生态道路——美国曼彻斯特城郊高速公路的影响范围

The Ecological Road-Effect Zone of a Massachusetts (U.S.A.) Suburban Highway

生态道路——美国曼彻斯特城郊高速公路的影响范围

RICHARD T. T. FORMAN* AND ROBERT D. DEBLINGER† *Harvard University, Graduate School of Design, Cambridge, MA 02138, U.S.A. Commonwealth of Massachusetts, Division of Fisheries and Wildlife, One Rabbit Hill Road, Westborough, MA 01581, U.S.A.

Abstract: Ecological flows and biological diversity trace broad patterns across the landscape, whereas trans­portation planning for human mobility traditionally focuses on a narrow strip close to a road or highway. To help close this gap we examined the “road-effect zone” over which significant ecological effects extend out­ward from a road. Nine ecological factors—involving wetlands, streams, road salt, exotic plants, moose, deer, amphibians, forest birds, and grassland birds—were measured or estimated near 25 km of a busy four-lane highway west of Boston, Massachusetts. The effects of all factors extended100 m from the road, and moose corridors, road avoidance by grassland birds, and perhaps road salt in a shallow reservoir extended out­wards >1 km. Most factors had effects at 2–5 specific locations, whereas traffic noise apparently exerted ef­fects along most of the road length. Creating a map of these effects indicates that the road-effect zone aver­ages approximately 600 m in width and is asymmetric, with convoluted boundaries and a few long fingers. We conclude that busy roads and nature reserves should be well separated, and that future transportation systems across landscapes can provide for ecological flows and biological diversity in addition to safe and ef­ficient human mobility.

Zona Ecológica de Efecto Carretero en una Autopista Suburbana de Massachusetts (U.S.A.)

Resumen: Los flujos ecológicos y la diversidad biológica trazan patrones amplios a lo largo del paisaje, mientras que la planeación del transporte para mobilidad humana se enfoca tradicionalmente en una franja angosta cercana a las carreteras o autopistas. Para ayudar a cerrar esta brecha examinamos la “zona de efecto carretero” sobre la cual se extienden efectos ecológicos significativos que parten de una carretera. Nueve factores ecológicos (involucrando humedales, arroyos, sal de la carretera, plantas exóticas, alces, vena-dos, anfibios, aves de bosque y aves de pastizal) fueron medidos o estimado cerca de 25 km de una autopista muy utilizada de cuatro carriles ubicada al Oeste de Boston, Massachusetts. Los efectos en todos estos factores se extendiéron>100 m de la autopista, los efectos en corredores de alces, de evasión de la carretera por aves de pastizales y la sal proveniente de la carretera en un reservorio somero se extendiéron >1 km. La mayoría de los factores tuvieron efectos en 2-5 localidades específicas, mientras que el ruido del tráfico tuvo efectos a lo largo de casi la totalidad de la extensión de la autopista. La creación de un mapa de estos efectos indica que la zona de efecto carretero promedia 600 m en amplitud y es asimétrica con límites intrincados y algunos relieves geográficos intrincados y largos. Concluímos que las carreteras muy transitadas y las reservas de la naturaleza deben estar muy bien separadas, y que los futuros sistemas de transportación pueden proveer flu­jos ecológicos y diversidad biológica a lo largo del paisaje, además de una mobilidad humana segura y efi­ciente.

Introduction tern of water superimposed on patches of different vegetation and land use, which in turn differentiate into The 9.4 million km² of the United States resemble a large, distinctive physiographic zones. Over the last four brightly colored mosaic or mural, with a fine-scale pat-centuries a 6.2 million–km network of public roads, now used by 200 million vehicles, has been superimposed on this land to provide safe and efficient human Paper submitted February 8, 1999; revised manuscript accepted July 16, 1999.mobility (National Research Council 1997). The effect of this engineering marvel on the foundation of the United States—the ecological infrastructure—must be consider­able. Recent evaluations indicate that one-fifth of the United States land area, and nearly the same proportion of The Netherlands, is directly affected ecologically by the road system (M. Reijnen et al. 1995; R. Reijnen et al. 1995, 1996; Forman & Deblinger 1998; Forman 1999).

Determining the ecological “road-effect zone” for roads is a central part of such evaluations (Forman et al. 1997; Forman & Alexander 1998; Forman & Deblinger 1998; Forman 1999). This zone is the area over which signifi­cant ecological effects extend outward from a road. It is many times wider than the road surface plus roadsides, or verges (Bennett 1991; Reck & Kaule 1993; Reijnen 1995; Forman 1995). The road-effect zone is highly asymmetric, due to nature’s directional flows and the spatial patterns on opposite sides of a road, and has con­voluted boundaries (Fig. 1).

We used transportation planning as the context for presenting and illustrating the road-effect zone. Whereas ecological flows and biological diversity manifest broad patterns across the landscape (Turner 1989; Forman 1995; Harris et al. 1996) and road networks affect the fragmentation and human use of entire landscapes (Dale et al. 1994; Reed et al. 1996), transportation planning has traditionally focused meticulously on the narrow area of the road bed and its immediate surrounding strip (Stoeckeler 1965; Directorate-General for Public Works and Water Management 1995; National Research Coun­cil 1997). It is time to close the gap between nature’s broad landscape processes and the detailed focus of road construction and upgrading projects.

