A188-0018
Real-time Forecast Evaluation of ECMWF Atmospheric Model Ensemble Exntended over Germany in 2020
Real-time Forecast Evaluation of ECMWF Atmospheric Model Ensemble Exntended over Germany in 2020
Tuesday, 15 December 2020
Poster
Abstract:
Several regions in Germany are facing an abnormal to exceptional drought
according to UFZ German Drought Monitor since the beginning of 2020
(https://www.ufz.de/droughtmonitor). An early warning of
evolving and significant drought can help to adapt to these extreme
hydro-climatic events. In this study, we investigate how subseasonal
forecasts issued by the European Centre for Medium-Range Weather
Forecasts (ECMWF) can help to predict drought events in Germany. We aim
to investigate precipitation and maximum 2-m air temperature forecasts
over Germany in real-time. The extended-range individual forecast runs
from the ECMWF latest atmospheric model (ENS extended) are pre-processed
and evaluated for every forecasts initialised between mid March and
November 2020. The observational reference is derived from a
high-density station network collected by German Weather Service (DWD)
interpolated to native model resolution by applying External Drift
Krigging (EDK). Trend preserving bias correctiuon will be also perform
to allow the comparison with DWD observations and further use of the
forcings in hydrological forecast. Several deterministic and
probabilistic verification metrics will be evaluated to understand the
model performance in lead-times of zero to three weeks. In addition,
some drought measures including consecutive dry days will be calculated
to analyse the performance of ECMWF model outputs in real-time. The
forecast information and their validations are generated as a component
of a hydroclimate forecasting systems developed by UFZ for the Modular
Observation Solutions for Earth Systems- MOSES project
(https://www.ufz.de/moses/). Real-time forecast products will
inform MOSES researchers to plan campaigns to investigate the evolution
of heatwaves and drought events during the next years.
according to UFZ German Drought Monitor since the beginning of 2020
(https://www.ufz.de/droughtmonitor). An early warning of
evolving and significant drought can help to adapt to these extreme
hydro-climatic events. In this study, we investigate how subseasonal
forecasts issued by the European Centre for Medium-Range Weather
Forecasts (ECMWF) can help to predict drought events in Germany. We aim
to investigate precipitation and maximum 2-m air temperature forecasts
over Germany in real-time. The extended-range individual forecast runs
from the ECMWF latest atmospheric model (ENS extended) are pre-processed
and evaluated for every forecasts initialised between mid March and
November 2020. The observational reference is derived from a
high-density station network collected by German Weather Service (DWD)
interpolated to native model resolution by applying External Drift
Krigging (EDK). Trend preserving bias correctiuon will be also perform
to allow the comparison with DWD observations and further use of the
forcings in hydrological forecast. Several deterministic and
probabilistic verification metrics will be evaluated to understand the
model performance in lead-times of zero to three weeks. In addition,
some drought measures including consecutive dry days will be calculated
to analyse the performance of ECMWF model outputs in real-time. The
forecast information and their validations are generated as a component
of a hydroclimate forecasting systems developed by UFZ for the Modular
Observation Solutions for Earth Systems- MOSES project
(https://www.ufz.de/moses/). Real-time forecast products will
inform MOSES researchers to plan campaigns to investigate the evolution
of heatwaves and drought events during the next years.