Effects on Ecological Conditions Ecological Condition Single-Segment Scale Intermediate Scale National or Regional Scale ABIOTIC Hydrology/Geomorphology Stream networks Fewa Many None Sediment production Fewa Many None Changes in waterflow Fewa Fewa None Chemical Characteristics Mineral nutrients Many Few None Heavy metals Many None None Organic (water and sediment) Few None None De-icing salt Many Few None Volatile organic carbons (air) Fewa None None Other Disturbances Landslides Fewa Few None Light Few None None BIOTIC Genetic Structure Barrier to movement Few Few None Functioning Isolated populations Few None None Composition Filtering effect Few None None Species/Populations Structure

OCR for page 62
Assessing and Managing the Ecological Impacts of Paved Roads Wildlife population structure Several Few None Functioning Additional habitat Several Few Few Reduced habitat quality Many Few Few Dispersal corridor Several Several None Movement barrier Several Few Few Distribution Several Several Few Composition Species richness Several Several None Road mortality Many Several Few Nonnative plants in roadsides and adjacent landscapes Many Several Nonea Ecosystem Structure Fragmentation Few Few None Functioning Pollutants Several None None Composition Environmental characteristics Several Few None aCategories thought to have more information available but not in readily accessible formats.

OCR for page 62
Assessing and Managing the Ecological Impacts of Paved Roads Road and Traffic Density The density of the road network, the volume of traffic on a road, the road’s location, topography, and other factors have major roles in the intensity of associated environmental effects of roads. A few studies have correlated the density of the road network to their environmental effects (Findlay and Houlahan 1997, Carr and Fahrig 2001). Reduced construction of new roads reduces habitat fragmentation, suggesting that, in general, less habitat fragmentation occurs in a less-dense road network with high traffic volumes than in a dense network with low traffic volumes per road mile. The direct effects of traffic density, defined in this report as the number of vehicle miles traveled on a given stretch of roadway in a given time, have been more widely studied. In general, an increase in traffic density correlates with an increase in the atmospheric deposition and the aquatic concentration of vehicle-emitted chemicals, such as heavy metals (Bocca et al. 2003, Fakayode and Olu-Owolabi 2003), particulate matter (Boudet et al. 2000), and organic pollutants (Ellis et al. 1997, Forman and Alexander 1998, Viskari et al. 2000, Ilgen et al. 2001, Latha and Badarinath 2003). Only areas with road density less than 0.72 km/km2 (1.16 mi/mi2) seem to support vibrant populations of wolves (Canis lupus) in Minnesota (Mech, et al. 1988, Fuller 1989), Wisconsin (Thiel 1985, Mladenoff et al. 1999), the western part of the Great Lakes region of the United States (Mladenoff et al. 1995), and Ontario (Canada) (Jensen et al. 1986). An exception to the trend is an established wolf population in a fragmented area of Minnesota with a road density of 1.42 km/km2 (2.29 mi/mi2) (Merrill 2000). Increased traffic density also has been shown to reduce amphibian population (Fahrig et al. 1995, Carr and Fahrig 2001). Road Surfaces Because construction of asphalt concrete and hydraulic cement concrete road surfaces involves many of the same techniques, the differences in the direct effects of roadway construction using these materials are minimal. Well-constructed asphalt concrete and hydraulic cement concrete pavements are impervious; therefore, both are likely to exhibit similar runoff. Work completed in National Cooperative Highway Re-

OCR for page 62
Assessing and Managing the Ecological Impacts of Paved Roads search Program Project 25-09 and reported in 2001 in National Cooperation Highway Research Program (NCHRP) Report 448 (Nelson et al. 2001) concluded that most materials, including asphalt concrete and hydraulic cement concrete, used in the construction and repair of highways “behave in a benign fashion in the environment. On the highway surface, leaching is slow, transport is rapid, and dilution is great….” Further, the results of Project 25-09 show no significant practical differences in the potential impact of runoff from asphalt concrete and hydraulic cement concrete pavement surfaces. Engineering Structures The impact of engineering structures is generally consistent with the functioning of the structure itself. Concrete barriers, right-of-way fences, noise barriers, and perhaps to a lesser extent, guardrails are designed to serve as barriers for people and noise, but they also function as barriers to flora and fauna. These barriers may result in habitat fragmentation and species isolation (Forman and Alexander 1998). Wildlife underpasses and overpasses, long-span bridges, and culverts can help to mitigate the adverse impacts of habitat fragmentation. Examples of the use of these structures for ecological improvements are identified in Chapter 4 of this report. Poorly designed engineering structures can often hinder the ecological improvements for which they were designed. In an aquatic culvert system, for example, several key design characteristics ensure effective utilization by the target species (see Box 3-1 under Biotic Consequences). Abiotic Consequences Abiotic conditions that can be influenced by roads include hydrological, geomorphological, and chemical characteristics and such disturbances as landslides, noise, and light. In this section, the committee considers only changes to the abiotic conditions themselves, and examples of each are provided below. How these abiotic changes affect the biota is considered in later sections.

