Why is uncertainty important?

A motivating example

The problem is illustrated by an example from practise based on a study conducted by the County of Copenhagen (Copenhagen County, 2000; Refsgaard et al., 2000). The County of Copenhagen is the authority responsible for water resources management in the county where the city of Copenhagen abstracts groundwater for most of its water supply. According to a new Water Supply Act the county had to prepare an action plan for protection of groundwater against pollution. As a first step, in 2000, the county asked five groups of Danish consulting firms to conduct studies of the aquifer’s vulnerability towards pollution in a 175 km² area west of Copenhagen, where the groundwater abstraction amounts to about 12 million m³/year. The key question to be answered was: “which parts of this particular area are most vulnerable to pollution and need to be protected?” The five consultants were selected from among the most well reputed consulting firms in Denmark, and they were known to have different views and preferences on which methodologies are most suitable for assessing vulnerability. As the job was one of the first consultancy studies in a new major market for preparation of groundwater protection action plans it was considered a prestigious job to which the consultants generally allocated some of their most qualified professionals.

The five consultants used significantly different approaches. One consultant based his approach on annual fluctuations of piezometric heads assuming that the larger the fluctuations the more interaction between aquifer and surface water systems and, hence, the larger vulnerability. Several consultants used the DRASTIC multi-criteria method (Aller et al., 1987), but modified it in different ways by changing weights and adding new, mainly geochemically oriented, criteria. One consultant based his approach on advanced hydrological modelling of both groundwater and surface water systems using the MIKE SHE code. Two other consultants used simpler groundwater modelling approaches. The three consultants applying modelling used simulated recharge as inputs to their respective DRASTIC approach. Thus, the five consultants used five different conceptual models to describe the possibility of groundwater pollution in the area. In addition, their different interpretations and interpolations made from a common data base resulted in significantly different figures; for example, for areal means of precipitation and evapotranspiration and the thickness of various geological layers (Refsgaard et al., 2000). Due to lack of concentration data in the aquifer system the methods could, mostly, not be tested against field data and their use could therefore be characterised as non-documented extrapolation. Such lack of rigorous validation tests is common in studies dealing with aquifer vulnerability towards pollution.

The conclusions of the five consultants regarding vulnerability to nitrate pollution are shown in Fig. 1. It is seen that the five estimates differ substantially from each other. In the present case, no data exist to validate the model predictions, because the five models have been used to make extrapolations towards unobservable futures. Thus, it is not possible, from existing field data, to tell which of the five model estimates are more reliable. The differences in prediction originate from two main sources: (i) data and parameter uncertainty and (ii) conceptual uncertainty. However, despite the significant data and parameter uncertainty, the main cause of the differences lies in the different conceptual models that were used by the five consultants.

Usually a water manager commissions only one study and bases his decisions on the conclusions from that study. The uniqueness of the present study was that five consultants were asked to answer the same question on the basis of the same data. In this respect the differences between the five best estimates are striking and clearly do not provide a sound basis for deciding anything about which areas should be protected. A worrying question, which is left unanswered, is whether the basis for decisions is just as poor in the many other cases where only a single conceptual model has been used and where, subsequently, a lot of money has been used to prepare and implement action plans.

In this case the uncertainty was so large that a basis for making a rationale decision did not exist. The conceptual uncertainty was discovered by chance, and the level of the uncertainty was a large, and uncomfortable, surprise to the water manager.

Fig. 1 Model predictions on aquifer vulnerability towards nitrate pollution for a 175 km² area west of Copenhagen (Copenhagen County, 2000).