A landscape ecology approach is promising for setting mitigation priorities for the ecological effects of the ex­isting road network, which was built largely before the recent, rapid growth in ecological knowledge. Successful mitigation and compensation projects scattered across the United States, such as salamander tunnels in Massa­chusetts and underpasses for ground water and wildlife in Florida, serve as useful pilot projects (Evink et al. 1996; Jackson 1996; Forman et al. 1997). Several Euro­pean nations and Australia have begun to implement widespread environmental mitigation for roads (Langton 1989; Saunders & Hobbs 1991; Pfister & Keller 1995; Canters 1997). The Netherlands, where transportation pol­icy explicitly meshes human mobility and natural processes in the landscape, has completed an array of mitigation and compensation projects, both in new road construction and for existing roads (Directorate-General for Public Works and Water Management 1995; Road and Hydraulic Engi­neering Division 1995; Forman & Hersperger 1996; Canters 1997; Friedman 1997a, 1997b; Forman & Alexander 1998).

Figure 1. Hypothetical diagram illustrating curvilinear nature of road effects for various ecological features such as high-density deer areas (road mortality), grassland bird nesting areas (noise disturbance), moose corridors (road mortality), and various wetland and waterway alterations including exotic plant invasion, salt intrusion, and channelization.

In the context of protecting species, habitats, and nat­ural processes, our broad objective was to examine the road-effect zone concept based on a particular example. Our specific goals were (1) to compare the distances and spatial patterns of ecological factors that extend out­ward considerable distances from a highway and (2) to combine these results to map the shape of the road-effect zone for a highway.

Highway, Landscape, and Ecological Factors

In 1997 we selected and studied an area in the middle suburbs (Lincoln, Concord, Acton) and outer suburbs (Boxborough, Littleton) west of Boston, Massachusetts (U.S.A.). Route 2 was chosen as a reasonably typical case of a major four-lane artery radiating outward from a U.S. city through an area experiencing rapid growth in houses and vehicular traffic. This 24.5-km stretch of highway begins 10 km west of the urban area (Cambridge city limit) and follows a relatively straight west-northwest di­rection. It connects two major “ring roads” of the urban region, the inner-loop Interstate highway 95 (I-95, also Route 128) and the outer-loop Interstate 495 (I-495). This portion of Route 2 was constructed in 1935, and much of the length was upgraded between 1987 and 1993. Interstate highway 95 (Route 128) was completed about 1950 and I-495 about 1970.

The section of highway we studied had four traffic lanes with a central divider. Fifty percent of the highway section was divided by a steel fence barrier on posts (guardrail), 25% was divided by continuous concrete barriers, 5% was divided by a median strip of mowed grass, and 20% of the highway section had no divider. Grassy roadsides along most of the highway generally extended a few meters beyond the paved road surface, and 40% of the section length had a steel fence barrier along the roadside. Thirty-three roads connected with Route 2 over its 24.5-km length. Overall, the eastern por­

tion (kilometers 0–6) and central portion (kilometers 6–18) had small road connections, the central area had cross­roads with stoplights, and the western portion (kilome­ters 18–25) was a controlled-access highway with only entrance-and-exit interchanges.

Traffic volume for the eastern 17 km of Route 2 was ap­proximately 50,000 vehicles per day (Monday through Friday; March data from kilometer 12.5; Concord Town Records). Traffic volume was 34,000 at kilometer 18 (Acton Town Records) and 42,000 at kilometer 24 (Little­ton Town Records). The traffic speed limit was generally 45 mph (73 km/hour) in the eastern and central portions and 55 mph (89 km/hour) in the western portion. The in­ner-loop highway, I-95, had >150,000 vehicles/day; the outer loop, I-495, had >50,000 vehicles/day.

The surrounding land was covered with glacial deposits and mainly sandy soils (Schofield & Baron 1993). Swamps, ponds, and streams were abundant, and Route 2 crossed a reservoir and two rivers. Deciduous forest with evergreens (predominately Acer, Quercus, Betula, Fraxinus, Pinus) generally appeared as the background matrix, some of which was wetland. Along 15% of the roadside the highway crossed or was adjacent to seven open-field areas ranging in size from 10–40 ha. Residential neighborhoods covered about one-third of this suburban landscape.

Two river crossings and three railroad crossings had underpasses through which wildlife moved to cross Route 2. The Sudbury River, with its elongated National Wildlife Refuge, flows northward as a major vegetation corridor crossing Route 2 at kilometer 10. The road crossed over 13 streams (first- to third-order). One me­dium-sized airport, Hanscom Air Force Base 3 km north of Route 2 at kilometer 4.5, contained a large open area partially managed for grassland birds. New Hampshire, 50 km to the north, was the source of moose (Alces al­ces) that periodically dispersed to the Route 2 area.

Road (or highway) refers to the area of paved road sur­face plus any median area, maintained roadsides, and ad­jacent up- or downslope roadbanks. In effect, the road is the area extending outward that receives road mainte­nance, or the area where the adjoining topographic con­tour is altered by the existence of the road, whichever is further. Additional information on Route 2, the sur­roundings, and mapping are presented by Forman and Deblinger (1998).

