Current Problems in Medium Range Forecasting at ECMWF: Model Aspects

Abstract. For three small, mountainous catchments in Germany two medium-range forecast systems are compared that predict precipitation for up to 5 days in advance. One system is composed of the global German weather service (DWD) model, GME, which is dynamically downscaled using the COSMO-EU regional model. The other system is an empirical (expanded) downscaling of the ECMWF model IFS. Forecasts are verified against multi-year daily observations, by applying standard skill scores to events of specified intensity. All event classes are skillfully predicted by the empirical system for up to five days lead time. For the available prediction range of one to two days it is superior to the dynamical system.

Download Full-text

Recent studies on troposphere delay modeling for SLR

10.5194/egusphere-egu2020-9378 ◽

2020 ◽

Author(s):

Mateusz Drożdżewski ◽

Janina Boisits ◽

Florian Zus ◽

Kyriakos Balidakis ◽

Krzysztof Sośnica

Keyword(s):

Mapping Function ◽

Terrestrial Reference Frame ◽

Optical Observations ◽

Weather Forecasts ◽

Horizontal Gradients ◽

Medium Range ◽

Space Geodetic Techniques ◽

Combined Solution ◽

Parameterized Model ◽

Ecmwf Model

Recent studies on troposphere delay in Satellite Laser Ranging (SLR) show that the compliance of horizontal gradients of troposphere delay reduces the observation residuals, as well as improves the consistency between SLR results and other space geodetic techniques, all of which are essential for the realization of the terrestrial reference frame. In this work, we examine 3 novel approaches of troposphere delay modeling in SLR, with respect to the standard Mendes-Pavlis approach. We test Potsdam Mapping Function (PMF) with mapping function coefficients and linear horizontal gradients which are based on ERA5 reanalysis provided by the European Centre for Medium-Range Weather Forecasts (ECMWF) model with improved time and spatial resolution in comparison to ERA-Interim reanalysis. We also test a solution based on Vienna Mapping Function 3 for optical observations (VMF3o) which considers the separation of the mapping functions for hydrostatic and non-hydrostatic delays and horizontal gradients. Eventually, we test a solution based on Mendes &#8211; Pavlis model with a parameterized model for horizontal gradients based on the 16-year time series of horizontal gradients from PMF. To conduct this experiment, we use SLR observations to passive geodetic satellites LAGEOS-1 and LAGEOS-2. From differences of residual standard deviations for all proposed solutions, we observe an improvement of the SLR observation residuals, for low elevation angles above 10% and improvement of the consistency between estimated pole coordinates and the combined solution IERS-14-C04 series with respect to the currently recommended solutions that neglect the horizontal gradients in SLR solutions.

Download Full-text

PREDICTION OF CUMULONIMBUS (CB) CLOUD BASED ON INTEGRATED FORECAST SYSTEM (IFS) OF EUROPEAN MEDIUM-RANGE WEATHER FORECAST (ECMWF) IN THE FLIGHT INFORMATION REGION (FIR) OF JAKARTA AND UJUNG PANDANG

Jurnal Sains & Teknologi Modifikasi Cuaca ◽

10.29122/jstmc.v21i2.4100 ◽

2020 ◽

Vol 21 (2) ◽

Author(s):

Achmad Fahruddin Rais ◽

Fani Setiawan ◽

Rezky Yunita ◽

Erika Meinovelia ◽

Soenardi Soenardi ◽

...

Keyword(s):

Brightness Temperature ◽

Cloud Cover ◽

Convective Cloud ◽

Weather Forecast ◽

Forecast System ◽

Medium Range Weather Forecast ◽

Medium Range ◽

Ujung Pandang ◽

Categorical Verification ◽

Ecmwf Model

This study was focused on cumulonimbus (Cb) cloud prediction based on Integrated Forecast System (IFS) European Medium-Range Weather Forecast (ECMWF) model in the Flight Information Region (FIRs) Jakarta and Ujung Pandang. The Cb cloud prediction was calculated using convective cloud cover (CC) of the precipitation product. The model predictability was examined through categorical verification. The Cb cloud observation was based on brightness temperature (BT) IR1 and brightness temperature difference (BTD) IR1-IR2. The results showed that CC 50%' predictor was the best predictor to estimate the Cb cloud. The study in the period other than 2019 is suggested for the next research because Indian Ocean Dipole (IOD) is extreme that may affect the Cb cloud growth in the study area.

Download Full-text

A Weather Forecast Model Accuracy Analysis and ECMWF Enhancement Proposal by Neural Network

Sensors ◽

10.3390/s19235144 ◽

2019 ◽

Vol 19 (23) ◽

pp. 5144 ◽

Cited By ~ 5

Author(s):

Jaroslav Frnda ◽

Marek Durica ◽

Jan Nedoma ◽

Stanislav Zabka ◽

Radek Martinek ◽

...

