The potentiality of neural networks for retrieving vertical profiles of atmospheric temperature and vapor is investigated. Inputs to the net are the brightness temperatures measured by a ground-based multichannel microwave radiometer and the surface measurements of temperature and relative humidity. A neural network algorithm has been tested comparing its retrievals with those obtained by means of a linear statistical inversion applied on the same data sets. The analysis has been limited to the case of profiles with clouds to test the ability of the neural network to face nonlinear problems. The technique has proven to be flexible, showing a good capability of exploiting information provided by other instruments, such as a laser ceilometer. A fault tolerance evaluation has also been considered, which showed interesting properties of robustness of the algorithm.
[1]
James H. Churnside,et al.
Temperature Profiling with Neural Network Inversion of Microwave Radiometer Data
,
1994
.
[2]
Geoffrey E. Hinton,et al.
Learning internal representations by error propagation
,
1986
.
[3]
D. A. Merritt,et al.
An Automatic Profiler of the Temperature, Wind and Humidity in the Troposphere.
,
1983
.
[4]
Hans J. Liebe,et al.
Propagation Modeling of Moist Air and Suspended Water/Ice Particles at Frequencies Below 1000 GHz
,
1993
.
[5]
Heekuck Oh,et al.
Neural Networks for Pattern Recognition
,
1993,
Adv. Comput..