As long linear ecosystems, rivers and streams are particularly vulnerable to fragmentation. There is growing concern about the role of road crossings – and especially culverts – in altering habitats and disrupting river and stream continuity. Most of the culverts currently in place were designed with the principal objective of moving water across a road alignment. Little consideration was given to ecosystem processes such as the natural hydrology, sediment transport, fish and wildlife passage, or the movement of woody debris. It is not surprising then that many culverts significantly disrupt the movement of aquatic organisms. Survival of individual animals, facilitation of reproduction, and the maintenance of population continuity are important functions of movement at a population level. Dispersal of individuals provides a mechanism for regulating population density. These dispersing individuals maintain gene flow among populations and may supplement populations where recruitment is unable to keep pace with the loss of individuals. For many small species (especially invertebrates), dispersal of individuals provides a mechanism for colonizing habitat, allowing local populations to come and go as habitat is created or eliminated, while maintaining viable regional populations. Much attention has been focused on passage for migratory fish, especially in the northwestern U.S. In some cases, considerable resources have been invested in projects addressing fish passage only to find that accommodations made for adults did not address the needs of juvenile fish. Long-term conservation of fish resources will depend not only on passage for both adult and juvenile fish but also on maintenance of healthy stream and river ecosystems. Essential to this approach is a focus on habitat quality and strategies for aquatic organism passage based on communities rather than individual species. Without an ecosystem-based approach to river and stream crossings we will be at risk of facilitating passage for particular fish species while at the same time undermining the ecological integrity of the ecosystems on which these fish depend. Stream simulation is an approach to culvert design that both avoids flow constriction during normal conditions and creates a stream channel within culverts that resists scouring during flood events. Designing culverts to avoid channel constriction and to maintain appropriate channel conditions within the structure is a relatively simple and effective approach for accommodating the normal movements of aquatic organisms and preserving (or restoring) many ecosystem processes that maintain habitats and aquatic animal populations. Road networks and river systems share several things in common. Both are long, linear features of the landscape. Transporting materials (and organisms) is fundamental to how they function. Connectivity is key to the continued functioning of both systems. Ultimately, our goal should be to create a transportation infrastructure that does not fragment or undermine the essential ecological infrastructure of the land.

As long linear ecosystems, rivers and streams are particularly vulnerable to fragmentation. There is growing concern about the role of road crossings – and especially culverts – in altering habitats and disrupting river and stream continuity. The River and Stream Continuity Project began in the year 2000 with a startup grant from the Massachusetts Watershed Initiative. The University of Massachusetts took the lead in convening a group of people from a variety of agencies and organizations who were concerned about the impact of road-stream crossings on fish and other aquatic organism passage. In 2005, three of the organizations/agencies that were key players in initiating and implementing the project joined to create the River and Stream Continuity Partnership. Founding members of the

Partnership include:

• UMass Extension ( University of Massachusetts Amherst)

• Massachusetts Riverways Program (MA Department of Fish and Game) and

• The Nature Conservancy

Members of the Partnership have made a commitment to the ongoing implementation of the River and Stream Continuity Project, including updates and revisions to the MA River and Stream Crossing Standards, coordination and implementation of volunteer assessments, management of the Continuity database, and projects to upgrade or replace substandard crossing structures.

Since its beginning, the River and Stream Continuity Project has:

• Developed “Massachusetts River and Stream Crossing Standards” to facilitate river and stream continuity as well as fish and wildlife passage. These standards are referenced in federal and state regulations and policies affecting road-stream crossings.

• Created a field protocol for volunteer assessment of road-stream crossings, including data forms, instructions, and training materials.

• Developed a system for scoring crossing structures for their effects on river and stream continuity and aquatic organism passage based on volunteer assessments.

• Created an online database for data on road-stream crossings collected by volunteers. All crossings are geo-refer enced and information from the database can be easily used in a GIS to depict the location and score of all as sessed structures in participating states.

• Developed a statewide GIS coverage prioritizing all mapped stream segments in Massachusetts into three catego ries based on information about their importance for fish and wildlife.

