Past 2008 JCSDA Seminars

Over the past 10 years, and most likely for the next 10 years, the Numerical Weather Prediction (NWP) community has faced, and will be facing, an unprecedented volume of new satellite data available for assimilation into NWP forecast systems. Simultaneously, the NCEP Environmental Modeling Center (EMC) is redesigning the operational suite of forecast systems, aka the NCEP Production Suite, to provide improved information to users and, simultaneously, a software suite capable of supporting a broader diversity of forecast models. This seminar will present a strategic path for the future wherein all these factors are considered.

It is known that operational mesoscale forecast models do not perform well on propagating summer rainfall over the central United States. Such precipitation characteristics are coupled with subsynoptic-scale perturbations embedded in the midtropospheric flows. Analysis of the North American Mesoscale model (NAM) forecasts found that the model tends to generate the perturbations with a propagation speed that is too slow. The speed bias results in displaced rainfall forecasts.

The GFS assimilation of FORMOSAT-3/COSMIC data in an experimental run during summer 2006 was evaluated. The diagnostic analysis was focused on the global stationary wave structure in the Northern Hemisphere. Results show that large impacts of the FORMOSAT- 3/COSMIC observations are mainly distributed over the major mountain ranges and the western tropical Pacific warm pool. Water vapor flux convergence is found to be enhanced over the warm pool region, resulting in more precipitation in the GFS forecasts.

The Air Force Weather Agency is actively collaborating with NASA Goddard Space Flight Center (GSFC) Hydrological Sciences Branch personnel to further develop the Land Information System (LIS) as a replacement to AFWA Agriculture Meteorological (AGRMET) model. The higher spatial resolution, modular design, and configurable grid capability in LIS will arm AFWA with an enhanced surface modeling system to help support global and regional DoD joint service surface characterization requirements and NWP surface layer initialization needs. Since 2005, AFWA has sponsored several LIS science and infrastructure advancement projects including precipitation analysis improvements, Ensemble Kalman Filter data assimilation module integration, LIS and Weather Research and Forecasting (WRF) coupling evaluation, and CRTM interface design. AFWA is also working with the NASA GSFC Snow Team to advance AFWA's global snow measuring capability, using newer satellite systems and more complex data merging techniques to better capture global snow cover and depth measurements. Finally, AFWA is embarking on a new plan to greatly improve its cloud analysis system, which will further improve the resolution and capabilities of the cloud analysis used to calculate the surface energy budget. The infrastructure advancements, along with our strong working relationship with the NCEP land team, will ultimately lead to a much improved AFWA surface characterization system supporting the nation's armed services.

Tropical cyclone (TC) intensity forecasting is a challenging problem in both the research and operational communities. With the advanced research version of the WRF model, several case studies are conducted to investigate two main problems: 1) What are the factors limiting the TC intensity forecast? 2) To what extent can data assimilation helps improve the TC intensity forecast? To achieve the above goals, high resolution numerical simulations are performed. Comprehensive satellite and in-situ data sets, collected from the NASA Tropical Cloud Systems and Processes (TCSP) Experiment, are assimilated into the WRF model with its 3DVAR system. The results show that the forecast of TC intensity is highly sensitive to the physical parameterizations in the WRF model. It is also indicated that the WRF model has a problem capturing the rapid intensity change of TCs. The QuikSCAT ocean surface winds, GOES-11 AMVs, dropsonde data, and airborne Doppler radar data from the TCSP mission show significant impacts on the storm vortex structure and environmental features. The enhanced data has greatly improved the intensity, track, and precipitation forecasts of TCs.

The Weather Research and Forecasting (WRF) model and its variational assimilation system (WRF-Var) are widely used by both the research community and some operational centers. A general satellite radiance assimilation framework has been developed in the WRF-Var system over the past three years. The WRF-Var radiance assimilation capability was designed to meet the requirements of both basic research and operational applications,and will be available to the research community along with the community WRF system.

