Previous work has pointed to the physical mechanisms behind the nocturnal offshore propagation of convection south-west of Sumatra. Low-level moisture flux convergence due to the land breeze front controls the progression of a squall line away from the coast overnight. However, the diurnal convection over the mountains occurs on only 57% of days in December-February (DJF) and propagates offshore on only 49% of those days. We investigate day-to-day variability in dynamical and thermodynamical conditions to explain the variability in diurnal convection and offshore propagation, using a convection-permitting simulation run for 900 DJF days. A convolutional neural network is used to identify regimes of diurnal cycle and offshore propagation behaviour. The diurnal cycle and offshore propagation are most likely to occur ahead of an active Madden-Julian Oscillation, or during El Niño or positive Indian Ocean Dipole; however, any regime can occur in any phase of these large-scale drivers, since the major control arises from the local scale. When the diurnal cycle of convection occurs, low-level wind is generally onshore, providing convergence over the mountains; and low-level humidity over the mountains is high enough to make the air column unstable for moist convection. When this convection propagates offshore, mid-level offshore winds provide a steering flow, combined with stronger convergence offshore due to the land breeze or convection-triggered cold pools. Low-level moisture around the coast also means that, as the convection propagates, the storm-relative inflow of air into the system adds greater instability than would be the case on other days.

Yutong Lu

and 4 more

Mesoscale convective systems (MCSs) are active in East China during the summer, causing significant precipitation and extreme weather. Increasing MCS frequency and intensity with climate change highlights the need for better simulation and forecasting. Traditional global and regional models with coarse resolution unable to explicitly resolve convection fail to represent MCSs and their precipitation accurately. This study conducted a 22-year (2000–2021) JJA simulation at a convection-permitting resolution (4 km grid spacing) using the WRF model (WRF-CPM) over East China. The WRF-CPM model’s ability to reproduce MCSs was evaluated against satellite infrared-retrieved cloud top temperature, IMERG V06 precipitation, and global reanalysis data ERA5. Results show that WRF-CPM captures the observed MCS frequency and precipitation patterns but overestimates them in most areas. The model also accurately simulates the eastward propagation of MCSs, albeit at a slightly faster speed and longer duration. MCSs in WRF-CPM exhibit realistic life cycles in terms of cloud top temperature, convective core area, and precipitation. WRF-CPM tends to overestimate rainfall frequency over 20 mm/h while underestimates rainfall per MCS, possibly due to an overestimated number and area. The model captures the diurnal cycle of MCSs well in most of East China, though it shows a 2-hour delay in southeast China and fails to reproduce the midnight peak to the east of Tibetan Plateau, probably because of model’s limited ability to represent thermal diurnal variation over complex topography. WRF-CPM captures the shear effect on MCS precipitation, indicating increased precipitation with stronger shear and higher total column water vapor.

Bobby Antonio

and 7 more

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.

Leo Saffin

and 6 more

The international field campaign for EUREC4A (Elucidating the role of clouds and circulation coupling in climate) gathered observations to better understand the links between trade-wind cumulus clouds, their organization, and larger scales, a large source of uncertainty in climate projections. A recent large-eddy simulation (LES) study showed a cloud transition that occurred during EUREC4A (2nd February 2020), where small shallow clouds developed into larger clouds with detrainment layers, was caused by an increase in mesoscale organization generated by a dynamical feedback in mesoscale vertical velocities. We show that kilometer-scale simulations with the Met Office Unified Model reproduce this increase in mesoscale organization and the process generating it, despite being much lower resolution. The simulations develop mesoscale organization stronger and earlier than the LES, more consistent with satellite observations. Sensitivity tests with a shorter spin-up time, to reduce initial organization, still have the same timing of development and sensitivity tests with cold pools suppressed show only a small effect on mesoscale organization. These results suggest that large-scale circulation, associated with an increased vertical velocity and moisture convergence, is driving the increase in mesoscale organization, as opposed to a threshold reached in cloud development. Mesoscale organization and clouds are sensitive to resolution, which affects changes in net radiation, and clouds still have substantial differences to observations. Therefore, while kilometer-scale simulations can be useful for understanding processes of mesoscale organization and links with large scales, including responses to climate change, simulations will still suffer from significant errors and uncertainties in radiative budgets.