Problem: predict thermodynamic state of the atmosphere (an open spatiallly-distributed thermodynamic system)
making the use of (incomplete indirect) data acquired with a stationary upward-looking ground-based radiometer.
Reformulating the problem:
a) parametrize a large stochastic system/data series
b) using observation of one stochastic vector process, predict another vector process.
Solution: neural network trained with avalable data sets/corresponding snapshots of the dynamic system states;
incorporating (when possible) theoretical models of links between observed processes and system states of interest.
Predicting cloud/fog onsets with ground-based
(FTIR-) radiometer data via COIF-NN
COIF-NN - cloud onsets' instant foreteller - neural network
Training set for the COIF-neural network has been composed with avaiable
downwelling infra-red radiation spectra (ground-based AERI ),and
radiosonde observations (RAOB) of atmospheric temperature and humidity profiles (ARM Archive)
Validation of time/altitude of cloud onsets has been performed with laser backscatter (LIDAR Ceilometer) data.
EXAMPLES OF PREDICTION
- Barrow , October 4, 2001
Figure 1. Lidar data & Predicted gap evolution
"Gap"= difference between actual- and dewpoint-temperature profiles: an onset occurs when the "gap" reaches "0"
Prediction has started at 0:29 GMT, driven by the operator of evolution which has been estimated over the time window (Oct. 3) 21:00 - (Oct. 4) 0:22 GMT. Note, that with respect to the lidar backscatter data a cloudbase was detected at altitude 1.1 km about 3 hours GMT.
Figure 2. Prediction of IR- spectrum evolution (the fisrt 4 spectral parameters are displayed)
Figure 3. Predicted evolution of IR-spectrum
aka Геннадий Рыжиков