Downloads: 473

Files in This Item:
File Description SizeFormat 
frk_043_162.pdf1.11 MBAdobe PDFView/Open
Title: 均質な模型林冠下での落葉量推定に関する検討
Other Titles: Studies on Estimation of Leaf Fall under Model Canopy
Authors: 斎藤, 秀樹  KAKEN_name
四手井, 綱英  KAKEN_name
Author's alias: Saito, Hideki
Shidei, Tsunahide
Issue Date: 15-Mar-1972
Publisher: 京都大学農学部附属演習林
Journal title: 京都大学農学部演習林報告
Volume: 43
Start page: 162
End page: 185
Abstract: リタートラップに関する基礎的な問題を明らかにするために, 均質な模型林冠をつくり人為的に落葉させて実験し, 次のことがわかった。1. 従来から使われてきた正方形, 円形, 正三角形の3種類のトラップによる落葉推定量には, 統計的にみたばあい差は認められなかった。作り易い形のトラップを使えぱ良いであろう。2. 最小トラップの大きさは, 測定する落葉の最大長の約2倍の大きさ (正方形では一辺の長さ, 円形では直径) で, これ以上の大きいトラップを使う必要がある。3. 極端に細長い長方形トラップを使うと, 仮りに狭い方の巾が落葉の大きさに対して2倍以上の巾があっても, この長い周辺長のために入る量が僅かだが少なくなる。たて対よこの比が4 : 1以上の細長い形はさけた方が良い。4. 標準偏差Sと平均量xとは両対数グラフ上で実験的に直線になった。この関係はトラップの形, 大きさ, 葉の形 (針葉形や広葉形) および一回の落葉量などに関係なく成立した。このs - x関係で一番バラツキの大きい上限式は同様, s=A・x_h (A, h定数) で求まった。この式をつかうと実験的にh<1だから, xを大きくする, すなわちトラップを大きくしてsを小さくし, 精度を高くしてトラップの大きさ (正方形トラップの一辺長 ; lcm) や, 全トラップによる採集面積z (m_2) を, 設置するトラップ数 (n) と関係づけて求めることが出来た。すなわち, l=A_l・n_hl z=A_z・n_h_Z で, 定数A_l, h_l ; A_z, h_zは葉の大きさ (樹種) ごとに決まる値である (表5-2 - 3)。実験的にh<1であるから, h_l, h_zは負の値となり, 大きいトラップを少数設けるか, 逆に小さいトラップを多数設けるかである。すなわち, 採集面積はトラップ数を増やすことにより, 少なくすることが出来る。lとzの計算した値を表5-4に例示した。自然の林分でも落葉が林床で均質に分布するばあいには, これを適用することが出来よう。また, 測定の計画の段階においてほぼ測定精度を予測し得ると思われる。
The purpose of this work is for the solution of basic problems on the measuring of leaf fall and the authors tried to examine it with the litter trap under the uniform model canopy. 1. The difference among the leaf fall caught by the traps of three different types (square, circle and triangle) was not recognized. Consequently, any type of them could be used in accordance with convenience. 2. The minimum size of a trap (LT) required the front (side length in square or diameter in circle trap) whose length is twice as large as the maximum length (Lmax) of the leaves (LT/Lmax≧2). 3. Even the area of a rectangular was same as that of a circle when the width of the rectangular was extremely narrow (short side length/Lmax≧2), minus effect on the value of measured leaf fall was recognized. It is better not to use the rectangular trap whose circuit length is more than about 1.4 times of the circle trap whose area is same as that of the rectangular trap. 4. The relationship between the standard deviation (s) and the mean value of leaf fall (x¯) was experimentally drawn with the straight regression on logarithmic graph. This shows that this relationship is not affected by the difference of the trap types, trap sizes, leaf types (needle or broad leaf) and the amount of leaf fall. On the upper limit of the extend on the graph the relationship between s and x¯ was, [Figure omitted] in which x¯ stands for the weight or number of leaves. And h was experimentally samller than 1.0 (0.8 in hinoki cypress and 0.65 in without hinoki cypress). Because of h<1, s results in relatively smaller as x¯ becomes greater, that is, trap size becoming larger. So it was easy to calculate the size of a square trap (l cm × l cm) or the total catching area of all the traps (z) by combinning with a number of traps (n). The relations are as follows; [Figure omitted] The constants, Al, hl, Az and hz were determined in accordance with the leaf size or species (Tab. 5-2, 3). The constants hl and hz were minus values, because h is experimentally smaller than 1.0, so it is necessary to use a small number of traps with a large size traps, or to use a large number of traps with small size traps. The required total catching area of all the traps is allowed to be smaller with a large number of traps. The values of l and z were shown in Tab. 5-4. In the condition of homogenious leaf fall on the floor, Tab. 5-4 will be able to use for the determination of trap size and the number of traps at the beginning time of the investigation.
URI: http://hdl.handle.net/2433/191514
Appears in Collections:第43号

Show full item record

Export to RefWorks


Export Format: 


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.