OCR for page 62
Assessing and Managing the Ecological Impacts of Paved Roads Hydrological and Geomorphological Changes Landscape changes result when roads alter the hydrological and geomorphological aspects of watersheds and landscapes. They can cause important changes (some for short periods, others for longer periods) in fluvial dynamics, sediment production, and chemical balances, which can adversely affect floodplain functioning and alter ecological conditions in aquatic and riparian areas (Figure 3-1). Roads also affect water movements, sedimentation, and transport of pollutants. Because they often interrupt or otherwise alter sheet flow and FIGURE 3-1 Road affecting four aspects of stream connectivity. (a) Upstream-downstream (1), floodplain-stream (2), forest-stream (3), and surface-subsurface water connections (4). (b) The connections severed or disrupted by a road in the floodplain. Source: Forman et al. 2003. Reprinted with permission; copyright 2003, Island Press.

OCR for page 62
Assessing and Managing the Ecological Impacts of Paved Roads example, of 43 species of woodland breeding birds, 26 species (60%) showed reduced densities near highways (Reijnen and Foppen 1994). Traffic noise explained the most variation in bird density in relation to roads in a regression model. This effect also occurred for grassland birds (Reijnen et al. 1996) and is more important in years with a low overall population size (Reijnen and Foppen 1995). An analysis of the total effect of The Netherlands’s most dense network of main roads on “meadow birds” showed a possible population decrease of 16%, attributable to reduced habitat quality and traffic noise (Reijnen et al. 1997). It is not known how well wildlife can acclimate to constant noise, for example, along a roadside, and acclimation no doubt varies among organisms. Furthermore, breeding activities of species, such as birds and amphibians that rely on vocalization, may be particularly susceptible to disruption by noisy conditions. Existing noise barriers in suburban settings are designed to protect humans from noise. As such, they can serve as barriers to animal movement. Generally, noise impacts decline with distance to the road. Range of Occurrence of Effects Roads interact with plants, animals, water, sediment, and other ecological attributes in ways that extend beyond the road edge (Table 3-3). The distance from a road that ecological effects can be detected is called the “road-effect zone” (Forman et al. 2003) or “zone of influence” (NRC 2003). The effect of distance varies, depending on the organism, location, and disturbance type, and generally increases with traffic volume (Clark and Karr 1979, Reijnen and Foppen 1994, Nellemann et al. 2001). Aquatic environments and organisms are highly sensitive to roads and traffic densities (Eaglin and Hubert 1993, Vos and Chardon 1998, Turtle 2000). For example, wetland species diversity is negatively correlated with paved-road density up to 2 km from wetlands (Findlay and Houlahan 1997). The effects of roads on wetland diversity take about 3-4 decades to be fully realized (Findlay and Bourdages 2000). Road-effect zones occur due to disturbances from high-volume traffic, which can reduce the habitat quality near roads (Table 3-3). Breeding densities and distribution of many bird species are reduced adjacent to busy roads. Animals avoid roads by a distance that increases with increasing traffic volumes. This road-avoidance zone contributes to the road-effect zone. Similar distance effects of roads occur with chemical