• Conducted volunteer assessments of road-stream crossings in Massachusetts, Connecticut, Rhode Island, Vermont and New Hampshire.

• Initiated demonstration projects to mitigate known barriers to aquatic organism passage on high-priority streams.

• Developed workshops, presentations and other educational material on the subject of river and stream continuity and the Massachusetts River and Stream Crossing Standards.

Sedimentation from unpaved roads is a primary threat to water quality in Ozark streams. TNC has partnered with various organizations including the US Forest Service (USFS), the Watershed Conservation Resource Center (WCRC), and others to develop a multi-phased approach to address the impacts of unpaved roads on these priority watersheds.

The first step in the approach utilizes advanced GIS/GPS technologies to develop a detailed vehicle-based road inventory of the target watershed or subwatershed. Sub-meter differential GPS with customized data dictionaries are used to characterize the location and function of sediment-producing and conveying features of the road infrastructure, including the road surface, prism and slope, ditches, bars, lead-offs, culverts, crossings, and outlets. The road inventory yields a comprehensive geodatabase and map series of the mapped features.

A stratified random sample of the inventoried road network is then measured to generate sediment yield predictions on ten percent of the road network. Detailed field measurements are collected with differential GPS and customized data dictionaries. The data are entered into the Water Erosion Prediction Project (WEPP) model, a process-based erosion prediction model developed by multiple federal agencies over the past 20 years. With sediment yields predicted for sample sites, erosion predictions are then extrapolated for the entire study watershed using the road inventory geodatabase.

Once sedimentation yields are predicted for each road segment in the entire study area, priority sub-watersheds are identified in the GIS using watershed sediment accumulation tools. These sub-watersheds with high potential for sediment yield may be compared to species inventory data, stream bank erosion surveys, and other land use information to set priorities for conservation planning and prioritization efforts. Priority infrastructure maintenance improvements are also identified through features that were flagged in the road inventory geodatabase as needing repair or replacement.

Road maintenance workshops are held with USFS engineers, county road crews, and other partners to transfer the inventory information, present the findings of the study and to demonstrate best management practices for road maintenance.

Since 2004, TNC and its partners in the Arkansas have worked in three priority Ozark watersheds to inventory over 600 miles of unpaved roads and 3000 associated point features in an area greater than 900 square miles. The area comprises over thirty 6th level (12-digit) HUCs.

Roadways affect wildlife habitat disproportionate to the area of land they occupy while impacting wildlife directly through direct loss of habitat, road mortality and disruption of movement. Roadways indirectly impact wildlife by isolating populations and disrupting gene flow and metapopulation dynamics. A variety of strategies have been used with mixed success to mitigate the impacts of transportation systems on wildlife. Underpasses are commonly used to facilitate movement of wildlife across roadways in Europe, Australia, Canada and the U.S.

Through 2005, 460 terrestrial and 300 aquatic crossing structures have been identified throughout the United States but only a small portion of these crossings have monitoring incorporated into their project design. Most monitoring is limited to usage of the passage structures with little data collected on movement through the adjacent landscape. Monitoring of the passage structures helps determine wildlife use of the structures but is limited in the ability to determine landscape level impacts.

A variety of techniques are utilized in monitoring passageway effectiveness, primarily camera traps and track beds. Building on prior studies, the Bennington Bypass project takes a broad, multi – taxa approach to monitoring crossing structures on a newly constructed highway in southern Vermont. We are utilizing a variety of techniques to assess movements of an array of species at the passage structure and in the surrounding landscape.

Techniques utilized in our study include: small mammal trapping, track beds/plates, remote camera sensing, snowtracking, road kill surveys, roadside track beds, amphibian recording devices, snake pit tagging and observational studies. We are also using this broad approach to monitoring as an opportunity to test and refine many of the techniques used in the study. By monitoring a wide variety of animal movements rather than focusing exclusively on wildlife use of the passages, we expect to more accurately assess the effectiveness of the mitigation structures. We anticipate that the results from this work will assist in developing monitoring protocols for future studies in Vermont and throughout the United States.

Background

Results

Conclusions