Mesoscale convective systems (MCSs) are difficult to forecast due to their inherent‐60 unpredictability and development from scales that are subgrid in typical global models. Here the impacts of model representation of convection on MCS structure and downstream forecast evolution are examined using two configurations of the Met Office Unified Model: the convection‐permitting (4.4‐km grid spacing) limited‐area Euro4 and convection‐parametrizing (25‐km grid spacing) Global configurations. MCSs are associated with a characteristic potential vorticity (PV) structure: a positive PV anomaly in the mid‐troposphere and negative PV anomalies above and to the side of it. Convection‐permitting models produce larger‐amplitude MCS PV anomalies than convection‐parametrizing models. These differences are shown to persist after coarse graining the output from a Euro4 simulation to the 25 km grid spacing of the Global configuration for a case study from July 2012, and are largest in magnitude and extent in the upper troposphere. The effect of the poor representation of this PV structure by convection‐ parametrizing models on forecasts is investigated by adding “MCS perturbations”, calculated as differences between the coarse‐grained Euro4 and the Global outputs, to five‐day Global configuration forecasts. Upper‐level MCS perturbations lead to greater forecast differences than those at middle levels, though using perturbations at all levels yields the greatest impact. For the first 36 hr, differences grow on the convective scale related to the MCS and its influence on a developing UK cyclone, despite perturbation amplitudes initially reducing. Subsequently, differences grow rapidly onto the synoptic scale and by five days impact the entire Northern Hemisphere. MCS perturbations slow the eastward movement of Rossby waves due to ridge amplification. Thus, perturbing convection‐parametrizing models to include PV anomalies associated with MCSs produces synoptic‐scale forecast differences implying that the misrepresentation of the PV structures associated with MCSs are a potential source of forecast errors.

Convection-permitting ensemble prediction systems (CP-ENS) have been implemented in the mid-latitudes for weather forecasting timescales over the past decade, enabled by the increase in computational resources. Recently, efforts are being made to study the benefits of CP-ENS for tropical regions. This study examines CP-ENS forecasts produced by the UK Met Office over tropical East Africa, for 24 cases in the period April-May 2019. The CP-ENS, an ensemble with parametrized convection (Glob-ENS), and their deterministic counterparts are evaluated against rainfall estimates derived from satellite observations (GPM-IMERG). The CP configurations have the best representation of the diurnal cycle, although heavy rainfall amounts are overestimated compared to observations. Pairwise comparisons between the different configurations reveal that the CP-ENS is generally the most skilful forecast for both 3-h and 24-h accumulations of heavy rainfall (97th percentile), followed by the CP deterministic forecast. More precisely, probabilistic forecasts of heavy rainfall, verified using a neighbourhood approach, show that the CP-ENS is skilful at scales greater than 100 km, significantly better than the Glob-ENS, although not as good as found in the mid-latitudes. Skill decreases with lead time and varies diurnally, especially for CP forecasts. The CP-ENS is under-spread both in terms of forecasting the locations of heavy rainfall and in terms of domain-averaged rainfall. This study demonstrates potential benefits in using CP-ENS for operational forecasting of heavy rainfall over tropical Africa and gives specific suggestions for further research and development, including probabilistic forecast guidance.

Africa is poised for a revolution in the quality and relevance of weather predictions, with potential for great benefits in terms of human and economic security. This revolution will be driven by recent international progress in nowcasting, numerical weather prediction, theoretical tropical dynamics and forecast communication, but will depend on suitable scientific investment being made. The commercial sector has recognized this opportunity and new forecast products are being made available to African stakeholders. At this time, it is vital that robust scientific methods are used to develop and evaluate the new generation of forecasts. The GCRF African SWIFT project represents an international effort to advance scientific solutions across the fields of nowcasting, synoptic and short-range severe weather prediction, subseasonal-to-seasonal (S2S) prediction, user engagement and forecast evaluation. This paper describes the opportunities facing African meteorology and the ways in which SWIFT is meeting those opportunities and identifying priority next steps.Delivery and maintenance of weather forecasting systems exploiting these new solutions requires a trained body of scientists with skills in research and training; modelling and operational prediction; communications and leadership. By supporting partnerships between academia and operational agencies in four African partner countries, the SWIFT project is helping to build capacity and capability in African forecasting science. A highlight of SWIFT is the coordination of three weather-forecasting “Testbeds” – the first of their kind in Africa – which have been used to bring new evaluation tools, research insights, user perspectives and communications pathways into a semi-operational forecasting environment.

