Search

Abstract The Northwest European shelf experienced unprecedented surface temperature anomalies in June 2023 (anomalies up to 5 °C locally, north of Ireland). Here, we show the shelf average underwent its longest recorded category II marine heatwave (16 days). With state-of-the-art observation and modelling capabilities, we show the marine heatwave developed quickly due to strong atmospheric forcing (high level of sunshine, weak winds, tropical air) and weak wave activity under anticyclonic weather regimes. Once formed, this shallow marine heatwave fed back on the weather: over the sea it reduced cloud cover and over land it contributed to breaking June mean temperature records and to enhanced convective rainfall through stronger, warmer and moister sea breezes. This marine heatwave was intensified by the last 20-year warming trend in sea surface temperatures. Such sea surface temperatures are projected to become commonplace by the middle of the century under a high greenhouse gas emission scenario.

Abstract The urban heat island (UHI) effect exacerbates near‐surface air temperature (T) extremes in cities, with negative impacts for human health, building energy consumption and infrastructure. Using conventional weather models, it is both difficult and computationally expensive to simulate the complex processes controlling neighbourhood‐scale variation of T. We use machine learning (ML) to bias correct and downscale T predictions made by the Met Office operational regional forecast model (UKV) to 100 m horizontal grid length over London, UK. A set of ML models (random forest, XGBoost, multiplayer perceptron) are trained using citizen weather station observations and UKV variables from eight heatwaves, along with high‐resolution land cover data. The ML models improve the T mean absolute error (MAE) by up to 0.12°C (11%) relative to the UKV. They also improve the UHI diurnal and spatial representation, reducing the UHI profile MAE from 0.64°C (UKV) to 0.15°C. A multiple linear regression performs almost as well as the ML models in terms of T MAE, but cannot match the UHI bias correction performance of the ML models, only reducing the UHI profile MAE to 0.49°C. UKV latent heat flux is found to be the most important predictor of T bias. It is demonstrated that including more heatwaves and observation sites in training would reduce overfitting and improve ML model performance.