No abstract

No abstract

Demand for information about physical climate risk is growing, particularly for the power generation sector, given its size and pronounced exposure to climate hazards. However, quantifying physical climate risks for a large number of assets remains challenging. Here we introduce a scalable and transparent methodology that enables multi-hazard physical climate risk assessments for any thermal or hydro power generation project. The methodology relies on basic power plant type and geolocation data inputs, publicly-available climate datasets, and hazard- and technology-specific vulnerability factors, to translate hazard severity into generation losses. We apply the methodology to the European Bank for Reconstruction and Development’s early 2021 thermal and hydro power generation portfolios of 80 assets. We show that under the Representative Concentration Pathway 4.5 scenario, those 80 power plants could experience a 4.0-10.9 TWh loss in annual generation (or 1.87-5.07% of total annual maximum generation) by 2030 compared to its baseline losses of 0.70–0.87 TWh (or 0.33–0.41%). One of the largest drivers of the increased risk is rising water temperatures, which is currently overlooked by mainstream climate risk disclosure guidelines.

This paper introduces a new method to quantify physical climate risks for power generation projects at the portfolio level. Co-developed by WRI and the European Bank for Reconstruction and Development (EBRD), the approach is designed to be flexible enough to work with portfolios with different levels of data availability, leverage the latest science in climate and hydrology, and use machine-learning techniques such as recurrent neural networks.