P4 - Advanced seasonal forecasts for the Horn of Africa by optimized initialization and perturbation

Status: Ongoing

Duration: 2021 - 2025

Keywords: Meteorology, Physics, Earth System Science

Description

Seasonal forecasts are fundamental to prepare agriculture with respect to extreme events such as droughts or extreme precipitation. Unfortunately, the IPCC AR5 confirmed an enhanced vulnerability of Ethiopia with respect to these events. Recently, the global seasonal forecast systems such as GloSea5 (MacLachlan et al. 2015) and SEAS5 (Johnson et al. 2019) or have been considerably improved, for example due to the coupling with high-resolution ocean circulation models, advanced model physics for the representation of land-atmosphere (L-A) feedback, and better use of remote sensing data for model initialization. Nevertheless, deficiencies remain that can be related to limited spatial resolution and the lack of land surface initialization data with respect to soil moisture and vegetation states.

Within the first phase of CLIFOOD, a new nested, ensemble-based model system was set up, which operates over the Horn of Africa at much higher spatial resolution on the km-scale (Mori et al. 2019). First results demonstrate a promising improvement of the prediction of extreme precipitation statistics. Also, a new definition of climate regions over Ethiopia was developed and proposed, which will support the set-up, operation, verification, and dissemination of seasonal forecast projects (Ware et al. 2020, submitted). For this purpose, satellite-based precipitation products (CHIRPS) and the Ethiopian National Meteorological Agency (NMA) weather station data were extensively used. The NMA temperature data also confirmed the rising temperatures over Ethiopia in the recent decades with a typical increase of 2°C since 1990. So far, these activities were related to seasonal hindcasts, model design and verification as well as to the use of data sets in the past for climatological analyses. Now, it is very reasonable to continue and advance this research activity on seasonal forecasting within CLIFOOD in order to work towards real seasonal predictions with high spatial resolution.

We continue and refine the development of our seasonal forecast system in order to achieve the following objectives:

1) Improvement of the quality of the forecast products by advanced initialization and perturbation of land surface variables and parameters with satellite observations

2) Increase the speed of the simulation and the processing of the results towards seasonal predictions

3) Intensify educational activities at HU with respect to climate change, agro-meteorology, remote sensing as well as weather and climate physics

4) Develop a strategy to disseminate weather forecast and seasonal prediction results to farmers in strong collaboration with HU and NMA.

Involved persons

Prof. Dr. Volker Wulfmeyer, Dr. Kirsten Warrach-Sagi
Tamene Mekonnen Adgeh

Involved institutions

Institute of Physics and Meteorology, University of Hohenheim

Involved persons

Involved institutions

Sponsors