FIGURE LEGENDS
Figure 1. Location of the study area in eastern Nagano and western Gunma Prefectures in central Japan. In this area, we collected skull specimens of sika deer (Cervus nippon ) that were hunted by hunters or in animal control kills. Between late December and early May, female and male deer were killed from 2012 to 2020 and in 2020, respectively. We also collected data for location of the deer (latitude and longitude) for each individual. Female and male deer specimens were collected by shooting and trapping and by trapping only, respectively. Each point indicates the location where a female (circle) or male (rhombus) deer was killed.
Figure 2. Relationship between the accessibility to agricultural crop fields and the δ15N values for a) female and b) male deer. Accessibility (∑j exp(−αdij )Ajwas expressed using an accessibility index defined by the incidence function model (Hanski 1994). Accessibility on the x -axis was calculated based on α , as estimated by the maximum likelihood method. The accessibility to agricultural crop fields was positively related to the δ15N values (coefficient = 0.55; 95% CI = 0.29 to 0.80) only in the female model, and therefore a regression line is only shown for females.
Figure 3. Relationship between the distance from the location of deer killed to agricultural crop fields (d ) and the likelihood of crop consumption (exp(−αd )) for females. The likelihood of crop consumption completely depended on the agricultural crop field area in each 1 × 1 km mesh where d is zero and decreases rapidly with increasing d (see Results). The solid line shows the mean and dashed lines indicate the 95% CI.