T-PARC (Summer Phase)

Sharanya J. Majumdar (RSMAS/U. Miami) Christopher S. Velden (CIMSS / U. Wisconsin) Section 4.7, THORPEX/DAOS WG Fourth Meeting 27-28 June 2011.

• • • •

Objective: To improve 1-3 day forecasts by obtaining targeted observations in regions with high sensitivity.

During the field phase, a team identified potential opportunities to collect targeted observations: – Cases selected 2-3 days prior to observation time.

– Common verification regions, Guam, Taiwan, and Japan – Individually selected verification regions: calculations performed through ECMWF/Met Office PREVIEW DTS Final flight paths chosen one day prior, based on targeted observation guidance and team consensus.

Post field phase: Data denial experiments; observation impact experiments; different events considered.

MWR Special Collection http://journals.ametsoc.org/page/Cyclone_Predictability IWTC-VII, La Réunion, France 15-20 November 2010 Special Focus 1a: Targeted observations for TC track forecasting. C.-C. Wu and Sharan Majumdar

Outline •

Tropical Cyclone Track – Aircraft: Dropwindsonde and Wind Lidar data – Satellite: AMVs and radiances

•

Other forecasts – Mid-latitudes – Downstream impacts

DLR Falcon 20

Tropical Cyclone Track

US Air Force WC-130 US NRL P-3 DOTSTAR Astra jet F. Harnisch

Period: 2008090900-2008091812 and 2008092412-2008092900 M. Weissmann

• Harnisch and Weissmann (MWR 2010) • Separation of dropwindsonde observations into 3 subsets: → typhoon vicinity: largest improvements of ECMWF track → remote sensitive regions: small positive to neutral influence → typhoon center and core: overall neutral influence Weissmann et al. (MWR 2011) NCEP and WRF/3dVar: Improvement from 20-40% Comparably low influence in ECMWF and JMA. • Lower forecast errors without dropsondes in ECMWF & JMA • More extensive use of satellite data and 4d-Var?

• Chou et al. (MWR 2011) for Mean 1–5 day NCEP track forecast error is reduced by 10–20% DOTSTAR and T-PARC cases (not as beneficial in ECMWF) The different behaviour of the models emphasizes that the benefit depends strongly on the quality of the first-guess field and the assimilation system

• YH Kim et al. (APJAS, 2010) : 17-22% improvement to short-range track forecasts.

Mid-tropospheric data most effective (WRF/3dVar).

• Jung et al. (APJAS, 2010) (WRF/3dVar).

: observations over ocean more important than over land.

Dropwindsondes most important at times they were launched. Otherwise, QuikSCAT and SATEM data were most important.

Observations in sensitive areas improved the forecast • HM Kim (WRF/EnKF).

et al.

(2011) : Positive impact of dropwindsondes can be found in ensemble forecasts • NOTE: Radiances not assimilated in these studies

IR Satellite Image 09/11/08 1830 LT 16W

Airborne Doppler Wind Lidar

Tokyo

Sinlaku

M. Weissmann Okinawa

Weissmann et al. (QJRMS, in review)

• • • • 2500 high-density, high-accuracy wind profiles measured from DLR Falcon during Typhoon Sinlaku.

Data denial – ECMWF track forecasts improved by ~50 km for 1-5 days.

– NOGAPS track forecasts did not improve (due to bogus?) – Improvement in 500 hPa and 1000 hPa Z.

Adjoint method – Total relative DWL contribution 2x as large in NOGAPS as ECMWF. – Impact per ob is comparable to other platforms (higher in NOGAPS) Atmospherics Dynamics Mission Aeolus (ADM-Aeolus) lidar instrument planned for launch by ESA (2013?)

