Combining Ex-ante LCA with Scenario-Discovery and conditional probabilities to assist decision-making under distinct degrees of Incertitude

Description

Platform presentation.
Short Abstract:

In times of ecological emergency, prioritizing investments and allocation of scarce time and resources to specific technological directions is crucial. Ideally, prioritizing should be done before initiating uncertain exploration of novel directions. For instance, should we invest resources in looking for new microalgal compounds to enhance fish health in European aquaculture ? Ex-ante LCA can assist such policy-making. However, providing estimates of environmental impacts at the very beginning of a technological exploration which is inherently chaotic comes with substantial degrees of incertitude.

While the ex-ante LCA community often addresses uncertainty in its models, theoretical advances within Post-Normal Science have highlighted the need to differentiate the different degrees of incertitude applying to models and data. In this context, “uncertainty” is a specific degree of incertitude characterized by the incapacity to propose reasonable probabilities for events. If the knowledge about the probabilities is sufficient, the incertitude belongs instead to the domain of risk. More than a semantic debate, the presentation of uncertainty as risk, as often done in ex-ante LCA, entails caveats and overconfidence in the decision-making process.
A way of dealing with parameters for which no distribution can be proposed is to assign values to them within what-if scenarios. A conditional probability of impact can therefore be expressed within these scenarios, as a projection of knowledge conditional to the realization of specific values for uncertain parameters. However, generating scenarios for early-stage developments is challenging and the risk of overlooking important potential configurations is high. Scenario-discovery is an alternative in which algorithms discover scenarios of interest for the decision makers by exploring the output space of a model. The decision makers can eventually focus on estimating the likelihood of the discovered scenarios only.
In our work, we combine ex-ante LCA with the distinction of degrees of Incertitude, conditional probability, and scenario-discovery in a case with substantial incertitude. We demonstrate the approach on a case in which policy-makers should decide whether resources should be invested in looking for new microalgal compounds to enhance fish health in European aquaculture. With this work, we build a bridge between Post-normal Science concepts and methodologies and LCA to provide sound decision-making assistance.

Period	4 May 2023
Event title	SETAC Europe 33RD Annual Meeting: Data-Driven Environmental Decision-making
Event type	Conference
Conference number	33
Location	Dublin, IrelandShow on map
Degree of Recognition	International