1. 1/31 Text NWP upgrades Track forecasting –Consensus –Ensembles Intensity forecasting –Objective aids Structure forecasting Web sites NWP for Tropical Cyclone forecasting - what is new for 2015? Rewa, 1993/94 http://www.bom.gov.au/cyclone/history/rewa.shtml

2. 2/31 Model upgrades : ECMWF http://www.ecmwf.int/en/forecasts/documentation-and- support/changes-ecmwf-model/cycle-41r1 May 2015 upgrade – resolution still at ~16km Improvement in average position error; BUT wind speed in tropics shows a 'deterioration' Slight improvement in intensity at all lead times; to day 3 this reduces the slight positive bias; from day 5 + the pre-existing bias towards over-deepening has increased slightly. Ocean wave upgrades; Future upgrades flagged; Ensembles 51 members run at 32km.

3. 3/31 Model upgrades : GFS Global model run now at ~13km resolution to +240h Large supercomputing investment in GFS next upgrade in November? Availability of GFS ensembles for SH ? BUT… new model not quite as good as previous version for TCs …mmmm http://www.nco.ncep.noaa.gov/pmb/changes/ Courtesy: V. Tallapragada, NCEP

4. 4/31 Model upgrades : HWRF http://www.emc.ncep.noaa.gov/gc_wmb/vxt/HWRF/index.php Run for all systems globally can now run 7 TCs at once – even for tropical lows Nested within GFS with variable resolution – higher for core June 2015 upgrade – now at max 2km resolution – higher than any other available operational model (18/6/2km). Results correct GFS upgrade 'weakness' Intensity results encouraging.

5. 5/31 Model upgrades : HWRF http://www.emc.ncep.noaa.gov/gc_wmb/vxt/HWRF/index.php Blanca 2015 3km run 2km run, more symmetric and smaller sized storm Courtesy: V. Tallapragada, NCEP

6. 6/31 Model upgrades : ACCESS ACCESS-G upgrade to APS-2 for 2015/16 season resolution 40km will improve to 25km; - Possible concern that model may 'overheat' development. ACCESS-TC remains a variable domain at 12km resolution for 2015/16 season New supercomputer will herald opportunities for upgrades in 2016 & 2017 http://www.bom.gov.au/australia/charts/viewer/

7. 7/31 Models : others UK, JMA, COAMPS, GFDN/L, NAVGEM ***UK: major 2014 upgrade narrows the gap to EC/GFS JMA: global ~21km resolution; trailing other globals in Aust region COAMPS: June 2015 upgrade Experimental CTCX Nested in NAVGEM GFDL: nested in GFS – NOW available in SH !! Skill here? GFDN: GFDL version nested in NAVGEM NAVGEM: US Navy Global model superceding NOGAPS still showing some skill

8. 8/31 Verification: Track JTWC 2014 in WPAC Consensus beat JTWC at all forecast times Meso-models, except HWRF, continue to lag global models (GFS/EC/UK) GFS/EC/UK performed similarly, except at Day 5 where GFS had ~50nm larger errors Courtesy: JTWC JTWC CON GFS UK EC HWRF

9. 9/31 Track Forecasting – The BoM Consensus approach Standard members – nine models EC + GFS + HWRF + UK + ACCESS-TC (Tier 1) + COAMPS(TX) + JMA + GFDL* + NAVGEM (Tier 2) ACCESS-G when ACCESS-TC unavailable long term guidance; 2015 change: no longer having 2 x EC in standard consensus; For tropical lows greater selective approaches; Using previous runs of EC/GFS/HWRF/UK case by case basis *GFDL replaces GFDN (awaiting NRL advice) Occasional erratic behaviour by JMA, GFDN and COAMPS Ensemble mean a possibility GFS/JMA ensemble mean in JTWC

10. 10/31 The BoM Consensus approach Non-SELECTIVE (NCON) – robust most of time TC Christine: spread in shifted models Tier 1 models consistent cf climatology (grey) cf previous forecast (dashed)

11. 11/31 Input to Forecast Track Map TC Christine: Uncertainty and gales shape watch/warning areas

12. 12/31 The BoM Consensus approach SELECTIVE (SCON) Will Quang hit Exmouth?? COAMPS (green) + UK/GFDL (to east) others inc. EC/GFS/JMA/A-TC further west; Question: Should we 1. discard COAMPS as the outlier? OR 2. Just take the consensus of all? Or 3. …? Use your pen to tick your preference.

