Improve Hurricane Structure Monitoring and Intensity Forecast Using NPP ATMS and GCOM-W

AMSR2

Fuzhong Weng (PI)NOAA/NESDIS/Center for Satellite Applications and Research

Xiaolei ZouFlorida State University

Vijay Tallapragada and Andrew Collard NOAA/NWS/Environmental Modeling Center

And

STAR/EMC Satellite Data Assimilation Team Members (Ben Zhang, Lin Lin, Tong Zhu, Greg Grawsowski, and In-Hyuk Kwon)

GOSE-R AWG Annual Meeting June 23-26, 2008<Soundings> AWG Annual 2

Proposed Tasks

1. Advanced Microwave Sounder Technology (ATMS) Backus-Gilbert Re-sampling

2. ATMS Algorithm for Hurricane Warm Core Monitoring

3. AMSR2 Algorithm for Sea Surface Temperature and Wind Speed in Storm Conditions

4. Direct Assimilation of Microwave Cloudy Radiances in Hurricane Model

GOSE-R AWG Annual Meeting June 23-26, 2008<Soundings> AWG Annual 3

Project Milestones and Deliverables (2012-2013)

• Generate the ATMS re-sampling weights using B-G  methods and understand the noise reduction in association with each averaging algorithm

• Preprocess  the ATMS data granules into new granule files with new B-G weights and generate the BUFR ATMS files for hurricane applications

• Retrieve atmospheric temperature profiles from ATMS under hurricane conditions

• Revise AMSR-E SST and SSW algorithms for AMSR2 applications and demonstrate the products over hurricane conditions (Delay to the second year)

By the end of the first year, we deliver the ATMS preprocessor  for generating a high resolution of ATMS SDR data with a nadir resolution of 30 km and AMSR2 SST and SSW algorithm software  

GOSE-R AWG Annual Meeting June 23-26, 2008<Soundings> AWG Annual 4

Project Milestones and Deliverables(2014-2015)

• Produce operational ATMS resampled data  and distribute to user community for demonstrating the values of ATMS oversampling data

• Derive AMSR2 surface temperature and wind speed products using a physical retrieval algorithm and validate the products with insitu data

• Collocate  ATMS and AMSR2 data and generate the retrievals of atmospheric and surface parameters using a combined set of AMSR2 and ATMS channels (delayed to the third year)

• Produce a composite analysis of hurricane vortex structure for hurricane model initialization.

•  Collocation of retrieved variables should be done within the DA framework

By the end of the second year, we deliver AMSR2 SST and SSW algorithm and products in hurricane conditions and also a composite vortex dataset 

GOSE-R AWG Annual Meeting June 23-26, 2008<Soundings> AWG Annual 5

Project Milestones and Deliverables(2015-2017)

• Characterize ATMS temperature sounding channel bias in presence of clouds and define the best predictors for bias correction with respect to GFS and HWRF

• Characterize the forward model errors for simulating ATMS radiances in cloudy conditions using the collocated Cloudsat and precipitation radar data with ATMS

• Conduct the experiments for assessing the impacts of cloud radiance assimilation in GFS and HWRF

By the end of third year, we will deliver a bias correction algorithm and forward radiative transfer model errors, with a focus on assimilating cloudy radiances in hurricane conditions

ATMS Deconvolution from Low to High Resolution

Raw 23 Tb (5.2 degree) Resampled 23 Tb( 2.2 degree)

ATMS Convolution from High to Low Resolution

Raw 89 GHz Tb (2.2 degree) Resampled 89 Tb( 5.2 degree)

NPP ATMS and VIIRS Imager and Products

Warm Core Cross section along 26.0 N VIIRS 0.64 µm visible and 11.45 µm IR images at 18:33 UTC, 28 Aug 2012

METAR, MSL Pressure, and Buoys information included

Hurricane Sandy Warm Core Anomaly Ascending 1730 UTC, 29 October 2012

Cross section along Longitude 72.9 WCross section along Latitude 38.1 N

At 1800 UTC Oct 29 Max Wind: 90 MPH, Min Pressure: 940 hPa

Sandy Max Wind, Warm Core and MSLP

Sandy Max Wind, Warm Core and MSLP

14

Passive Microwave Imager: an example of GCOM-W1 AMSR-2 Instrument

• Deployable main reflector system with 2.0m diameter.

