Large-scale spatial distribution of deer browsing damage to young tree plantations.

The economic loss caused by herbivore browsing in forest plantations is a concerning problem in many areas around the world. Information on the spatial distribution of browsing damage is important for forest owners when selecting locations for new plantations, because planting trees in areas of high browsing pressure increases economic losses. Although it is difficult to survey browsing damage across large areas, sporadic sampling data on browsing damage are often collected by foresters, governments, and researchers. Thus, in this study, we applied a generalized additive model (GAM) for analysis of sporadic data to reveal large-scale spatial variation in deer (Cervus nippon) browsing damage. A map of browsing pressure produced by a GAM that used years after planting (plantation age) and location as independent factors showed a few areas of high browsing pressure. In addition, browsing damage increased with increasing plantation age, and plantation stands aged 2+ years showed more browsing damage. Areas with high browsing damage estimated based on plantation stands aged 2+ years generally coincided with areas of high deer abundance, with some exceptions. Thus, this model reflects large-scale browsing damage relatively well and will help forest owners to avoid locating new plantations in areas of high browsing pressure. [ABSTRACT FROM AUTHOR]