Analysis of neural network capabilities in IPDA spaceborne lidar sensing of CO2 using heterogeneous a priori data
G.G. Matvienko, A.Ya. Sukhanov, S.V. Babchenko
V.E. Zuev Institute of Atmospheric Optics of Siberian Branch of the Russian Academy of Science, 1, Academician Zuev square, Tomsk, 634055, Russia
Keywords: атмосфера, космический лидар, углекислый газ, парниковый газ, нейронная сеть, atmosphere, spaceborne lidar, carbon dioxide, greenhouse gas, neural network
Abstract
A possibility of retrieving the columnar concentration of carbon dioxide using a neural network is analyzed, as well as the concentration profile when sounding from a space orbit of 450 km and from an aerostat at altitudes of 23 and 10 km. Possibilities of using a priori data on temperature, pressure, and reflected and scattered signals are considered. The errors of retrieval of the columnar CO2 are 0.15% and 0.5% at altitudes lower than 2 km for lidar with a telescope diameter of 1 m and laser pulse energy of 50 μJ at a resolution of 60 km.
|