OCR for page 62
Assessing and Managing the Ecological Impacts of Paved Roads TABLE 3-3 Examples of the Extent to Which Road-Induced Effects Penetrate Adjacent Habitat Road Effect Distance Reference Heavy metals In soils and plants near roads 50-100 m Ministry of Transport, Netherlands 1994 Chemical pollution Oxides of nitrogen changing plant communities 200 m Angold 1997 Animal distribution Pink-footed geese (Anser brachyrhunchus) and graylag geese (A. anser) 100 m Keller 1991 150 m Ortega and Capen 1999 Territory size of ovenbirds (Seiurus aurocapillus)     Traffic noise 200-1,200 m Van der Zande et al. 1980 Breeding bird density 40-1,500 m Reijnen et al. 1996 Road lights Breeding bird density 200-250 m De Molenaar et al. 2000 Avoidance zone Caribou (Rangifer tarandus) 5,000 m 200 m Nellemann and Cameron 1998 Rost and Bailey 1979 Deer (Odocoileus hemionus) and elk (Cervus elaphus) 1,000 m   Grizzly bears (Ursus arctos) and black bears (U. americanus)   Kasworm and Manley 1990 Increase in edge species Component of bird community 100 m Ferris 1979 Road density Wetlands species richness 2,000 m Findlay and Houlahan Moor frog (Rana arvalis) presence 750 m 1997 Leopard frog (R. pipiens) distribution 1,500 m Vos and Chardon 1998, Carr and Fahrig 2001 Early melting of permafrost 100 m Walker et al. 1987 pollution, nonnative plant species, and other wildlife species’ distributions. The road-effect zone is reduced on low-volume roads. Scale of Effects Most of the literature reviewed by the committee focused on the effects of roads on species and populations of wildlife and plants. The

OCR for page 62
Assessing and Managing the Ecological Impacts of Paved Roads need to assess project-level effects of road building and expansion on species and their populations as part of policy guidelines (the National Environmental Policy Act and the Endangered Species Act) has been the catalyst for most single-segment and intermediate-scale studies (Evink 2002). Although current research is making valuable contributions, its ultimate impact is limited by low funding, inadequate coordination across research entities, and short-term or project-specific focus (TRB 2002a). Species and populations have been relatively well researched at the intermediate scale, particularly with respect to the effect of roads on species composition. Little information has been reported at the national scale on species richness, nonnative plants at roadsides, and road-related mortality of wildlife. These effects, although varied, are widespread geographically. Thus, a synthesis of available information or metaanalysis could provide valuable insight into the extent of effects at a broader scale than is known today. The committee found that very little research covers long periods, and almost no research has addressed large spatial scales of road effects. Few studies were found that describe the effects of roads on ecosystems, and most of those were carried out at the single-segment scale. Investigations of the effects at larger scales are scarce. A national-scale assessment estimated that one-fifth of the U.S. land area is directly affected ecologically by public roads (Forman 2000), even though the paved road network covers less than 1% of the U.S. land area. With increasing availability of digital biophysical and land-use data, geographic information system (GIS) tools and applications are becoming widely used among resource managers and transportation planners for amassing information and modeling the potential effects of roads at multiple scales (Dale et al. 1994, Tinker et al. 1998, Vos and Chardon 1998, Clevenger et al. 2002a). The increased use of GIS will probably facilitate more GIS-based studies that evaluate the ecological effects of roads at regional and national scales. Ecologists have long conducted studies on species diversity patterns at broad spatial and temporal scales (Brown and Lomolino 1998). Yet, attempts to understand how geographical and environmental features structure genetic variation at the population and the individual levels are new (Manel et al. 2003). These approaches focus on processes at fine spatial and temporal scales by detecting genetic discontinuities and correlating these with such environmental features as barriers, including highways (Gerlach and Musolf 2000, Conrey and Mills 2001, Thompson 2003). The new genetic approaches and techniques combined with in

OCR for page 62
Assessing and Managing the Ecological Impacts of Paved Roads creasing interest in how highways affect population viability is likely to result in more research in the coming years. The committee selected several of the most common ecological effects of roads and plotted the extent of their effects with respect to spatial and temporal scales (Figure 3-5). The abiotic consequences of altered water flow and sediment deposition are relatively fast-acting and essentially limited to single-segment and watershed scales. However, the impact of chemical pollutants, both organics and heavy metals, can be long lived (such as contaminated drinking-water sources) and far reaching (such as atmospheric deposition) due to sediment and particulate transport, sediment accumulation, and bioaccumulation. For the biotic component, reduced genetic structure related to barrier effects on animal movement probably would be manifested over a longer period (months and decades) and have effects at a broader spatial scale (watershed and eco-regions) than most abiotic effects. Proliferation of invasive exotic plants and landscape fragmentation due to roads occur over longer periods and are pervasive, affecting entire nations and continents as well as smaller-scale areas. FIGURE 3-5 Spatial and temporal dimensions of ecological effects of roads.

