Aerosol size distribution using sun-photometer AOD data of five wavelengths and artificial neural network

Authors:
Hamed Parsiani;Andres Bonilla
Affiliations:
Department of Electrical and Computer Engineering, University of Puerto Rico, Mayagüez, Puerto Rico;Department of Electrical and Computer Engineering, University of Puerto Rico, Mayagüez, Puerto Rico
Venue:
WSEAS TRANSACTIONS on SYSTEMS
Year:
2008

Citing 3
Cited 0

The optimal multi-layer structure of backpropagation networks

NN'06 Proceedings of the 7th WSEAS International Conference on Neural Networks
Neural network solution to the cogeneration system by using coal

ICC'08 Proceedings of the 12th WSEAS international conference on Circuits
Null steering beamformer based on RBF neural networks

ICCOM'08 Proceedings of the 12th WSEAS international conference on Communications

Quantified Score

Hi-index	0.00

Visualization

Abstract

Aerosol size distribution (ASD) is an integral parameter in regional atmospheric models [1]. The LIDAR laboratory at UPRM will provide Puerto Rico a means of measuring ASD, and therefore improve these models. This project intends to develop a method of obtaining ASD with the use of a Sun-photometer data, local CIMEL data obtained from AERONET (Aerosol Robotic Network) [2], and an artificial neural network (ANN). The Sun-photometer used is an instrument that measures Aerosol Optical Depth (AOD) at five wavelengths of 380, 440, 500, 675, and 870 nms. A feed-forward, back-propagation artificial neural network was used to map the underlying pattern between AOD and ASD in southwestern Puerto Rico. After training, the network will produce ASD outputs based on new AOD inputs measured locally at UPRM with the Sun-photometer. It was determined that accurate predictions of ASD (MSE of 10-4) could be made depending on the size of the AOD data pool selected.