Flying over Lake Victoria: first aircraft observations of the lake–land breeze circulation

By Beth Woodhams for GCRF African SWIFT

Flying an aircraft just 30m above the world’s most dangerous lake may sound more like a daredevil stunt than a scientific experiment, but the HyVic pilot flight campaign did just that! Using a special aircraft kitted out with scientific instruments, SWIFT scientists flew over Lake Victoria—a Lake in East Africa known for its frequent and deadly night-time storms—to measure temperature, wind and moisture over the lake in unprecedented detail. These first ever detailed data of the atmosphere  above the lake, collected by this ‘flying laboratory’, are now helping to improve understanding of why storms form on this lake, why they are so difficult to predict with weather models and the implications for climate change projections.

Figure 1. (Left) A satellite view of Lake Victoria in East Africa. The lake itself is approximately the size of the Republic of Ireland. (Right) The FAAM aircraft in flight (Image Credit:

The HyVic flights were possible due to an opportunity provided by the NERC MOYA project, and by collaboration and coordination between projects and institutions: flights and analysis were funded by the WMO-led HIGHWAY project of WISER through UKAID; HyCRISTAL; NCAS; and the UK Met Office, alongside GCRF African SWIFT. The campaign was facilitated by the Uganda National Meteorological Authority.

The importance of forecasting storms over Lake Victoria

Lake Victoria, the world’s largest tropical lake, is a hotspot for severe convective storms which contribute to an estimated 5,000 fatalities every year and generate severe flooding in coastal cities. The lake storms are not only hazardous to those on the lake but are the major source of water to the lake. The lake and storms also modify rainfall patterns over hundreds of kilometres surrounding the lake, affecting the 40 million people who live in the lake basin.

However, forecasting severe weather in East Africa remains a great challenge for weather models (numerical weather prediction), despite the introduction of new cutting-edge modelling by the UK Met Office (convection-permitting (CP) forecasts; Chamberlain et al., 2014; Woodhams et al., 2018). Forecasting is also hampered by a lack of knowledge of processes controlling the lake storms, in part due to the difficulty of obtaining observations over the lake itself.

As climate change is expected to increase the intensity of storms, there is a growing need for improved predictions. Future changes in rainfall are also expected to affect lake water levels, with impacts on coastal cities, hydroelectric power and ecosystems. Improved understanding of the storms therefore benefits not only weather forecasts, but also the climate predictions needed for climate change adaptation for this rapidly changing population centre. 

A lake–land breeze circulation—driven by temperature gradients between the lake and land—has previously been shown to play a role in the frequent nocturnal storms over Lake Victoria, although past studies generally looked at the mean diurnal cycle of storms in coarse resolution models (e.g., Song et al., 2004; Anyah et al., 2006). Detailed observations of the system, in particular over the lake itself, are lacking. One previous study associated with SWIFT used high-resolution simulations to investigate the lake–land breeze circulation and associated storms for individual case studies (Woodhams et al., 2019). This study confirmed the importance of the lake–land breeze circulation, but indicated that the circulation and associated convection may be modified under different synoptic conditions. A bulge of moist air forming above the lake surface overnight and extending from the surface to ~1.5 km was identified in the simulations (Figure 2). It was hypothesised that the properties of this feature may affect storm formation. However, model simulations—even at high-resolution—are far from perfect and observations are required to substantiate any findings. In situ observations around Lake Victoria are limited and upper-air observations over the lake itself present an exceptional challenge.

Figure 2. Schematic of the formation of the moisture bulge over Lake Victoria, taken from Woodhams et al. (2019)

Planning and executing the flights

In January 2019, a team of scientists from the University of Leeds and the UK Met Office travelled to Entebbe, Uganda for the HyVic pilot flight campaign, with the goal of collecting unprecedented observations of the Lake Victoria lake–land breeze circulation using the Facility for Airborne Atmospheric Measurements (FAAM) BAe-146 aircraft. The campaign consisted of an evening and morning flight to sample the lake and land breeze components of the circulation respectively. In particular, these flights aimed to investigate some of the features simulated in Woodhams et al. (2019), such as the moisture bulge; characterise the lake and land breeze fronts; and collect observations to be used for model verification.

Figure 3. Team HyVic on board the FAAM Aircraft

Figure 4. Team HyVic studying aeronautical charts of the Lake Victoria Basin based on the latest weather forecasts

For some of the scientists, this was the first time taking part in a field campaign, let alone guiding a huge aircraft. Talking with pilots, instrument scientists, local meteorological services and experienced FAAM users allowed the HyVic scientists to create their flight plans. In the days and hours before the flight, it was important to monitor the weather forecasts and adjust plans accordingly. On board the flights themselves, the HyVic team worked in the roles of ‘mission scientists’, monitoring the live data feeds and communicating flight adjustments to the pilots based on the available data.

