Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Climate Adaptation with Reinforcement Learning: Economic vs. Quality of Life Adaptation Pathways

Miguel Costa, Arthur Vandervoort, Martin Drews, Karyn Morrissey, Francisco C. Pereira

TL;DR

The paper tackles climate adaptation under deep uncertainty by formalizing RL-based pathways within an IAM that links rainfall/flood dynamics to transport outcomes. It introduces eight actionable adaptation measures and a tunable reward function $R = \sum_i \beta_I I_i + \beta_D D_i + \beta_C C_i + \beta_Q Q_i + \beta_A A_i + \beta_M M_i$ to explore how different priorities—such as QoL versus economic protection—shape long-term strategies. Empirical results for Copenhagen show QoL-focused policies can drive greater total investment and more evenly distributed spending, highlighting the strong influence of normative choices on policy pathways. The publicly available framework enables policymakers to balance short- and long-term objectives under uncertainty and to compare trade-offs across objective functions. Future work aims to scale to the full city and analyze distributional impacts across socio-demographic groups.

Abstract

Climate change will cause an increase in the frequency and severity of flood events, prompting the need for cohesive adaptation policymaking. Designing effective adaptation policies, however, depends on managing the uncertainty of long-term climate impacts. Meanwhile, such policies can feature important normative choices that are not always made explicit. We propose that Reinforcement Learning (RL) can be a useful tool to both identify adaptation pathways under uncertain conditions while it also allows for the explicit modelling (and consequent comparison) of different adaptation priorities (e.g. economic vs. wellbeing). We use an Integrated Assessment Model (IAM) to link together a rainfall and flood model, and compute the impacts of flooding in terms of quality of life (QoL), transportation, and infrastructure damage. Our results show that models prioritising QoL over economic impacts results in more adaptation spending as well as a more even distribution of spending over the study area, highlighting the extent to which such normative assumptions can alter adaptation policy. Our framework is publicly available: https://github.com/MLSM-at-DTU/maat_qol_framework.

Climate Adaptation with Reinforcement Learning: Economic vs. Quality of Life Adaptation Pathways

TL;DR

The paper tackles climate adaptation under deep uncertainty by formalizing RL-based pathways within an IAM that links rainfall/flood dynamics to transport outcomes. It introduces eight actionable adaptation measures and a tunable reward function to explore how different priorities—such as QoL versus economic protection—shape long-term strategies. Empirical results for Copenhagen show QoL-focused policies can drive greater total investment and more evenly distributed spending, highlighting the strong influence of normative choices on policy pathways. The publicly available framework enables policymakers to balance short- and long-term objectives under uncertainty and to compare trade-offs across objective functions. Future work aims to scale to the full city and analyze distributional impacts across socio-demographic groups.

Abstract

Climate change will cause an increase in the frequency and severity of flood events, prompting the need for cohesive adaptation policymaking. Designing effective adaptation policies, however, depends on managing the uncertainty of long-term climate impacts. Meanwhile, such policies can feature important normative choices that are not always made explicit. We propose that Reinforcement Learning (RL) can be a useful tool to both identify adaptation pathways under uncertain conditions while it also allows for the explicit modelling (and consequent comparison) of different adaptation priorities (e.g. economic vs. wellbeing). We use an Integrated Assessment Model (IAM) to link together a rainfall and flood model, and compute the impacts of flooding in terms of quality of life (QoL), transportation, and infrastructure damage. Our results show that models prioritising QoL over economic impacts results in more adaptation spending as well as a more even distribution of spending over the study area, highlighting the extent to which such normative assumptions can alter adaptation policy. Our framework is publicly available: https://github.com/MLSM-at-DTU/maat_qol_framework.

Paper Structure

This paper contains 12 sections, 1 equation, 2 figures.

Table of Contents

  1. Introduction
  2. Modelling Framework
  3. Rainfall Projection Model and Flood Model
  4. Transportation Model
  5. Trip Routing:
  6. Accessibility:
  7. Transport-related Impacts
  8. Direct Infrastructure Impacts:
  9. Indirect Transport Impacts:
  10. Quality of Life Impacts:
  11. Reinforcement Learning
  12. Preliminary Experiments & Discussion

Figures (2)

  • Figure 1: Integrated Assessment Model using RL to learn the best sequence of transport-related adaptation policies for rainfall events in Copenhagen from 2023-2100.
  • Figure 2: Top row: The proportion of money spent by TAZ by 2100 for the EC and QoL policies. Bottom row: Adaptation pathways per TAZ for the EC (left) and the QoL policies (right).