OCR for page 62
Assessing and Managing the Ecological Impacts of Paved Roads Figure 3-5 reveals some of the spatial and temporal scales at which these effects most often occur. (See Figure 6-4 for another plot of spatial scales by the U.S. Department of Defense.) The figure suggests that ecological effects of roads can alter ecological processes over scales that range from minutes to centuries (time) and from meters to thousands of kilometers (in space). That is, ecological effects occur at scales that extend much longer in time and broader in space than those scales currently being used in assessment, planning or management. This issue is addressed in subsequent chapters. The figure also suggests that there is no “correct” scale for understanding how roads interact with ecosystems, but rather multiple, overlapping ranges of scales that correspond to specific ecological structures and processes under consideration. Figure 3-5 also indicates that ecological processes occur at particular scale ranges, and hence assessments conducted at different scales could miss some ecological effects of roads. An example was discussed in a recent National Research Council report (NRC 2003). In that case, (unpaved) roads and the traffic on them affected the movements of caribou, especially of females. As a result, the caribou were more likely to encounter insects in years favorable to insects, and the interaction between roads and insects resulted in a subtle but measurable reduction in carbon productivity in those years. Local assessments did in fact fail to identify that effect; the assessment that was required extended over a far greater area than the direct road-effect zone for caribou. Other cases might include birds that migrate between wintering and breeding areas over many thousands of kilometers. A road that affects a resting area on a migration route might affect the entire population of the species, but conducting the assessment at a scale of hundreds of meters—or even at the scale of a county—could easily miss that population effect. A very similar example would be a road that crossed a stream that provided habitat for a migratory species of fish (see for example Box 3-1 and the entire discussion in the section “Roads as Barriers” in this chapter). An assessment at the scale of the barrier would most likely miss any population effect. Cross-Scale Effects This section indicates that ecosystems are scale variant; that is, the cross-scale biological structures and processes cannot be easily aggregated from one scale to another but are dependent on the scale of focus.

OCR for page 62
Assessing and Managing the Ecological Impacts of Paved Roads Scale variance is in contrast to many physical processes, such as waterflow dynamics and large-scale fire patterns that are scale invariant (Gunderson and Snyder 1994). Scale invariance means that structures and processes are self-similar as the scales change. A broad set of human-induced ecological changes, such as forest-pest outbreaks, algae blooms, salinization, and grassland to shrubland conversion, are all examples of scale-variant phenomena. The property of scale variance has great implications for the ability to assess and manage across a wide range of scales. The major conceptual framework for understanding ecological cross-scale structure and dynamics is hierarchy theory. Ecologists (Allen and Starr 1982, O'Neill et al. 1986, Allen and Hoekstra 1992) built on the seminal work of Simon (1962) to develop a theoretical base that emphasizes a pattern of aggregations (hierarchical levels, or “holons”) nearly separable across scales. Hierarchical levels can be identified by a stronger set of interactions within a hierarchical level than among levels. These hierarchical levels correlate to scales, and each level has characteristic spatial and temporal domains; that is, each level (see leaf, tree, or forest level in Figure 1-9) has a characteristic turnover time and spatial domain. Hierarchical levels can be identified in ecological systems, but how do they interact? The nature of ecological interactions at various scales has been the subject of much scientific debate. The focus has been on the interaction between processes and their associated structures that operate for long periods and over large spatial scales and processes that are faster and smaller. They also have been cast as “top-down” versus “bottom-up” control. An example of top-down control, sometimes called “hierarchical control,” is how altered soil types and microclimates associated with road rights-of-way determine the suite of plant and animal species that thrive. Because roads affect variables that change slowly, such as geological formations, soil composition, and topography, they often produce top-down effects. However, top-down and bottom-up effects often occur together and interact; this complex and dynamic set of ecological interactions has been called “panarchy” by Gunderson and Holling (2002). Examples of such interactions include disturbance dynamics, such as forest fires or forest-pest outbreaks (Gunderson and Holling 2002). Peterson (2002) demonstrated how a road network can disrupt spatial patterns and succession in southern U.S. forests. Other types of surprising ecological behavior appear to arise from such panarchical interactions.

