Coastal sand dunes used to be extensive in Japan. However, most of them shrunk or disappeared because of anthropogenic land development. This study integrated a time series of geographic information for Tottori Prefecture, Japan, from 1818 to 2000. The information compiled using GIS includes aerial photographs, old topographic maps, and pictorial drawings. Then we analyzed the changes of the land cover on the coastal sand dunes, especially vegetation, and reconstructed the historical transition of the coastal sand dunes as a habitat for flora and fauna. The area of large and medium-sized coastal sand dunes was 1,893 ha, and the area of small-scale coastal sand dunes was 115 ha in 1818. However, the area of large and medium-sized coastal sand dunes was 141 ha, and that of small-scale coastal sand dunes was 25 ha in 2000. The remained coastal sand dunes were partial beach areas and foredunes. The presence of vegetation indicates a healthy condition for vegetation on coastal sand dunes, and zonation is usually established if the width of the coastal sand dune is more than 100 m. In 1818, this condition was existed in many places and generally maintained until 1900. Some sand dune areas at that time included wetlands, which must have provided high biodiversity. Although these wetlands had almost disappeared by 1952, each coastal sand dune had a broad part with a width over 100 m, and vegetation zonation was probably established there. Such wide parts significantly decreased by 2000. Even if wide parts are fragmented, rare species such as Rosa rugosa and Orobanche coerulescens, are found and vegetation zonation is detectable. Such places should be designated as protected areas.

We compared the insect assemblages with vegetation composition in the understory of a hinoki (Chamaecyparis obtuse) plantation before, immediately after, and one year after different types of thinning in order to clarify whether changes in the physical environments or in the understory vegetation composition through thinning affects the diversity of insect assemblages. We established four experimental treatments; ordinary (point) thinning (PT: scattered section on thinned trees), line thinning (LT: one line cut with three lines retained), point and line thinning (PLT: point thinning was applied to the three retained lines) and control (NT: no thinning). Immediately after thinning, the number of insects increased in the all thinned plots, in spite of the understory vegetation which was decreased by the forest-floor disturbance associated with thinning. This result suggested that the insects responded to the change in the microclimate on the forest floor rather than the changes in understory vegetation composition. One year after thinning, the number of insects increased in the thinned plots presumably affected by the recovered understory vegetation in these thinned plots. Regarding the different thinning types, the increase rate of the individual numbers was highest in LT, followed by PLT and PT immediately after thinning, and in PLT, followed by PT and LT one year after thinning, suggesting the more intensive and spatially aggregated canopy gaps in PLT promoted the increase in the individual number. From these results, it was concluded that the changes in the insect assemblages varying with different thinning methods might be influenced mainly by the degree of changes in microenvironments in the understory immediately after the thinning, but possibly reflected the degree of vegetation recovery during one year after thinning.

In this study, we clarified the drey site selection of Japanese squirrels living in cool temperate zone at three spatial scales (landscape, stand, and tree). The survey was conducted in a forest area of the Shonai region, Yamagata Prefecture. From March to May 2020, we searched for globular nests along seven survey routes. We also conducted a plot survey from August to November 2020 to investigate the forest structure of some drey trees found in the route survey. To examine selectivity at the landscape scale, we analyzed the relationship between drey sites and environmental factors (topography, distance from bodies of water, and forest type). At the stand-scale, we analyzed whether there were differences in mean tree height, mean diameter at breast height (DBH), and number of standing trees between drey and non-drey plots. At the tree scale, we analyzed whether there were differences in tree height, DBH, and species composition between drey and non-drey trees. The results of the landscape-scale analysis showed that Japanese squirrels significantly selected deciduous and evergreen coniferous forests as drey sites, with evergreen coniferous forest being relatively more important. Plots with fewer standing trees were selected as drey sites at the stand-scale. Trees with higher height and a larger DBH were significantly selected at the tree scale. The drey selection of Japanese squirrels differs depending on the spatial scale; therefore it is important to evaluate drey site selection at multiple scales.

In order to understand changes in the use of bamboo forests by wild boar, the count of the bamboo shoot feeding marks by wild boars, the position of culm, diameter of culm and a stand age-structure survey were conducted in Awaji Island from 2017 to 2018. Edible bamboo shoots appear on the ground in spring, but we found that wild boar has dug the soil and eaten bamboo shoots before the end of October. In November the number of eaten bamboo shoots accounted 0.1 to 0.7/100m²/day. The total number of the eaten bamboo shoots reached 10 to 15 /100m² during this month and this value was almost equal to the number of current-year culm appeared in the next spring. It is considered that wild boars severely affected to the regeneration of bamboo forest.