...
The month before the event had been wet and the ground was already saturated in the area, as seen in the figures below (analysis of soil moisture level 1,2,3 with 3 days apart, starting from 10 May).
Gallery includeLabel soil_an sort comment title Soil moisture level 1,2,3 analysis
Above Below is the radar sequence from the Romanian Met Service for a part of 14 May, showing the persistent rainfall over Serbia and parts of Romania.
Gallery includeLabel obs_precip sort comment
The figures above below show the observed precipitation for the 14,15, 16 and 17 May (06-06) and the HRES forecast (day 2-3). The forecast underestimated the precipitation in the worst affected areas.
The figure below shows the 5-day accumulation (13-18 May 00UTC). For the climatology of 5-day accumulations (bottom, left) unfortunately no station appears over Serbia but I found some stations in Austria, Slovakia, Poland and Romania where 5-day totals exceeded 99th percentiles of the 15-year observed climate. On the plot below you can see two examples.
Gallery includeLabel obs_precip sort comment title Observed and forecast precipiation (06-06)
The table below shows the observed precipitation for Loznica (13262, 44.3N, 19.1N), Beograd (13274, 44.9N, 20.3E), Valjevo (13269, 44.2N, 19.5E) and Tuzla (14557, 44.5N, 18.7E). The precipitation is the 24-hour observed at 06UTC. The total for the 5-day period (12 May - 17 May 00UTC) was for Loznica 219 mm, Beograd 190 mm, Valjevo 199 mm and Tuzla 249 mm., where the main part fell between 13 May 06UTC and 16 May 06UTC.
| Loznica | Beograd | Valjevo | Tuzla | |
|---|---|---|---|---|
| 13 May | 0 | 0 | 6 | 0 |
| 14 May | 50 | 21 | 38 | 49 |
| 15 May | 110 | 108 | 108 | 95 |
| 16 May | 53 | 45 | 44 | 90 |
| 17 May | 6 | 16 | 3 | 15 |
| 18 May | 0 | 0 | 3 | 0 |
The figure below (top, left) shows the 5-day accumulation (13-18 May 00UTC). For the climatology of 5-day accumulations (bottom, left) unfortunately no station appears over Serbia but I found some stations in Austria, Slovakia, Poland and Romania where 5-day totals exceeded 99th percentiles of the 15-year observed climate. On the plot below you can see two examples.
The connection to the cyclone, high wind gusts where measured, especially in the Slovakian mountains with wind gusts of more than 40 m/s. The plot below shows the maximum observed wind gusts for the 15 May.
...
Gallery includeLabel z500_hres sort comment title z500 and t850 HRES forecasts (all valid 15 May 00UTC)
The nest next plots show the same forecasts but for MSLP and precipitation.
Gallery includeLabel msl_hres sort comment title MSLP and precipiation HRES (all valid 15 May 00UTC)
As seen abovein the previous section, the precipitation in the worst affected areas seems to have been underestimated.
The figures above below show the precipitation verification for 14 May 06UTC - 15 May 06UTC for forecast from 12, 13 and 14 may 00UTC. Only the very last forecast captured the observed rainfall amounts reasonable.
The figures above below show the precipitation verification for 15 May 06UTC - 16 May 06UTC for forecast from 13, 14 and 15 may 00UTC. Here the extreme precipitation over the orography in Austria and Slovakia was underestimated.
The plots below show the accumulated precipitation 12 May 06UTC to 17 May 06UTC for forecast with different resolutions (T1279, T639 and T319) initialised 12 May 00UTC. For this forecast it is a clear difference between the resolutions.
The next figure above shows the 3-day accumulated precipitation in the box 44N-45N, 18E-20E from different initial dates (from different resolutions). In the deterministic forecasts a hint of the extreme rainfall started to appear on 6 May. The T319 forecast was very consistent from 6 May 12UTC (probably with a large portion of change). We do not yet have any good gridded precipitation data set, but from the synop stations in the area the observed precipitation for the period should be 170-200 mm.
3.3 ENS
On 14th May in Serbia and Bosnia heavy rain between 40 and 130 mm just in 24h was recorded. From the few stations in Serbia displayed on the plot below we can see that 24-hour accumulations exceeded the 99th percentile of the observed 15-year climate. The most extreme precipitation was observed in Loznica, a station in western Serbia at the border with Bosnia where 129 mm were recorded on 14th May whilst the climate maximum is just 68.4 mm. Comparing the forecast with observations we can notice that 7-days in advance the tail of the forecast consistently predicted extreme values of the rainfall but the forecast was uncertain and therefore EFI values were positive but not high whilst SOT was positive indicating a possibility of an extreme event. Approaching the day of interest forecast CDFs became more extreme ending up with really extreme CDF in the most recent forecast on the day of the extreme rainfall. We can also notice that the forecast underestimated the rainfall totals even for the most extreme members but at the same time such an extreme rain as observed in Loznica seem locakized. All the other stations around measured rainfall amounts between 40 mm and just above 100 mm which is close to the rainfall in the most extreme ensemble members.
...
The plots below show the accumulated precipitation 12 May 06UTC to 17 May 06UTC for forecast (control) from different centre (ECMWF, UKMO, NCEP and CMC) initialised 12 May 00UTC. Most rain was produced by UKMO and CMC. Note that NCEP clearly underestimated the precipitation for this initial time.
The figure above below shows the 3-day accumulated precipitation in the box 44N-45N, 18E-20E from different initial dates (for control forecasts from different centres). In general UKMO seems to be able to produce most rain for the region.
The plots below show CDF for 5-day precipitation 12 May 06UTC to 17 May 06UTC for Beograd (point for 44.9N, 20.3E), for different centres obtained from the TIGGE archive. The observed value for the period was 190 mm.
...
The plots below show CDF for 3-day precipitation 13 May 06UTC to 16 May 06UTC for Beograd, for different centres. The observed value for the period was 174 mm. Here the results are quite mixed
Gallery includeLabel tigge_cdf_3d sort comment title TIGGE CDF for Beograd (3-day)
The plots below show CDF for 3-day precipitation 13 May 06UTC to 16 May 06UTC for Loznica (point for 44.3N, 19.1E), for different centres. The observed value for the period was 213 mm. Here UKMO outperformed all other centres.
Gallery includeLabel tigge_cdf_3d_loz sort comment title TIGGE CDF for Loznica (3-day)
...
To conclude, the results are quite mixed comparing different centres. The outstanding results are the high precipitation for 3-days from UKMO for Loznica.
The plots below show the accumulated precipitation 14 May 06UTC to 15 May 06UTC for forecast (control) from different centre (ECMWF, UKMO, NCEP and CMC) initialised 14 May 00UTC.For being short forecasts, the results are quite different from the 4 centres.
3.6 EFAS
Gallery includeLabel efas_severe sort comment title EFAS
The figures above show the sites where the alert levels are reached in the EFAS forecasts for different initial date of the forecasts. The colour of the circles corresponds to different return periods (purple>20 years). Already on the 8 May several stations had a high level forecasted and the signal got persistent from the 10 May.
The figures above show the modelled flow for river Sava, close to Beograd. The peak flow appear around 18-19 May. Already the longest forecast covering this date (9 May) had a number of ensemble members with return period >20 years. After the rain fell 14-15 May the uncertainty got minimal (as the flood model for the initial conditions is driven by observed precipitation).
4. Experience from general performance/other cases
- The flooding in central Europe 2013. Technical Memorandum soon finished about the case. This case agrees about the underestimation of the extreme precipitation and the dependence of resolution.
5. Good and bad aspects of the forecasts for the event
...