Radiance assimilation capabilities in the WRF-Var - the fast radiative transfer model, bias correction algorithm, quality control, and observation error tuning - will be described. Both the RTTOV and CRTM radiaitve transfer systems are incorporated into the WRF-Var system. Case study results on assimilating AMSU-A observations to improve Katrina track and intensity analysies and forecasts will be presented. Extended experiments over different regions to assess radiance assimilation impact yield encouraging results. Preliminary findings on cloud/rain affected radiance assimilation using CRTM will also be shown. The presentation will conclude with a demonstration of radiance assimilation with the WRF- 4DVAR system.

The atmospheric data assimilation system used by the Global Modeling and Assimilation Office (GMAO) uses the GEOS-5 finite volume atmospheric model and the Gridpoint Statistical Interpolation (GSI) analysis scheme developed at NCEP. The system is now being used to generate products input to NASA instrument team algorithms and also to generate MERRA, an atmospheric reanalysis for the satellite era. The GEOS-5 DAS is also used to contribute to satellite data assimilation issues relevant to the JCSDA. For example, the adjoint system developed for the DAS has been used to investigate observation impacts and work has begun to investigate the impact of cloud-cleared radiances on forecast skill. This presentation will highlight some recent results and also some preliminary results from a newly developed 4DVAR version of GEOS-5.

The Orbiting Carbon Observatory (OCO) is currently under development by the NASA Jet Propulsion Laboratory to identify and characterize natural CO₂ sinks. This Earth System Science Pathfinder mission is scheduled for launch in December 2008. During its nominal two-year operational lifetime, OCO will make space-based measurements of CO₂ and molecular oxygen (O2) over the sunlit hemisphere of the Earth. These data will be analyzed with remote sensing algorithms to retrieve estimates of the column- averaged CO₂ dry air mole fraction, XCO₂ with the accuracy and sampling resolution needed to characterize surface sources and sinks of CO₂ on regional scales over the entire globe. The observatory consists of a dedicated spacecraft bus that carries and points a single instrument. This instrument incorporates 3 high-resolution grating spectrometers that make coincident measurements of reflected sunlight in near-infrared CO₂ and molecular oxygen (O2) bands. The pre-flight qualification and calibration testing of the OCO instrument has just been completed. These tests describe the instrument's radiometric, spectral, and spatial performance. The end-to-end instrument performance was verified by recording atmospheric solar spectra with the flight instrument and comparing these results to spectra recorded simultaneously from a collocated ground-based high-resolution Fourier transform spectrometer. This comparison indicates that the instrument meets or exceeds its design objectives and will provide excellent data for XCO₂ retrievals.

This presentation introduces the audience to some basic concepts, terminology, and practices related to the verification of weather forecasts. To convey the broad scope of the topic, objective verification of both deterministic and probabilistic forecasts is discussed. Anomaly correlations and phase errors are computed for verifying the Hydrometeorological Prediction Center's (HPC) deterministic forecasts of mean sea level pressure. HPC quantitative precipitation forecast verification exemplifies the use of 2 X 2 contingency tables applied to deterministic forecasts. Finally, verification of HPC's probabilistic heat index forecasts demonstrates use of the Brier score and the attribute diagram.

Based on both the National Center for Atmospheric Research Advanced Research Weather Research and Forecasting (ARWRF)- Variational and Joint Center for Satellite Data Assimilation Global Statistical Interpolation data assimilation systems, Advanced TIROS Operational Vertical Sounder and Special Sensor Microwave Imager Sounder radiance data were assimilated into the ARWRF mesoscale forecasting system. A series of experiments were designed to access the model forecast accuracy over North America, and Southwest and East Asia. The statistical results show that the satellite data assimilation improves the initial conditions and reduces the model errors somewhat.

The Strategic Satellite Plan is the first NOAA plan to assess, formulate and ascribe a notional architecture of satellites, sensors and ground architecture to support NOAA's observation requirements. This briefing will discuss the analyses accomplished, the priorities, and the projected program through FY2020. It outlines a plan to satisfy requirements, trade studies that need to be conducted, a notional set of satellite systems and partnerships to accomplish the mission.