Ensemble forecasts are run operationally to determine the forecast uncertainty arising from initial condition, model physics and boundary condition uncertainty. However, global configuration ensembles, which use a convection parametrization scheme, may miss uncertainty because of the misrepresentation of intense convection by such schemes. Here, the impacts of the misrepresentation of mesoscale convective systems (MCSs) on downstream ensemble forecast skill and evolution are determined for a case study. MCS perturbations (calculated‐15 from the difference between outputs from convection‐parametrizing and convection‐permitting Met Office model configurations) are added to six members of a global configuration ensemble created by downscaling forecasts from the global version of the Met Office Global and Regional Ensemble Prediction System. For the first 36 hours, differences grow on the convective scale related to the MCSs, leading to systematic deepening of a developing UK cyclone, although there is damping of the perturbations found in root mean square difference calculations between the forecasts with and without the perturbations (particularly in mean sea level pressure). Subsequently, differences grow rapidly to the synoptic scale, and by five days impact the entire Northern Hemisphere. The MCS perturbations can have systematic effects on ensemble forecasts (e.g., a systematic displacement of a downstream cyclone is found), but, for this case, there is no discernible change in forecast skill as measured by the root mean square error of the ensemble means and the effects of the MCS perturbations are smaller than those generated by the initial condition perturbations. The spread of the combined ensemble (the two ensembles with and without the MCS perturbations) is larger than that of the individual ensembles. Thus, perturbing convection‐parametrizing models to include potential vorticity anomalies associated with MCSs represented in convection‐permitting forecasts, or idealized representations of them, produces alternative realizations from those generated by initial condition perturbations and has the potential to be useful operationally.

<p>Convection-permitting ensemble prediction systems (CP-ENS) have been implemented in the<br>mid-latitudes for weather forecasting timescales over the past decade, enabled by the increase in<br>computational resources. Recently, efforts are being made to study the benefits of CP-ENS for<br>tropical regions. This study examines CP-ENS forecasts produced by the UK Met Office over<br>tropical East Africa, for 24 cases in the period April-May 2019. The CP-ENS, an ensemble with<br>parametrized convection (Glob-ENS), and their deterministic counterparts are evaluated against<br>rainfall estimates derived from satellite observations (GPM-IMERG). The CP configurations have<br>the best representation of the diurnal cycle, although heavy rainfall amounts are overestimated<br>compared to observations. Pairwise comparisons between the different configurations reveal that<br>the CP-ENS is generally the most skilful forecast for both 3-h and 24-h accumulations of heavy<br>rainfall (97th percentile), followed by the CP deterministic forecast. More precisely, probabilistic<br>forecasts of heavy rainfall, verified using a neighbourhood approach, show that the CP-ENS is<br>skilful at scales greater than 100 km, significantly better than the Glob-ENS, although not as good<br>as found in the mid-latitudes. Skill decreases with lead time and varies diurnally, especially for<br>CP forecasts. The CP-ENS is under-spread both in terms of forecasting the locations of heavy<br>rainfall and in terms of domain-averaged rainfall. This study demonstrates potential benefits in<br>using CP-ENS for operational forecasting of heavy rainfall over tropical Africa and gives specific<br>suggestions for further research and development, including probabilistic forecast guidance.</p>