论文部分内容阅读
泥石流是特殊的流域侵蚀作用 ,同其他流域过程一样 ,密切联系着流域的形态特征。一般说来 ,泥石流都发生在小流域 (10 2 km2 以下 ) ,而经典的流域形态研究所涉及的流域范围却达到 10 7km2 的尺度。我们想知道泥石流小流域是否具有特殊的数量特征。通过流域特征量的统计 ,我们看到 ,与一般流域相比 ,泥石流小流域的特征参数之间的关系形式上相同 ,而在数值上不同 ,这从一个方面肯定了流域演化存在着普遍规律 (如自组织临界性 ) ,同时也证明泥石流是流域演化历史的“特殊一幕”。这样 ,泥石流的区域演化问题就真正同一般的流域系统演化问题联系起来了 :流域系统演化的一系列方法将有助于我们进一步建立泥石流的系统动力学。具体说来 ,本文根据中国泥石流编目数据库的资料 ,对近 6 0 0 0个泥石流流域的特征参量进行了统计 ,结果显示 ,各参量分布都服从广义的Gamma分布 :g(x ,λ ,ρ ,κ) =λρ(λx) ρκ - 1exp(- (λx)ρ) /Γ(κ)当 ρ =1或κ =1时 ,即分别为其特例Gamma分布和Weibull分布。这种分布的一个重要性质是 ,在形如y =axn 的幂函数变换下 ,分布形式保持不变 ,只是分布参数以一定方式改变 :g(y) =g(y ,λn/a ,ρ/n ,κ)这实际上就是更一般形式的标度不变性。特征参量分布的一致性同时也就意味?
Debris flow is a special basin erosion, which, like other watershed processes, is closely related to the basin’s morphological characteristics. In general, debris flows occur in small watersheds (below 102 km 2), whereas the catchment watersheds covered by the classical watershed morphology study reach a scale of 10 7 km 2. We would like to know whether the debris flow catchment has a special quantitative feature. Through the statistics of watershed features, we can see that the relationship between the characteristic parameters of debris flow and small watershed is identical in form but different in value, which confirms the general law of the evolution of the watershed from one aspect Such as self-organized criticality), but also proved that debris flow is a “special scene” in the evolutionary history of the basin. In this way, the problem of regional evolution of debris flow is really connected with the evolution of the general basin system: a series of methods of evolution of the basin system will help us to further establish the system dynamics of debris flow. Specifically, according to the data of China’s debris flow cataloging database, the statistical parameters of nearly 600 debris flow basins are calculated. The results show that all the parameters follow the general Gamma distribution: g (x, λ, ρ, κ = λρ λx ρκ - 1exp (- (λx) ρ) / Γ (κ) When ρ = 1 or κ = 1, they are the special Gamma distribution and Weibull distribution respectively. An important property of this distribution is that the distributional form remains unchanged under the power function transformation of the form y = axn, except that the distribution parameters change in some way: g (y) = g (y, λn / a, ρ / n, κ) This is actually a more general form of scale invariance. Consistency of the characteristic parameter distribution also means that?