More than 20 ecological effects of roads have been de­scribed (Walker & Walker 1991; Reck & Kaule 1993; Forman 1995; Forman et al. 1997; National Research Council 1997). These extend outward for widely varying distances. The road-effect zone, however, is determined only by those factors that extend outward the farthest distances. Therefore factors such as the effects of dust or road salt on vegetation, the presence of heavy metals, microclimatic changes, and small mammal crossings, whose effects in most areas extend outward only tens of meters, were omitted from the list of key factors. We also eliminated road access for long-distance hunting and other human activities or disturbances (Dale et al. 1994; Forman 1995) because numerous smaller connect­ing roads permit such access to nearby forested land. Short-term ecological effects of the road construction process (Ashley & Robinson 1996) were also omitted.

The following therefore remained as potential key fac­tors that may extend outward from the highway >100 m (in some cases perhaps for kilometers). Water- and plant-related factors included (1) altered streams and wetland drainage, (2) road salt reaching surface-water bodies, and (3) habitat invasion by exotic species planted on roadsides. Key animal habitat and movement patterns affected by roads included (1) large mammals such as moose and deer (Odocoileus virginianus), (2) forest and grassland birds, and (3) amphibians.

We measured some factors directly, whereas others were estimated based on evidence in other studies. Data sets ranged from reasonably thorough to embryonic, so some details should be interpreted with caution. We in­cluded best estimates of all factors that could be esti­mated because for the road-effect zone it is important to see the entire picture.

Road Effects on Water and Plants

Altered Streams and Wetland Drainage

To determine the number, location, and distance from the highway of all channelized or rerouted streams and wetlands at least partially drained due to the construc­tion of Route 2, we examined all streams crossing under the road and recorded the length of any channelization (straightening of the natural stream curviness) upslope and downslope. Almost all streams flowed through wet­lands, and stream curviness was easy to observe in No­vember 1997 with a light snow cover on the ground. In­termittent channels were measured in addition to year-round streams.

Thirteen streams and three intermittent channels crossed under the road, which is generally routed across higher terrain. Ten of the streams were first-order, two second-order, and one third-order (based on U.S. Geological Sur­vey 7.5> topographic quadrangle map). One stream and one intermittent channel showed evidence of alteration predating the 1935 construction of Route 2 and were ex­cluded from our analysis.

Almost all streams and intermittent channels crossing the road were channelized for some distance either up-slope, downslope, or both. Two-thirds of the streams were channelized upslope and one-third downslope. Both intermittent channels upslope and no channels downslope were straightened. The length of channel­ized streams varied from 30 m to 400 m upslope and from 30 m to 500 m downslope of the road. Only two streams were channelized 100 m or more away from the road (Fig. 2). Apparently no stream was rerouted >30 m for construction of the road.

Almost all wetlands adjacent to the road were swamps dominated mainly by red maple (Acer rubrum). A few were marshes with cattail (Typha latifolia) and purple loosestrife (Lythrum salicaria). The presence of a chan­nelized stream either upslope or downslope was used as an indicator of at least partial wetland drainage (Stoeck­eler 1965).

Nine wetlands were crossed by or adjacent to the 25 km of Route 2. All of these showed evidence of wetland drainage, indicating that the wetlands are smaller now than before road construction. Wetland drainage effects extended outward from the road for distances varying from 50 m to 500 m. Five wetlands were affected >100 m from the road (Fig. 2).

Thus, two streams were channelized and five wet­lands partially drained to a distance of >100 m from Route 2. These measures of stream and intermittent channel straightening, plus wetland drainage, are con­servative estimates of road effects on the aquatic and wetland ecosystems. Measuring attributes such as change in sediment load, fish populations, and streamflow re­gime would likely show road effects extending consider­ably farther than those we recorded. Similarly, water ta­ble and wetland soil measurements would probably show wetland effects extending farther from the road.

Road Salt and Surface-Water Bodies

To estimate where salt spread on the highway in winter may affect natural ecosystems, especially the many sur­face-water bodies near Route 2, we consulted local natu­ral-resource specialists and searched for evidence from the literature. Visible damage to natural woody vegeta­tion by salt appeared to be localized within 10 m or so of the road, even on downslope sides and in open areas. Road salt is also known to move in groundwater and pol­lute wells in the sandy, gravelly glacial till of Massachu­setts and elsewhere in New England (National Research Council 1991), but local natural-resource people knew of no reports or evidence of this effect along the Route 2 study segment. Thus, the primary potential widely spread­ing ecological effects of road salt are believed to be on shallow ponds, lakes, and reservoirs, streams parallel and adjacent to the road, and vernal pools near the road (National Research Council 1991).

The only evidence we found of road salt pollution along Route 2 was from the Cambridge Reservoir at kilometer 0.5 (Fig. 2). Within approximately 1 km of Route 2 the reservoir was only about 1.0–1.5 m deep. Elevated salt lev­els in this supply of city drinking water have been reported for many years. The primary source is considered to be I-95 along its eastern border and the secondary source Route 2, where “Low Road Salt” signs are often displayed in winter.

Figure 2. Locations of major water- and plant-related effects of divided highway Route 2 in suburbs west of Bos­ton, Massachusetts. Top map is eastern half and bottom map is western half of study area. Wetland drainage and stream and intermittent-channel straightening that extend at least 100 m from the highway are marked. Streams and channels estimated to be affected by road salt run parallel and adjacent to the road for at least 100 m. All lo­cations of woody non-native species apparently invading from roadsides into natural woods are marked. Adapted from Forman and Deblinger (1998).