Keyword(s):

Neural Network ◽

Atmospheric Condition ◽

Weather Forecast ◽

Horizontal Resolution ◽

Sensor Data ◽

Limited Area ◽

African Countries ◽

Weather System ◽

Medium Range ◽

Ecmwf Model

This paper presents a neural network approach for weather forecast improvement. Predicted parameters, such as air temperature or precipitation, play a crucial role not only in the transportation sector but they also influence people’s everyday activities. Numerical weather models require real measured data for the correct forecast run. This data is obtained from automatic weather stations by intelligent sensors. Sensor data collection and its processing is a necessity for finding the optimal weather conditions estimation. The European Centre for Medium-Range Weather Forecasts (ECMWF) model serves as the main base for medium-range predictions among the European countries. This model is capable of providing forecast up to 10 days with horizontal resolution of 9 km. Although ECMWF is currently the global weather system with the highest horizontal resolution, this resolution is still two times worse than the one offered by limited area (regional) numeric models (e.g., ALADIN that is used in many European and north African countries). They use global forecasting model and sensor-based weather monitoring network as the input parameters (global atmospheric situation at regional model geographic boundaries, description of atmospheric condition in numerical form), and because the analysed area is much smaller (typically one country), computing power allows them to use even higher resolution for key meteorological parameters prediction. However, the forecast data obtained from regional models are available only for a specific country, and end-users cannot find them all in one place. Furthermore, not all members provide open access to these data. Since the ECMWF model is commercial, several web services offer it free of charge. Additionally, because this model delivers forecast prediction for the whole of Europe (and for the whole world, too), this attitude is more user-friendly and attractive for potential customers. Therefore, the proposed novel hybrid method based on machine learning is capable of increasing ECMWF forecast outputs accuracy to the same level as limited area models provide, and it can deliver a more accurate forecast in real-time.

Download Full-text

Predictability of the Strong Ural blocking Event in January 2012 in the Subseasonal to Seasonal Models of Europe and Canada

Atmosphere ◽

10.3390/atmos11050538 ◽

2020 ◽

Vol 11 (5) ◽

pp. 538

Author(s):

Dong Chen ◽

Shaobo Qiao ◽

Shankai Tang ◽

Ho Nam Cheung ◽

Jieyu Liu ◽

...

Keyword(s):

Climate Change ◽

East Asia ◽

Development Stage ◽

Weather Forecasts ◽

Vorticity Advection ◽

Medium Range ◽

Ecmwf Model ◽

Temperature Tendency ◽

Blocking Event ◽

Vorticity Tendency

The occurrence of a Ural blocking (UB) event is an important precursor of severe cold air outbreaks in Siberia and East Asia, and thus is significant to accurately predict UB events. Using subseasonal to seasonal (S2S) models of the European Centre for Medium-Range Weather Forecasts (ECMWF) and the Environment and Climate Change Canada (ECCC), we evaluated the predictability of a persistent UB event on 18 to 26 January 2012. Results showed that the ECCC model was superior to the ECMWF model in predicting the development stage of the UB event ten days in advance, while the ECMWF model had better predictions than the ECCC model for more than ten days in advance and the decaying stage of the UB event. By comparing the dynamic and thermodynamic evolution of the UB event predicted by the two models via the geostrophic vorticity tendency equation and temperature tendency equation, we found that the ECCC model better predicted the vertical vorticity advection, ageostrophic vorticity tendency, the tilting effect, horizontal temperature advection, and adiabatic heating during the development stage, whereas the ECMWF model better predicted the three dynamic and the two thermodynamic terms during the decaying stage. In addition, during both the development and decaying stages, the two models were good (bad) at predicting the vortex stretching term (horizontal vorticity advection), with the PCC between both the predictions and the observations larger (smaller) than +0.70 (+0.10) Thus, we suggest that the prediction of the persistent UB event in the S2S model might be improved by the better prediction of the horizontal vorticity advection.

