Adaptive use of research aircraft data sets for hurricane forecasts |
| |
| Authors: | M K Biswas T N Krishnamurti |
| |
| Institution: | (1) Department of Meteorology, Florida State University, Tallahassee, FL, USA |
| |
| Abstract: | Summary This study uses an adaptive observational strategy for hurricane forecasting. It shows the impacts of Lidar Atmospheric Sensing
Experiment (LASE) and dropsonde data sets from Convection and Moisture Experiment (CAMEX) field campaigns on hurricane track
and intensity forecasts. The following cases are used in this study: Bonnie, Danielle and Georges of 1998 and Erin, Gabrielle
and Humberto of 2001. A single model run for each storm is carried out using the Florida State University Global Spectral
Model (FSUGSM) with the European Center for Medium Range Weather Forecasts (ECMWF) analysis as initial conditions, in addition
to 50 other model runs where the analysis is randomly perturbed for each storm. The centers of maximum variance of the DLM
heights are located from the forecast error variance fields at the 84-hr forecast. Back correlations are then performed using
the centers of these maximum variances and the fields at the 36-hr forecast. The regions having the highest correlations in
the vicinity of the hurricanes are indicative of regions from where the error growth emanates and suggests the need for additional
observations. Data sets are next assimilated in those areas that contain high correlations. Forecasts are computed using the
new initial conditions for the storm cases, and track and intensity skills are then examined with respect to the control forecast.
The adaptive strategy is capable of identifying sensitive areas where additional observations can help in reducing the hurricane
track forecast errors. A reduction of position error by approximately 52% for day 3 of forecast (averaged over 7 storm cases)
over the control runs is observed. The intensity forecast shows only a slight positive impact due to the model’s coarse resolution.
Corresponding author’s address: T. N. Krishnamurti, Department of Meteorology, Florida State University, Tallahassee, FL 32306-4520,
USA |
| |
| Keywords: | |
| 本文献已被 SpringerLink 等数据库收录！ |
|