Four first-order streams flowed parallel and adjacent to the road (within 30 m) for a distance of at least 100 m (distances arbitrary but consistent with the terrain; stream directions not realigned by road construction). Three such intermittent channels were also present. Road salt can be expected to reach these parallel streams and channels in quantities sufficient to affect down­stream aquatic ecosystems (Isabelle et al. 1987; Fennes­sey 1989; National Research Council 1991). In addition to the four first-order streams, we included two stream segments that were close to a salt storage area at kilome­ter 6 (Fig. 2). A parallel adjacent fourth-order stream and the three intermittent channels were not mapped be­cause it was less clear what significant ecological effects would be expected.

Three vernal pools with official state certification were located within 200 m of Route 2, and 15 such pools occurred within 1 km of the highway (Fig. 2). Salt, which moves in groundwater through porous glacial till and reaches a shallow pool or pond, can be expected to cause elevated pool salinity and affect sensitive species, espe­cially in spring. At least two of the vernal pools are down-slope of Route 2 and are most likely to suffer salt effects.

Thus, one reservoir is known to have widely extend­ing road salt effects, and we can hypothesize ecological effects on at least four streams and two vernal pools. The distance that effects from Route 2 extend is un­known, but they may be in the range of 200–1500 m.

Habitat Invasion by Planted Roadside Exotics

To determine where and how far non-native (exotic) species planted along the roadside spread into the ad­joining natural forest (Saunders & Hobbs 1991; Tyser & Worley 1992), we drove slowly along the entire route and recorded locations of all exotic woody plantings vis­ible in the roadbed (planted by highway personnel, community groups, or adjacent landowners). Based mainly on location within a roadside, we excluded plants that apparently colonized naturally. For each roadside planting encountered we recorded the adjoin­ing land use: natural woods, open, or built. Of the ap­proximately 100 roadside locations with planted exot­ics, we examined on foot >50 of the most likely spots for species invasions. The remaining locations were sur­rounded by built area and thus were excluded. Wher­ever natural woods (upland or wetland) were next to a planted woody exotic (22 locations), we examined the nearest 1 ha of woods on foot to determine the presence or absence and the distance of seedlings or saplings ap­parently originating from the planted exotic. Each woods survey lasted 15–20 minutes during 1–12 Octo­ber 1997, when peak autumn colors facilitated plant identifications.

Forty-four woody exotic species were planted at the approximately 100 locations along the 25-km suburban highway (Forman & Deblinger 1998). Exotics included a few species native to the United States but not growing naturally near Route 2. The most frequently planted spe­cies were yew (Taxus sp.), Norway spruce (Picea abies), blue spruce (Picea pungens), and arbor vitae (Thuja oc­cidentalis).

Both the average and variance in number of planted roadside species decreased with distance from city boundary (Forman & Deblinger 1998), but both in­creased with local building density near the highway (within 200 m of the road surface in a 1-km road seg­ment; measured from aerial photographs). Few exotic species were planted in roadsides where nearby build­ing density was>60 buildings/km². No correlation was found between the species richness of roadside plant­ings and local road density (also measured within 200 m) near the highway.

Evidence of invasion into natural ecosystems by these planted roadside exotics was present in 11 woods (Fig. 2). This represented 50% of the cases in which woods adjoined a planted woody exotic. Six invading species were recorded: Norway maple (Acer platanoides), little-leaf linden (Tilia cordata), red pine (Pinus resinosa), red spruce (Picea rubens), euonymous (Euonymous al­ata), and privet (Ligustrum vulgare). The percent of woods invaded did not correlate with distance from ur­ban center (measured from Cambridge boundary). Also there was no correlation with nearby building density or nearby road density. The distance a species invaded into woods from the border of the road surface ranged from 10 m to 120 m, although more intensive sampling might have detected invasions to greater distances.

Thus, 11 woods are apparently being invaded from roadside plantings of non-native woody species along the 25 km of suburban Route 2. The distance presently affected is up to about 120 m from the highway.

Road Effects on Wildlife and Species Pattern

The overall effects of a road on animal populations appear to be (1) a barrier effect that blocks movement routes and subdivides species into smaller subpopulations, (2) ani­mal avoidance of nearby habitat due to traffic noise, and (3) roadkills (Forman & Alexander 1998). For Route 2 we provide estimates for the location, distance, and impor­tance of (1) the blocking of wildlife corridor routes of moose and salamander, (2) roadkills of deer, and (3) avoid­ance of nearby habitat by forest and grassland birds.

Species such as white-tailed deer may adapt behavior­ally to roads by avoiding nearby habitats where traffic noise inhibits predator detection and by selecting partic­ularly advantageous routes and times to cross roads. On the other hand, some species, such as the spotted sala­mander (Ambystoma maculatum) are genetically and physiologically “programmed” to migrate at a certain time from uplands to wet areas for breeding; en route they head directly across a road (Langton 1989; Fahrig et al. 1995). Therefore, transportation planning should pro­vide for both species with considerable ability and those with little ability to avoid roads.

