Status:Finalized Material from: Ivan
Discussed in the following Daily reports:
Picture
1. Impact
A slow-moving low severely affected Afghanistan and Pakistan between 23rd and 25th February 2015. The combination of the slow movement of this storm and moisture from the Arabian Sea led to rain and snow in Afghanistan, Pakistan and far northwest India. This storm triggered numerous avalanches and areas of flooding across the region. Avalanches have closed roads, knocked down power lines and killed more than 160 people across northern Afghanistan.
2. Description of the event
This plot shows ECMWF analysis of 500 hPa geopotential and observed 24-h precipitation totals valid for 23 Feb (top right) and 24 Feb (bottom right). The most affected region in Afghanistan is also highlighted.
3. Predictability
3.1 Data assimilation
3.2 HRES
In the short range when we rely more and more on the HRES, we can see that both O-suite and E-suite (Cycle 41r1) managed to give a good quantitative forecast. The areas with the largest forecast amounts agree pretty well with observations. Some amounts above 100 mm/24h in the forecast can be seen in the most severely affected area to the NE of Kabul.
3.3 ENS
The EFI for snowfall provided very strong signal of abnormal weather even 7 days in advance. Values of the EFI for snowfall above 0.8 in 6-day forecast for example do not happen often. The strongest signal is placed in the most affected area where deadly avalanches were reported due to heavy snowfall. On the plot below the EFI forecast is shown for day 4 (T+72-96) and day 6 (T+120-144).
The EFI and CDFs for two of the locations with the largest amounts reveal that even 6-7 days in advance EFI reached very high values meaning that some severe weather is likely. Moreover, the ENS was very consistent implying high confidence.
3.4 Monthly forecasts
3.5 Comparison with other centres
4. Experience from general performance/other cases
5. Good and bad aspects of the forecasts for the event
1. Northern Afghanistan and Pakistan were badly affected by heavy snowfall and excessive rain causing avalanches and flooding that caused many deaths.
2. Our ENS managed to provide a strong signal of extreme weather 6-7 days in advance.
3. HRES also provided a very good quantitative forecast of precipitation amounts.