The paper is about the dimensionality reduction in remote sensing inversion problems using neural networks. The application domain is the estimation of ozone concentration profiles from the radiance measurements provided by the instrument Global Ozone Monitoring Experiment (GOME) on board of ESA satellite ERS-2. The complete dimensionality reduction technique includes the use of radiative transfer modeling with the final purpose of extracting the GOME spectral ranges most crucial for the inversion. The retrieval performance has been assessed by comparison with collocated profiles obtained with lidar measurements.
Computational Science (CS) and Computational Intelligence (CI) both share the same objective: finding solutions to difficult problems. However, the methods to the solutions are different. The main objective of this book series, "Transactions on Computational Science and Computational Intelligence", is to facilitate increased opportunities for cross-fertilization across CS and CI. This book series will publish monographs, professional books, contributed volumes, and textbooks in Computational Science and Computational Intelligence. Book proposals are solicited for consideration in all topics in CS and CI including, but not limited to,
The Global Ozone Monitoring Experiment (GOME) is a spectrometer boarded on the European Space Agency (ESA) Remote Sensing Satellite (ERS)-2, which measures, in a nadir-viewing geometry, the solar radiation that is scattered by the earth’s atmosphere. Measurements are performed in the wavelength range of 240–790 nm, with a spectral resolution of 0.2–0.4 nm. Vertical ozone profiles can be retrieved from observations of the backscattered light using neural network inversion schemes, which are able to provide real-time estimations with an accuracy that is comparable with the accuracy obtained by means of traditional optimal estimation (OE) methods, which are high-computation demanding and generally offline.
In this study, neural network–estimated profiles have been compared to the ozone profiles that are retrieved by the Improved Limb Atmospheric Spectrometer (ILAS) boarded on the Japanese Advanced Earth Observing Satellite (ADEOS). From November 1996 to June 1997 more than 3000 coincident profiles have been found in the northern and southern high-latitude regions of the globe. The relative difference between GOME and ILAS profiles over the entire dataset and over subsets corresponding to different measurement and seasonal conditions have been calculated, and the results have been critically discussed.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.