The performance of the HadGEM3 as seasonal forecasting and research tool for southern African climate variability.
The predictability of southern African climate variability will be investigated by using the output from the HadGEM3 model. This model will be tested on its ability to predict the seasonal-to-interannual variability of the region at forecast lead-times that may benefit the users of such forecasts. The forecast products to be investigated are seasonal rainfall totals, surface temperatures as well as their extremes, and intra-seasonal characteristics. The HadGEM3 will also be included in the current multi-model forecast systems of the CSIR (Council for Scientific and Industrial Research), and its contribution towards the improvement of multi-model forecast skill over southern Africa will be assessed.
The operational forecast skill of the HadGEM3 will also be compared with local and international global models. Forecast comparisons will be conducted on raw model output, as well as on statistically corrected forecasts to station level using model output statistics (MOS). In addition, HadGEM3 data will be post-processed and analysed using linear and non-linear statistical analysis tools such as canonical correlation analysis (CCA) and self-organizing maps (SOMs). The ability of the HadGEM3 to produce weather and climate characteristics within the season will also be assessed. For example, how the model is able to simulate tropical-temperate cloud bands and other major synoptic systems over southern Africa.
The HadGEM3 as coupled model will also be employed to improve on the existing global SST forecasting systems. Such systems are of significant importance in the region since skilfully predicted SST anomalies will improve on the 2-tiered AGCMs currently being run operationally in the region.
Dr Landman has been working as a weather and climate scientist for 24 years. He is a Principal Researcher at the CSIR, where he also serves as a Fellow of the Natural Resources and the Environment Operational Unit. He specializes in the development of operational prediction systems for short-range weather to long-range seasonal time scales, and has published extensively in a good number of peer reviewed journals on prediction systems and on southern African climate variability. He has BSc, BSc (Honours) and MSc in Meteorology degrees from the University of Pretoria, and a PhD (Science) degree from the University of the Witwatersrand. He worked as a Postdoctoral Fellow for two years at the IRI. He is a Professor Extraordinary at the University of Pretoria, he serves as member of the CLIVAR Working Group on Seasonal to Interannual Prediction, and has a B-rating from the National Research Foundation of South Africa.
Areas of particular interest and expertise:
This research project aligns with CSRP Theme 1: Evaluation of HadGEM3 representation and predictability of climate variability
Willem Landman is working in collaboration with Jeff Knight of the Met Office.
Last updated: 12 May 2016