Evaluation of the Optical Autocovariance Wind Lidar OAWL

Evaluation of the Optical Autocovariance Wind Lidar (OAWL) and its Potential Impact on Numerical Weather Prediction: A collaboration between: • • • NOAA/AOML: R. Atlas, T. Vukicevic, L. Bucci NOAA ESRL: Mike Hardesty, Y. Xie Ball Aerospace and Technologies Corp: S. Tucker, C. Weimer Simpson Weather Associates: G. D. Emmitt, S. Wood, S. Greco NASA/GSFC/SSAI: Eric Kemp, Jossy Jacob JCSDA: L. P. Riishojgaard. M. Masutani, S. Casey, J. Woollen

Outline • • • Objectives of OSSEs Methodology and earlier results for winds Need for regional, as well as, global OSSEs Current hurricane related OSSEs Description of the OAWL OSSEs that are beginning at this time

OBSERVING SYSTEM SIMULATION EXPERIMENTS OBJECTIVES: 1. To provide a QUANTITATIVE assessment of the potential impact of proposed observing systems on earth system science, data assimilation, and numerical weather prediction. 2. To evaluate new methodology for the processing and assimilation of remotely sensed data. 3. To evaluate tradeoffs in the design and configuration of proposed observing systems (e. g. coverage, resolution, accuracy and data redundancy). 4. Can also be used to determine the ability of existing observing system to detect climatic trends and to optimize the global observing system for climate monitoring and other applications.

EARLY SIMULATION STUDIES 1. • • 2. PROVIDED AN ANALYSIS OF GARP DATA REQUIREMENTS “USEFUL” RANGE OF PREDICTABILITY NEED FOR REFERENCE LEVEL DATA RELATIVE USEFULNESS OF ASYNOPTIC vs SYNOPTIC DATA ASSIMILATION INDICATED THAT • ALL THREE OF THE PRIMARY VARIABLES (TEMPERATURE, MOISTURE, WIND) COULD BE DETERMINED IF A CONTINUOUS TIME HISTORY OF ANY ONE OF THESE VARIABLES WERE INSERTED INTO A GENERAL CIRCULATION MODEL. (“CHARNEY CONJECTURE”)

LIMITATIONS OF EARLY STUDIES • MOST IMPORTANT IS THE USE OF THE SAME MODEL FOR NATURE AND ASSIMILATION / FORECASTING “IDENTICAL TWIN EXPERIMENTS” • MODEL DEPENDENCE OF RESULTS • TREATING OBSERVATIONAL ERRORS AS RANDOM AND UNCORRELATED

Previous OSSEs 1. EVALUATED THE RELATIVE IMPACT OF TEMPERATURE, WIND AND MOISTURE DATA - These experiments showed wind data to be more effective than mass data in correcting analysis errors and indicated significant potential for spacebased wind profile data to improve weather prediction. The impact on average statistical scores for the northern hemisphere was modest, but in approximately 10% of the cases a significant improvement in the prediction of weather systems over the United States was observed. 2. EVALUATED THE RELATIVE IMPORTANCE OF UPPER AND LOWER LEVEL WIND DATA. - These experiments showed that the wind profile data from 500 hpa and higher provided most of the impact on numerical forecasting. 3. EVALUATED DIFFERENT ORBITAL CONFIGURATIONS AND THE EFFECT OF REDUCED POWER FOR A SPACE-BASED LASER WIND SOUNDER (LAWS). - These experiments showed the quantitative reduction in impact that would result from proposed degradation of the LAWS instrument. 4. DETERMINED DRAFT DATA REQUIREMENTS OF SPACE-BASED LIDAR WINDS. -These experiments evaluated different coverages, resolutions, and accuracies for lidar wind measurements to estimate both research and operational requirements for the Global Tropospheric Wind Sounder (GTWS) Mission.