Distant areas, including large patches of natural vege­tation commonly containing large populations, are often key sources of and targets for animals approaching or crossing highways (e.g., deer and fisher [Martes pen­nanti]; Harris et al. 1996; Findlay & Houlahan 1997). Distinct, major wildlife routes (e.g., of moose and sala­mander) may cross highways (Bennett 1991; Harris et al. 1996), and habitat near roads sometimes contains con­centrated resident populations (e.g., of deer).

Large Mammals

Moose populations in northern New England have in­creased exponentially in the past 15 years. Although a small resident population is beginning to emerge in cen­tral and western Massachusetts, moose in eastern Massa­chusetts are usually young males dispersing southward from New Hampshire in search of females during the breeding season in September–October or young fe­males dispersing to new home ranges in May–June (Vecellio et al. 1993). Perhaps due to their large size and relatively low metabolic rate, moose may move through the landscape with little regard to human activity.

Fourteen individual moose have been sighted in our study area since 1992 (Forman & Deblinger 1998). They all crossed major highways, including I-495, and appar­ently tended to cross in railroad and river underpasses (Fig. 3). These records illustrate that important moose routes in a suburban landscape with fragmented habitat are railroad, river, and powerline corridors (Fig. 3). Adja­cent wetlands and lakes probably enhance the corridor quality. The easternmost moose sighted at kilometer 1 was a yearling female that traveled along a powerline right-of-way and railroad tracks into downtown Boston.

White-tailed deer density is generally highest in the few remaining large tracts of protected forest, a pattern partly associated with recently reduced hunting in this suburban landscape (Deblinger et al. 1993). Two such areas of deer concentration exist next to Route 2, where relatively natural forest land is also adjacent to rivers and wetlands (Fig. 3). Both forest areas are bisected by rail­road beds that deer use to cross Route 2 as they move into suburban development areas. Locations of deer hit by vehicles along Route 2 tend to be concentrated near these adjacent protected areas of high deer density (1995–1996 records of Concord Police Department).

Forest and Grassland Birds

Forest-interior birds such as Ovenbirds (Seiurus auro­capillus), cuckoos (Coccyzus spp.), and Pileated Wood­peckers (Dryocopus pileatus) have specific habitat re­quirements that may be affected by highway noise. Detailed forest-bird data near Route 2 have not been compiled. Therefore, to estimate the effect of this high­way on forest birds we extrapolated from the most com­plete data set available from another temperate area.

Dutch studies have determined road “effect-distances”— (distance from a road for which a significant change, such as reduction in bird density, is recorded)—next to main roads with the same traffic volume as that of Route 2

Figure 3. The road-effect zone of Route 2, plus locations of moose, deer concentrations, grassland birds, traffic noise effects, and other highway effects in the surrounding landscape. The road-effect zone is indicated by short dashes plus dotted areas. Dotted areas are major wildlife corridors and locations of widely extending water and plant effects (Fig. 2). No bird data are available for the Western and peripheral portions of the map. Adapted from Forman and Deblinger (1998).

(Foppen & Reijnen 1994; Reijnen 1995; M. Reijnen et al. 1995; R. Reijnen et al. 1995). Evidence indicates that traf­fic noise rather than visual disturbance, air pollutants, or predators along roads is the primary cause for avian com­munity changes. A major effect of traffic noise is hypothe­sized to be its interference with bird communication dur­ing incubation and fledgling phases of reproduction.

For forest birds as a whole and for the most sensitive species, effect-distances in woodland extend hundreds of meters from a busy road (Fig. 4). The population den­sity of the most sensitive forest-interior species is re­duced in woods to approximately 650 m from a main road. In this zone the population is one-third lower than that at greater distances. We used this 650 m for the most sensitive species, rather than the distance for all species combined, in estimating the road-effect zone for Route 2. The most sensitive species tend to be of pri­mary conservation interest.

Grassland birds are those species found only or pre­dominantly in large open areas such as meadow, hay-field, cultivation (tillage), pasture, or some combination thereof. Three of the five key grassland species in the Route 2 area are state-listed threatened species: Upland Sandpiper (Bartramia longicauda), Grasshopper Spar­row (Ammodramus savannarum), and Henslow’s Spar­row (A. henslowii). The other two species, Meadowlark (Sternella magna) and Bobolink (Dolichonyx oryzi­vorus), are often found in somewhat smaller open areas than those of the preceding species. Hanscom Air Force Base (2–3 km north of Route 2 in Lincoln, Concord, and Bedford), a medium-sized airport with approximately 146 ha of grassland, serves as the major local area for grassland bird populations. The Upland Sandpiper is re­ported only from Hanscom and one site (Lincoln) 3 km south of Route 2, the Grasshopper Sparrow from Hanscom and one location (Concord) 1.5 km south of Route 2; and Henslow’s sparrow from one site (Lincoln) 2 km north and one site (Lincoln) 3 km south of Route 2.

Grassland birds are intensively observed in the area of kilometers 0–15 (Lincoln, Concord, eastern Acton). Breeding-season (15 May to 1 August) records of several experienced ornithologists, naturalists, and officials for approximately 1993–1998 were accumulated and mapped for Meadowlarks and Bobolinks. Open areas observed were grouped into three categories: regular breeding (evi­dence of breeding in 3 or more years), occasional sight­ing (present in 1 or 2 years), and no sighting. Size of open area does not correlate with distance from Route 2 (two of the largest open areas are adjacent to the highway).

