Vegetation Science
Online ISSN : 2189-4809
Print ISSN : 1342-2448
ISSN-L : 1342-2448
Volume 27, Issue 2
Displaying 1-4 of 4 articles from this issue
  • Tamotsu HATTORI, Noriko MINAMIYAMA, Yasuhiko OGAWA
    Article type: Article
    2010 Volume 27 Issue 2 Pages 45-61
    Published: December 25, 2010
    Released on J-STAGE: January 06, 2017
    JOURNAL FREE ACCESS
    The Man'yoshu, which was edited in the eighth century, is an excellent anthology of old Japanese poems about harmonious co-existence between nature and mankind. Based on the names and descriptions of plants, plant communities and environmental conditions written in the Man'yoshu, we could deduce the vegetational land-scape in the Asuka and Nara periods. The number of plants and plant communities described in the Man'yoshu include 145 species and 19 communities, respectively. The combination of two or more plant species in a single poem was considered a part of the species composition of an actual plant community. The combination of plant species and environmental conditions in a poem was considered to correspond to the actual relationship between plant species and environmental conditions. These interpretations confirmed that poems of the Man'yoshu were written realistically in expression of nature and mankind. At Sato (mostly cultivated and developed open areas) and Yama (mountains near a village), the vegetational landscapes in the Asuka and Nara periods were similar to these in recent years. Contrarily, at Okuyama (mountains far from a residential area) the vegetational landscape in the Asuka and Nara periods, composed of natural lucidophyllous-coniferous mixed forests, was different from that in the present day, which is composed of secondary and artificial forests.
    Download PDF (2117K)
  • Toshiyuki YAMASHITA, Satoshi ITO, Kumiko OHTSUKA
    Article type: Article
    2010 Volume 27 Issue 2 Pages 63-71
    Published: December 25, 2010
    Released on J-STAGE: January 06, 2017
    JOURNAL FREE ACCESS
    We investigated forest vegetation and habitat condition in lower warm-temperate evergreen forests with special reference to Quercus hondae, an endangered evergreen oak species, in southern and eastern Kyushu, Japan. TWINSPAN was applied to 77 dataset of stands comprising 65 published datasets and 12 new datasets obtained from surveys conducted for this study. In the first division, the 77 stands were divided into two groups: Q. hondae-dominated forests and Q. gilva-dominated forests, distinguished by Persea japonica and Arachniodes sporadosora as the indicator species, from Q. glauca-dominated forests. In the second division, Q. hondae-dominated forests were distinguished by Q. glauca and Castanopsis cuspidata from Q. gilva-dominated forests. Comparisons of the TWINSPAN results with previously published phytosociologial studies for this region showed that the Q. hondae-dominated forests and Q. gilva-dominated forests had similar compositions in terms of the character species for the Lasiantho-Quercetum gilvae. However, the coexistence of Q. hondae with Castanopsis cuspidata and the comparison with the character species of the upper unit suggested that Q. hondae forests would more appropriately be classified to the Symploco glaucae-Castanopsietum sieboldii. CCA was applied to the 12 stands investigated within the lower slopes of a valley in Miyazaki Prefecture, which is the center of the distribution of Q. hondae. The results indicate that habitat of Q. hondae forests are more likely to be established in thinner soil on steeper slopes than that of Q. gilva forests, and in thicker soil on gentle slopes than that of Q. glauca forests. These habitat were also confirmed by the CCA scores of each dominant species; Q. hondae had a score intermediate to that of Q. gilva and Q. glauca. The results suggest that the difference in habitats of Q. hondae and Q. gilva might reflect the other factors beside soil moisture.
    Download PDF (1074K)
  • Shigeyuki OGAWA, Susumu OKITSU
    Article type: Article
    2010 Volume 27 Issue 2 Pages 73-81
    Published: December 25, 2010
    Released on J-STAGE: January 06, 2017
    JOURNAL FREE ACCESS
    Betula spp. (birch) are key forest species in deciduous broad-leaved forests on the Pacific side of eastern Japan. However to date, there have only been a few reports on the site environments and maintenance of Betula forests. Field surveys were conducted to study the site environments and maintenance of Betula (Betula davurica, B. platyphylla var, japonica, and B. grossa) forests in the Soto-Chichibu Mountains, central Japan. Betula forests were distributed on the landslide sites in the area dominated by Quercus serrata. Those landslide sites were formed by the landslide event in 1948. Therefore, it is considered that Betula trees were able to invade within open sites of landslide, and form their own forests. The formation of Betula forests depends on the formation of open site due to geomorphic disturbances of landslide. Then, the Soto-Chichibu Mountains often experience geomorphic disturbances of landslide by mechanism and process of landforms of Mikabu green rocks, resulting in the formation of open site by landslide event. For that reason, we consider that Betula forests will be maintained by frequent geomorphic disturbances in the future.
    Download PDF (1279K)
  • Noritoshi KAMAGATA, Keitarou HARA
    Article type: Article
    2010 Volume 27 Issue 2 Pages 83-94
    Published: December 25, 2010
    Released on J-STAGE: January 06, 2017
    JOURNAL FREE ACCESS
    This research aims to evaluate the suitability of object-based image analysis as a tool for vegetation mapping. The research was implemented in the Tanzawa Mountains of the western Kanto Region, in central Honshu Island, Japan. This site is a typical mountainous area usually subject to misinterpretation due to shadows, where even visual interpretation using a stereoscopic viewer is often unable to accurately distinguish among vegetation types in the valleys. Our research performed object-based image analysis on Very High Resolution (VHR) remote sensing data. The classification accuracy was assessed by comparing the results with a physiognomical master vegetation map that was generated from phytosociological vegetation maps produced by the Japanese Ministry of the Environment. The ability of the analysis to extract boundaries was evaluated by comparing both the outline boundary and the individual lines of the results with those on the physiognomical master vegetation map. The overall classification accuracy was 72.7%, and the Kappa Index 0.526. The outline boundary extraction scored 76.8%, 63.8% and 49.6% respectively for completeness, correctness and quality; while the line extraction scored 78.7%. These results indicate a high classification accuracy and a high level of ability to extract boundaries, and demonstrate that object-based analysis using VHR remote sensing data is a promising tool for vegetation mapping in mountainous regions.
    Download PDF (3235K)
feedback
Top