Citation
Abstract
We examine twenty years of weather analysis data (WDA) acquired from the Deep Space Network (DSN) sites as well as the European Centre for Medium-Range Weather Forecasts (ECMWF) fifth generation ECMWF atmospheric reanalysis (ERA5) data referenced to those sites spanning from 2001 to 2020. Annual and seasonal statistics of atmospheric attenuation and atmospheric noise temperature were estimated from the data using the established model formulation of the Radiocommunication Sector of the International Telecommunications Union (ITU-R). We also conducted comparisons of these statistics with those extracted from water vapor radiometers (WVR) that reside or have resided at the sites for overlapping years. This article attempts to characterize the statistical behavior of the extracted atmospheric attenuation and atmospheric noise temperature over a 20-year period. We find that the data types show no significant trends within the 20-year period and that there are some site-dependent differences in the statistics with respect to those of other data sources (e.g., WVR, ITU). These data sets were developed as alternatives to ones used in the machine learning (ML) forecast models.
Details
- Volume
- 42-231
- Published
- November 15, 2022
- Pages
- 1–19
- File Size
- 1.7 MB