Figure 3. (continued)

Within 1 km of Route 2, Meadowlarks and Bobolinks are only occasionally sighted and apparently do not or only rarely breed (Fig. 3). No open area within this 1-km distance is a regular breeding site for either species. In contrast, regular breeding occurs at two sites in the 1–3 km zone and at four sites in the 3–5 km zone. Caution in interpreting this grassland bird pattern is warranted due to the scattered bird records, the scarcity of open areas by the highway, and the diverse management regimes of open areas. Nevertheless, the data are consistent with the hypothesis that grassland birds are reduced in den­sity and species number for hundreds of meters (or per­haps 1-2 km) from Route 2.

We turn again to the more complete European data set, which shows a decrease in density (Fig. 4) and diver­sity of grassland birds by main roads (Van der Zande et al. 1980; R. Reijnen 1995; M. Reijnen et al. 1995; R. Reijnen et al. 1996). The most sensitive grassland species is reduced in density to a distance of 930 m from a main road, a figure we used to calculate the road-effect zone.

Amphibians

Salamanders migrate to and breed synchronously in tem­porary ponds in early spring. Such mass migrations be­tween home range and breeding location, when bisected by roadways, can result in significant mortality (Fahrig et al. 1995; Ashley & Robinson 1996). In Amherst, Massachusetts (110 km to the west), for example, road closure during the migratory period reduced mortality, and after two amphib­ian tunnels with associated drift fences were placed under the road, mortality decreased further (Langton 1989; Jack­son 1996).

Three state-certified vernal pools are within 200 m of Route 2 (Fig. 3). Although amphibian movement has not been studied here, the highway is considered a likely barrier to movement in the vicinity of these locations.

Road-Effect Zone

The preceding results for nine diverse ecological vari­ables are synthesized to delineate the road-effect zone for Route 2.

Some factors, such as invading roadside plants, pro­duced effects at several distinct locations, but appar­ently for Route 2 the effects extended outward only about 100 m. Salamander migration routes may be blocked by a highway one hundred to a few hundred meters from a vernal pool. Road salt effects on aquatic

systems may extend outward some 200–1500 m, also at distinct locations. In contrast, traffic noise apparently af­fected bird communities in forest and open areas for hundreds of meters on both sides and along a large pro­portion of the highway. For mammals such as deer, fisher, and black bear (Ursus americanus) the highway eliminated suitable habitat and interrupted major travel corridors between large core patches of suitable vege­tated habitat. These core-habitat areas were located up to several kilometers from the highway (for instance, the only recent bear record in the vicinity of kilometers 0–18 along Route 2 was in a 400-ha forest centered 5 km north of kilometer 9).

The road-effect zone for Route 2 (Fig. 3; Forman & Deb-linger 1998; Forman 1999) had highly convoluted boun­daries as a result of the sequence of affected ecological factors along the road and their differing effect-distances (Forman et al. 1997; Forman & Alexander 1998). The road-effect zone at most points was asymmetric partially because of directional water and wind flows across the zone. Topographic patterns and cut roadbanks also caused asymmetry in ecological effect distances. Never­theless, the different arrangement of suitable habitats and land uses generally present on opposite sides of the highway, especially of major natural-vegetation patches and corridors, was a major cause of the asymmetric pat­terns of ecological effects.

The maximum distance that direct ecological effects extended outward from Route 2 (edge of road surface) averaged just over 300 m (Fig. 3). Therefore the average width of the road-effect zone for this suburban highway exceeded 600 m, with the area affected being approxi­mately 0.6 km² per kilometer of road length. Of course, variability from side to side and from location to location along the route was high and of considerable ecological in­terest. Variability appeared higher in the middle suburbs (kilometers 0–18) than in the outer suburbs. We hypothe­sized that this was the result mainly of more land-use het­erogeneity (i.e., a more even mix of woods, built areas, and open areas) in the middle suburbs, compared with more extensive forest cover in the outer suburbs.

Two important ecological factors could not be esti­mated for this study, but in time they should be added. First, highways that eliminate suitable habitat and inter­rupt major travel corridors between large patches of suitable vegetated habitat for large mammals may pro­duce effects extending for kilometers from the highway (Dale et al. 1994; Forman 1995; Harris et al. 1996; Find­lay & Houlahan 1997). The effects of interrupted wildlife corridors would typically appear as “fingers” extending outward from the road-effect zone boundary (similar to fingers representing the effect of road sediments on downstream fish). Second, the road subdivides previ­ously large connected populations into smaller subpopu­lations (Langton 1989; Reed et al. 1996; Forman & Alex­ander 1998) that tend to be isolated on habitat patches

Figure 4. Birds in (a) forest and (b) pasture relative to distance from main roads with traffic noise. In for­est the pattern for all species combined is compared with that for the sensitive Cuckoo; all species in grass­land are compared with the sensitive Black-tailed God­wit. Based on songbird studies at 69 locations (54 in woodland, 15 in open grassland) by highways with 50,000–60,000 vehicles/day in The Netherlands (M. Reijnen et al. 1995; R. Reijnen 1995; R. Reijnen et al. 1995, 1996). Adapted from M. Reijnen et al. 1995 (Forman & Deblinger 1998). and that may be distant from the road. The subpopula­tions are considered less stable and more vulnerable to local disappearance or extinction (Wiens 1996).

