Title :
Surface winds from the SSM/I using neural networks
Author :
Krasnopolsky, V.M. ; Breaker, L.C. ; Gemmill, W.H.
Author_Institution :
Environ. Modeling Center, Nat. Center for Environ. Prediction, Camp Springs, MD, USA
Abstract :
An improved neural network (NN) wind speed retrieval algorithm which covers the entire range permitted for retrievals both with respect to moisture and wind speed is presented. Improvements in the NN approach have permitted the development of a wind speed retrieval algorithm which can retrieve wind speeds up to ~25 m/sec for LWP concentrations <0.5 kg/m2, with a bias of ~0.2 m/s and an rms error of ~1.75 m/s. Also, results are presented which demonstrate significant correlation (correlation coefficient ~0.75) between wind directions retrieved, using a NN, and buoy wind direction. Although the F10 SSM/I data set which was used in this study was noisy, higher quality data are now being used to develop a prototype for a NN wind direction retrieval algorithm for the SSM/I
Keywords :
atmospheric boundary layer; atmospheric techniques; geophysical signal processing; image processing; remote sensing by radar; wind; 0 to 25 m/s; SSM/I; correlation; neural network; surface winds; wind speed retrieval algorithm; Electronic mail; Ink; Moisture; Neural networks; Noise generators; Predictive models; Sea surface; Springs; Transfer functions; Wind speed;
Conference_Titel :
Geoscience and Remote Sensing Symposium, 1996. IGARSS '96. 'Remote Sensing for a Sustainable Future.', International
Conference_Location :
Lincoln, NE
Print_ISBN :
0-7803-3068-4
DOI :
10.1109/IGARSS.1996.517862
