Form NRESS300 version 3.0 APR 09 PI name oreste reale organization name universities space research association columbia NASA proposal number 17TASNPP170045 proposal title using airs and cris data…

The National Aeronautics and Space Administration (NASA) awarded a $663,831.74 grant to the Universities Space Research Association (USRA), a nonprofit organization, to conduct research on using data from the Atmospheric Infrared Sounder (AIRS) and the Cross-track Infrared Sounder (CrIS) to improve the assimilation of hyperspectral sensor data in NASA's Global Earth Observing System (GEOS-5) global data assimilation and forecasting system, with a focus on understanding and predicting tropical cyclone processes. This 3-year successor effort builds on USRA's prior work demonstrating the benefits of assimilating cloud-cleared AIRS data. The research aims to develop a comprehensive methodology for adaptively assimilating cloud-cleared radiances from multiple hyperspectral sensors to enhance global weather forecasts and tropical cyclone prediction. No set-aside was used for this grant, which was awarded on May 15, 2019 and has a completion date of November 30, 2021.

Form NRESS300 version 3.0 APR 09 PI name oreste reale organization name universities space research association columbia NASA proposal number 17TASNPP170045 proposal title using airs and cris data to understand processes affecting tropical cyclone structure in a global data assimilation and forecasting framework. Section VII project summary a three-year successor proposal previous award NNX14AK19G is presented. The underlying motivation is to improve the assimilation of airs and cris radiances in a global data assimilation and forecast system the NASA GEOS5 and its successors. The final goal is a comprehensive methodology to adaptively assimilate cloud-cleared radiances from all hyperspectral sensors improving our understanding of tropical cyclone TC processes and our predictive ability. Past published work authored by this team has demonstrated the usefulness of assimilating cloudcleared airsderived information. More recent work under the currently funded effort has shown that airs data may negatively interact with each other and with other hyperspectral sensors because of suboptimal sampling and correlated errors in data assimilation DA errors are assumed to be uncorrelated. The current approach in DA aimed at reducing both computational costs and the effects of error correlation resorts to a data subsampling on regular grids called thinning the sampling resulting from this approach is suboptimal because impacts from different data are dependent upon the meteorology. IT is well understood that modest improvements in forecast skill can be achieved by the addition of observations in regions where the observational network is already dense, initial conditions are accurately known, or where error growth is small. Consequently, adaptive strategies based on ingesting more observations in areas where observational errors are large or where forecast errors grow most rapidly, have been proposed.unfortunately, strategies of adaptive thinning which take into account correlation errors in different regions are exceptionally complex and hard to implement. This team has been developing a simple, TC-centered adaptive thinning strategy for both airs and cris clear-sky radiances demonstrating that increased data coverage around TCS and decreased data coverage elsewhere can improve both global forecast skill and TC representation. Moreover this team has worked extensively on the assimilation of airs cloud-cleared radiances demonstrating that they are a substantially superior product compared to clearsky radiances, as long as they are more aggressively thinned globally. The main goal of this proposal is to provide a route towards the operational use of airs and cris cloudcleared radiances to improve the global analyses and forecasts with focus on TCS. In particular, the methodology will consist of an extensive suite of observing system experiments oses assimilating airs and cris radiances in the NASA GEOS5 and its successors in a variety of thinning configurations. In addition, the gmao expertise in applying the adjoint of the forecast and analysis system to assess observation impact will benefit the research. With this tool the amount by which each individual observation reduces or increases a defined measure of forecast error over a given area is calculated. This will provide an estimate of the fractional contributions of different sensors in the same thinning configurations. The combination of oses and adjoint models to estimate observational impacts and guide experiment design has been demonstrated to be particularly effective and will allow the design of a comprehensive thinning strategy for all hyperspectral sensors, including iasifocusing on cloudcleared radiances. This proposal will also benefit through interaction and feedback with the science and data production teams the overall validation of airs and cris products. This proposal addresses the science data analysis element of the a.37 the science of terra aqua and suomi NPP call an