Lon., longitude; Lat., latitude; Popd., population density; Popc., population count; *p<0.05, *** p<0.001.

Fig. 1. Relationship between local extinctions and temperature, population, geographical distribution from AD1700 to 2000. The first set of variables are temperature and geographical distribution (a). The second set includes the population and geographical distribution (b). Local extinctions are dichotomous events (0,1), and temperate is represented by oxygen isotopes.

Fig. 2. Extant occurrence records, distribution range and extinct records of Chinese pangolins across China. We used 50 km buffer zones to exclude extant occurrence records and identified extinct records.

Fig. 3. Extinction-risk assessment of Chinese pangolins across China predicted by MaxEnt during 1970-2000.

Fig. 4. Variable contribution rate of the first two principal components (a) and representation of principal components to variables (b). There were 314 extinction events identified. Through PCA, variances of the first three principal components were greater than 1, and the three together accounted for 79.64% of the variance.

Fig. 5. Percentage of extant Chinese pangolin populations confronting moderate and high extinction risk.