Evaluation of Potential Impacts of Doppler Lidar Wind Measurements on High-impact Weather Forecasting: A Regional OSSE Study Zhaoxia Pu and Lei Zhang University of Utah Collaborate with Bruce Gentry NASA GSFC Belay Demoz Howard University Acknowledgements: Michiko Masutani, EMC/NCEP, Joint OSSEs Las Peter Riishojgaard, JCSDA Robert Atlas, NOAA/AMOL NASA/HQ 13 th IOAS-AOLS 89 th AMS Annual Meeting, Phoenix, AZ January 15, 2009

About this talk Observing system simulation experiments (OSSEs) with Doppler Wind Lidar (DWL) measurements Problems and challenges in regional OSSEs

Backgorund “Tropospheric winds are the number one unmet measurement for improving weather forecasts.” -- National Research Council NASA has supported a “Lidar wind science” program to help develop next generation space-based Doppler Wind Lidar (DWL) measurements. It is expected that the future DWL will provide much denser (high horizontal and vertical resolution) wind profile observations than the currently available rawinsonde networks It has also been well recognized that the future DWL should be operated in adaptive observation mode in order to save power resources.

General concept of OSSEs (Courtesy of R. Atlas 2008)

Previous OSSEs with Doppler Lidar Wind (DLW) Liu and Kalnay (2007), adaptive targeting strategies (simple global model) R. Atlats et al. (2008), Impact of DLW on hurricane track forecast (NASA global model) M. Masutani (2008), Evaluation of the impact of DLW with NCEP GFS D. Emmitt (2008), Various experiments with DLW The objective of OSSEs Access the impact of the data from future instruments Define minimum requirements for future instruments in order to meet a certain goals in weather/climate prediction. Most of previous OSSEs have been done with global models Impact of DLW data on high-impact weather system (e.g. hurricane) has been examined ( Atlas et al. 2008; Emmitt et al. 2008). Positive impact has been shown on hurricane track forecast.

Issues and Research Objectives For these high-impact weather systems, we not only care about their track, but are also concerned about their intensity and structures For instance Hurricane intensity forecast remains a challenging problem (Roger et al. 2006) Then: To what extent can next generation DLW observing system improve the hurricane intensity forecast?

Do ECMWF global natural runs represent hurricane structure? Case: 00UTC October – 00UTC October 4, 2005 SLP Track 3-h rainfall (T511) Tropical cyclones were found to be reasonably well represented in the Natural Run for a model of intermediate resolution.

Due to the relatively low resolution of the Natural 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. Are ECMWF natural runs adequate for verification purposes? 3-h rainfall T511 WRF 9-km WRF 3-km

Regional OSSEs Regional Natural Run: Regional numerical simulation with Advanced Research Version of Weather Research and Forecasting (WRF) model Nested inside of ECMWF T799 natural run Resolution: 27km, 9km and 3km Physics option: BMJ cumulus, LIN microphysics, MYJ PBL Simulation of the data (25-km resolution) CTRL run: WRF simulation nested inside of T511 Resolution: 27km, 9km and 3km Physics option: KF cumulus, WSM6 microphysics, YSU PBL Data Assimilation : WRF 3DVAR Verification

Data Impact Regional OSSE results

3-h precipitation at 48 h fcst CTRLExpTruth

Regional OSSE results Minimum central SLPMaximum surface wind

Some issues with regional OSSEs Forecast is very sensitive to physics Parameterization options Can be good or bad

Some issues with regional OSSEs

Exp.1 T799 IC and T799 BC Exp.2 T799IC and FNL BC Some issues with regional OSSEs Forecast is very sensitive to the boundary conditions

Concluding remarks Future DLW measurements could be a very useful data source for improving the forecast of high-impact weather systems (e.g., hurricane) More work needs to be done to explore the instrument strategies In order to improve the intensity forecast There are some issues (such as the impact of model physics, boundary conditions, and resolution) that need to be studied carefully with regional OSSEs in order to make a realistic assessment of the impact. The study is ongoing…