Previous OSSEs (continued) 5. DEVELOPED AND TESTED IMPROVED METHODOLOGY FOR ASSIMILATING SATELLITE SCATTEROMETER DATA. - Applying this methodology resulted in the demonstration of the first significant positive impact of real scatterometer data in 1983. 6. DEVELOPED AND TESTED DIFFERENT METHODS FOR ASSIMILATING SATELLITE SURFACE WIND SPEED DATA. - This led to assimilation of SSM/I wind speed data to improve ocean surface wind analyses. 7. EVALUATED THE QUANTITATIVE AND RELATIVE IMPACT OF ERS AND NSCAT YEARS PRIOR TO THEIR LAUNCH. - These results were confirmed after the launch of both instruments. 8. EVALUATED THE QUANTITATIVE IMPACT OF AIRS SOUNDING DATA AND THE IMPORTANCE OF CLOUD-CLEARING. These results were also confirmed by later data impact experiments with real AIRS data.

Quick. OSSEs 1. Quick. OSSEs follow a similar methodology, but are generally performed for much shorter periods and often as single forecasts after a limited data assimilation. 2. They have the advantages of being cheaper and faster to perform, and can sometimes be used to answer questions relating to a particular storm or to demonstrate potential. 3. There utility is very limited in that they typically cannot be used to establish statistically significant quantitative results.

Why do we need regional OSSEs? • For many high-impact weather systems (e. g. , hurricanes etc. ), we are not only care about their tracks, but also concern about their intensities and structures • Resolutions of global models are commonly too coarse to resolve the intensity and structure of those high impact systems -- A specific example: hurricane intensity forecasts remain one of the major challenges in operational forecasts

Are ECMWF nature runs adequate for verification purposes? T 511 3 -h rainfall WRF 9 -km WRF 3 -km Due to the relatively low resolution of the Nature Run in comparison with the scale of the storm inner core structure, the data may only be useful for the OSSEs that focus on tropical cyclone track forecasting rather than intensity forecasting.

Regional OSSEs “Regional Nature Run”

ECMWF Nature run hurricanes

High Resolution Hurricane Nature Run: WRF Simulation Embedded Inside the ECMWF Nature Run 60 levels; 3 km resolution; double-moment microphysics; advanced radiation schemes. R I ECMWF T 511 Nature Run 3 km WRF-ARW Nature Run

Further improvements from 1 km resolution: 3 km simulation nested to 1 km for 18 hours • More realistic distribution of precipitation • 20 grid points between each arrow shown above

OBSERVING SYSTEM SIMULATION EXPERIMENTS Objectives for Hurricanes: 1. Evaluate the potential impact of new (proposed) observing systems on hurricane track and intensity predictions. 2. Evaluate tradeoffs in the design and configuration of proposed observing systems (e. g. coverage, resolution, accuracy and data redundancy). 3. Optimize sampling strategies for current and future airborne and spacebased observing systems. 4. Evaluate and improve data assimilation and vortex initialization methodology for hurricane prediction.

Earlier OSSEs for Hurricanes 1. Global OSSEs Using 3&1/2 month fv. GCM Nature run at. 5 deg resolution - aimed at evaluating the potential impact of Doppler Lidar winds on hurricane track prediction 2. Global Quick OSSE using. 25 deg fv. GCM 5 -day forecast as Nature - aimed at evaluating impact of wind profile observations on the forecast track for an Ivan like hurricane and in testing hypotheses relating to hurricane track forecasting 3. Regional Quick OSSE using mm 5 nature run - aimed at evaluating the potential impact of HIRAD on hurricane surface wind analyses 4. Regional Quick OSSEs using WRF ARW 3 -5 day forecasts as nature runs -aimed at evaluating potential value of AIRS, Doppler Wind Lidar or other data for hurricane intensity forecasting.

