A general mathematical framework for representing soil organic matter dynamics
A general mathematical framework for representing soil organic matter dynamics
复制标题
DOI：
10.1890/15-0361.1
复制
发表时间：
2015-11
影响因子：
6.1
通讯作者：
C. Sierra;Markus Müller
中科院分区：
文献类型：
--
作者：
C. Sierra;Markus Müller
我们在这里提出了一个表示土壤有机质动态的通用数学框架。该框架用动力系统的语言来表达，并概括了以前的建模方法。它基于关于土壤有机质分解的六项基本原理：(1) 质量平衡，(2) 分解的基质依赖性，(3) 腐烂速度的异质性，(4) 有机质的内部转化，(5) 环境变异效应，以及 (6) 基质相互作用。我们展示了先前提出的大多数模型如何是这个通用模型的特殊情况。这种方法提供了根据模型所包含的主要原理或概念对模型进行分类的工具。它还有助于先验地识别不同模型或模型组的一般行为。所提出的数学表示的另一个重要特征是可以开发任何细节级别的特定模型。这一特性被描述为建模层次结构，其中高度抽象的通用模型可以容纳特定建模目标的模型结构的特定实现。该框架还允许我们研究模型组的一般属性，例如它们的定性行为、应用的时间尺度以及它们的动态稳定性。例如，我们发现模型渐近稳定的条件，即长期收敛到稳定的稳定状态，但可能会在有或没有振荡的情况下接近该状态。我们还扩展了包含时间依赖性的模型的动态稳定性概念，并且不收敛到固定的稳态，而是收敛到状态空间中的稳定区域。作为动态稳定性概念应用的一个例子，我们展示了该框架如何帮助解释土壤变暖实验中土壤呼吸通量的适应。
We propose here a general mathematical framework to represent soil organic matter dynamics. This framework is expressed in the language of dynamical systems and generalizes previous modeling approaches. It is based on a set of six basic principles about the decomposition of soil organic matter: (1) mass balance, (2) substrate dependence of decomposition, (3) heterogeneity of the speed of decay, (4) internal transformations of organic matter, (5) environmental variability effects, and (6) substrate interactions. We show how the majority of models previously proposed are special cases of this general model. This approach provides tools to classify models according to the main principles or concepts they include. It also helps to identify a priori the general behavior of different models or groups of models. Another important characteristic of the proposed mathematical representation is the possibility to develop particular models at any level of detail. This characteristic is described as a modeling hierarchy, in which a general model of a high degree of abstraction can accommodate specific realizations of model structure for specific modeling objectives. This framework also allows us to study general properties of groups of models such as their qualitative behavior, timescale of application, and their dynamic stability. For instance, we find conditions under which models are asymptotically stable, i.e., converge to a stable steady state in the long term, but may approach this state with or without oscillations. We also expand the concept of dynamic stability for models that include time dependencies and do not converge to a fixed steady state, but rather to a region of stability in the state-space. As an example of the application of the concept of dynamic stability, we show how this framework helps to explain the acclimation of soil respiration fluxes in soil-warming experiments.