Download Full-text

An Assessment of the Impact of Aeolus Doppler Wind Lidar Observations for Use in Numerical Weather Prediction at ECMWF

10.5194/egusphere-egu2020-5340 ◽

2020 ◽

Author(s):

Michael P. Rennie ◽

Lars Isaksen

Keyword(s):

Numerical Weather Prediction ◽

Positive Impact ◽

Weather Prediction ◽

Systematic Errors ◽

Forecast Errors ◽

Horizontal Line ◽

Numerical Weather ◽

Medium Range ◽

Ecmwf Model ◽

The Impact

The European Space Agency&#8217;s Aeolus mission, which was launched in August 2018, provides profiles of horizontal line-of-sight (HLOS) wind observations from a polar orbiting satellite. &#160;The European Centre For Medium-Range Weather Forecasts (ECMWF) began the operational assimilation of Aeolus Level-2B winds on 9 January 2020 in their global NWP (Numerical Weather Prediction) model, 1 year and 4 months after the first Level-2B wind products were produced in near real time via ESA&#8217;s ground processing segment. &#160;This achievement was possible because of the production of good data quality, which was met through a close collaboration of all the parties involved within the Aeolus Data Innovation and Science Cluster (DISC) and via the great efforts of ESA, industry and ground processing algorithms pre- and post-launch. Through the careful assessment of the statistics of differences of the Aeolus winds relative to the ECMWF model, the Level-2B Rayleigh winds were found to have large systematic errors. &#160;The systematic errors were found to be highly correlated with ALADIN&#8217;s (Atmospheric Laser Doppler Instrument) primary mirror temperatures, which vary in a complex manner due to the variation in Earthshine and thermal control of the mirror. &#160;The correction of this source of bias in the ground processing is underway, therefore in the meantime a bias correction scheme using the ECMWF model as a reference was developed for successful data assimilation; the scheme will be described. &#160; We will present the results of the Aeolus NWP impact assessment which led to the decision to go operational. &#160;Aeolus&#8217; second laser (FM-B, available since late June 2019) provides statistically significant positive impact of moderate to large amplitude, of similar magnitude to some other important and well-established observing systems (such as IR radiances, GNNS radio occultation and Atmospheric Motion Vectors). &#160;Observing System Experiments demonstrate reduction of forecast errors in geopotential and vector wind of around 2% in the tropics and 2-3% in the southern hemisphere for short-range and medium range forecasts (up to day 10). &#160;This positive impact is particularly impressive given that Aeolus provides less than 1% of the total number of observations assimilated, showing the value of direct wind observations for global NWP.

Download Full-text

Mid-winter lower stratosphere temperatures in the Antarctic vortex: comparison between observations and ECMWF and NCEP operational models

Atmospheric Chemistry and Physics ◽

10.5194/acp-7-435-2007 ◽

2007 ◽

Vol 7 (2) ◽

pp. 435-441 ◽

Cited By ~ 20

Author(s):

M. C. Parrondo ◽

M. Yela ◽

M. Gil ◽

P. von der Gathen ◽

H. Ochoa

Keyword(s):

Weather Forecasting ◽

Temperature Fields ◽

Temperature Dependent ◽

Transport Models ◽

Medium Range ◽

Operational Models ◽

Environmental Prediction ◽

Ecmwf Model ◽

The Antarctic ◽

Good Agreement

Abstract. Radiosonde temperature profiles from Belgrano (78° S) and other Antarctic stations have been compared with European Centre for Medium-Range Weather Forecasting (ECMWF) and National Centers for Environmental Prediction (NCEP) operational analyses during the winter of 2003. Results show good agreement between radiosondes and NCEP and a bias in the ECMWF model which is height and temperature dependent, being up to 3°C too cold at 80 and 25–30 hPa, and hence resulting in an overestimation of the predicted potential PSC areas. Here we show the results of the comparison and discuss the potential implications that this bias might have on the ozone depletion computed by Chemical Transport Models based on ECMWF temperature fields, after rejecting the possibility of a bias in the sondes at extreme low temperatures.

Download Full-text

Dynamically vs. empirically downscaled medium-range precipitation forecasts

Hydrology and Earth System Sciences Discussions ◽

10.5194/hessd-6-3517-2009 ◽

2009 ◽

Vol 6 (2) ◽

pp. 3517-3542

Author(s):

G. Bürger

Keyword(s):

Dynamical System ◽

Lead Time ◽

Regional Model ◽

The Other ◽

Medium Range Forecast ◽

Mountainous Catchments ◽

Skill Scores ◽

Medium Range ◽

Ecmwf Model ◽

Empirical System

Abstract. For three small, mountainous catchments in Germany two medium-range forecast systems are compared that predict precipitation for up to 5 days in advance. One system is composed of the global German weather service (DWD) model, GME, which is dynamically downscaled using the COSMO-EU regional model. The other system is an empirical (expanded) downscaling of the ECMWF model IFS. Forecasts are verified against multi-year daily observations, by applying standard skill scores to events of specified intensity. All event classes are skillfully predicted by the empirical system for up to five days lead time. For the available prediction range of one to two days it is superior to the dynamical system.

