2012年-世界发展银行全球_Investment_Decision_Making_Under_Deep_Uncertainty___Application_to_Climate_Change_41页_1mb

Summary of "Investment Decision Making Under Deep Uncertainty: Application to Climate Change"

Core Content

This paper explores the challenges of making investment decisions under deep uncertainty, particularly in the context of climate change. It highlights the limitations of traditional decision-making approaches and proposes alternative methodologies that can better handle the uncertainties associated with climate projections.

Main Viewpoints

• Deep Uncertainty: The paper defines deep uncertainty as a situation where:

  • Multiple possible future worlds exist without known relative probabilities.
  • There are divergent but equally valid world-views, including values used to define success.
  • Decisions are interdependent and evolve over time.

• Challenges in Climate Change Projections:

  • Climate models provide statistical information but not deterministic forecasts.
  • There is significant uncertainty in regional climate projections due to differences in models and natural variability.
  • Historical data is less reliable for planning under future climate change due to the non-linear and site-specific nature of ecological responses.

• Decision-Making Methodologies:

  • Cost-Benefit Analysis (CBA) under Uncertainty: Accounts for uncertainty in decision-making but may not be suitable for highly uncertain environments.
  • CBA with Real Options: Allows for flexibility and adaptation over time, suitable for projects with long lifetimes and high uncertainty.
  • Robust Decision Making: Focuses on strategies that perform well across a wide range of possible futures.
  • Climate Informed Decision Analysis: Integrates climate projections into decision-making, emphasizing the need for expert judgment alongside models.

• Key Recommendations:

  • No single methodology is universally optimal; instead, a menu of approaches is needed.
  • Decision-makers should consider the project's lifetime, cost, and vulnerability when choosing a methodology.
  • Complementing model results with expert knowledge is crucial, as models may share biases and uncertainties.

Key Information

Uncertainty Sources

• Policy Uncertainty: Linked to future greenhouse gas emissions, which depend on demographic, socio-economic, technological, and policy factors.
• Epistemic Uncertainty: Arises from our incomplete understanding of the climate system and its interactions with other systems.
• Aleatory Uncertainty: Due to natural variability and the chaotic nature of the climate system.

Downscaling Techniques

• Statistical Downscaling: Uses historical data to relate large-scale climate drivers to local phenomena. It is computationally efficient but may not account for future climate changes accurately.
• Regional Climate Models (RCMs): Provide higher resolution but are still influenced by global climate models (GCMs) and cannot resolve all uncertainties related to natural variability.

Sectoral Exposure

• Several sectors are highly exposed to climate change and require long-term planning:
  • Water infrastructures: 30–200 years, high exposure.
  • Land-use planning: >100 years, high exposure.
  • Coastline and flood defences: >50 years, high exposure.
  • Building and housing: 30–150 years, moderate exposure.
  • Transportation infrastructure: 30–200 years, moderate exposure.
  • Urbanism: >100 years, moderate exposure.
  • Energy production: 20–70 years, moderate exposure.

Limitations of Climate Models

• Climate models are not deterministic and cannot forecast future climate with certainty.
• The uncertainty in model outputs is influenced by natural variability, model biases, and emission scenarios.
• Ensembles of simulations are needed to distinguish between natural variability and model-based uncertainty.

Expert Judgment and Decision-Making

• Expert knowledge is essential in interpreting model results, especially when the future may lie outside the projected range.
• Subjective probabilities based on expert judgment are often necessary in the absence of reliable model-based probabilities.

Conclusion

• The paper concludes that while climate change introduces significant uncertainty, it is not insurmountable.
• A combination of methodologies and expert knowledge is necessary for effective decision-making.
• Capacity building in developing countries, such as through local expertise centers, is more efficient than relying solely on costly downscaling exercises.
• The paper emphasizes the importance of robustness in decision-making, especially for long-term and high-impact investments.