Improved targeting methods for TCs

Var-ETKF Moist Adjoint Majumdar et al. (2011, QJRMS) Mesoscale SVs Doyle et al. (2011, CISE) Ensemble sensitivity Kim et al. (2011, WAF, in press) Mahajan and Hakim (2011)

TC track: impact of satellite data

JAMC 2011, in press

JAMC 2011, in press

JAMC 2011, in press

JAMC 2011, in press

Hourly AMVs: reduce mean 3-5 day track forecast errors by 6-10% Rapid-scan: further reduces the 3-5 day NOGAPS track forecast errors

Assimilation of AMVs on the mesoscale

• • • Framework: NCAR Data Assimilation Research Testbed (DART) Data assimilation: Ensemble Kalman Filter (EnKF) Model: Advanced Research WRF (WRF-ARW) • • • • CIMSS: Cooperative Institute for Meteorological Satellite Studies ; JMA: Japan Meteorological Agency Ensemble members: 32; Case: Typhoon Sinlaku (2008) Assimilation cycle started Sep. 1 st , 2008. (one week before genesis) 9km moving nest grid with feedback to 27km grid in the forecasts when TC is present.

Deterministic: ECMWF 1.125°x1.125° (Baseline)

Analysis Track and Intensity CIMSS JMA Best Track CIMSS Structure CTL CIMSS 09/09:00Z 09/10:00Z Upper-lev Div (left) Azi-mean Vort (Right) 09/11:00Z CTL CIMSS

Prior Post

Analysis increment – Theta

Targeting Typhoon season with extra-satellite data

Selective data thinning experiments • Cntrl : 1.25

o Global • SV-Sat: 1.25

o Global and 0.625

o in SV areas.

• Drop : 1.25

o Global +Targeted Dropsondes • SV-Sat-Drop: Targeted Dropsondes+ SV areas 0.625

o Additional information • All experiments are run at T799TL95/159/255 L91 (12-hour 4D-Var) • 06-30 September 2008 • Verification and SV-target region 10-50N, 110-180E • 20 Leading T95L62 SV • SVs area cover 20% of the target region C. Cardinali

Sinlaku 09-19 September: mean track error km 09 + 10 11 Sept

C. Cardinali

cntrl cntr cntrl

Forecast Sensitivity to Obs: SV-Sat+Drop

• Extra-satellite data gave a more consistent impact due to homogeneous coverage and data diversity (moist, temperature, cloud, precipitation and surface wind) C. Cardinali

Sinlaku

50 45 40 35 30 25 20 15 10 5 0 AN T+12 Forecast error and Verifying analysis T+24

9 0 0 9 12 10 0 0 10 12 110 0 1112 12 0 0 12 12 13 0 0 13 12 14 0 0 14 12 15 0 0 15 12 16 0 0 16 12 17 0 0 17 12

Influence on ECMWF midlatitude forecasts

Pacific; lead time:96 h 100 50 0 -50 -100 -150 910 913 916 only 2 Sinlaku flights 919 date 922 925 928 improved track forecast --> improved first-guess for subsequent days --> improved mid-latitude forecast overall neutral influence of observations during ET, although these were partly guided by SV calculations optimized for the Pacific M. Weissmann

Downstream Impacts

• • Aberson (MWR 2011, in press) Dropwindsonde data provide global improvements to NCEP GFS TC track forecasts of about 10% through 72 h, but decreasing at longer forecast lead times.

T-PARC Accomplishments • • • • •

Demonstrated utility of coordinated aircraft missions, and dropwindsonde and DWL data Benefits of higher spatial and temporal density of satellite winds and radiances Improvements to forecasts downstream, although targeting strategy not essential here Accelerated use of TIGGE: full fields and CXML database Large number of peer-reviewed publications

•

Recommendations (from IWTC-VII)

Aircraft observations are limited (particularly in NW Pacific): make improved use of existing observations. Satellite radiance data, and AMVs. Special rawinsonde launches?

• Given that observations / models / DA evolve, need to frequently review targeted observing programs.

• Explore new strategies, in basic research, OSEs and OSSEs.

• Consider new observing platforms e.g. UAS, wind lidars.

• Coordinate use of observations (e.g. EURORISK PREVIEW) • Explore tropical cyclone formation, structure and intensity.