2023-10-04T06:10:22Z
https://soar-ir.repo.nii.ac.jp/oai
oai:soar-ir.repo.nii.ac.jp:02001460
2023-05-19T00:03:01Z
1628:1629:1684378378459
起伏ある林床の微地形の最小2 乗法による判定
Classification of Undulating Forest Floor Microtopography by a Least Squares Method
荒瀬, 輝夫
中野, 正基
微地形
最小2乗法
2次多項式
林床
クロモジ
Microtopography
Least squares method
Quadratic polynomial
Forest floor
Lindera umbellata
Undulating forest-floor microtopography was examined in a natural Lindera umbellata habitat using a least squares method. Centered at each tree, relative elevation was measured at five points (0, 50 and 100 cm from the tree) along longitudinal and lateral base lines. A quadratic polynomial with the relative elevation as the dependent variable and distance from the tree as the independent variable was used to model the microtopography with the sign of the coefficient of the binominal term indicating convex or concave microtopography. Regarding the approximation accuracy for the base line directions, a standard error was slightly larger (mostly up to 15 cm) while R2 was larger (mostly over 0.9) for longitudinal measurements, and standard error was smaller (mostly up to 10 cm) while R2 was lower (less than half were over 0.9) for lateral measurements. Results of this least squares method were compared to those of the topographic position index (TPI), which is based on differences in relative elevation; 92.8% of ‘convex-convex’ (longitudinal-lateral directions) microtopography cases were evaluated as ‘convex’ in TPI and 78.9% of the ‘concave-concave’ cases were evaluated as ‘concave’ in TPI. However, this least squares method detected ‘convex-concave’ and ‘concave-convex’ microtopographies, which cannot be described by TPI. From analysis using quantification theory type one (multiple regression analysis with dummy data as independent variables) to explain the number of L. umbellata trees by various topographical features, ‘mid-slope’ and ‘valley’ in larger-scale topographies, and ‘concave’ in microtopographies for both the longitudinal and lateral directions were estimated with negative weighting; microtopography in the present study could be related to the degree of soil humidity in the L. umbellata habitat.
Article
環境科学年報 44 : 1-7, (2022)
journal article
信州大学環境科学研究会
2022-03
application/pdf
環境科学年報
44
1
7
0915-7492
AN1054947X
https://soar-ir.repo.nii.ac.jp/record/2001460/files/44-01_Arase-Nakano.pdf
jpn