Title: Analysis tools for online evaluation of the operational hurricane forecasts using satellite data
Presenting Author: Svetla M. Hristova-Veleva
Organization: Jet Propulsion Laboratory
Co-Author(s): Mark Boothe, S. Gopalakrishnan, Ziad Haddad, Brian Knosp, Bjorn Lambrigtsen, Peggy Li, Michael Montgomery, Noppasin Niamsuwan, Vijay Tallapragada, Simone Tanelli, Joseph Turk, Tomislava Vukicevic

Abstract:
To achieve the HFIP goals of improving the forecast accuracy of hurricane intensity, track and impact at landfall we first need to understand whether the models properly reflect the physical processes and their interactions. Some of the important questions to be answered are: Is the representation of the precipitation structure correct? Is the storm scale and asymmetry reflected properly? Is the environment captured correctly? Is the interaction between the storm and its environment represented accurately? Such studies require the use of large amounts of satellite data, coming from diverse instruments in order to create robust statistics. Due to the complexity of the remote sensing data and the volume of the respective model forecast this in-depth evaluation is usually limited to a number of case studies. With the goal to facilitate model evaluation that goes beyond the comparison of "Best Track" metrics, we are working on providing fusion of models and observations by bringing them together into a common system and developing online analysis and visualization tools. To help support the HFIP objectives, this NASA/ESTO/AIST-11-funded project is developed in close collaboration with our colleagues from NOAA/EMC and NOAA/AOML/HRD to bring the operational and research versions of HWRF forecasts into the satellite database. We will start by presenting how the operational HWRF forecasts are integrated. In particular, we focus on projecting the model fields into the observational space. For that purpose, we integrate the operationally-produced synthetic brightness temperatures (based on the use of CRTM and provided courtesy of EMC) with the brightness temperatures observed by a number of satellites (TMI, AMSR2, SSMI and SSMIS). In addition, we are working on bringing into the system the synthetic brightness temperatures that are produced by a different satellite simulator (NEOS3, developed under AIST-08, with PI S. Tanelli) while still using the same HWRF input. By providing this alternative way of modeling the synthetic brightness temperatures, we aim at shedding light on the uncertainty that comes from the forward modeling itself, an uncertainty that is not well understood or even explored, even though it is rather significant. A major focus of our efforts is to develop and implement a set of online analysis tools. In this presentation we will describe our current system that allows the user to select a region of interest and to perform statistical analysis of the brightness temperatures inside that region. Such an approach allows us to go beyond the point-to-point comparison and to obtain robust statistics that reflect important features of the storm structure and go beyond using only 'Best Track' data for model evaluation. Finally, we will outline our goals for the coming year, namely the development of additional analysis tools and the investigations of the impact of the microphysical assumptions. We will also present ideas for future efforts. The work described here was performed at the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration.