Analysing complex behaviour of hydrological systems through a system dynamics approach

作者：Khan Shahbaz^*; Luo Yufeng; Ahmad Aftab

来源：Environmental Modelling & Software, 2009, 24(12): 1363-1372.

DOI：10.1016/j.envsoft.2007.06.006

• System dynamics
• Causal loop diagrams
• Dynamic simulation
• Irrigation
• Salinity
• Soil water balance
• Groundwater
• Surface-groundwater interaction
• Aerobic rice
• China
• Vensim

The interaction among various water cycle components consists of complex, non-linear, and bidirectional (interdependent) biophysical processes which can be interpreted using feedback loops in a system dynamics (SD) environment. This paper demonstrates application of an SD approach with two case studies using a specialised software tool, Vensim. The first case study simulates water balance in a rice field system on a daily basis under aerobic conditions with provision of supplemental irrigation on demand. A physically based conceptual water balance model was developed and then implemented using Vensim to simulate the processes that occur in the field water balance system including percolation, surface runoff, actual evapotranspiration, and capillary rise. The second case study simulates surface-groundwater dynamic interactions in an irrigation area where river water and groundwater are two key sources of irrigation. The modelled system encompasses dynamically linked processes including seepage from the river, evaporation from a shallow watertable, groundwater storage, and lateral flow from upland to lowland areas. The model can be applied to simulate responses of different irrigation management scenarios, to develop strategies to improve water use efficiency and control watertable, to prevent salinization in upland, and to reduce the cost of groundwater abstraction in lowland areas. The discussed applications of the SD approach conclude that it helps to conceptualize and simulate complex and dynamic water system processes deterministically which are otherwise partly simulated by conventional hydrologic and stochastic modelling approaches. It is recognised that conceptualization and implementation phases of this approach are challenging, however, the latter is greatly assisted by modern computer softwares.

• 出版日期2009-12
• 单位武汉大学