Download Full-text

The Skill of ECMWF Medium-Range Forecasts during the Year of Tropical Convection 2008

Monthly Weather Review ◽

10.1175/2010mwr3217.1 ◽

2010 ◽

Vol 138 (10) ◽

pp. 3787-3805 ◽

Cited By ~ 17

Author(s):

Arindam Chakraborty

Keyword(s):

Surface Temperature ◽

Diurnal Cycle ◽

Lead Time ◽

Tropical Convection ◽

Precipitation Forecast ◽

Lead Times ◽

Near Surface ◽

Forecast Lead Time ◽

Medium Range ◽

Ecmwf Model

Abstract This study uses the European Centre for Medium-Range Weather Forecasts (ECMWF) model-generated high-resolution 10-day-long predictions for the Year of Tropical Convection (YOTC) 2008. Precipitation forecast skills of the model over the tropics are evaluated against the Tropical Rainfall Measuring Mission (TRMM) estimates. It has been shown that the model was able to capture the monthly to seasonal mean features of tropical convection reasonably. Northward propagation of convective bands over the Bay of Bengal was also forecasted realistically up to 5 days in advance, including the onset phase of the monsoon during the first half of June 2008. However, large errors exist in the daily datasets especially for longer lead times over smaller domains. For shorter lead times (less than 4–5 days), forecast errors are much smaller over the oceans than over land. Moreover, the rate of increase of errors with lead time is rapid over the oceans and is confined to the regions where observed precipitation shows large day-to-day variability. It has been shown that this rapid growth of errors over the oceans is related to the spatial pattern of near-surface air temperature. This is probably due to the one-way air–sea interaction in the atmosphere-only model used for forecasting. While the prescribed surface temperature over the oceans remain realistic at shorter lead times, the pattern and hence the gradient of the surface temperature is not altered with change in atmospheric parameters at longer lead times. It has also been shown that the ECMWF model had considerable difficulties in forecasting very low and very heavy intensity of precipitation over South Asia. The model has too few grids with “zero” precipitation and heavy (>40 mm day−1) precipitation. On the other hand, drizzle-like precipitation is too frequent in the model compared to that in the TRMM datasets. Further analysis shows that a major source of error in the ECMWF precipitation forecasts is the diurnal cycle over the South Asian monsoon region. The peak intensity of precipitation in the model forecasts over land (ocean) appear about 6 (9) h earlier than that in the observations. Moreover, the amplitude of the diurnal cycle is much higher in the model forecasts compared to that in the TRMM estimates. It has been seen that the phase error of the diurnal cycle increases with forecast lead time. The error in monthly mean 3-hourly precipitation forecasts is about 2–4 times of the error in the daily mean datasets. Thus, effort should be given to improve the phase and amplitude forecast of the diurnal cycle of precipitation from the model.

Download Full-text

Some Characteristics of Model-Predicted Precipitation during the Summer Monsoon over India

Journal of Applied Meteorology ◽

10.1175/jam-2198.1 ◽

2005 ◽

Vol 44 (3) ◽

pp. 324-339 ◽

Cited By ~ 13

Author(s):

B. K. Basu

Keyword(s):

Summer Monsoon ◽

Lead Time ◽

Precipitation Pattern ◽

Spatial Average ◽

Spatial Averaging ◽

Forecast Lead Time ◽

Medium Range ◽

Ecmwf Model ◽

Rainy Days ◽

Range Forecasts

Abstract For the summer monsoon seasons of 1995, 1996, and 1997 the day-1 to day-4 forecasts of precipitation from both the National Centre for Medium Range Weather Forecasting (NCMRWF) and the European Centre for Medium-Range Forecasts (ECMWF) models reproduce the main features of the observed precipitation pattern when averaged over the whole season. On average, less than 30% of all rain gauge stations in India report rain on a given day during the monsoon season. The number of observed rainy days increases to 41% after spatial averaging over ECMWF model grid boxes and to 50% after spatial averaging over NCMRWF model grid boxes. The NCMRWF model forecasts have 10%–15% more rainy days, mostly in the light or moderate precipitation categories, when compared with the spatial average of observed values. Seasonal accumulated values of all of India’s average precipitation show a slight increase with the forecast lead time for the NCMRWF model and a small decrease for the ECMWF model. The weekly accumulated values of forecast precipitation from both models, averaged over the whole of India, are in good phase relationship (∼0.9 in most cases) with the observed value for forecasts with a lead time up to day 4. Values of statistical parameters, based on the frequency of occurrence in various classes, indicate that the NCMRWF model has some skill in predicting precipitation over India during the summer monsoon. The NCMRWF model forecasts have higher trend correlation with the observed precipitation over India than do the ECMWF model forecasts. The mean error in precipitation is, however, much less in the ECMWF model forecasts, and the spatial distribution of seasonal average medium-range forecasts of ECMWF is closer to that observed along the west coast mountain ridgeline.

Download Full-text