Ungulate substrate use in fauna passages

This dataset describes the substrate used by ungulates when crossing through fauna passages (n=7) in boreal Sweden and Finland, as well as the trajectories of the ungulates when crossing. Data of crossing events in the Excel file and the spatial maps are linked by the "Count" number. For each crossing event, date and time, any occurrence of snow, group size, and the animal´s direction of movement are given. As additional information, details of the seven fauna passages (location, dimensions, disturbances, and the monitoring period) are provided.

Methods
We selected fauna passages for the study on the following criteria: i) minimum width 20 m and containing a variety of substrates (≥2) to allow animals to actively select substrate, ii) passage monitored for at least 1 year, iii) camera trap set up allowed equal (unbiased) coverage of the entire passage width. We monitored ungulates in the passages using motion-triggered cameras (Browning 2017 Spec Ops Advantage Trail in the Swedish passages, Uovision UM785-3G in the Finnish passages) with an infrared (IR) motion sensor, IR flash, and IR night vision. The trigger speed of cameras was 0.4 sec (Browning) or 0.6 sec (Uovision), cameras were set to rapid fire 3–5 images when being triggered, thereafter with a delay of 5 sec (Browning) or 60 sec (Uovision) to the next triggering. We mounted the cameras in the middle of the passages, at approximately 1 m above ground, attached to pre-existing fence poles or, in a few cases, new poles. We placed the cameras on one or both sides of the passage and facing the passage center (see spatial maps), and in one case, we also mounted cameras near a structure’s center pillar and facing the sides, accordingly aiming at covering the entire width of the passage. We revisited cameras once per 1–3 months in order to check functionality and download images. Ungulate use was recorded for a minimum of one year.

We defined a crossing event as all consecutive recordings of ungulates (either single individuals or groups of conspecifics) at a passage separated by at least 10 min from the next recording. This time limit was set to avoid the most immediate temporal dependence between observations. We used only events that we judged showing animals successfully crossing the passage, i.e., coming in from one side and leaving towards the other. For each event, we noted the ungulate species and any occurrence of snow cover (as viewed from images), the date and time (start of event), the direction of movement (north or south), and the number of individuals involved (group size). We excluded a small number of crossing events where species could not be definitively identified (roe/white-tailed deer; Finnish data). For each event, we recorded the substrate where the animal crossed a predefined imaginary line over the passage in front of the cameras (hatched line in spatial maps). In the case of groups, we recorded only the first crossing individual, since we assumed that the other individuals in the group were likely to follow the trail of the first rather than make an independent decision. 

Substrates were defined as follows: 1. Grassy/sandy = spontaneous herbaceous or graminoid vegetation on sandy soil, 2. Stony/rocky = stones or rocks (> ca 5 cm) sometimes with sparse herbaceous vegetation, 3. Fiber mat = artificial coco liner, ca 5 cm thick, rolled out on stony/rocky substrate, 4. Shrub = woody vegetation ca 0.5–2 m height, mainly with ground covered by low vegetation), 5. Water = river or stream; open water, ice covered in winter, 6. Road = local gravel or dirt road. Substrates, grassy/sandy, and road were only represented in the overpasses, while stony/rocky, fiber mat, and water were only represented in the underpasses. As a complement, we mapped each trajectory on a spatial map.