Skip to end of metadata
Go to start of metadata

 Status:Ongoing analysis

Material from: Mohamed, Linus



1. Impact

In the first week of July, the West and South West of Japan was dramatically affected by heavy rainfall that led to widespread flooding and landslides causing  the loss of more than 100 lives and the destruction of thousands of houses.  The amounts of rain exceeded daily totals of 580 mm in places. The the city of Uwajima for instance received 364 mm in just two hours on 8 July. The maximum of the rain occurred from 5 to 8 July.

2. Description of the event

The plots below show daily rainfall in observations from 6, 7 and 8 July.

3. Predictability


3.1 Data assimilation


3.2 HRES

The plots below show the observed rainfall on 6 July (first panel) and HRES forecasts from different initial times valid on 6 July.

The plots below shows 72h accumulated precipitation from the HRES based on 4 July,  2 July and 1 July VT: 05 to 08 July 2018.

3.3 ENS

The plots below show EFI and SOT for 3-day precipitation (5-7 July).

The plots below show probability of precipitation exceeding 200mm/72h from 03/07 and 01/07 VT: 05 to 08 July 2018.

The plot below shows the evolution of forecasts for 3-day accumulated precipitation (5-7 July) in the box 30N-35N, 130E-133E. The plot includes ENS (blue box-and-whisker), HRES (red dot) and model climate (red box and whisker). For this box the mode picked up the signal from 29 June 00UTC (last bad forecast) to 12 UTC and further on to 1 July 00UTC (first good forecast).

3.4 Monthly forecasts

The plots below show precipitation anomalies for 2- 8July in monthly forecasts. As expected from the forecast evolution plot above, there was not any signal beyond 29 July, and even a dry anomaly was present in longer 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

  • Reasonable ENS performance in medium range (from 1 July onwards)
  • No signal in monthly forecasts

6. Additional material