Valuable insights into the thermospheric mass density and horizontal winds can be obtained from satellites equipped with accelerometers. To derive these quantities, radiation pressure must be accurately modeled and removed from the calibrated accelerometer measurements. However, the documented surface reflection and absorption coefficients, as well as the satellite’s thermal properties, are often inaccurate or, in some cases, even absent. This study presents a method for optimizing these parameters jointly with the accelerometer scale factors. Focusing on GRACE data from 2009, a case where radiation pressure was dominant over aerodynamic force, enabled us to refine the radiation pressure model without detrimental effects from errors in aerodynamic force modeling. We evaluated three variants of estimating the scale factor: estimating no accelerometer scale factors, only the y-axis scale factor, or both the y- and z-axis scale factors. We use the difference between the measured and modeled accelerations (the residual) as our target functional. Estimating both scale factors yielded the lowest residual for both GRACE satellites, even though the radiation pressure model was tuned using GRACE-A data only. After the optimization, we observed a systematic feature in the cross-track residuals within the geographical domain, which strongly correlates with the magnetic field vector experienced by the spacecraft. While its cause remains unknown, we introduced an empirical correction that effectively removed the feature and significantly increased consistency between GRACE-A and GRACE-B. Overall, we were able to reduce the RMS of the residuals by more than 13% in the cross-track direction and 32% in the radial direction, indicating a significant increase in modeling accuracy. The presented method provides a generalizable approach that can also be applied to future satellite missions with accelerometers.

The latest characterization of turbulent spot growth accounting for wall temperature and compressibility effects along with a formulation of the transitional Reynolds stresses are assessed in the present γ-α transition model. Additionally, previous model developments to reproduce the normal intermittency profile are now rederived based upon the exact Klebanoff profile. Furthermore, destruction and production terms of the intermittency equation are further refined assuring a more generic transition model. The construction of this transition model is primarily set up to be independent from the used turbulence model, allowing the most suitable to be selected by the user for the envisaged application. The newly developed γ-α model is validated with the classical SST turbulence model for subsonic (with and without pressure gradient) and supersonic (with and without shock wave boundary layer interaction) test cases and has shown to outperform the classical γ-Reθt model in onset location prediction, skin friction slope and overshoot.

As the instruments of satellites are getting more sophisticated, they are also getting more sensitive to external influences. Deposition of gases due to the low-temperature environment can be one of these external influences. For the European-Chinese SMILE mission, a Soft X-Ray Imager (SXI) is present which has CCD’s exposed to very low temperature. As water vapour is present in the air, as well as water absorbed into the paint of the enclosure, ice deposition could happen under specific conditions. To assess the outgassing of PU1 paint, an experimental outgassing study has been performed at room temperature. Additionally, numerical simulations are performed both in the continuum and the molecular regime to assess the proper functioning of the venting paths. A semi-theoretical correlation is provided which allows calculating the mass flow leaving the instrument based on the internal pressure and temperature. For the specific application of the SXI instrument, the venting rate of the enclosed air is assessed. Also, the outgassing of water vapour should not result in pressure conditions which could trigger ice deposition on the cold CCD surface. Based on the performed simulations and experiments, no issues are expected for the SXI.

The Gravity Field and Steady-State Ocean Circulation Explorer (GOCE) satellite, which operated at an altitude of ∼250km, provided neutral thermosphere mass density and crosswind observations in the dawn-dusk sectors throughout most of its operational lifetime (2009–2013). As a result of its Sun-synchronous orbit, GOCE’s large solar panels remained at a near-perpendicular angle to the incoming solar radiation, leading to a significant radiation pressure acceleration. In this research, we focused on revisiting and reprocessing GOCE thermosphere mass density and crosswind data. We selected the coefficients describing the thermo-optical surface properties and employed a high-fidelity satellite geometry in a ray-racing simulation. Additionally, we distinguished between the solar flux in the visible and infrared bands and introduced a model for the satellite’s thermal emission. The availability of the in situ thermistor measurements allowed for the validation of the thermal model. Moreover, we replaced the Level-1b ion thruster data with raw telemetry, filling multiple data gaps. We analysed how incremental improvements in the radiation pressure modelling affected the observed crosswind speed. By replacing the panel model with the high-fidelity satellite geometry, the crosswind speed decreased up to 5 ms⁻¹. The biggest difference reduction of 40ms⁻¹resulted from introducing the thermal model. Splitting the solar flux further decreases the observed crosswind speed by up to 8ms⁻¹. The reduction in crosswind speed was most prominent during the first years of the mission when the solar activity was low. We compared the newly processed GOCE zonal wind data with respect to the most recent previous release. We observed a median absolute deviation decrease of 10 ms⁻¹around the south magnetic pole in the dawn sector. The yearly consistency of low-latitude zonal winds did not change significantly. The main obstacle in quantifying the improvement compared to the previous crosswind dataset stemmed from the fact that the previous and new datasets were generated with different crosswind estimation algorithms. The difference in thermosphere density compared to previously published datasets is minor since the effect of radiation pressure is most prominent in the cross-track direction. Finally, we verified the assumption about the energy accommodation coefficient of 0.82 and concluded that it remains valid after implementing the radiation pressure modelling improvements.