Uncertainty and risk in water quality modelling and management

Abstract

The case is presented for increasing attention to the evaluation of uncertainty in water quality modelling practice, and for this evaluation to be extended to risk management applications. A framework for risk-based modelling of water quality is outlined and presented as a potentially valuable component of a broader risk assessment methodology. Technical considerations for the successful implementation of the modelling framework are discussed. The primary arguments presented are as follows. (1) For a large number of practical applications, deterministic use of complex water quality models is not supported by the available data and/or human resources, and is not warranted by the limited information contained in the results. Modelling tools should be flexible enough to be employed at levels of complexities which suit the modelling task, data and available resources. (2) Monte Carlo simulation has largely untapped potential for the evaluation of model performance, estimation of model uncertainty and identification of factors (including pollution sources, environmental influences and ill-defined objectives) contributing to the risk of failing water quality objectives. (3) For practical application of Monte Carlo methods, attention needs to be given to numerical efficiency, and for successful communication of results, effective interfaces are required. A risk-based modelling tool developed by the authors is introduced.

A simple model to separately simulate point and diffuse nutrient signatures in stream flows

“…Deterministic models have advantages for management, as the simulation of real-life processes is useful for indicating the sources of pollutants, and therefore, where measurements can be taken, and where ameliorative action can be implemented. However, many authors have argued for the use of simpler models that include some estimate of uncertainty (Beck 1987;Melntyre et al 2003;Reckhow 1994;Young et al 1996). While, in the past, water quality models have tended to ignore uncertainty, there has been an increasing momentum to incorporate uncertainty into water quality models, which is partly a reflection of the maturation of the science underlying water quality modelling (Beck 1987).…”

Regionalization of models for operational purposes in developing countries: an introduction

“…doi: 10.2166/nh.2011.007 There is therefore an urgent need to roll-out scientific developments of programmes such as UNESCO FRIEND (Flow Regimes from International Experimental Network Data) and HELP (Hydrology for the Environment, Life and Policy) as well as the PUB (Predictions in Ungauged Basins; Sivapalan et al 2003) initiative of IAHS (International Association of Hydroiogical Sciences). Equally important is the need to incorporate issues of predictive uncertainty in the estimates, as well as how these translate into risks associated with decision making (Bogardi & Kundzewicz 2002;Mclntyre et al 2003;Pappenberger & Beven 2006). As part of any uncertainty estimation is related to the quantity and quality of the available data, it is inevitable that the information used for decision making in developing countries is likely to be very uncertain.…”

Receiving water quality assessment: comparison between simplified and detailed integrated urban modelling approaches

“…In addition, information about model boundary conditions, such as sources of pollution, often suffer from the same shortcomings, especially for distributed variables, which are difficult to measure (pollution runoff, sediment quality, etc.). In summary, lack of data to support model identification is a major cause of model uncertainty (McIntyre et al 2003). Therefore, uncertainty analysis must be performed for integrated urban water quality models in order to be able to quantify the level of reliability of the model results and hence the level of safety in the case of model employment for risk analysis.…”

“…The evaluation of parameter uncertainties is necessary to estimate their impact on model performance and for their calibration (Beck 1987). Indeed, as pointed out by McIntyre et al (2003), uncertainty identification in many contemporary models, such as WASP5 (Ambrose et al 1993), MIKE11 (Havnø et al 1995) and CE-QUAL (Cole & Wells 2000), is difficult because they are relatively complex and often linked to computationally intensive hydrodynamic, among other, modules.…”