Status:Ongoing analysis Material from: Linus,Tim H.
1. Impact
In the middle of December the post-landfall cyclone related to TC Jasper caused severe rainfall over Queensland, Australia. Following on from the passage westwards the tropical cyclone, across northeast Australia, legacy moisture and slow-moving low level circulations have had a profound impact on local communities, as up to 2m of rain fell in a few days (see here for news/discussion: https://www.theguardian.com/australia-news/2023/dec/19/cyclone-jasper-how-did-it-cause-so-much-rain-and-could-global-heating-be-to-blame). Data from the wettest site, with a complete record, that I could find on the BOM website - namely "Whyanbeel Valley" - is shown below. It lies about 70km NW of Cairns, which was also badly affected. Here you can see 699mm for the most recent 24h period.
https://en.wikipedia.org/wiki/Cyclone_Jasper
https://en.wikipedia.org/wiki/2023_Cairns_floods
2. Description of the event
In this evaluation we will focus on the rainfall over three periods: 5-day rainfall 13-18 December plus 1-day rainfall on 13 and 17 December. The focus region is the coastline of Queensland (land points inside 15.5S - 17.5S, 145E- 146E).
The plots below show analyses of MSLP and 6 hour rainfall from 11 December 00UTC to 18 December 00UTC, every 12th hour. TC Jasper made landfall on 13 December, but later stalled over Queensland, while the rainfall continued. On 17 December the rainfall intensified.
Observed rainfall from Whyanbeel valley:
3. Predictability
3.1 Data assimilation
The plot below show first guess (left) and analysis (right) of MSLP and departures of surface pressure observations for the LWDA cycle on 13 December 00UTC. This data assimilation cycle seems to have severe problems with the TC.
3.2 HRES
The plots below show 5-day precipitation (13 December 00UTC - 18 December 00UTC) in concatenated short forecasts (first plot) and ENS control forecasts with different lead times. Note that sea points have been masked out. The box marks 15.5S - 17.5S, 145E- 146E.
The plots below show 1-day precipitation (13 December 00UTC - 14 December 00UTC) in concatenated short forecasts (first plot) and ENS control forecasts with different lead times. Note that sea points have been masked out. The box marks 15.5S - 17.5S, 145E- 146E. The last forecast (from 13 Dec 00UTC) had the precipitation shifted to the south.
The plots below show 1-day precipitation (17 December 00UTC - 18 December 00UTC) in concatenated short forecasts (first plot) and ENS control forecasts with different lead times. Note that sea points have been masked out. The box marks 15.5S - 17.5S, 145E- 146E. The extreme rainfall on 17 December seems to only been captured 48 hours in advance.
3.3 ENS
The plots below shows EFI and SOT for 5-day precipitation (13 December 00UTC - 18 December 00UTC) from different initial dates.
The plot below shows the forecast evolution plot for 5-day precipitation (13 December 00UTC - 18 December 00UTC) in land-points inside the box outlined in the plots above. Concatenated 6-hour forecasts - green dot, ENS control –red, ENS blue box-and-whisker, Model climate – cyan box-and-whisker. Ensemble mean as black diamonds. Triangle marks the maximum in the model climate based on 1800 forecasts.
The predictability for this period to be wetter than normal in the region was relatively good with a clear signal from 5 December and onwards. This coinside with the genesis of TC Jasper.
The plots below shows EFI and SOT for 1-day precipitation (13 December 00UTC - 14 December 00UTC) from different initial dates.
The plot below shows the forecast evolution plot for 1-day precipitation (13 December 00UTC) in land-points inside the box outlined in the plots above. Concatenated 6-hour forecasts - green dot, ENS control –red, ENS blue box-and-whisker, Model climate – cyan box-and-whisker. Ensemble mean as black diamonds. Triangle marks the maximum in the model climate based on 1800 forecasts.
The problem for the forecast on 13 December is obvious here, and is probably related to the data assimilation issue discussed in Section 3.1.
The plots below shows EFI and SOT for 1-day precipitation (17 December 00UTC - 18 December 00UTC) from different initial dates.
The plot below shows the forecast evolution plot for 1-day precipitation (17 December 00UTC) in land-points inside the box outlined in the plots above. Concatenated 6-hour forecasts - green dot, ENS control –red, ENS blue box-and-whisker, Model climate – cyan box-and-whisker. Ensemble mean as black diamonds. Triangle marks the maximum in the model climate based on 1800 forecasts.
The predictability of this phase seem to have been relatively low as the signal started to appear on 14 December, probably related to the propagation of the tropical cyclone (see below).
The plots below show the tropical cyclone track for TC Jasper for the operational ECMWF forecasts from 15 December 00UTC (first plot) to 5 December 00UTC (last plot). The symbols shows the position on 15 December 00UTC (hourglass for BestTrack). HRES (red), ENS CF (blue), ENS PF (grey) and BestTrack (black).
Central pressure (top) and maximum wind speed (bottom) for HRES (red), ENS CF (blue), ENS PF (grey) and BestTrack (black).
3.4 Monthly forecasts
The plots below show forecasts of weekly precipitation anomalies for 11 December to 18 December. The wet signal along the coast of Queensland started to appear 2-3 weeks in advance.
3.5 Comparison with other centres
4. Experience from general performance/other cases
5. Good and bad aspects of the forecasts for the event
- Early signal for genesis of TC Jasper
- Problematic forecast jsut before landfall (13 December)
- Erroneous westward propagation after landfall also in short forecasts