Journal of Geophysical Research Atmospheres · 2003 · 48 citations · 22 references
Natural SciencesEarth SciencesMeteorologyAtmospheric ScienceNumerical Weather PredictionWeather ForecastingData AssimilationHydrometeorological ModelingHydrometeorological Observation SystemsNumerical Weather Prediction Data AssimilationAtmospheric Moisture ProfilesGlobal Navigation Satellite System Radio OccultationHydrometeorological Data AssimilationHydrostatic EquationPressure Profile AnalysisTemperature Profile Retrieval
A 1‐D Var retrieval study of simulated radio occultation measurements is presented. Temperature and a water vapor profile are retrieved along with a reference pressure to generate the pressure profile by applying the hydrostatic equation. High‐resolution European Center for Medium‐Range Weather Forecasts (ECMWF) atmospheric fields are used by a ray tracing tool to calculate the exact positions of the tangent point. The 1‐D atmospheric profiles following the calculated tangent point trajectory in the 3‐D ECMWF fields are used to simulate bending angle measurements with a 1‐D forward model. Assimilation of these bending angles in a 1‐D Var tool employing the same 1‐D forward model is performed. We analyze the sensitivity of the retrieval to changes in vertical resolution, horizontal smearing of the tangent point trajectory, and the assumption of hydrostatic equilibrium for a nonvertical atmospheric scan. We find that retrievals calculated without adequate vertical resolution can have significant errors in temperature and water vapor. Errors in the retrieval by assuming hydrostatic equilibrium for a nonvertical scan generally cause only minor errors in the retrieved profiles. A study into the occurrence of rays curving down toward the Earth surface indicates that about 5–10% of the profiles could experience so‐called critical refraction at altitudes between 0.5 km and 2 km in case of the applied high‐resolution ECMWF data.
22