The Use of Artificial Neuron Networks for Retrieval Temperature and Humidity Sounding of the Atmosphere According to the Data of the MTVZA-GY Microwave Radiometer Installed on the Meteor-M No. 2-2 Satellite.
Saved in:
| Title: | The Use of Artificial Neuron Networks for Retrieval Temperature and Humidity Sounding of the Atmosphere According to the Data of the MTVZA-GY Microwave Radiometer Installed on the Meteor-M No. 2-2 Satellite. |
|---|---|
| Authors: | Filei, A. A.1 (AUTHOR) andreyvm-61@mail.ru, Andreev, A. I.1 (AUTHOR), Uspensky, A. B.2 (AUTHOR) |
| Source: | Izvestiya, Atmospheric & Oceanic Physics. Dec2021, Vol. 57 Issue 12, p1515-1526. 12p. |
| Subject Terms: | *Microwave radiometers, *Numerical weather forecasting, *Artificial neural networks, *Atmosphere, *Atmospheric layers, *Humidity |
| Abstract: | We consider the application of artificial neural networks for remote temperature and humidity profiles sounding of the atmosphere according to the data of the MTVZA-GY microwave radiometer installed on the Meteor-M No. 2-2 satellite. The satellite estimates of temperature and humidity profiles were compared with radiosonde data and with the products of numerical weather prediction models. In comparison with radiosonde data, the root-mean-square error of the temperature profile estimates does not exceed 3.0 K in the near-surface layer and is within 2 K in the rest of the atmospheric layer of 1000–10 hPa. The maximum root-mean-square error in the estimation of relative humidity profiles when compared with radiosonde data is 37% in the tropopause area, and when compared with the numerical weather predictive model data, does not exceed 20% in the entire atmospheric column. [ABSTRACT FROM AUTHOR] |
| Database: | Energy & Power Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | We consider the application of artificial neural networks for remote temperature and humidity profiles sounding of the atmosphere according to the data of the MTVZA-GY microwave radiometer installed on the Meteor-M No. 2-2 satellite. The satellite estimates of temperature and humidity profiles were compared with radiosonde data and with the products of numerical weather prediction models. In comparison with radiosonde data, the root-mean-square error of the temperature profile estimates does not exceed 3.0 K in the near-surface layer and is within 2 K in the rest of the atmospheric layer of 1000–10 hPa. The maximum root-mean-square error in the estimation of relative humidity profiles when compared with radiosonde data is 37% in the tropopause area, and when compared with the numerical weather predictive model data, does not exceed 20% in the entire atmospheric column. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 00014338 |
| DOI: | 10.1134/S0001433821120070 |
