Abstract
Harvest mouse detectability represents a real issue for surveys, as usual ground-trap monitoring is known to fail to detect this species, especially in summer. The present study proposes to test ground versus aerial live-trapping efficiency for harvest mouse in summer, with a trapping design using paired ground and aerial traps over a 2-year survey in a reedbed. Over 10,720 trap-nights, the harvest mouse represented 85% of the 1078 small mammals captured, and it was the only species among five which was more often trapped above ground (n = 702) than on the ground (n = 213). Capture probability was significantly higher in aerial than in ground traps, with a slightly positive effect of leaf litter and, surprisingly, a negative effect of dense vegetation cover. Following these results, recommendations are made for harvest mouse monitoring.