THE USE OF COLLABORATION, MONITORING, ADAPTIVE MANAGEMENT, AND PHASED CONSTRUCTION ON STATE HIGHWAY 9, COLORADO

Transcription

1 PROCEEDINGS OF THE 2017 INTERNATIONAL CONFERENCE ON ECOLOGY AND TRANSPORTATION THE USE OF COLLABORATION, MONITORING, ADAPTIVE MANAGEMENT, AND PHASED CONSTRUCTION ON STATE HIGHWAY 9, COLORADO Julia Kintsch ( , Principal & Senior Ecologist, 812 Aspen Rd., ECO-resolutions, Golden, Colorado 80401, USA Patricia Cramer ( , Independent Researcher, Logan, Utah, USA Michelle Cowardin ( , Wildlife Biologist, Colorado Parks and Wildlife, Hot Sulphur Springs, USA Paige Singer ( , Conservation Biologist, Rocky Mountain Wild, Denver, Colorado, USA Bryan Roeder ( , Environmental Research Manager, Colorado Department of Transportation, Denver, Colorado, USA ABSTRACT The State Highway 9 (SH 9) Colorado River South Wildlife & Safety Improvement Project is the culmination of a comprehensive and collaborative effort by the Colorado Department of Transportation (CDOT), Colorado Parks and Wildlife (CPW), and the privately-owned Blue Valley Ranch, as well as a number of other public and private partners. It was designed to improve safety while still providing permeability for wildlife along a 10.4-mile stretch of SH 9 between Kremmling and Silverthorne in Grand County. The project is located in a valley with a high winter concentration of mule deer, as well as other wildlife, such as elk, moose, pronghorn, black bear, bobcat, fox, coyote, and mountain lion, resulting in many wildlife-vehicle collisions. Construction was phased over two years, allowing for a research project to monitor and evaluate how well initial efforts and infrastructure performed in Phase 1. This created a unique opportunity for the researchers to assist the partners in pursuing an adaptive management approach prior to the construction of Phase 2. Phase 1, completed in 2015, included one wildlife overpass, three underpasses and six miles of continuous wildlife exclusion fencing (2.4 meters) on both sides of the highway. Pre-completion camera trap monitoring commenced in December 2015 at 27 locations, including structures, escape ramps and double-cattle guards. Phase 2 was completed in 2016, and included a second overpass, two additional underpasses and the continuation of the wildlife exclusion fencing through the project area. Camera traps were placed on the remaining structures in Postconstruction monitoring will continue at 44 locations through 2020, including all structures, the southern terminus of the wildlife fencing, and at select escape ramps, pedestrian gates, and double cattle guards. Wildlife-vehicle collision (WVC) crash and carcass data collected pre- and post-construction are also used to evaluate the efficacy of the project.

2 Kintsch, Cramer, Cowardin, Singer and Roeder 2 Pre-completion monitoring documented initial wildlife responses to the Phase 1 mitigation features. The camera data provided a basis for the researchers to work with CDOT project engineers to integrate modifications into the Phase 2 project designs, specifically with regards to the double cattle guards and escape ramps, in an effort to improve the performance of these mitigation features. The primary alterations incorporated into Phase 2 included: a new round-bar double cattle guard design at select locations; adjusted escape ramps locations to correspond with areas where mule deer that were trapped in the right-of-way tended to congregate; escape ramps with a decreased slope; and escape ramps without guide rail fencing placed on the slope. These alterations were possible because of the interagency and cross-disciplinary relationship already established at the project level, which provided the collaborative framework for CDOT engineers, CPW, and the research team to engage. The concurrence in this project of phased construction, cross-disciplinary collaboration, and pre-completion monitoring proved fruitful for an adaptive management approach. INTRODUCTION The State Highway 9 (SH 9) Colorado River South Wildlife & Safety Improvement Project resulted in the installation of seven large wildlife crossing structures between Kremmling and Green Mountain Reservoir in Grand County, Colorado. The project was designed to improve driver safety while providing permeability for wildlife. State Highway 9 runs north-south through the Lower Blue Valley, a broad sagebrush valley between the Gore Range to the west and the Williams Fork Mountains to the east. The Blue River also runs from south to north through the valley, west of the highway, to its confluence with the Colorado River. The Lower Blue Valley supports a high winter concentration of mule deer, as well as other wildlife, such as elk, moose, pronghorn, black bear, bobcat, fox, coyote, and mountain lion throughout the year. These concentrations of wildlife have resulted in numerous wildlife-vehicle collisions (WVC), particularly during the winter months. Wildlife-vehicle collisions accounted for 35% of all reported crash types from , and were the largest cause of vehicular accidents. In this five-year time span, 133 collisions with wildlife were reported to law enforcement, including 17 resulting in injuries and three resulting in human fatalities (CDOT 2012), equating to a rate of 2.4 reported WVC crashes per mile per year. Although these numbers are high, they underestimate the full extent of the conflict between traffic and wildlife on SH 9. More comprehensive winter carcass counts conducted by Blue Valley Ranch from the winter of 2007/8 through winter 2011/12 recorded 274 WVC carcasses, more than double the number of reported accidents. To meet the objective of reduced WVC and allowing for wildlife movement across the highway, two wildlife overpasses and five arch underpasses were constructed and connected with 10.4 miles of 8-foot high wildlife fencing. Other mitigation features include deer guards installed at all road intersections and private driveways; wildlife escape ramps; and pedestrian walk-through gates to provide a pathway for people through the wildlife fence. The project includes drainage culverts, including several medium-sized (8 box or pipe culverts) that are integrated into the fencing and may provide passage for small or medium-sized fauna. The wildlife mitigations are the culmination of a comprehensive and collaborative effort by the Colorado Department of Transportation (CDOT), Colorado Parks and Wildlife (CPW), and the privately-owned Blue Valley Ranch (BVR), as well as many other public and private partners.

