Department of Civil and Environmental Engineering, Imperial College London, London, UK.
Sci Total Environ. 2013 Feb 15;445-446:79-93. doi: 10.1016/j.scitotenv.2012.12.022. Epub 2013 Jan 11.
Urban drainage system models can be useful to assess and manage system performance and to plan its development. However, due to data and computational costs, sophisticated, high-resolution contemporary models of the sewer system may not be applicable. This constraint is particularly marked in developing country mega-cities where catchments can be large, data tend to be scarce, and there are many unknowns, for example regarding sources, losses and wrong connections. This paper presents work undertaken over the last 7 years to develop a suitable monitoring and modelling framework to support operation and development of the wastewater system of Bogotá (Colombia). Components of the framework covered here are: (a) the flow and water quality database, (b) a wastewater pollution load generator, and (c) a semi-distributed sewer network model, which aims at a complexity that matches the information available from the previous two components. Results from a catchment within Bogotá, area 150 km(2) and with 2.5 million inhabitants, show that the model outputs capture the scale and dynamics of the observed concentrations and loads at various points on the sewer system. However uncertainty is high because much of variability of observed dry weather flow profiles is apparently random. Against this variability, the effects of in-sewer processes were not identifiable except where backwaters caused particularly high retention times. Hence the work has resulted in an operational model with a scientifically justified, yet useful, level of complexity for Bogotá. More generally, the work demonstrates the value of monitoring and modelling programmes, including having modellers actively involved in monitoring specification and operations; and the insights into suitable level of model complexity that may be gained by uncertainty and sensitivity analysis.
城市排水系统模型可用于评估和管理系统性能，并规划其发展。然而，由于数据和计算成本的原因，复杂的、高分辨率的现代下水道系统模型可能并不适用。这一限制在发展中国家的特大城市中尤为明显，因为这些城市的汇水面积可能很大，数据往往稀缺，而且存在许多未知因素，例如污染源、流失和错误连接。本文介绍了过去 7 年来为开发适合监测和建模框架以支持波哥大（哥伦比亚）废水系统运行和发展而开展的工作。这里介绍的框架组件包括：（a）流量和水质数据库；（b）废水污染负荷生成器；（c）半分布式下水道网络模型，其目的是使模型的复杂性与前两个组件提供的信息相匹配。对波哥大一个 150 平方公里、有 250 万居民的集水区的结果表明，模型输出能够捕捉到下水道系统各个点观测到的浓度和负荷的规模和动态。然而，不确定性很高，因为大部分观测到的旱季流量曲线的变异性显然是随机的。与这种变异性相比，除了死水造成特别高的保留时间之外，下水道内的过程影响无法识别。因此，这项工作产生了一个具有科学依据、但对波哥大来说具有有用的复杂性水平的运行模型。更一般地说，这项工作展示了监测和建模计划的价值，包括让建模人员积极参与监测规范和运行；以及通过不确定性和敏感性分析可以获得的关于合适模型复杂性水平的见解。
