Forecasting method of beech seed crop had already been established in Hokkaido, in which two conditions were required for good crop in terms of female flower production, i.e. > 500 /m² and > 20 times of the previous year (in order to escape from natural enemy). Since flower production and predator activity may depend on climatic environments, the values of these two conditions cannot be equal at all other stands in Japan. In Yamagata prefecture, for example, it has been recently pointed out that more than 350 flowers/m² would be enough for a good seed crop (90-350 for medium crop and < 90 for bad crop). In 2013, seed production showed large variation among 16 stands in Yamagata prefecture. Thus, it seemed to be a suitable year to verify the proposed conditions. Seed trap census revealed that the proposed conditions could predict the crop at each of stand with high percentage of accuracy. Thus forecasting methods which may contribute to reforestation or wildlife management can be established for beech stands in Yamagata.

View full abstract

To determine seed dispersal agents of Hovenia dulcis, which does not produce pulp-coated seeds but does produce seeds with edible peduncles, we counted the number of animals recorded by camera traps under the canopy of H.dulcis trees in Yamagata prefecture, northern Japan. In addition, we counted the number of H.dulcis seeds contained in fecal samples. Japanese macaque (Macaca fuscata) was recorded most frequent by camera traps in all animal species. H.dulcis seeds were found in all fecal samples of the Masked palm civet (Paguma larvata) and Raccoon dog (Nyctereutes procyonoides) collected. The results suggest that these animals are main seed dispersal agents. Subsequently, we carried out germination experiments in the laboratory and field, in which we compared germination of the following: ingested seeds (seeds defecated in feces after feeding fruits to mammals), fallen seeds (seeds collected from the fruits that had fallen on the ground). In both experiments, ingested seeds and fallen seeds rarely germinated, suggesting that other factors affect seed germination in H.dulcis. This indicates that these frugivorous mammals contribute to dispersion of H.dulcis seeds.

View full abstract

Species composition and biomass of shrubs was investigated at the forest floors of 8 stands suffered mass mortality caused by Oak wilt diseases at Shonai district in Yamagata. Another 8 stands without mass mortality were also investigated. All of these understories were covered with dense shrubs, while very few oak seedlings could be found. Relative light levels estimated by rPPFD at 0, 0.5 and 1.5 m above ground were approximately 2, 4 and 8%, respectively. Thus regeneration of oak trees seems to be difficult due to sever suppression by dense shrubs. Removing the shrubs from the forest floor would be effective to enhance seedling growth, because the rPPFD were significantly improved over 20% after the operation. Three kinds of the removed shrubs, i.e., Lindera umbellate, Clethra barbinervis and Camellia rusticana were chipped and given to cattle as feed. Even though the cattle did not prefer the shrubs very much, chips were available as feed. As is the case in pruning branch of the fruit tree, silage fermentation might improve the quality as feed of wood chips and may offer one option contributing to forest restoration and steady supply of food.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract