Existing weather models are known to have poor skill at forecasting rainfall over East Africa, where there are regular threats of drought and floods. Improved forecasts could reduce the effects of these extreme weather events and provide significant socioeconomic benefits to the region. We present a novel machine learning-based method to improve precipitation forecasts in East Africa, using postprocessing based on a conditional generative adversarial network (cGAN). This addresses the challenge of realistically representing tropical rainfall, where convection dominates and is poorly simulated in conventional global forecast models. We postprocess hourly forecasts made by the European Centre for Medium-Range Weather Forecasts Integrated Forecast System at 6-18h lead times, at $0.1^{\circ}$ resolution. We combine the cGAN predictions with a novel neighbourhood version of quantile mapping, to integrate the strengths of machine learning and conventional postprocessing. Our results indicate that the cGAN substantially improves the diurnal cycle of rainfall, and improves predictions up to the $99.9^{\text{th}}$ percentile ($\sim 10 \text{mm}/\text{hr}$). This improvement extends to the 2018 March–May season, which had extremely high rainfall, indicating that the approach has some ability to generalise to more extreme conditions. We explore the potential for the cGAN to produce probabilistic forecasts and find that the spread of this ensemble broadly reflects the predictability of the observations, but is also characterised by a mixture of under- and over-dispersion. Overall our results demonstrate how the strengths of machine learning and conventional postprocessing methods can be combined, and illuminate what benefits machine learning approaches can bring to this region.

Observed rapid Arctic warming and sea-ice loss are likely to continue in the future, unless and after greenhouse gas emissions are reduced to net-zero. Here, we examine the possible effects of future sea-ice loss at 2°C global warming above pre-industrial levels on winter temperature extremes across the Northern Hemisphere, using coordinated experiments from the Polar Amplification Model Intercomparison Project. 1-in-20-year cold extremes are simulated to become less severe at high- and mid-latitudes in response to Arctic sea-ice loss. 1-in-20-year winter warm extremes become warmer at northern high latitudes due to sea-ice loss, but warm by less than cold extremes. We compare the response to sea-ice loss to that from global SST change also at 2°C global warming. SST change causes less severe cold extremes and more severe warm extremes globally. Except northern high latitudes, the response to SST change is of larger magnitude than that to Arctic sea-ice loss.

Recent extreme weather in the UK highlights the need to understand the potential for more extreme events in the present-day, and how such events may change with global warming. We present a methodology for more efficiently sampling extremes in future climate projections. As a proof-of-concept, we use the UK’s most recent set of national Climate Projections (UKCP18). UKCP18 includes a 15-member perturbed parameter ensemble (PPE) of coupled global simulations, providing a range of climate projections incorporating uncertainty in both internal variability and forced response. However, this ensemble is too small to adequately sample extremes with return periods over 100 years, which are of interest to policy-makers and adaptation planners. To better understand the statistics of these events, we use distributed computing to run three ~1000-member initial-condition ensembles with the atmosphere-only HadAM4 model at 60km resolution on volunteers’ computers, taking boundary conditions from future extreme winters within the UKCP18 ensemble. We find that every UKCP18 extreme winter is captured within our ensembles, and that two of the three ensembles are conditioned towards producing extremes by the boundary conditions. Our ensembles contain several extremes that would only be expected to be sampled by a UKCP18 PPE of over 500 members, which would be prohibitively expensive with current supercomputing resource. The most extreme winters simulated lie above those for UKCP18 by 0.85K for daily maximum temperature and 37% of the present-day average for precipitation (UK winter means).

Rapid Arctic warming and decline in sea ice have been observed in recent decades. These trends will likely continue, potentially changing winter extremes elsewhere in the Northern Hemisphere. We use coordinated Polar Amplification Model Intercomparison Project (PAMIP) experiments to decompose the Northern Hemisphere winter cold temperature responses to future Arctic sea-ice loss and sea surface temperature (SST) change, separately, at 2C global mean warming. Cold extremes (20-year return period) will generally become warmer at high- and mid-latitudes due to Arctic sea-ice loss, with the largest warming in East Canada. SST change will warm cold extremes everywhere, overwhelming simulated sea ice-induced cooling responses in, e.g., southwestern United States. In general, the SST-induced changes dominate over sea ice-induced changes, with exceptions in East Canada, Nunavut (Canada) and North Pacific Russia. Our results suggest that if climate models do not adequately capture the sea-ice and SST components, cold extremes will be biased.

The simultaneous occurrence of extremely wet winters at multiple locations in the same region can contribute to widespread flooding and associated socio-economic losses. However, the change in the spatial extent of precipitation extremes is largely overlooked in climate change assessments. Employing new multi-thousand-year climate model simulations, we show that under both 2.0°C and 1.5°C warming scenarios, wintertime precipitation extreme extents would increase over about 80-90% of the Northern Hemisphere. Stabilising at 1.5°C rather than 2°C would reduce the average magnitude of the increase by 1.6-2 times. According to the climate model, the increased extents are caused by increases in precipitation intensity increasing rather than changes in the spatial organisation of the events. Relatively small percentage increases in precipitation intensities (e.g., by 4%) can drive disproportionately larger, by 1-2 orders of magnitude, growth in the spatial extents (by 97%).