Conclusion

We envision a future transportation system that provides effectively for both (1) natural processes and biodiver­sity and (2) safe and efficient human mobility. Transpor­tation planning should take a broad perspective on the spatial patterns and ecological flows extending horizon­tally across the landscape. The road-effect zone we de­scribe is the minimum distance from a road or proposed road appropriate in planning.

Busy roads should be kept well away from nature re­serves and similar protected areas. Natural areas near such roads may be impoverished in species sensitive to road and traffic disturbance. Road ecology is one of the great frontiers awaiting science and society. A range of solutions to the problem of roads exists, and research should develop others to effectively address the most extensive ecological factors underlying the road-effect zone. Research, pilot projects, monitoring, and public education are needed to attain this vision of a future transportation system.

Acknowledgments

We are deeply grateful to J. W. Brain, S. F. Ells, C. F. Gib­son, A. L. Jones, E. Mallett, D. H. Monahan, T. Tidman, and R. K. Walton for grassland bird data; to P. J. Barosh for information on geology and hydrology; to T. Hayes, G. McGeon, D. H. Monahan, and T. Tidman for moose data; to R. Reijnen for highlighting the importance of traffic noise on avian populations; to T. Early for prepar­ing the figure; and to K. Bramley, D. Graaskamp, and S. McRae for preparing maps.

Literature Cited

Ashley, E. P., and J. T. Robinson. 1996. Road mortality of amphibians, reptiles and other wildlife on the Long Point causeway, Lake Erie, Ontario. Canadian Field-Naturalist 110:403–412.

Bennett, A. F. 1991. Roads, roadsides and wildlife conservation: a re­view. Pages 99–117 in D. A. Saunders and R. J. Hobbs, editors. Na­ture conservation 2: the role of corridors. Surrey Beatty, Chipping Norton, New South Wales, Australia.

Canters, K., editor. 1997. Habitat fragmentation & infrastructure. Min­istry of Transport, Public Works and Water Management, Delft, The Netherlands.

Dale, V. H., S. M. Pearson, H. L. Offerman, and R. V. O’Neill. 1994. Re­lating patterns of land-use change to faunal biodiversity in the Cen­tral Amazon. Conservation Biology 8:1027–1036.

Deblinger, R. D., D. W. Rimmer, J. J. Vaske, G. M. Vecellio, and M. P. Donnelly. 1993. Ecological benefits and hunter acceptance of a controlled deer hunt in coastal Massachusetts. Northeast Wildlife 50:11–21.

Directorate-General for Public Works and Water Management. 1995. Nat­uur over wegen (Nature across motorways). Delft, The Netherlands.

Evink, G. L., P. Garrett, D. Zeigler, and J. Berry, editors. 1996. Trends in addressing transportation related wildlife mortality. Publication FL–ER–58–96. Florida Department of Transportation, Tallahassee.

Fahrig, L., J. H. Pedlar, S. E. Pope, P. D. Taylor, and J. F. Wegner. 1995. Effect of road traffic on amphibian density. Biological Conservation 73:177–182.

Fennessey, T. W. 1989. Guidelines for handling acid-producing materi­als on low-volume roads. Transportation Research Record 1291: 186–189.

Findlay, C. S., and J. Houlahan. 1997. Anthropogenic correlates of spe­cies richness in southeastern Ontario wetlands. Conservation Biol­ogy 11:1000–1009.

Foppen, R., and R. Reijnen. 1994. The effects of car traffic on breeding bird populations in woodland. II. Breeding dispersal of male wil­low warblers (Phylloscopus trochilus) in relation to the proximity of a highway. Journal of Applied Ecology 31:95–101.

Forman, R. T. T. 1995. Land mosaics: the ecology of landscapes and re­gions. Cambridge University Press, Cambridge, United Kingdom.

Forman, R. T. T. 2000. Estimate of the area affected ecologically by the road system in the United States. Conservation Biology 14:31–35.

Forman, R. T. T., and L. E. Alexander. 1998. Roads and their major ecolog­ical effects. Annual Review of Ecology and Systematics 29:207–231.

Forman, R. T. T., and R. D. Deblinger. 1998. The ecological road-effect zone for transportation planning, and a Massachusetts highway ex­ample. Pages 78–96 in G. L. Evink, P. Garrett, D. Zeigler, and J. Berry, editors. Proceedings of the international conference on wild­life ecology and transportation. Publication FL-ER–69–98. Florida Department of Transportation, Tallahassee.

Forman, R. T. T., and A. M. Hersperger. 1996. Road ecology and road density in different landscapes, with international planning and mitigation solutions. Pages 1–22 in G. L. Evink, P. Garrett, D. Zei­gler, and J. Berry, editors. Trends in addressing transportation re­lated wildlife mortality. Publication FL–ER–58–96. Florida Depart­ment of Transportation, Tallahassee.

Forman, R. T. T., D. S. Friedman, D. Fitzhenry, J. D. Martin, A. S. Chen, and L. E. Alexander. 1997. Ecological effects of roads: toward three summary indices and an overview for North America. Pages 40–54 in K. Canters, editor. Habitat fragmentation and infrastructure. Min­istry of Transport, Public Works and Water Management, Delft, The Netherlands.

