Usefulness of Information for Drought Management Decisions: Data and User Perspectives

This thesis examines how information can benefit drought management decisions at the basin scale. Three perspectives are tested. The question is first approached from a data-centric perspective, assessing the usefulness of remotely sensed datasets to detect early stages of drought and determining how much time can be gained to inform operational land and water management practices. A user-centric approach is then followed, identifying through semi-structured interviews the information use and needs voiced by farmers and reservoir operators to support water allocation decisions during droughts, and quantifying the value of (new) information to support these decisions through modelling. Lastly, the user-centric and data-centric perspectives are combined to assess the usefulness of seasonal forecasts of water availability to support water allocation decisions in irrigated agriculture through a risk-based approach and the factors that have an impact on that usefulness.

The assessment from the user-centric perspective provided essential knowledge on the decisions, the courses of action available to decision makers, and the conditions that determine the selection of one of the available courses of action at each decision point. This knowledge is necessary to assess whether the information can change the decision outcome, which is a prerequisite for its usefulness. The user perspective also helped confirm there is a perceived need for additional information among the decision makers interviewed. The assessment from the data-centric perspective showed the ability of the selected datasets to provide key information required by the decision makers in a timely manner, while the combined perspective allowed to demonstrate the capacity of information available from seasonal forecasts to actually impact the outcome of the decision. These three perspectives show that there are multiple factors that need to be considered when assessing the usefulness of information. Notably, these are (i) the ability of information to provide either the observations or predictions that are needed by the decision maker at the time when they are needed, and (ii) the capacity of the decision maker to change the course of action as a result of the available information. The results show that both of these factors depend on the options available to the decision maker. These may differ for different individuals, depending also on the level of risk aversion decision makers have as well as their technical capacities, and the context of the decision. Changes in the market value of goods, or weather variability, may also impact the usefulness of information for the decisions analysed in this research.

Bringing these perspectives and their respective methods together contributes to fill the gap between technical and human-centred approaches to assess the usefulness of information for drought management decisions.