3 Kintsch, Cramer, Cowardin, Singer and Roeder 3 The project was constructed over two years. Phase 1 encompassed the northern portion of the project area, from MP Phase 1 construction was completed in November 2015, and included four wildlife crossing structures (three large arch underpasses and one overpass), 8-foot high (2.4 meters) wildlife exclusion fencing on both sides of the highway, 34 wildlife escape ramps, 12 deer guards where county roads and driveways bisect the fencing, and 2 pedestrian walk-through gates. CDOT and CPW are supporting a five-year research study to evaluate how well the wildlife mitigation achieves the goals of reducing WVC and providing safe passages for wildlife above or beneath the highway. Wildlife monitoring commenced in the Phase 1 segment in December 2015 to evaluate wildlife responses to the mitigation features over the first winter following construction in that segment. Preliminary results and observations from these first four months of monitoring were used to evaluate the mitigation designs and recommend adjustments to the siting and design of mitigation features for the Phase 2 segment (MP ), which includes a second overpass, two wildlife underpasses, continued wildlife exclusion fencing, and additional escape ramps, deer guards and pedestrian walk-through gates. Rather than awaiting lessons learned only at the completion of the monitoring study, this project incorporated preliminary lessons learned into the second phase of construction. This approach allowed for adaptive management between the two construction phases and set the stage for experimental research studies of different mitigation designs within a single project area. METHODS Interagency Collaboration CDOT and CPW began collaborating on the State Highway 9 Colorado River South Wildlife & Safety Improvement Project during the design phase in the fall of CDOT relied on CPW to provide information and recommend designs for the wildlife features including the crossing structures, wildlife fence, escape ramps, pedestrian gates and deer guards. The agencies worked together for two years to develop the most appropriate design for SH 9 based on funding considerations and targeted wildlife species. CPW reached out to local consultants, and other DOT and wildlife agency staff in western states that had completed similar wildlife mitigation projects for additional guidance and advice on design features. The collaboration between CDOT and CPW extended from the design phase through construction and on into a 5-year monitoring project that will continue through During the 2-year construction phase, the CPW biologist attended the weekly construction meetings where CDOT and the construction managers discussed the project progression, issues and changes. Mitigation Effectiveness Monitoring A five-year monitoring effort was launched in Fall 2015, following the construction of the Phase 1 segment. The researchers used motion-triggered cameras to monitor wildlife activity at wildlife crossing structures, wildlife escape ramps, deer guards, pedestrian walk-through gates and the southern terminus of the wildlife fencing. Forty motion-triggered Reconyx Professional Series cameras (PC800 and PC900) were deployed at 24 monitoring locations in the Phase 1 segment. Cameras were installed on T-posts using a U-bolt system and Reconyx security boxes. Where cameras were placed in areas with human activity or visible from the roadside, the cameras were mounted inside metal utility boxes to disguise the camera. All cameras were code-locked and secured with master locks and/or cable locks. The cameras were motion-triggered and took