Friedman, D. S. 1997a. Nature as infrastructure: the national ecological network and wildlife-crossing structures in The Netherlands. Report

138. DLO Winand Staring Centre, Wageningen, The Netherlands. Friedman, D. S. 1997b. Walking on the wild side. Landscape Architec­ture September:52–57.

Harris, L. D., T. S. Hoctor, and S. E. Gergel. 1996. Landscape processes and their significance to biodiversity conservation. Pages 319–347 in

O. Rhodes Jr., R. Chesser, and M. Smith, editors. Population dynamics in ecological space and time. University of Chicago Press, Chicago.

Isabelle, P. S., L. J. Fooks, P. A. Keddy, and S. D. Wilson. 1987. Effects of roadside snowmelt on wetland vegetation: an experimental study. Journal of Environmental Management 25:57–60.

Jackson, S. 1996. Underpass systems for amphibians. Pages 255–260 in

G. L. Evink, P. Garrett, D. Zeigler, and J. Berry, editors. Trends in addressing transportation related wildlife mortality. Publication FL– ER–58–96. Florida Department of Transportation, Tallahassee.

Langton, T. E. S., editor. 1989. Amphibians and roads. ACO Polymer Products, Shefford, Bedfordshire, United Kingdom.

National Research Council. 1991. Highway deicing: comparing salt and calcium magnesium acetate. Special report 235. Transportation Re­search Board, National Research Council, Washington, D.C.

National Research Council. 1997. Toward a sustainable future: address­ing the long–term effects of motor vehicle transportation on cli­mate and ecology. National Academy Press, Washington, D.C.

Pfister, Von H. P., and V. Keller. 1995. Strassen und Wildtiere. -Sind Grunbrucken eine Losung? Bauen fur die Landsirtschaft 32:26–30. Reck, H., and G. Kaule. 1993. Strassen und Lebensraume: Ermittlung

Forman & Deblinger

und Beurteilung strassenbedingter Auswirkungen auf Pflanzen, Tiere und ihre Lebensraume. Forschung Strassenbau und Strassen­verkehrstechnik, Heft 654. Herausgegeben vom Bundesminister fur Verkehr, Bonn-Bad Godesberg, Germany.

Reed, R. A., J. Johnson-Barnard, and W. L. Baker. 1996. Contribution of roads to forest fragmentation in the Rocky Mountains. Conserva­tion Biology 10:1098–1106.

Reijnen, M. J. S. M., G. Veenbaas, and R. P. B. Foppen. 1995. Predicting the effects of motorway traffic on breeding bird populations. Minis­try of Transport and Public Works, Delft, The Netherlands.

Reijnen, R. circa 1995. Disturbance by car traffic as a threat to breed­ing birds in The Netherlands. Thesis. Institute for Forestry and Na­ture Research (IBN-DLO), Wageningen, The Netherlands.

Reijnen, R., R. Foppen, C. ter Braak, and J. Thissen. 1995. The effects of car traffic on breeding bird populations in woodland. III. Reduc­tion of density in relation to the proximity of main roads. Journal of Applied Ecology 32:187–202.

Reijnen, R., R. Foppen, and H. Meeuwsen. 1996. The effects of car traf­fic on the density of breeding birds in Dutch agricultural grass­lands. Biological Conservation 75:255–260.

Road and Hydraulic Engineering Division. 1995. Wildlife crossings for roads and waterways. Ministry of Transport, Public Works and Wa­ter Management, Delft, The Netherlands.

Saunders, D. A., and R. J. Hobbs, editors. 1991. Nature conservation 2: the role of corridors. Surrey Beatty, Chipping Norton, New South Wales, Australia.

Schofield, E. A., and R. C. Baron, editors. 1993. Thoreau’s world and ours: a natural legacy. North American Press, Golden, Colorado.

Stoeckeler, J. H. 1965. Drainage along swamp forest roads: lessons from Northern Europe. Journal of Forestry 63:771–776.

Turner, M. G. 1989. Landscape ecology: the effect of pattern on pro­cess. Annual Review of Ecology and Systematics 20:171–197.

Tyser, R. W., and C. A. Worley. 1992. Alien flora in grasslands adjacent to road and trail corridors in Glacier National Park, Montana (U.S.A.). Conservation Biology 6:253–262.

Van der Zande, A. N., W. J. ter Keurs, and W. J. Van der Weijden. 1980. The impact of roads on the densities of four bird species in an open field habitat—evidence of a long distance effect. Biological Conser­vation 18:299–321.

Vecellio, G. M., R. D. Deblinger, and J. E. Cardoza. 1993. Status and management of moose in Massachusetts. Alces 29:1–7.

Walker, D. A., and M. D. Walker. 1991. History and pattern of distur­bance in Alaskan arctic terrestrial ecosystems: a hierarchical ap­proach to analysing landscape change. Journal of Applied Ecology 28:244–276.

Wiens, J. A. 1996. Wildlife in patchy environments: metapopulations, mosaics, and management. Pages 53–84 in D. R. McCullough, edi­tor. Metapopulations and wildlife conservation. Island Press, Washington, D.C.