Significance of woody browse preferences in evaluating the impact of sika deer browsing on tree seedlings

Abstract

The growing populations of sika deer (Cervus nippon yesoensis) in Hokkaido, Japan, are having increasingly negative impacts on forests. Although the percentage or frequency of browsed twigs and seedlings is a useful index of the impact of deer on forests, differences in the browse preferences of deer among tree species should also be considered. We evaluated the woody browse preferences of sika deer and examined a method for evaluating deer impacts based on the occurrence of browsing scars on seedlings, together with browse preferences. We also validated the percentages of browsed seedlings irrespective of browse preferences as an index of deer impact. We measured the heights of seedlings 30–200 cm tall in five 5 × 20 m plots located in Abies sachalinensis plantations in each of seven tracts and recorded browsing damage that occurred on leader shoots of seedlings. Browse species preference was evaluated using a generalized linear model (GLM) that included the seven tracts, species, and seedling height as explanatory variables. The results of the GLM suggest that differences in browse preferences among species are continuous and cannot be categorized discretely. The tract coefficients are considered to be an index of deer impact levels and can be estimated from the percentage of browsed seedlings for a species if the species coefficient is given. Simple percentages of browsed seedlings can also be used as an index of impact level, regardless of browse preferences, though these may be inaccurate, especially when the site is dominated by significantly preferred or avoided species.