4 Kintsch, Cramer, Cowardin, Singer and Roeder 4 photos day and night with a rapid-fire setting and no down time. Cameras were set to take bursts of 10 photos per trigger, and to continue triggering for as long as the moving object is in front of the camera. Exceptions were at deer guards with heavy vehicle traffic, where cameras were set to 3 or 5 photos per trigger and, in some cases, were scheduled to trigger only between before dusk to after dawn At each monitoring location, cameras were set up to maximize capture rates and wildlife responses to the mitigation features. At crossing structures, cameras were placed to capture wildlife behavior at the entrance of the structure to distinguish successful through-passage from repels and parallel movements. Two cameras were placed at each arch underpass, at opposite corners. In addition, a habitat camera was placed on one side of each underpass, feet (15-30 meters) from the structure entrance, facing away from the road out into the habitat. The North Underpass has steep entrance ramps leading to the top of the structure, so in addition to the two cameras on top of the structure, additional cameras were placed at the bottom of the ramps on either side of the structure. These entrance cameras were more likely to capture repels and parallel movements, while the structure cameras could be used to confirm throughpassage. Habitat cameras were placed on both sides of the overpass to capture wildlife movements in the adjacent habitat. Cameras at other monitoring locations were positioned to capture specific wildlife behaviors. At deer guards and pedestrian walk-through gates, cameras were placed to capture wildlife behavior in front of the guard or walk through gate (e.g., approaches, repels and breaches). Two cameras were set up at each monitored escape ramp, one at the base of the ramp to capture wildlife approaching the ramp or walking around the ramp; and one on the habitat side to capture wildlife at the top of the ramp, including successful jump downs as well as jump up attempts from the habitat side onto the ramp. At the south fence end, cameras were positioned to capture both wildlife movements into and out of the fenced right-of way, as well as movements that occurred beyond the fence end. Cameras were visited every 4-5 weeks to exchange memory cards and batteries. Photo data were systematically processed to identify events every time a camera is triggered. RESULTS It proved to be vital for CPW to participate in the weekly meetings to build a strong and trusting relationship with CDOT and the construction crew. This relationship allowed for design changes to be discussed and adapted during construction improving the effectiveness of the project. Specifically, the commencement of monitoring activities following construction of the Phase 1 segment, but prior to construction in the Phase 2 segment created a unique opportunity for the research team and CPW to recommend adaptations to the mitigation designs based on the preliminary results and observations from the first winter of monitoring (2015/16). Precompletion monitoring documented initial wildlife responses to the Phase 1 mitigation features. The camera data provided a basis for the researchers to work with CDOT project engineers to integrate modifications into the Phase 2 project designs, particularly with regards to the deer guards and escape ramps, in an effort to improve the performance of these and other features of the mitigation. Adaptive Management Recommendations

5 Kintsch, Cramer, Cowardin, Singer and Roeder 5 The following alterations were recommended and implemented as a result of the preliminary monitoring results and observations by the researchers during the first winter following the construction of Phase 1: Deer Guards Deer breaches observed in Phase 1 were hypothesized to be the result of 1) deer being able to easily walk on the flat bars, and 2) snow getting trapped between the flat bars and creating a packed surface for deer to walk across. As a result of these observations, CDOT developed a round-bar guard design (Fig. 1) that was installed at three locations in Phase 2 and was used to replace the flat-bar guards at two locations in Phase 1. This will enable the research team to study the relative effectiveness of the two designs in preventing incursions into the fenced right-of-way by deer and other wildlife. Escape Ramps Based on observations in Phase 1, the research team made several suggestions regarding escape ramp design and placement: 1) the initial design of ramps with a 2:1 slope in the Phase 1 segment may inhibit wildlife from ascending the ramps (Fig. 2); 2) ramps placed at low points relative to the roadway, where deer trapped in the fenced right-of-way tend to congregate, may be more effective than ramps placed at high points relative to the roadway; and 3) the perpendicular rail fence on the ramps may not be functioning as intended and, instead, may obscure wildlife visibility at the top of the ramps. As a result of these observations, all escape ramps in Phase 2 were constructed with a 3:1 slope instead of a 2:1 slope, and two new 3:1 slope ramps were constructed in Phase 1 to provide a comparison to the existing 2:1 slope ramps. In addition, ramp locations in the Phase 2 segment were reassessed and moved to low spots, where possible, and six of the new ramps were constructed without guide rail Figure 2. Escape ramp in the Phase 1 segment with 2:1 slope and perpendicular rail fence. Figure 1. CDOT s round bar deer guard design with angle iron support beams. fence. These alterations will allow the research team to evaluate these design modifications during post-construction monitoring. Small Culverts Several small 8 x8 (2.4-m) box culverts in Phase 1 were not tied into the wildlife fencing and instead were fenced out. These culverts were not identified as wildlife culverts in the design plans and not recognized as having value for wildlife. However, small culverts may provide passage for small- and medium-sized wildlife in the project area. The researchers recommend that small culverts in Phase 2 be tied into the wildlife fencing and that the fencing around select