• Frequency channel set is identical to that of AMSR-E except 7.3GHz channel for RFI mitigation.

• 2-point external calibration with the improved HTS (hot-load).

AMSR2 Channel Set

Center Freq.[GHz]

Band width [MHz]

Polarization

Beam width [deg](Ground res. [km])

Sampling interval

[km]

6.925/7.3 350

VandH

1.8 (35 x 62)

10

1.7 (34 x 58)

10.65 100 1.2 (24 x 42)

18.7 200 0.65 (14 x 22)

23.8 400 0.75 (15 x 26)

36.5 1000 0.35 (7 x 12)

89.0 3000 0.15 (3 x 5) 5

Stowed

AMSR2 characteristics

Scan Conical scan

Swath width 1450km

Antenna 2.0m offset parabola

Digitalization 12bit

Incidence angle nominal 55 degree

Polarization Vertical and Horizontal

Dynamic range 2.7-340K

Deployed

GCOM-W AMSR-2 provides higher space resolutions compared its precursor on EOS-Aqua (AMSR-E) and better design for mitigating radio frequency interference in land remote sensing application

Information Content from GCOM-W1 AMSR2

Duration 9 - 21 February 2012MSPD=115 mph  MSLP=932 (hPa)JAXA launched GCOM-W1 on Oct 18, 2011 with AMSR2 on board and NESDIS is developing

NOAA unique AMSR2 products for user community.

AMSR-E Ocean Products: Theoretical Base

Information Content from GCOM-W1 AMSR2 Hurricane Sandy-10-28-2012 06 UTC

Duration 9 - 21 February 2012MSPD=115 mph  MSLP=932 (hPa)

JAXA launched GCOM-W1 on Oct 18, 2011 with AMSR2 on board and NESDIS is developing NOAA unique AMSR2 products for user community.

SST SSW

Statement of Problems in GSI

• NCEP GSI (3DVar data assimilation system) is being used by community for both global and regional model analysis but its interface is not designed well for different model configurations

• In 2011 and 2012 version of Hurricane Weather Research Forecast (HWRF) model, most of satellite data are not used in HWRF analysis process due to its model top setup

• Analyses show GSI quality controls for satellite water vapor sounding data are also problematic (lots of bad data sneak into the analysis process).

• Bias correction schemes for satellite data developed for the global model applications have not been fully vetted for regional model applications

Pre

ssur

e (h

Pa)

ATMS Weighting Function

NCEP HWRF Top

STAR HWRF Top

ATMS Weighting Functions

Our approach: Raise the model top to allow for more satellite data assimilated into hurricane forecast model

O-B (MHS Channel 3 at 1800 UTC 05/22/08)

MHS Ch 3 Passing GSI QC MHS Ch3 Affected by Clouds over Oceans

MHS Ch3 Removed by New Cloud Algorithms WRF Cloud Liquid Water

Issues on GSI QC for SSMIS Imaging Channel (10/27/2012)

Yellow: SSMIS clear data (CLW<0.05mm) not passing QC Cyan: SSMIS cloudy pixels (CLW >0.05mm) passing QC

HWRF Model and Data Assimilation System

HWRF Model:

• 2012 NCEP-Trunk version 934

• Three telescoping domains:Outer domain: 27km: 75x75o; Inner domain: 9km ~11x10o

Inner-most domain: 3km inner-most nest ~6x6o

Revised Model Level and Top:

• Vertical levels: 61

• Model top: 0.5 hPa

Data Assimilation System:

• HWRF 6 hour forecasts

• GSI (3DVAR)

• The Hurricane Weather Research and Forecasting (HWRF) Model dynamical core is designed based on the WRF model using NCEP Non-Hydrostatic Mesoscale Model (NMM) core with a movable high-resolution nested grid (telescopic)

• Regional-Scale, Moving Nest, Ocean-Atmosphere Coupled Modeling System. Horizontal resolution: 27 km outer grid, 9 km inner grid, 42 vertical levels

• Non-Hydrostaticsystem of equations formulated on a rotated latitude-longitude, Arakawa E-grid and a vertical, pressure hybrid (sigma_p-P) coordinate.

• Advanced HWRF 3D Variational analysis that includes vortex relocation, correction to winds, MSLP, temperature and moisture in the hurricane region and adjustment to actual storm intensity.