13. 13/31 Track Forecasting – The BoM Consensus approach SELECTIVE (SCON) Quang: bias to UK (to east) sfc wind comparison at +66h UK stronger being steered by deeper NW'ly flow than other models

14. 14/31 The BoM Consensus approach SELECTIVE (SCON) Olwyn – strange shapes! When to be selective: High spread, can explain model behaviour pre-Olwyn: bias to EC and UK Uncertainty bulge for other scenarios that remain possible

15. 15/31 Track Forecasting – The BoM Consensus approach SELECTIVE (SCON) Ex-TC Lam: high spread in models – EC Vs GFS/UK !

16. 16/31 Track Forecasting – The BoM Consensus approach SELECTIVE (SCON) Track map: high uncertainty – watch Qld and NT side of Gulf but EC treated as 'unlikely' Note: when being selective ensure reasoning is documented

17. 17/31 Track Forecasting - ensembles EC strike probability TC Ita EC ensemble Large spread Can assist with uncertainty area No change in approach

18. 18/31 Track Forecasting - ensembles Filtering, clustering and super-ensembles Filter on position/intensity Cluster techniques Useful SOMETIMES Bang for buck? Super-ensembles coming…availability

19. 19/31 Track Forecasting – BEWARE Don't ever be complacent Sometimes nature doesn't go by the NWP rules

20. 20/31 Track Forecasting – BEWARE Don't ever be complacent Changes in track close to coast – timing consequences for landfall storm tide e.g. Dylan, Marcia, Yasi…Trochoidal motion

21. 21/31 Intensity Forecasting Verification: JTWC 2014 in WPAC Statistical-Dynamical models (S5XX/YY) perform best HWRF good to +48h then COAMPS/GFDN better GFDL similarly to other meso-models (not shown) Courtesy: JTWC S5XX S5YY

22. 22/31 HWRF improvements Courtesy: V. Tallapragada, NCEP Better

23. 23/31 Intensity forecasting still tricky but… Inputs: manual (+- DT/24h); Objective Aids: S5XX, S5YY; Models: esp. HWRF (ensembles not yet that useful) model sfc wind patterns Recognition of satellite signatures for rapid changes Basic process remains same. Source: Sampson&Knaff, IWTC http://www.wmo.int/pages/prog/arep/wwrp/new/documents/Topic2.7_AdvancesinIntensityGuidance.pdf

24. Objective Intensity Guidance: S5XX and SFYY S5XX has been better in SH and IO but S5YY expected to be better now. S5YY LGEN/DSHN – SHIPS/LGEM using NAVGEM LGEA/DSHA – SHIPS/LGEM using GFS track and wind fields, and NAVGEM thermal fields CHIPS GFDN COAMPS-TC HWRF S5XX: From STIPS (no LGEM) Source: Sampson&Knaff, IWTC http://www.wmo.int/pages/prog/arep/wwrp/new/documents/Topic2.7_AdvancesinIntensityGuidance.pdf WARNING ACRONYMS!!!!!

25. 25/31 Intensity forecasting: Rapid Intensification (RI) index SHIPS: gives probability of a 30kn/24h intensity based upon 9 predictors. Calibration ~15-20% consider R ~ 30% RI is definite. Fluctuations in output Source: B. Sampson

26. 26/31 Intensity forecasting : visualising Intensity plots make it easier for comparison

27. 27/31 Structure forecasting – the poor cousin Greater emphasis for future because it affects: Watch/warning areas; Onset times of gales so preparation activities; Wave generation and storm surge; Model guidance graphic and numeric - 2016 for TCModule

28. 28/31 Web sites: HWRF http://www.emc.ncep.noaa.gov/gc_wmb/vxt/HWRF/index.php

29. 29/31 Web sites CIRA TC page additions http://rammb.cira.colostate.edu/products/tc_realtime/index.asp Model plots

30. 30/31 Web sites Norwegian for ECMWF wind, rain http://www.yr.no/kart/

31. 31/31 Summary Forecasting - it is all getting easier! Questions? j.courtney@bom.gov.au

32. 32/31 Web sites NOAA NESDIS TC page additions http://www.ssd.noaa.gov/PS/TROP/index.html Fix data

33. 33/31 The BoM Consensus approach Non-SELECTIVE (NCON) – robust most of time TC Christine: Model position check

34. 34/31 Track Forecasting – The BoM Consensus approach SELECTIVE (SCON) TC Quang: spread in models – UK (eastern outlier) Vs

35. 35/31 Track Skill has improved NWP + consensus

36. 36/31 First SHIPS - Atlantic then STIPS for NWPac and SH Uses model parameters that affect intensity (shear, SST, RH, etc) based on historical cases and combines multi-models to determine Vmax Initialised by JTWC analysis, 1minute winds. Available via JTWC track file Enhanced SHIPS = LGEM exp email has more info Knaff and Sampson, AMOJ 2009. Intensity guidance: statistical –dynamical aids