OCR for page 62
Assessing and Managing the Ecological Impacts of Paved Roads Scales of the U.S. Road System and Ecological Effects The physical structure of the U.S. road system covers few thousands of kilometers. The replacement time of roads is on the order of multiple decades, although this value can vary as a function of road type. The presence of roads across the landscape generates a variety of effects and interactions with ecological systems. The effects fall into three categories: (1) effects that are fixed in scale; that is, the domain of the change is fixed in space and time with sharp boundaries; (2) effects that generate or initiate cross-scale interactions; and (3) effects that constrain or limit cross-scale interactions. In this context, cross-scale interactions are those that traverse hierarchical levels. Each of the categories is described in the following paragraphs. Many ecological effects of roads are spatially small. Most of the documented effects occur at the road-segment level, which includes the road, roadside, and a zone described as the effective road-impact zone (Forman et al. 2003). Generally, the zone ranges from a few meters to a few kilometers, depending upon the type of impact. Many effects are confined to the road and shoulder zone. Altered physical and chemical soil conditions from construction, management (fertilization or salt applications), or vehicle exhausts are found primarily in a narrow zone around roads. Some of the effects on biota, such as changes to populations (increased mortality) or community composition, occur primarily within this zone or within an area of a few hundred meters perpendicular to the road segment. Some road effects cross scales of space and time. These effects include the abiotic and the biotic components of ecological systems. One example is the set of effects on hydrological systems, where sediments, nutrients, and heavy metals are introduced into riparian systems. Changes in inputs have created shifts in biogeochemical cycles, resulting in changes in species distribution and abundance. Another cross-scale effect occurs when roads serve as ecological corridors and increase dispersion. One example attributed to roads is the spread of exotic organisms, both plants, such as kudzu, and animals, such as armadillos (Taulman and Robbins 1996). Many ecological effects of roads eventually influence structures and processes at longer and broader scales than first imagined. The spread of kudzu (or any other invasive organism) is a good example of an unintended, broader scale effect. Originally intended as an ornamental vegetation cover for stabilizing steep roadsides, the vine has spread

OCR for page 62
Assessing and Managing the Ecological Impacts of Paved Roads across much of the southeast, invading areas never imagined in original assessments. Similar arguments could be made for the nutrients, such as nitrogen from automobile exhausts, that have been observed to spread via atmospheric transport and accumulate in wetland and estuarine areas. Cumulative effects have been described in two ways. One type of cumulative effect is the cross-scale effect described in the previous paragraph. These effects accumulate over time, space, or both. They can manifest as a cumulative change in an ecosystem structure or function, such as the increase in the amount of heavy metals in soils adjacent to roadways or the increase in species or populations, such as scavengers that eat organisms killed on the roadway. Spatial accretions occur as structures increase in distribution, such as the spread of kudzu. The other type of cumulative effect involves synergistic interaction among key structures or functions associated with a road. For example, caribou migration in Alaska was differentially affected by the combination of roads and oil pipelines (NRC 2003). The implications of the latter type of effect for assessment and environmental review are discussed in the next section on cumulative effects. The final set of effects occurs when roads decrease the scale of ecological structures or processes. Often, they are barriers to landscape-scale phenomena. The restriction of wildlife migration or dispersion has long been recognized as a road effect. Fragmentation of populations due to roads is another such effect. Broad-scale disturbances, such as fire, that are critical to many types of ecosystems (prairies in the Midwest and pine forests in the eastern and western United States) are limited in spatial extent by roads that act as fire breaks. In some fire-adapted ecosystems where fire is heavily managed, rules of smoke management restrict when and how prescribed fires may be set because of the need to prevent visual hazards on roads caused by the smoke. Shifts in the timing and extent of fires can generate large-scale changes in ecosystems. Cumulative Effects Even though the awareness of the ecological effects of roads has grown steadily over the past few decades, only a small body of literature addresses cumulative effects associated with roads. In evaluating effects of oil and gas activity on Alaska’s North Slope, the NRC (2003) found that roads had a synergistic effect with pipelines and off-road vehicle

