 |
| Titel |
The skill of ECMWF long range Forecasting System to drive impact models for health and hydrology in Africa |
| VerfasserIn |
F. Di Giuseppe, A. M. Tompkins, R. Lowe, E. Dutra, F. Wetterhall |
| Konferenz |
EGU General Assembly 2012
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 14 (2012) |
| Datensatznummer |
250063043
|
|
|
|
|
|
| Zusammenfassung |
| As the quality of numerical weather prediction over the monthly to seasonal leadtimes
steadily improves there is an increasing motivation to apply these fruitfully to the impacts
sectors of health, water, energy and agriculture. Despite these improvements, the accuracy of
fields such as temperature and precipitation that are required to drive sectoral models can still
be poor. This is true globally, but particularly so in Africa, the region of focus in
the present study. In the last year ECMWF has been particularly active through
EU research founded projects in demonstrating the capability of its longer range
forecasting system to drive impact modeling systems in this region. A first assessment
on the consequences of the documented errors in ECMWF forecasting system is
therefore presented here looking at two different application fields which we found
particularly critical for Africa - vector-born diseases prevention and hydrological
monitoring.
A new malaria community model (VECTRI) has been developed at ICTP and tested for
the 3 target regions participating in the QWECI project. The impacts on the mean
malaria climate is assessed using the newly realized seasonal forecasting system
(Sys4) with the dismissed system 3 (Sys3) which had the same model cycle of the
up-to-date ECMWF re-analysis product (ERA-Interim). The predictive skill of
Sys4 to be employed for malaria monitoring and forecast are also evaluated by
aggregating the fields to country level. As a part of the DEWFORA projects, ECMWF
is also developing a system for drought monitoring and forecasting over Africa
whose main meteorological input is precipitation. Similarly to what is done for the
VECTRI model, the skill of seasonal forecasts of precipitation is, in this application,
translated into the capability of predicting drought while ERA-Interim is used in
monitoring. On a monitoring level, the near real-time update of ERA-Interim could
compensate the lack of observations in the regions. However, ERA-Interim suffers from
biases and drifts that limit its application for drought monitoring purposes in some
regions. |
|
|
|
|
|
|