Potentional Impact of new space-based observations on a Hurricane Track Prediction • Tracks • Green: actual track • Red: forecast beginning 63 hours before landfall with current data • Blue: improved forecast for same time period with simulated wind lidar

Current and planned OSSEs for hurricanes • OSSEs in support of HFIP to evaluate sampling strategies for hurricane reconnaissance, new observing systems, modeling and data assimilation and predictability. • OSSEs to determine potential impact of UAS and to optimize sampling strategies. • OSSEs to evaluate hyperspectral sounders • OSSEs to evaluate alternative wind lidar technologies: OAWL/FI vs. 3 d Winds Hybrid

Task 1. Construct a conceptual model for a spacebased OAWL • SWA and BATC arrive at consensus for simulation of OAWL/FI data. • Three levels of confidence will be attempted: spacebased system performance scaled from demonstrated performance with ground and/or airborne system; performance scaled with use of SOTA hardware and proven performance; and, performance using projected improvements in key hardware components (optics, detectors, etc).

Task 2. Generate a realistic data product • (~ 2 weeks of simulation) for a space-based OAWL/FI using the ECMWF’s T 511 Nature Run (NR) with the Doppler Lidar Simulation Model (DLSM) currently operating through the GSFC Software Systems Support Office (SSSO) portal. • Produce a similar data set for the GSFC’s fv. GCM NR.

Task 3. Compare the OAWL/FI data products with those previously generated for 3 D Winds using the same NR and DLSM • These comparisons will provide a first order sense of the advantages and disadvantages of the NASA/GSFC IDL-vetted OAWL/FI relative to the current “hybrid DWL” concept that went through the IDL and MDL. • Previous hybrid DWL design concepts have been assessed through global OSSEs done by the Joint Center for Satellite Data Assimilation (JCSDA) under NASA funding provided under the Wind Lidar Science element of ROSES 2007.

Task 4. Conduct an initial impact study • This initial OSSE will use the DLSM to generate an extended set of simulated OAWL/FI data for a one- three month period using the fv. GCM NR. • These impact studies will be conducted using the Sensor Web OSSE system at the SSSO.

Tasks 5. Conduct a global OSSE using the T 511 NR • A comprehensive OSSE using the current joint global OSSE system based on the T 511 NR will be conducted. These experiments will follow rigorous internationally accepted OSSE procedures and evaluate the relative impact of the OAWL and 3 DWINDS lidars on data assimilation and numerical weather prediction. • These experiments will be generated by the JCSDA, and evaluated collaboratively by both JCSDA and AOML.

Tasks 6. Conduct a regional OSSE using the 1 km resolution WRF ARW embedded in the T 511 NR • A comprehensive regional OSSE using a two week 1 km resolution NR will be conducted to evaluate the relative impact of the OAWL and Hybrid-system lidars on hurricane genesis, track and intensity forecasting. • Data assimilation and forecasts will be generated using the SOTA operational NOAA Hurricane forecast model (HWRF) and ensemble Kalman filter data assimilation system at AOML.

Backup Slides

III. WRF Model Configurations and Execution …after extensive testing the following options were implemented in WRF 3. 2. 1… Domain: • 27 km resolution grid covering tropical Atlantic. • Nesting to 9 km, 3 km, and 1 km. Inner grid 480 km x 480 km. • 60 vertical levels Physics: • Kain-Fritsch cumulus parameterization on 27 km and 9 km grids. • 6 -class, double-moment microphysics (WDM 6). • New RRTM advanced longwave and shortwave radiation schemes (radiation called every six minutes). • YSU PBL scheme with TC-relevant modifications to Ck and Cd. • Simple mixed layer scheme for ocean cooling under the storm.

Data Sets The output from these simulations has been archived. • All WRF state variables, diagnosed surface fields, and all physics tendencies (boundary layer, friction, cumulus, radiation). • 3 D fields saved every 30 minutes on the 27 km, 9 km and 3 km grids. • 3 D fields saved every 6 minutes on the 1 km grid. 1 hour = 16 GB. • gzipped total output ~ 2 TB After further validation, we plan to make the Net. CDF output files freely available on the internet. Anyone can study inner-core structure, rapid intensification, rainbands, etc. , without having to expend 96 CPUs x 3 weeks ~ 48000 CPU hours.