News
ERC-grant for Lieke Melsen
Wageningen hydrologist and modeller Lieke Melsen is to receive a 1.5-million euro grant from the European Research Council (ERC). She and one postdoc and two PhD students will study hidden choices and values in water management models underpinning the policies implemented by governments and water authorities.
‘Computer models are frequently used to support policy decisions, as was the case during the COVID pandemic’, Melsen explains. ‘Models are viewed as objective, mathematical entities, but in developing the models, choices may be based on values, interests, uncertainties and bias. These choices impact the outcomes of a model, as well as what and for whom it can be used. During the COVID pandemic, for example, a decision was made to base epidemiological models on the number of intensive care beds being occupied rather than on, to name a different perspective, the mental health of youngsters.’
Such decisions are also made for water management models, Melsen states. ‘The first choice is what question the client has, as this generally determines what question the model must answer.’ Modellers then make decisions when designing the model. Hence, her team will discuss with modellers what choices were made. ‘The modellers often come up with explanations such as “this is our best guess”, or “this is how we have always done it”.’
Once the model is completed, a second decision-making moment occurs during the discussion with the client or policymakers. Melsen: ‘Clients cannot look inside the model, but frequently have opinions about the outcomes.’ A model may yield politically undesirable outcomes or high costs. Discussions and negotiations may ensue, during which the stakeholders seek to find room for the modeller to tweak the model so that more favourable outcomes are produced.
Melsen’s team will investigate this process in three case studies. The Dutch case study involves a model that determines the future safety of water. The American case involves a model that calculates the water quality in a bay where there is abundant farming activity. The third case, in Australia, involves a model that determines the water quality of the Great Barrier Reef under the influence of coastal farming activities. Melsen: ‘We aim to investigate whether the choices that are made lead to tensions between the modellers with their expertise and interests and policymakers with possible economic, societal and electoral interests. Models are not neutral.’