This study examined factors controlling the vegetation in Kushiro mire, northern Japan, especially in the alder and reed communities. The alder community was classified into four types that were arranged by canonical correspondence analysis (CCA) as follows: Carex lasiocarpa type, Carex augustinowiczii type, Persicaria type, and Spiraea type. The four types exhibited differences in size (tree height) and in the ratio of sprouting stems. Although plants growing on the forest floor primarily determined the species composition of the community, shading by alder crowns had no marked effect on the community gradient. Therefore, hydrochemical variables could control the growth and occurrence of both the floor plants and alder tree size. The differences between reed community and the alder community sites were characterized as heavily eutrophic by the Kruskal-Wallis test. Overall, the chemical variables pH and P 2 O 5 and the hydrochemical variables maximum and range of water level were positively correlated with the CCA axis, and effectively explained the community gradient in relation to alder growth. These variables could be controlled by the inflow of neutral and turbid water from river floods or runoff, which would increase pH by replacing the acidic mire water, ash content, and P 2 O 5 by conveying suspended inorganic particles that adsorb phosphorus.

The interrelation among size, biomass, and sprouting of alder trees was studied to extract the most important hydrochemistrical factors controlling the growth of alder forest in Kushiro Mire, northern Japan. The gradient was mostly explained by chemical variables such as pH, ash content, and P 2 O 5 , which showed strong positive correlation with each other, and secondarily by fluctuation of the water table (WL, i.e., water level range). These variables are more important than other hydrochemical ones, because neutral and turbid flood water replaces acidic mire water and conveys fine sediment with adsorbed phosphorus, which in turn could regulate the pH and amount of phosphorus. Also, the number of sprouts showed negative correlation mainly with tree size and redox potential (Eh), which suggested a flooded environment. Because of this, the size of alder was suppressed by hydrochemical variables; however, alder individuals producing new sprouts were maintained. We conclude that variation in size, biomass, and sprouting of alder was mainly controlled by acidity and phosphorus availability, and was significantly influenced by water fluctuation.

Imaging using polarized neutrons is one of the most attractive techniques in the neutron imaging field, because of its capability to visualize magnetic field inside materials or spaces by analyzing neutron polarization. An advanced method, which can quantify the magnetic field by combining the time-of-flight method with a polarization analysis of pulsed neutrons, has been developed at J-PARC. To introduce this method to the compact accelerator-driven neutron source, we have started the magnetic imaging experiments at Hokkaido University Neutron Source (HUNS). Using an experimental system consisting of a pair of magnetic super-mirrors as a polarizer and an analyzer, a spin flipper, and a two-dimensional neutron detector, we obtained the polarization of 90% at the wavelength over 6 Å. The first demonstration experiments were performed for coil samples. As a result, an oscillatory behaviour of polarization depending on the wavelength due to the neutron spin's Larmor precession was clearly observed.

No abstract