@manual{81281,
  author = {Katrin Lonitz and C. Marquardt and Neill Bowler and Sean Healy},
  title = {Final Technical Note of 'Impact assessment of commercial GNSS-RO data'},
  abstract = {<p>
  <span dir="ltr" role="presentation">The addition of Spire and COSMIC-2 data is a large increase in the GNSS-RO data being assimi-</span><br role="presentation" /><span dir="ltr" role="presentation">lated at ECMWF and Met Office NWP systems. Both institutions have investigated the effect of Spire</span><br role="presentation" /><span dir="ltr" role="presentation">and COSMIC-2 data running Observing System Experiments in their state-of-the-art NWP systems.</span><br role="presentation" /><span dir="ltr" role="presentation">They found that the addition of COSMIC-2 and Spire observations is beneficial, with adding Spire</span><br role="presentation" /><span dir="ltr" role="presentation">on top of COSMIC-2 showing further improvements for temperature, humidity and wind. In conclu-</span><br role="presentation" /><span dir="ltr" role="presentation">sion, Met Office and ECMWF would assimilate the Spire data operationally if it becomes available.</span><br role="presentation" /><span dir="ltr" role="presentation">Additionally,</span><span dir="ltr" role="presentation"> </span><span dir="ltr" role="presentation">ECMWF</span><span dir="ltr" role="presentation"> </span><span dir="ltr" role="presentation">have</span><span dir="ltr" role="presentation"> </span><span dir="ltr" role="presentation">compared</span><span dir="ltr" role="presentation"> </span><span dir="ltr" role="presentation">the</span><span dir="ltr" role="presentation"> </span><span dir="ltr" role="presentation">impact</span><span dir="ltr" role="presentation"> </span><span dir="ltr" role="presentation">of</span><span dir="ltr" role="presentation"> </span><span dir="ltr" role="presentation">how</span><span dir="ltr" role="presentation"> </span><span dir="ltr" role="presentation">the</span><span dir="ltr" role="presentation"> </span><span dir="ltr" role="presentation">Spire</span><span dir="ltr" role="presentation"> </span><span dir="ltr" role="presentation">data</span><span dir="ltr" role="presentation"> </span><span dir="ltr" role="presentation">is</span><span dir="ltr" role="presentation"> </span><span dir="ltr" role="presentation">processed,</span><span dir="ltr" role="presentation"> </span><span dir="ltr" role="presentation">by</span><span dir="ltr" role="presentation"> </span><span dir="ltr" role="presentation">compar-</span><br role="presentation" /><span dir="ltr" role="presentation">ing bending angles provided directly by Spire with Spire bending angles produced at EUMETSAT.</span><br role="presentation" /><span dir="ltr" role="presentation">Furthermore, this extensive GNSS-RO dataset also provides the opportunity to investigate how the</span><br role="presentation" /><span dir="ltr" role="presentation">impact of GNSS-RO data scales with observation number.</span><span dir="ltr" role="presentation"> </span><span dir="ltr" role="presentation">For this reason, we have performed a set</span><br role="presentation" /><span dir="ltr" role="presentation">Ensembles of Data Assimilation (EDA) experiments.</span><span dir="ltr" role="presentation"> </span><span dir="ltr" role="presentation">We have found that increasing the number of</span><br role="presentation" /><span dir="ltr" role="presentation">GNSS-RO data reduces the ensemble spread,</span><span dir="ltr" role="presentation"> </span><span dir="ltr" role="presentation">especially at 100 hPa and higher and in the Tropics.</span><br role="presentation" /><span dir="ltr" role="presentation">In this region,</span><span dir="ltr" role="presentation"> </span><span dir="ltr" role="presentation">we also found a linear relationship between EDA spread and variance in first guess</span><br role="presentation" /><span dir="ltr" role="presentation">departures of radiosonde temperature observations.</span><span dir="ltr" role="presentation"> </span><span dir="ltr" role="presentation">For the Northern and Southern Hemisphere the</span><br role="presentation" /><span dir="ltr" role="presentation">decrease in EDA spread was less strong when more GNSS-RO data has been added, which might be</span><br role="presentation" /><span dir="ltr" role="presentation">linked inter alia to the heterogeneously distributed GNSS-RO data in the outer-tropics.</span></p>
},
  year = {2021},
  journal = {ESA Contract Report 4000131086/20/NL/FF/a},
  month = {11/2021},
  publisher = {ECMWF},
  url = {https://www.ecmwf.int/node/20240},
  doi = {10.21957/wrh6voyyi},
  language = {eng},
}