连续型地理物理过程模型与GIS的集成

Linking a GIS to a spatially distributed, physically-based process model offers many advantages.But many problems arise because the relationship between the reality being represented by the mathematical model and the data model used to organize the spatial data in the GIS has not been rigorously defined. In particular, while many physical process models are based on theories that assume continuity and incorporate physical fields as independent variables, current GISs can only represent continuous phenomena in a variety of discrete data models such as space-time cube, sequent snap shots. .This research aims at developing and outlining a strategy in which field variables are used to enable modelers to work directly with the spatial data as spatially continuous phenomena.The goal of this research is to better understand requirements of process modeling in order to extend GIS with functionality for process modeling.The following three tasks will be studied in this research: .(1) An abstraction from details of quantitative process modeling to identify generic process modeling functionality. Two generalized model of mathematical methods, differential equation and partial differential equation (PDE),are used to modeling the continuity of the geographic physical process model analysis modeling for general representation of geographic physical process..(2) A process decription language will be designed to develop a generic method to describe these processes. This language consists of:(a）a vocabulary consisting of mathematical operators to describe the model; (b) composition rules for combining the terms of the vocabulary to models;(c) a visual user interface to guide through the modeling procedure. process description language will help non-experts to perform complex geographic physical process modeling; .(3) Spatial Modeling System will be developed through the integration of GIS and Python opensource PDE module (Python Escript/Finley Libary)..Finanlly, classic process models of Hohai University（such as Xinanjiang Model for hydrological modeling,Plain River Network Model for hydrodynamic modeling,East China Sea Model for tide and wave hydrodynamical modeling）will be incorporated into the above-mentioned spatial modeling system to illustrate the effectiveness of proposed method.

地理物理过程是指遵循物理规律的地理过程，因为属于物理过程，相应地遵守质量守恒、能量守恒以及动量守恒定律等基本物理定律,如水的运动、潮流运动、台风演进、土壤侵蚀等连续时空运动过程。目前GIS还无法有效地应用于连续型地理物理过程的建模，其原因在于目前的GIS表达方法还无法考虑空间现象在空间和时间上的传播和扩散的连续时空运动特性。研究分三方面进行：(1)研究拟选择差分方程和偏微分方程(PDE)两种概化模型的数理方法来对连续型地理物理过程进行分析建模，实现地理物理过程的通用表达；(2)研究尝试开发地理物理过程的过程描述语言，用以辅助非专家的建模者进行复杂过程的建模；(3)研究拟通过集成目前Python开源偏微分方程求解模块来实现GIS与过程模型的集成,开发基于Python开源库的空间过程建模系统。最终以河海大学经典水文水动力过程模型的GIS建模为案例，证明上述方法的有效性。

近年来GIS软件发展迅猛，但是依然缺乏专用的过程模型来调控系统中地理对象的状态及演化。在理论方法上，本研究紧密结合实际空间建模需要，以空间模型特别是地理物理过程模型与GIS 集成的理论研究为基础，对美国SWAT水文学模型、中国新安江水文模型、意大利FEST-EWB水热平衡水文模型等经典过程模型进行各类形式化表达，使用Python开发区语言，结合模型建模工具和Python类库实现通用过程模型的形式化表达和原型开发。在实际应用中，研究已部分成功地用于水文水资源模拟、山洪分析评价、海岸带脆弱性评估。另外，研究还对未来的空间建模的进行了一些探讨，研究认为空间建模要主动适应大数据时代的挑战，不能单纯考虑复杂空间模型(大模型)的构建，还应注意大数据的推广和高性能计算的使用。