Evaluation of UK ERA5 Sea Surface Temperature for Extreme Value Analysis

Oral Presentation 

Sea surface temperature (SST) is defined as the temperature of the sea in the top 10 meters from the surface. Coastal nuclear power plants (NPPs) often use sea water for cooling and hence extreme high SST could reduce the efficiency of their cooling systems. Higher SST also has other impacts on NPPs, including an increased risk of overheating the plant. This would affect safe operations, the risk of heat-related equipment failures from increased heat stress, and thermal pollution from discharge compounding with pre-existing high SST potentially damaging marine ecosystems. Suitable observational SST datasets for extreme value analysis (EVA) can be scarce, and so spatially and temporally long-term reanalysis datasets, such as ERA5 [1], are explored for potential EVA application for NPPs. Robust extreme value analysis (EVA) allows EDF to build resilience to climate change and help the UK achieve Net Zero. Here we present an evaluation of the ERA5 reanalysis SST dataset against observational datasets recorded by Datawell Directional Waverider Mk III buoys off the English coast obtained from the National Network of Regional Coastal Monitoring Programmes [2]. For this purpose the ERA5 timeseries are cut to the period of the observations (2003-2022). We also use a full ERA5 timeseries (1959-present) when evaluating EVA outputs. The datasets are further evaluated by comparing block-maxima EVA results, using annual maximum SST sampled from the daily SST timeseries. Bias trends vary on a seasonal and interannual scale. ERA5 presents lower summer values in comparison to the observations. The observational datasets produce higher best estimate return levels than ERA5 datasets, despite the availability of longer time periods in ERA5 and considerable variability in the model parameters between datasets. The implications of this work will be discussed from an energy sector perspective.

[1] H. Hersbach et al, “The ERA5 global reanalysis,” Q J R Meteorol Soc, vol. 146, pp. 1999-2049, 2020. 

[2] Channel Coastal Observatory, “Southeast Regional Coastal Monitoring Programme”, copyright: New Forest District Council, 2023, Available: https://coastalmonitoring.org [Accessed 31 10 2023]