Abstract:
Heatwaves are large deviations of near-surface temperatures from the climatology, over a period of around a week to sometimes as long as a month. In addition to severely impacting human health, heatwaves also affect the economy and infrastructure. In the current warming scenario, heatwaves have been predicted to become more common due to projected changes in land surface properties and the large-scale circulation due to warming.
In this project, we study the effects of land surface properties and large-scale warming on heatwaves. We do this by setting up multiple model configurations with different land surface properties and large-scale warming. To sample heatwaves in a computationally efficient manner, we utilise the GKLT algorithm, a rare event sampling algorithm. Using the GKLT algorithm, we calculate the return times of extreme heatwaves. We find that the return times of temperature extremes change as the land surface properties and the large-scale circulation are changed. We see that the shapes of the extreme tails of the temperature distribution change between the different model configurations, leading to the changes in the return times observed. Finally, we perform a statistical analysis on the heatwaves sampled by the algorithm for the different configurations and study how the changes in return times can be understood in terms of the intensity, duration and number of heatwaves. We find that for the land surface configurations, the return times of the time-averaged temperature anomalies decrease uniformly across anomalies as the relative humidity over land is decreased. For the warming configurations, the return times of the time-averaged temperature anomalies decrease predominantly for the larger anomaly values as the atmosphere is warmed, while the return times of the lower anomaly values remain close to that of the baseline.
