Forecasting flooding from Cyclone Idai
Figure 1. NASA Earth Observatory image by Lauren Dauphin, using VIIRS data from the Suomi National Polar-orbiting Partnership.
Cyclone Idai was one of the worst weather disasters to hit the southern hemisphere, impacting almost 3 million people across Mozambique, Malawi and Zimbabwe during March 2019. We used our high-resolution modelling and forecasting framework to predict the flood impacts of the cyclone up to 7 days in advance.
The framework consists of a hydrological model that runs in near real-time, and is updated every day to estimate runoff and streamflow at 5km resolution and for a high-resolution river network across the region. The model is run forward in time using weather forecasts from the Global Ensemble Forecast System (GEFS) to estimate an ensemble of 10-day runoff and streamflow forecasts. The GEFS weather forecasts are adjusted to remove biases and downscaled to 5km resolution.
The model was able to forecast the spatial extent of the extreme streamflow conditions up to 7 days in advance (see Figure 2 below). This lead-time could provide governments with early warning to enable risk reduction measures to be put in place.
Figure 2. Monitoring (right) and forecasting (left) the flooding associated with Cyclone Idai.
