Extended Abstract (304K)P9.7
Evaluation of the oceanic cloud-top height diagnostic product: strategy of the verification methodology
Agnes Takacs, NCAR, Boulder, CO; and B. G. Brown, R. Hueftle, L. D. Holland, S. Madine, J. L. Mahoney, and M. Kay
The Oceanic Weather Product Development Team (OW PDT), sponsored by the FAA’s Aviation Weather Research Program (AWRP) has developed the oceanic cloud-top height diagnostic product (CTOP) to meet aviation requirements. The CTOP is based on a combination of GOES IR reflectance values with temperature and pressure profiles from the Global Forecast System (GFS) numerical weather prediction model.
The greatest challenge associated with creating accurate and timely oceanic weather hazard information is the lack of data available for algorithm development and verification. To complicate the problem, the observation datasets used for verification should be independent of those used by the algorithm. In a data sparse environment, this independence is difficult to achieve. Therefore, it is very important to discover creative approaches for extracting as much information as possible from the limited amount of global data. The AWRP’s Quality Assessment Product Development Team has investigated approaches for evaluating the performance of the CTOP algorithm. The overall verification strategy and initial results are described in this paper, and extensive results from the primary verification study are described in a companion paper.
Cloud top height is not observed directly, except by lidar observations over land. Therefore, to obtain matching datasets for verification, cloud-top height values were inferred using several different data sources that are not used by the algorithm to diagnose CTOP. A primary dataset for this evaluation is the GOES sounder-based cloud-top pressure product produced by NESDIS (CTP). Radiosonde and radar observations available over the CONUS coastal areas and islands are also used to create estimates of cloud-top height (RCT and ECT, respectively). These types of observations provide upper and lower bounds on the expected CTOP values. The RCT and ECT observations are compared to each other and to the NESDIS CTP and the CTOP values. These comparisons have aided in the interpretation of the results over open ocean regions, where the evaluation has been extended using only the NESDIS CTP for verification of CTOP. The latter verification technique and associated verification results are discussed in Part II of this paper.
Poster Session 9, Modelling and Verification, Poster Session
Thursday, 7 October 2004, 3:00 PM-4:30 PM
Previous paper Next paper