he construction of a new 10.4 km (6.5 mi) section of HWY Robert-Cliche (73) south of Québec City, Canada, integrated an unprecedented number of mitigation measures to maintain connectivity between a bisected white-tailed deer winter yard and minimize apprehended deer-vehicle collisions. In this paper we present mitigation measures planned and complete as well as the monitoring approach to document deer use and movements in the winter yard before, during and after the construction. Some preliminary results regarding the impact of this project on the deer winter use of the project area also will be presented and briefly discussed. We conducted 4 years (1999-2002) of winter track surveys along the projected centerline of the new highway section and aerial surveys done in mid-winter of 2003 and 2004 to document movements and to delineate boundaries of the Calway deeryard. Mitigation measures were then proposed and integrated in the project design for the bisected deeryard. It included wildlife fencing for more than half (6.2 km or 3.9 mi) of the new highway section and combining it with 5 underpasses: one concrete box culvert, two open-span bridges over two major rivers and 2 open-span bridges over 2 rural roads. Before and during construction deer were captured each year in January and fitted with radio-collars. Yearly aerial surveys were also conducted to determine spatial use in relation with the construction phases. Around 20 deer were radio-collared each winter and telemetry data showed that about one-third of deer were long distance migrants (> 10 km) between their winter and summer home range, another one-third were short distance migrants (1 to 10 km), whereas the remaining were yearly residents of their winter range. All radio-collared deer monitored for more than a year consistently traveled between the same winter and summer home ranges. However some marked deer moved elsewhere to winter. Two primary deer crossing structures were located at the Doyon Creek and Calway River and three secondary ones were available to deer. The design and specifications of three required underpasses were modified to facilitate use by deer. As of October 2006, four underpasses were completed, as well as 5.1 km (3.2 mi) of wildlife fencing and 21 jump-outs. An additional 6 escape ramps will be built before construction ends to allow trapped deer to escape from the fenced rights-of-way (ROW). Motorists were not yet allowed to use paved sections but they will be after project completion in fall 2007. During the 2006 spring migration, about twenty deer were trapped within the 1.6 km (1.0 mi) fenced section and did not find the hole at the jump-outs. Adjustments were made on existing ramps to allow the deer to see the opening and not be reluctant to jump out to the adjacent forest. Also, new drawings and specifications were made to eliminate fence angles and reduce the height and slope of the ramp for the remaining one to build. Weekly visits from January to March 2007 showed that numerous deer were using both primary and secondary deer crossing structures to access both sides of the deeryard. Data from the aerial survey showed that the fenced highway section induced a light shift in the spatial use of the deeryard during the 2007 winter. Telemetry data provided evidence that deer with split winter home ranges continued to use both sides of the new section of highway despite a 5.1 km stretch of deer-proof fencing.