New insights and looking to the future

The flight campaign was a success, observing the lake breeze front (a transition zone between the cool, moist lake air and warmer, drier air over land) in unprecedented detail. The front was shown to have a horizontal extent of ~5km, which has implications for its representation even in current ‘high-resolution’ models. At least one region of elevated moisture, coincident with cloud and increased turbulence, was observed over the lake surface during the early morning. Observations suggested that the bulge had some similar attributes to that simulated in Woodhams et al. (2019), formed of uplifted boundary-layer air as a result of near-surface convergence. Observations were also compared to bespoke 300 m resolution simulations using a convection-permitting configuration of the Met Office Unified Model, which successfully simulated the lake breeze and aspects of the morning circulation, but showed significant differences in the vertical extent of the boundary layer.

Overall, this campaign has provided a unique observational dataset which is enhancing our understanding of the dynamics of Lake Victoria’s lake–land breeze circulation. By providing evidence of the key processes numerical weather prediction models must represent to provide more skilful storm forecasts, this work will ultimately lead to improvements in forecasts and safety on the lake. Centres such as the Met Office employ a ‘seamless’ modelling strategy, which means that this analysis will pull through to our understanding of state-of-the-art climate change simulations, such as those recently run by the Met Office for Africa (Senior et al., 2021). The new data have brought new questions and so the success of HyVic with just two flight provides a mandate for an extended field campaign in the future.

Figure 5. A view of a storm approaching Lake Victoria at Mwanza, Tanzania

Paper Information

Woodhams, B.J., Barrett, P.A., Marsham, J.H., Birch, C.E., Bain, C.L., Fletcher, J.K., Hartley, A.J., Webster, S., Mangeni, S. (2021). Aircraft observations and sub-km modelling of the lake–land breeze circulation over Lake Victoria. Quarterly Journal for the Royal Meteorological Society, 148(743), 557-580.


Anyah, R. O., Semazzi, F. H., & Xie, L. (2006). Simulated physical mechanisms associated with climate variability over Lake Victoria basin in East Africa. Monthly weather review134(12), 3588-3609.

Chamberlain, J. M., Bain, C. L., Boyd, D. F. A., McCourt, K., Butcher, T., & Palmer, S. (2014). Forecasting storms over Lake Victoria using a high resolution model. Meteorological Applications21(2), 419-430.

Flohn, H., & Fraedrich, K. (1966). Tagesperiodische zirkulation und niederschlagsverteilung am Victoria-See (Ostafrika). Meteorologische Rundschau19(6), 157-165.

Semazzi, F. H. M. (2011). Enhancing safety of navigation and efficient exploitation of natural resources over Lake Victoria and its basin by strengthening meteorological services on the lake. North Carolina State University Climate Modeling Laboratory Tech. Rep.

Senior, C. A., Marsham, J. H., Berthou, S., Burgin, L. E., Folwell, S. S., Kendon, E. J., … & Willet, M. R. (2021). Convection-Permitting Regional Climate Change Simulations for Understanding Future Climate and Informing Decision-Making in Africa. Bulletin of the American Meteorological Society102(6), E1206-E1223.

Song, Y., Semazzi, F. H., Xie, L., & Ogallo, L. J. (2004). A coupled regional climate model for the Lake Victoria basin of East Africa. International Journal of Climatology: A Journal of the Royal Meteorological Society24(1), 57-75.

Woodhams, B. J., Birch, C. E., Marsham, J. H., Bain, C. L., Roberts, N. M., & Boyd, D. F. (2018). What is the added value of a convection-permitting model for forecasting extreme rainfall over tropical East Africa?. Monthly Weather Review146(9), 2757-2780.

Woodhams, B. J., Birch, C. E., Marsham, J. H., Lane, T. P., Bain, C. L., & Webster, S. (2019). Identifying key controls on storm formation over the Lake Victoria basin. Monthly Weather Review147(9), 3365-3390.

Woodhams, B. J., Barrett, P. A., Marsham, J. H., Birch, C. E., Bain, C. L., Fletcher, J. K., … & Mangeni, S. (2021). Aircraft observations of the lake‐land breeze circulation over Lake Victoria. Quarterly Journal of the Royal Meteorological Society.