Groundwater modeling to simulate groundwater/surface water interactions

计算机模型是城市地下水系统原位观测的有力工具。历史的地下水流模型代码是现实的物理耦合之间的折衷，以模拟水的存储和流量的交换。这些传统的模型方法需要繁琐的校准和验证程序，并且在评估模型适用性的拟合优度度量方面受到限制。基于脚本的方法来设置，运行和后处理地下水模型的最新进展已被证明可以提高效率，使地下水模型应用于复杂的城市地下水系统具有很大的灵活性。这些新的建模方法提供了模型开发的自由度，便于对数值解、边界条件和基于过程的校准指标进行动态修改，以确定最佳模型框架。本项目将应用基于脚本的地下水流模型和遥感技术来评估模型性能，以及城市概念模型设计来量化城市地下水系统中的水储存通量。工作的一个关键方面是一个基于蒙特卡洛的框架，以产生合奏的模型配方，以评估特设模型的性能。分析的结果将确定模型框架，可用于测试一系列雨水控制措施，影响水存储和级联变化的城市径流的分配。

Computer models are powerful tools to compliment in-situ observations in urban groundwater systems. Historical groundwater flow model codes were a compromise between realistic physical coupling to simulate exchanges in water stores and flows. These traditional model approaches require cumbersome calibration and validation procedures and are limited in goodness-of-fit metrics to evaluate model suitability. Recent advances in script-based methods to setup, run and post-process groundwater models have been shown to improve efficiency, allowing great flexibility in the application of groundwater models to complex urban groundwater systems. The freedom in model development offered by these new modeling approaches facilitates on-the-fly modifications to numerical solutions, boundary conditions and process-based calibration metrics to identify optimal model frameworks. This project shall apply script-based groundwater flow models and remote sensing techniques to appraise model performance, as well as urban conceptual model design to quantify water store fluxes within an urban groundwater system. A key aspect of the work is a Monte Carlo-based framework to produce ensembles of model formulations to assess ad-hoc model performance. Outcomes of the analysis will identify model frameworks that can be used to test a range of stormwater control measures that influence the portioning of water storage and cascading changes to urban streamflow.