• Uses SAS convection scheme, GFS/GFDL surface, boundary layer physics, GFDL/GFS radiation and Ferrier Microphysical Scheme.

• Ocean coupled modeling system (POM/HYCOM).

Control Experiment – L61

Conventional Data:

Radiosondes, aircraft reports (AIREP/PIREP, RECCO, MDCRS-ACARS, TAMDAR, AMDAR), Surface ship and buoy observations , Surface observations over land, Pibal winds,Wind profilers, VAD wind, Dropsondes

Satellite Instrument Data:

• AMSU-A (channel 5-14) from NOAA-18, NOAA-19 and METOP-A• HIRS from NOAA-19 and METOP-A • AIRS from EOS Aqua • ASCAT from METOP-A • GPSRO from GRAS/COSMIC

Sensitivity Experiments

Conv Only: No Satellite Radiance Data

ATMS: L61 + ATMS

CrIS: L61 + CrIS

IASI: L61 + IASI

1800 UTC 0000 UTC 0000 UTC day55-day Forecast

HWRF FST Turn on GSI

Period: 2012102218 ~ 2012102918, 29 cycles in total

HWRF GFS

Hurricane Sandy Tracks from NCEP GFS and HWRF Operational Forecasts

CONV OnlyL61

Impacts of Direct Assimilation of Operational Satellite Radiances in HWRF on Hurricane Sandy’s Track

L61:Control Run

Impacts of Direct Assimilation of Suomi NPP ATMS Radiances on Hurricane Sandy’s Track

L61+ATMS

Predicted vs. observed track for Hurricane Sandy during October 22 to 29. NCEP 2012 HWRF is revised with a high model and 6 forecast as background for direct satellite radiance assimilation in GSI. Control Run: All conventional data and NOAA/METOP/EOS/COSMIC. It is clearly demonstrated that assimilation of Suomi NPP ATMS radiance data reduce the forecast errors of Hurricane Sandy’s track . Slide courtesy: Fuzhong Weng, NOAA/STAR

Comparison of Temperature Increments from ATMS and AMSU-A

Shaded: ATMS Red contour: AMSU-ABlack contour: Conventional

ATMS and AMSU-A (NOAA-19) produce largest temperature innovation in storm regions in similar magnitudes and complementary in spatial coverage

L61+IASIL61 L61+CrIS

Impacts of Direct Assimilation of Hyperspectral Infrared Sounders on Hurricane Sandy’s Track

Multiple Forecasts of Max. Wind SpeedGFS HWRF

L61

34

CONVonly

Multiple Forecasts of Max. Wind SpeedL61

35

IASI

ATMS

CrIS

Multiple Forecasts of Min. Surf. Pres.

36

GFS HWRF

L61 CONV only

Multiple Forecasts of Min. Surf. Pres.

37

L61

IASI

ATMS

CrIS

48-hr forecast24-hr forecast

Track Forecast

Observation

ATMSIASI

L61

CrISObservation

ATMSIASI

L61

CrIS

96-hr forecast72-hr forecast

Track Forecast

Observation

ATMSIASI

L61

CrISObservation

ATMSIASI

L61

CrIS

Max. Wind Speed Forecast24-hr forecast 48-hr forecast

72-hr forecast 96-hr forecast

Observation

ATMSIASI

L61

CrISObservation

ATMSIASI

L61

CrIS

Observation

ATMSIASI

L61

CrISObservation

ATMSIASI

L61

CrIS

Min. Surf. Pres. Forecast24-hr forecast 48-hr forecast

72-hr forecast 96-hr forecast

Observation

ATMSIASI

L61

CrISObservation

ATMSIASI

L61

CrIS

Observation

ATMSIASI

L61

CrISObservation

ATMSIASI

L61

CrIS

Summary and Conclusions

• Our JPSS proving ground project is progressing very well and all the tasks in 2012-2013 are on track.

• Suomi NPP ATMS/CRIS added more values for improving hurricane monitoring and forecasts.

• ATMS is very unique for resolving hurricane warm core features through spatial oversampling and additional channels.

• AMSR2 provides SST and SSW information within the hurricane precipitation areas and the algorithms have been developed and tested for retrievals

• 2012 HWRF/GSI is re-configured with more vertical layers and higher model top for improving direct satellite radiance assimilation.

• Our control and sensitivity experiments show NPP ATMS improves Sandy’s track and intensity forecasts