6 Kintsch, Cramer, Cowardin, Singer and Roeder 6 culverts in Phase 1 be reconstructed to tie into the culverts. These alterations were implemented accordingly. Fence Gap at Overpass Structure To prevent mule deer from crossing the overpass outside of the wildlife fence, as documented by camera monitoring (Fig. 3), the researchers recommended that the 4 (1.2-m) gap between the fence and the structure edge be closed with wildlife fencing. These fence gaps were closed during Phase 2 construction. Figure 3. Mule deer crossing over the North Overpass outside of the wildlife exclusion fence. Fence Gaps Multiple fence gaps where the fencing did not come all the way down to the ground were observed in the Phase 1 segment. The researchers recommended that fence gaps be minimized to less than 6 (15-cm) throughout the project to prevent wildlife incursions into the right-of-way. CONCLUSIONS The partnership between CDOT and CPW for the SH 9 wildlife mitigation project has created a positive working relationship that allowed for the monitoring research team and CPW to provide specific recommendations for altering designs for Phase 2 construction based on observations and preliminary results from Phase 1. This adaptive management approach prior to the completion of construction was facilitated by several factors: an established relationship between agency partners at the project level; a phased construction timeline over a two-year period, which was split into a northern segment and a southern segment; and the commencement of monitoring research between the two construction phases. The combination of these factors allowed for the early integration of lessons learned from the first phase of construction into designs for the second phase of construction, and serves as a model for future wildlife highway mitigation projects, including new collaborative efforts between the two agencies. ACKNOWLEDGEMENTS State Highway 9 wildlife monitoring research is funded by CDOT Applied Research and Innovation Branch and CPW, with support from Muley Fanatics, the Rocky Mountain Elk Foundation and the Grand Foundation. The construction of the SH 9 Colorado River South Wildlife & Safety Improvement Project was funded by CDOT with generous support from Blue Valley Ranch, Grand County, Summit County, the Town of Kremmling, and the Town of Silverthorne. In addition, a local citizens committee, Citizens for a Safe Highway 9, and Grand Foundation collected donations from 215 private citizens.

7 Kintsch, Cramer, Cowardin, Singer and Roeder 7 BIOGRAPHICAL SKETCHES Julia Kintsch is Principal and Conservation Ecologist at ECO-resolutions, and a certified Senior Ecologist by the Ecological Society of America. Julia specializes in conservation planning, road ecology, and the collaborative processes needed to achieve conservation objectives across large landscapes. In 2008, Julia launched ECO-resolutions to work directly with transportation, wildlife and land management agencies in planning and designing for landscape connectivity, including wildlife linkage analyses, mitigation design, and effectiveness monitoring. Patricia Cramer is a wildlife researcher in transportation ecology. She works with state departments of transportation and wildlife agencies to help research and design the best wildlife crossing and mitigation measures, and to bring wildlife into the transportation planning process. Her work has earned awards from Utah Department of Transportation (2015), Federal Highways Environmental Excellence Awards (2013, 2011), Utah Wildlife Society (2013), the Mule Deer Foundation (2012), and the Denver Zoo (2010). Michelle Cowardin is a wildlife biologist with Colorado Parks and Wildlife, and the lead biologist on the State Highway 9 Wildlife and Safety Improvement Project. In addition to working extensively with the Colorado Department of Transportation during the design and construction phases, she is also one of the project leads for the five-year pre- and postconstruction monitoring study. Michelle has a B.S. in wildlife and fisheries science from Pennsylvania State University and a M.S. degree in Ecology from Colorado State University. Paige Singer is a conservation biologist with Rocky Mountain Wild where she specializes in transportation ecology and geospatial information systems analyses and leads the organization s Connected Landscapes program. She has been working with partners around state to collect and analyze data related to wildlife-road interactions, wildlife linkages and the effectiveness of wildlife crossing structures. Bryan Roeder is the Environmental Research Manager for the Colorado Department of Transportation. He has been with CDOT since Bryan specializes in road ecology but is also responsible for studies related to water and air quality, sustainability, the National Environmental Policy Act, and other areas of intersection between transportation planning and the environment. Bryan has a B.A. in Biology from Colorado College and a M.S. in Wildlife Sciences from New Mexico State University.