OCR for page 62
Assessing and Managing the Ecological Impacts of Paved Roads traffic. These factors accumulated to affect the habitat and behavior of animals, physically changed the environment next to the road, and increased access and social contacts among human communities. Caribou migration on and near roads, although they are gravel, is one example of a cumulative effect; roads with a parallel pipeline and those without a parallel pipeline had different migration patterns (Forman et al. 2003, NRC 2003). In addition, new roads often are associated with development of residential, commercial, and industrial activities. In some cases, the roads are built to support the new activities and, in other cases, the roads lead to the additional development. INFORMATION GAPS Historically, most studies of road effects have been carried out at the project level, with local studies focusing on specific transportation effects. Collaborative research among multiple government agencies has been lacking. States or provinces have had little coordinated formal data sharing to allow for information syntheses and analyses of effects at larger and perhaps more meaningful scales of evaluation. Defining the appropriate scale of research will depend on the ecological condition of interest. A watershed is one example of an appropriate spatial scale to assess water-quality issues. Transportation projects sponsored by the Federal Highway Administration (FHWA) and the Transportation Research Board (TRB) have stimulated and encouraged collaborative studies involving multiple state agencies with similar transportation problems that might be solved through large-scale ecological assessments and pooled-funding approaches (TPF 2004). Pooled-funding projects often focus on specific information needs defined by the collaborative state transportation agencies, which frequently represent diverse environments across the continent. Often, projects are not defined by appropriate scales or ecologically defined areas, such as specific eco-regions (for example, the northern Rocky Mountains). Reports have called for nationwide assessments and national syntheses on how wildlife respond to highway barriers, for mapping habitat linkages and landscape connectivity at regional and national scales, and for means of standardizing roadkill data collection and analyses (Evink 2002). Two reports (Evink 2002, TRB 2002a) highlight the need for more systems-level studies addressing long-term issues regarding re-

OCR for page 62
Assessing and Managing the Ecological Impacts of Paved Roads search in surface transportation and natural systems, in addition to continuing studies focusing on short-term, project-specific, transportation needs. General sources of research for transportation and ecology include FHWA, TRB, NCHRP, American Association of State Highway and Transportation Officials Center for Environmental Excellence, universities, and other agencies. As discussed above, these efforts have produced a substantial body of literature that documents the effects of roads and traffic on ecological conditions. However, almost no studies of ecological effects of roads have been conducted over long time periods (multiple decades) or at large spatial scales (spatial windows above tens of kilometers). Such studies should be a priority for research. Few, if any, studies directly address ecological effects of road density. The appropriate scale for research is not always known beforehand, and the ecological impacts of roads can go undetected if an arbitrary scale is chosen for the research. Some multiscale studies have shown that roads affect ecological condition at much larger scales than previously thought. Therefore, multiscale studies can uncover the ecological effects of roads and the scale at which roads affect ecological condition. Finally, much of the research on the ecological effects of roads can be found in reports that may not have been peer-reviewed or commercially published. For example, committee members are aware of studies documenting the effects of roads on sediment production in reports from the state departments of transportation, the Army Corps of Engineers, and the World Bank. Although included in some searchable databases, such as the Transportation Research Information Service, these reports are not included in scientific abstracting services (for example, Cambridge Abstracts) and, therefore, are generally less accessible to the academic research community. Future studies on the ecological impacts of roads should be published in the peer-reviewed venues. SUMMARY Roads influence ecological conditions across a range of organizational levels and scales. A large part of the scientific knowledge of ecological effects of roads has been based on short-term studies focused on narrowly defined objectives and has generally been related to specific construction or planning needs. As a result, more research is needed on ecological effects that occur over large areas or long periods. Ecological

OCR for page 62
Assessing and Managing the Ecological Impacts of Paved Roads conditions are not only affected by the construction of roads and road appurtenances (bridges and culverts) but also by the traffic on the roads and, at larger scales, by the increases in road density. The ecological effects of roads are much larger than the roads themselves, and the effects can extend far beyond ordinary planning domains. Few studies address the complex nature of the ecological effects of roads. For example, little is known about how roads impede access to foraging areas or key prey species, potentially resulting in cascading or other trophic effects. Studies assessing ecological effects are often based on small sampling periods and, therefore, do not adequately sample the range of variability in ecological systems. More research should be directed at identifying the appropriate scale at which roads affect ecological conditions. Information on the resiliency of biodiversity components to road-related disturbances is needed to better understand the effects of roads on ecological systems. Research on the ecological effects of roads over long periods or at large spatial scales and research on the complex nature and impacts of roads within ecologically defined areas, such as watersheds, eco-regions, or species’ ranges, should be a priority. Research on the local scale should continue, however, because the context of many transportation decisions is at the local scale with direct application, and studies that address the context are likely to be the most frequently used and have the largest influence.