Survey design
The MIREN road survey is conducted by region. In each region, participants select three sample roads that extend over a broad elevation gradient, ideally reaching elevations beyond the treeline (for examples, see Figure 1). We define a region as an area in the same biogeographical unit containing similar flora, geology and elevational ranges, usually with distances between roads of less than 150 km (Figure 2). Selected roads should begin at the bottom of the mountain region, in a valley, at sea level or where no further elevation change occurs, and reach the highest elevation typical for roads in the region. Roads can be gravel or paved but should be open for public vehicle traffic for at least some part of the year. Once roads have been selected, the elevational range of each road is divided into 19 equally wide elevational bands from the lowest to highest possible sampling location, giving a total of 20 sample sites per road located at the splits between elevational bands. Sample sites are determined prior to going into the field and located as precisely as possible using a global positioning system (GPS). At each sample site, three 2 m × 50 m plots are laid out in the form of a “T”: one plot (the top of the “T”) is parallel to the road. The other two plots extend end-to-end and perpendicular to the road, starting from the centre of the first plot, with midpoints at 25 m and 75 m from the roadside plot (Figure 3). The same plots are resurveyed every five years. If the plot locations have to be changed due to unforeseen circumstances, new sites are placed as near as possible and geolocated.